Sharon Gibbs with George Saliba TROLABES FROM [ERICAN fflSTORY S O N I A N I N S T I T U T I O N P R E S S SERIES PUBLICATIONS OF THE SMITHSONIAN INSTITUTION Emphasis upon publication as a means of "diffusing knowledge" was expressed by the first Secretary of the Smithsonian. In his formal plan for the Institution, Joseph Henry outlined a program that included the following statement: " I t is proposed to publish a series of reports, giving an account of the new discoveries in science, and of the changes made from year to year in all branches of knowledge." This theme of basic research has been adhered to through the years by thousands of titles issued in series publications under the Smithsonian imprint, commencing with Smithsonian Contributions to Knowledge in 1848 and continuing with the following active series: Smithsonian Contributions to Anthropology Smithsonian Contributions to Astrophysics Smithsonian Contributions to Botany Smithsonian Contributions to the Earth Sciences Smithsonian Contributions to the Marine Sciences Smithsonian Contributions to Paleobiology Smithsonian Contributions to Zoology Smithsonian Studies in Air and Space Smithsonian Studies in History and Technology In these series, the institution publishes small papers and full-scale monographs that report the research and collections of its various museums and bureaux or of professional colleagues in the world of science and scholarship. The publications are distributed by mailing lists to libraries, universities, and similar institutions throughout the world. Papers or monographs submitted for series publication are received by the Smithsonian Institution Press, subject to its own review for format and style, only through departments of the various Smithsonian museums or bureaux, where the manuscripts are given substantive review. Press requirements for manuscript and art preparation are outlined on the inside back cover. S. Dillon Ripley Secretary Smithsonian Institution SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY ? NUMBER 45 Planispheric Astrolabes from the National Museum of American History Sharon Gibbs with George Saliba SMITHSONIAN INSTITUTION PRESS City of Washington 1984 A B S T R A C T Gibbs, Sharon, with George Sahba. Planispheric Astrolabes from the Na? tional Museum of American History. Smithsonian Studies in History and Technol? ogy, number 45, 231 pages, 130 figures, 23 tables, 1984.?This monograph describes via catalog entries and comparative analysis what has for many years been one of the five largest collections of planispheric astrolabes in the world. Until 1974, when seven instruments that had been on long term loan were returned to their owner, the National Museum of American History of the Smithsonian Institution preserved 48 examples of functioning astrolabes. This is the first detailed discussion of all 48, together referred to as the collection. The majority of the instruments, including the seven no longer on loan to the museum, once were part of the collection of Samuel Verplanck Hoffman of New York City. An introductory chapter, using words and drawings, describes the basic elements of a planispheric astrolabe, thereby introducing terms that appear frequently in later sections. The section "Historical Perspective" emphasizes the information conveyed by the makers' names and dates inscribed on instruments in the collection. It places this information in the larger context of the history of the development of the astrolabe. Each of the functional elements incorporated into the astrolabes in the collection is discussed in detail in a chapter devoted to comparative analysis. That section illuminates dis? tinctions between European instruments and instruments made in India or in the Muslim world. In each section, the basic features of a functional element are described and any remarkable treatments of these features are noted. In addition, the traditional function of each element is specified, relying on instructions for its use prepared by Masha^'allah, al-Biruni, or Chaucer. Photographs illustrate each section. Complementing this comparative analysis is an illustrated catalog of the collection. It includes transcriptions and translations of inscriptions that appear on the instruments. Appended to the catalog are two sections that present and discuss the information conveyed by the gazetteers incorporated into many Muslim astrolabes and the star networks (or retes) included on all complete astrolabes in the collection. Finally, a third appendix describes the process used to prepare the ecliptic circle component of the astrolabe's star network. In doing so it conveys basic information about the construction of a planispheric astrolabe. Copyright ? Smithsonian Institution 1984 OFFFCIAL PUBLICATION DATE is handstamped in a limited number of initial copies and is recorded in the Institution's annual report, Smithsonian Year. SERIES COVER DESIGN: COVER: Front of astrolabe, CCA No. 39 (see page 73). Library of Congress Cataloging in Publication Data National Museum of American History (U.S.) Planispheric astrolabes from the National Museum of American History. (Smithsonian studies in history and technology ; no. 45) Bibliography: p. 1. Astrolabes. 2. Astrolabes?Catalogs. 3. National Museum of American History (U.S.) Catalogs. I. Gibbs, Sharon L. II. Saliba, George. III. Title. IV. Series. QB85.N37 1984 552'.4 83-600270 Contents Page Foreword v Preface viii Design of the Planispheric Astrolabe 1 Historical Perspective 12 Comparative Analysis 22 Catalog of the Collection 61 Appendix I: Gazetteers 190 Appendix II: Star Names 207 Appendix III: The Principle of Stereographic Projection 220 Notes 223 Literature Cited 227 Index 228 i l l Foreword The astrolabe of Islam was a traveler's instrument. Whether the owner was an astrologer, astronomer, geographer, surveyor, religious or civil official, his planispheric astrolabe was portable and designed for use across several lati? tudes. Travel has also been an important factor in accounting for the popu? larity of astrolabes and pseudo-astrolabes as collectors' items in recent decades. Americans and Europeans visiting the shop of an instrument maker in Isfahan have been attracted by the esthetic appeal of the twentieth century object, just as international visitors in the antique store of a scientific instrument dealer in Paris or London have been drawn by the mathematical and scientific elegance of the beautifully crafted, traditional, functional object. Curiosity about astrolabes is not a recent phenomenon, however. Several nineteenth-century studies of individual instruments attest to the interest taken by philologists and other scholars in these artifacts, which came to be treated as source documents for the study of Islamic civilization and its impact on Western learning. Numerous hobbyists and art collectors likewise acquired astrolabes out of interest in history, science, or art long before the ownership of a single such object (to be displayed in the living room) became a desirable goal in itself. Indeed, most of the major known private collections date back to the last century. The collection of astrolabes in the National Museum of American History reflects long-standing American interests in the subject. It was an American traveler in Europe who purchased the first astrolabe for the United States National Museum in 1888. He was Samuel P. Langley (1834-1906), noted contributor to the study of solar radiation and the early history of flight, and third Secretary of the Smithsonian Institution. He authorized purchase of this astrolabe (CCA No. 2569), along with several other scientific instruments, from Raoul Heilbrunner in Paris. Subsequent astrolabes came from owners representing a variety of economic classes, who had acquired the objects in the nineteenth and twentieth centuries, either as single objects of curiosity or in the course of building a collection. Among these owners were Americans of colonial ancestry as well as recent immigrants. Aside from purchases made possible in large part by corporate donors such as International Business Machines Corporation, the objects in the collection came through the gener? osity of individual donors. We owe particular gratitude to E. Nagel, N. Grossman, Lessing J. Rosenwald, and an anonymous donor. The bulk of the astrolabes in the National Museum of American History belonged to the Samuel V. Hoffman collection, however, and came to the Smithsonian Institution through the cooperation of his children, the late Edgar and Margaret Hoffman. v i SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y Samuel Verplanck Hoffman (1866-1942) of New York City was the son of Eugene Augustus Hoffman, the dean of the General Theological Seminary, and the grandson of Samuel Verplanck Hoffman, a well-to-do New York merchant and insurance executive. The family had considerable holdings, including a large amount of real estate in Manhattan, which S.V. Hoffman managed. A graduate of Stevens Institute of Technology, Hoffman had done graduate work at Columbia and, more intensively, at the Johns Hopkins University, where he studied astrophysics and chemistry. By the turn of the century, he had acquired a reputation as a knowledgeable collector, and gave advice on the subject to the Smithsonian Institution. Sections from his sundial collection were loaned to the Smithsonian on several occasions. He was closely affiliated with the New York Historical Society, serving as its president from 1903 to 1913, and as a trustee from 1913 until 1942. Samuel V. Hoffman bequeathed his fine library on early mathematical and astronomical instruments to the New York Historical Society, along with most of his collection of sundials, the mariner's astrolabe associated with Cham- plain, and seven planispheric astrolabes. He bequeathed the major part of the astrolabe collection and other assorted pieces to his children, who decided in 1959 that these objects should become part of the National Collections entrusted to the Smithsonian Institution. It is this collection that forms the core of the National Museum of American History's astrolabe holdings. The present study also includes the seven planispheric astrolabes that had been given to the New York Historical Society, since these were loaned to the Smithsonian Institution and displayed in this Museum until the New York Historical Society sold them at auction in 1975. They are identified by the NMAH numbers 322327, 322458-322461, 322463, and 322464 (CCA Nos. 65, 4, 58, 42, 55, 144, and 88, respectively. According to the latest astrolabe census, the collection of the National Museum of American History is now the fourth-largest museum collection in the world, ranking behind those of the Museum of the History of Science in Oxford, the Greenwich Maritime Museum, and the Adler Planetarium in Chicago. For that reason, it seemed particularly appropriate to subject the objects in this collection to close study. Although there have been a number of such studies of individual instruments, and good descriptive accounts of certain collections, the only publication that approached the detailed descrip? tion we found desirable is Gunther's classical two-volume work published in 1932. While many of our astrolabes were included in that study, the infor? mation given was often incomplete. Gunther's unsurpassed achievement in describing the world's known astrolabes is not denigrated by the observation that we have an advantage over him in working from the object rather than the photograph, and in being able to draw on subsequent scholarship such as that of Hartner, Michel, or Kunitzsch. We have been particularly fortunate in the fact that Sharon Gibbs was willing to undertake the arduous task of subjecting each of the astrolabes and its components to minute scrutiny and to describe her findings in meticulous detail. We are grateful to George Saliba for collaborating with her in the NUMBER 45 Vli translation of the Arabic inscriptions, and to David Pingree, who provided similar help with the Sanskrit. We hope that their work will encourage other subject matter specialists to provide similar descriptions. Having the results of such studies at our disposal should enable us to address the many unanswered questions concerning schools, traditions, and styles that make this subject such a fascinating field of inquiry. Uta C. Merzbach, Curator Division of Mathematics The National Museum of American History Preface The preparation of this monograph would not have been possible without the encouragement and cooperation of Dr. Uta Merzbach, who has, for the years of my association with them, been the custodian of the National Museum's collection of planispheric astrolabes. My gratitude to Dr. Merzbach extends also to her assistants, most notably Mrs. Ona Jordan, who has presided over many of my examinations of the objects and provided quick access to photographs and prints. This publication completes, after a long hiatus, work begun as a 1972-1973 post-doctoral research fellow at the Museum. The work was jointly sponsored by Dr. Merzbach and Silvio Bedini, then Assistant Museum Director. Mr. Bedini's interest and support have complemented Dr. Merzbach's encourage? ment and cooperation to lead to the publication of this volume. Some years after I had completed my Smithsonian fellowship I learned that Dr. George Saliba, now at Columbia University, had also studied those astrolabes in the collection with Arabic inscriptions. Dr. Saliba and I agreed that collaboration would enhance our efforts. The result is this volume and the inclusion (thanks to Dr. Saliba's contributions) of Arabic transcriptions, accompanying English translations, and pertinent linguistic commentary. Dr. Saliba's research is particularly evident in the appendices on star names and gazetteers. Dr. Saliba has frequently obtained the help of Mr. H. Javadi, Mr. A. Davaran, and A. Kholdani concerning the deciphering of some Persian inscriptions. Their generous assistance is gratefully acknowledged. He is especially grateful to Professor Owen Gingerich of Harvard University, who encouraged his work on this collection. I am grateful to Russell Cashdollar for turning my rough sketches into the professional line drawings that illustrate this publication. Finally, I owe a special debt of gratitude to an extraordinary typist, Rosemary Regan. Sharon Gibbs Washington June 1982 Vlll Planispheric Astrolabes from the National Museum of American History Sharon Gibbs with George Saliba Design of the Planispheric Astrolabe Information on how to make and use a plan? ispheric astrolabe is currently available from a variety of sources. A fourth-century work by Theon Alexandrinus "on the little astrolabe" ap? parently can be considered the prototype for all later contributions to this literature.^ While the remarks in this section follow the tradition estab? lished by Theon, they are not intended to be a major contribution to it. They are offered pri? marily as an introduction to terminology em? ployed in the catalog and the analysis of the collection of instruments which is the subject of this study. Some basic components characterize a plani? spheric astrolabe no matter where or when it was made. These are the body, the suspensory appa? ratus, the alidade, the pin, and the star map (or rete). The functions of these parts remain fairly constant over a large geographic area and a considerable period of time. Together, the body, the suspensory apparatus, the pin, and the ali? dade function as an observational instrument (see Figure 10); the star map, the pin, and the body Sharon Gibbs, National Archives and Records Service, Washington, D.C. 20408. George Saliba, Columbia University, New York, N. Y. 10027. function as a model of the heavens (see Figure 13). The body constitutes the bulk of the instru? ment. It is shaped like a thin disk. It has a definite front {wajh ov facies)'^ and back {zahr or dorsum). In all but a few cases, the front incorporates a raised rim {hajra or limbus) and a depressed central area {umm or mater). The various surfaces of the body of an astrolabe are seldom left uninscribed. This inscription (both form and content) may be used to identify both the date (see page 13) and place of origin of any given instrument. Tables 1, 2, and 3 list the scales that may appear on the back of an astrolabe, and indicate whether each can usually be found on an instrument inscribed with Arabic characters and made in the mashriq (East) or the maghrib (West), or on a European^ instrument inscribed with Latin characters. Fig? ures 1, 2, and 3 illustrate the typical placement of these scales. Note that, in all cases, at least one scale serves to measure the altitude of a celestial object when the astrolabe is suspended. Just as the scales and tables on the back of an astrolabe may be used to distinguish an Arabic from a European instrument, so too may the inscription on the front rim. Table 4 shows the SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y TABLE 1.?Scales that may appear on the back of a mashriqi astrolabe (refer to Figure 1 for an explanation of location of entries in "Placement" column; refer to "Comparative Analysis" for full explanation of scales) Title of table/scale Outer Circular Altitude Cotangent Arc Declination Azimuth of Qibla Altitude of noon sun Hours Prayers Angle Sine Cosine Box Tangent Cotangent Inner Circular Terms Zodiacal signs Faces Lunar mansions Inner Box Triplicities Placement Margin I, II III, IV Quadrant UR UR UR UL, UR UR Quadrant UL UL Quadrant LL(a), LR(d) LL(b),LR(c) Quadrant LL and LR V, VI LL and LR VII LL and LR VIII LL and LR IX Quadrant LL and LR X Units degrees of arc feet/fingers zodiacal divisions cities degrees of latitude 1/12 of daylight times of day 1? or 5? angles or 1/60 of quadrant radius feet LL, fingers, LR feet LL, fingers, LR zodiacal divisions zodiacal divisions zodiacal divisions zodiacal divisions zodiacal divisions Number of units (type) 90 (equal) 7-60 (unequal) 6-60 1-7 22-50 1-6 1-5 18-90 (unequal) 60 (equal) 7 LL, 12 LR 7 LL, 12 LR (equal) 60 (unequal) 12 (equal) 36 (equal) 28 (equal) 12+ planets and natures variety possible. Figures 4 and 5 illustrate the placement. To ensure the effective operation of the astrolabe as a model of the heavens (see below), the rim is interrupted by a hole or a projecting tooth. The umm or mater of an astrolabe, if it is engraved at all, may contain either a stereographic projection of the horizon coordi? nates for a given latitude (see Appendix III and Figure 6) or a list of important cities accompanied by geographic parameters (see Appendix I). The quality of an astrolabe as an observational instrument depends on the care with which it is fitted with a means of suspension (see Figure 7). An acceptable suspensory apparatus generally consists of a rigid appendage to the disk base, called the "throne" (kursi or armilla fixa); a loop of metal, the handle {^urwa or armilla reflexa), which passes through a hole pierced in the highest point of the throne and can move only in the plane of the disk base; and a metal ring {halqa or armilla rotunda) with a diameter large enough to encircle a thumb and pass through the handle attached to the throne. The preceding description FIGURE 1.?Back of an astrolabe made in the mashriq. Numbers and letters refer to scales listed in Table 1. refers only to the basic characteristics of each element of the appratus. Above all, the combi? nation must allow the instrument to hang plumb. Shape and the amount of decoration may vary TABLE 2.?Scales that may appear on the back of a maghribi astrolabe (refer to Figure 2 for the location of entries in the "Placement" column; the calendar scale may be centrally located, in which case the 365 days would divide it unequally) Tit le of table/scale Outer circular Altitude Cotangent Zodiacal signs Outer eccentric Calendar Arc Hours Box Tangent Cotangent Placement Margin 1,2 3,4 5 Margin 6 Quadrant UL Quadrant LL(a), LR(d) LL(b), LR(c) Units degrees of arc 1/12 of gnomon zodiacal divisions days of year 1/12 daylight feet feet Number of units (type) 90 (equal) 60 (unequal) 360 (equal) 365 (equal) 1-6 12 (equal) 12 (equal) FIGURE 2 (left).?Back of an astrolabe made in the maghrib. Numbers and letters locate scales listed in Table 2. FIGURE 3 (right).?Back of an astrolabe made in Europe. Numbers and letters refer to scales listed in Table 3. TABLE 3.?Scales that may appear on the back of a European astrolabe: summarizes the content of the back of a "standard" European astrolabe but does not characterize those instruments with universal projections (refer to Figure 3 for location of entries in "Placement" column; the calendar scale may be centrally located, in which case the 365 days would divide it unequally) Title of table/scale Outer circular Altitude Zodiacal signs Outer eccentric Calendar Arc Hours Hours Box Tangent Cotangent Placement Margin 1,2,3,4 5 Margin 6 Quadrant UL, U R UL, UR Quadrant LL(a), LR(d) LL(b), LR(c) Units degrees of arc zodiacal divisions days of year 1/12 of daylight 1/24 of day fingers or 1/60 side of box fingers or 1/60 side of box Number of units (type) 90 (equal) 360 (equal) 365 (equal) 1-12 1-12 1-12 1-60 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y TABLE 4.?Scales that may appear on the rim of an astrolabe (see Figures 4 and 5 for the placement of these scales) Title of scale Degrees Degrees Winds Hours Units 5? and 1? 5? and 1? 30? 15? Range 0?-360? 4 (0?-90?) 0?-360? 0?-360? Tradition mashriq maghrib mashriq European European FIGURE 5.?Placement of scales on the rim of an astrolabe made in Europe. FIGURE 4.?Variations in the placement of scales on the rim of an astrolabe inscribed in Arabic characters. FIGURE 6.?Astrolabe plate appropriate for 36? latitude. Numbers refer to projected lines listed in Table 5. according to the taste and ingenuity of the maker. As is true with the suspensory apparatus, the shape of the alidade {al-^idada or regula) may be dictated by stylistic convention. Utility demands, however, that its length be approximately equal to the diameter of the body; that it be wide enough at its center to be pierced by a small circular hole; and that each arm incorporate a beveled edge that coincides with a line (the linea fiduciae), which passes through the center of the NUMBER 45 FIGURE 7.?An astrolabe's suspensory apparatus. 0 :> O O o > y FIGURE 8.?Alidade with pin and ringlet. small central hole. An inscription in Arabic char? acters and engraved lines invariably decorate the alidade of an instrument made in the mashriq (see Figure 8). No alidade is complete unless it has been fitted with two sighting plates. These FIGURE 9.?The star map, '^ankabut, or rete, of an astrolabe. Numbers refer to stars listed on page 56. TABLE 5.?Coordinates that may be projected on the plate of an astrolabe (see Figure 6 for location of numbers, in parentheses here, specifying position of scales) Projected lines Colures (1) Tropics and equator (2) Horizon (3) Altitudes (4) Twilight Azimuths Equal hours Seasonal hours (8) Houses Prayers Number 2 3 1-many 15-90 1 36 12-15 12 12 4 Position quadrant boundaries UL,UR,LL,LR LL,LR UL,UR LL,LR UL,UR,* LL,LR LL,LR UL,UR,LL,LR LL,LR Tradition all all all all all all mashriq all European maghrib * On some astrolabes made in the Muslim world, the azimuths extend to the lower half of the plate. are usually pierced with two holes (centered on each plate's center line). The sighting plates are affixed perpendicular to each arm of the alidade at equal distances from its center. The complete alidade is connected to the body of the astrolabe by means of a broad-headed pin SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y {qutb or clavus), which passes through the central hole in the alidade and a hole in the center of the body. The diameter of the pin is such that the alidade moves easily, but not loosely, about it as an axis. The pin is a small but crucial component of the astrolabe's design. Without it the astrolabe could not function as an observational tool. In fact, the astrolabe's other function as a model of celestial motion is equally dependent for its op? eration on this pin. The openwork star map (^ankabUt or rete) is the basic component of the astrolabe that completes the model of celestial motion (see Figure 9). It can be viewed as a thin plate of diameter equal to the diameter of the area on the front of the astrolabe framed by the rim. It incorporates the stereographic projection of the ecliptic and of selected bright stars (see Appendices II and III). All of the metal that does not either define these projections or connect them to the center of the plate is cut away. The center itself is pierced with a small circular hole so that the star map may fit over the pin passing through the center of the body and may rotate freely about the pin as axis. The star map is of openwork construction to allow a view of a fixed stereographic projection engraved on the surface under it. This stereo? graphic projection of the horizon coordinates for a given latitude may appear on the umm or mater. If it does not, and occasionally even it it does, the astrolabe may be fitted with a series of solid plates (safiha or tympanum) of diameter equal to the diameter of the openwork star map. These plates are inscribed with stereographically projected ho? rizon coordinates for several latitudes (see Table 5 for a list of the coordinates that may appear). Each plate is modified with a notch or a tooth (see Figure 6) that fits into the corresponding tooth or notch incorporated into the rim, so that when the astrolabe is assembled, the plate is fixed inside the umm or mater. Most European instruments and some instru? ments inscribed with Arabic characters include a second rotating arm, similar to the alidade, but without sighting plates. This index arm is de? signed to rotate freely about the pin over the star map and rim on the front of the astrolabe. Once all the separate components of the astro? labe (front and back) are in place, a slot in the tip of the consolidating pin extends beyond the top layer just enough so that a wedge (faras or equus?literally "horse") may be fitted into it. When used to make celestial observations, the astrolabe is suspended from its ring, and its ali? dade is adjusted so that either a ray of sunlight passes through the two smaller holes in the sight? ing plates or a star is centered in the two larger holes. The fiducial edge of the upper alidade arm registers solar or stellar altitude (Figure 10). Measurements of solar or stellar altitude can be converted with the help of astronomical tables or a knowledge of spherical trigonometry into indi? cations of geographic position or time of day or night. Sixteenth- and seventeenth-century explor? ers chose to bring only the observational elements of the astrolabe to sea, and to use the resulting instrument in conjunction with printed tables to help them navigate. Those elements of an astrolabe that character? ize it as planispheric replace tabulated astronom? ical data as interpreters of measures taken by the alidade. Whereas astronomical tables are arith? metical models of heavenly motion, the astro? labe's rete, plate, and pin constitute a geometrical model of heavenly motion. In converting stellar altitude to time of night, for example, the astro? labe's geometrical model functions as outlined below. An engraved plate, appropriate for the ob? server's latitude, is fitted into the front central area of the astrolabe. The rete is positioned over the plate and the pin is replaced so that the rete may rotate around it. Assuming that one of the stars incorporated into the rete is visible above the horizon, its altitude is measured. The rete is then rotated until the representation of this star coincides with the projected circle on the plate, which represents its measured altitude. Without moving the rete, the position of that point in its FIGURE 10.?Noah's Ark, as shown in an Indian painting of the early seventeenth century depicting the use of an astro? labe (left middle) as an observational instrument. (Freer Gallery of Art, Smithsonian Institution, Accession No. 48.8). liKtegT" SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 11.?Astrolabe No. 186. The tip of the pointer identified as Cor Leonis lies in the middle of the third seasonal hour. NUMBER 45 FIGURE 12.?Astrolabe No. 262. Assuming that its index arm marks the position of the sun in the ecliptic, this astrolabe marks the middle of the third equinoctial hour after midnight. 10 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y ^./^. ^ '^,,A'L^;S hz>^X ?y^A ^ ? , J^^/.l; '^: ?^^i'&!^ FIGURE 13.?This sixteenth century miniature depicts an astrologer using an astrolabe (left middle) to prepare the horoscope of Alexander the Great. (Metropolitan Museum of Art. Accession No. 13.228.14 44157. Ms5). N U M B E R 45 11 yearly path (the ecliptic) that represents the sun is noted. It coincides with a seasonal hour ster? eographically projected onto the plate (see Figure 11) or an equinoctial hour radially projected onto the rim (see Figure 12).^ The same elements of the planispheric astro? labe's design that enable it to function effectively as a timekeeper also contribute to its usefulness as an astrological computer. By adjusting the rete so that it represents the heavens at the place and moment of an individual's birth, an astrologer equipped with an astrolabe (see Figure 13) can avoid many of the complicated calculations in? volved in casting a horoscope. He can read di? rectly from the adjusted instrument the ascend? ant, the midheaven, the descendant, and the nadir. If the plate includes projections of the boundaries of the astrological houses, he can com? plete the horoscope chart by reading the points of intersection between these boundaries and the ecliptic circle. In the absence of projected house boundaries he can estimate their position by using the projections of the second, fourth, sixth, eighth, and tenth seasonal hours. Figure 14 shows an astrolabe representing the heavens at noon on 9 December. The horoscopic chart indicated by this setting would incorporate the following astrolog- ically significant points: the ascendant, Pisces FIGURE 14.?Astrolabe No. 221. The configuration of rete and plate represent the configuration of the heavens on 9 December. 18?; the midheaven, Saggitarius 25?; the descen? dant, Virgo 18?; and the nadir, Gemini 25?. Historical Perspective PRE-FIFTEENTH CENTURY The earliest astrolabes in the collection (made between the eleventh and fourteenth centuries) are inscribed with Kufic characters.^ All of these examples (Nos. 4, 15, 144, 2572, 3643, and 4001)^ are signed and dated. Their makers worked in or near Isfahan, Kerman, Granada, and Seville. Thus, several hundred years and several hundred miles separate these instruments from their pro? totype, possibly constructed for Rhodes by Hip- parchus in the second century B.C. Reconstruction of the steps leading from Hipparchus's discussion of stereographic projection to Muhammad ben as-Sahli's construction of a planispheric astrolabe (No. 2572) in A.D. 1090 is not easily accomplished. There is some evidence that the theory was trans? mitted via a chain that included the following links: Claudius Ptolemy [A.D. 150], Theon Alex? andrinus [A.D. 375], and Severus Sebokht, Bishop of Kenneserin, Syria (A.D. 660). Some time near the end of the eighth century, the Jewish astronomer Masha^allah discussed both the construction and use of the astrolabe in a treatise that survived to be published in Europe in a Renaissance compendium of learning.^ Sim? ilar discussions were included in the works of a number of later Arab astronomers.^ For example, al-Biruni completed his contribution to this latter genre in A.D. 1040,^ '^ just a few years before Muh? ammad dated his astrolabe. As is true of planispheric astrolabes in general, only subtle differences exist between the instru? ment described by al-Biruni and the pre-fifteenth- century examples in the collection. Some of these differences may reflect the needs of users; some probably the taste of makers. Among the former might be included variations in the content of the umm (front) and the zahr (back). The latter are apparent in the size and shape of the kursi (sus? pensory apparatus) and the pattern of the "^ankabut (star map). Users of the two early astrolabes in the collec? tion made in Isfahan (No. 4) and Kerman (No. 15) apparently had need of a gazetteer inscribed in the umm. The practice of including such a table continued in and was confined to the eastern portion of the Muslim world^^ as long as astro? labes were made there. That users of these two early Persian astrolabes were astrologers is sug? gested by their apparent need for the inclusion of astrological tables in the content of the zahr. A table of lunar mansions is notably absent from both instruments, however. The omission is not surprising in light of the fact that al-Biruni's Book of Instruction in the Elements of the Art of Astrology does not discuss the astrological significance of this method of dividing the heavens. Makers of these two early Persian astrolabes have most clearly asserted their individuality in the design of the kursi. That of Hamid ben Mah- mud, on an instrument dated A.H. 547 [A.D. 1152]^' (No. 4), is quite high, and pierced to form a vinelike motif. Ja'^far ben '^Umar's kursi, on an instrument dated A.H. 774 [A.D. 1372] (No. 15), is lower, undecorated, and pierced by only two small holes on either side of the suspensory hole. The patterns of the "ankabHts of these astrolabes are quite similar to each other and to the illustra? tion of an "ankabUt in at least one surviving man? uscript of al-Biruni's discussion of the subject.^^ The only difference lies in Ja*^far ben "Umar's attempt to introduce decorative curves into the basic functional design. The two symmetric "dips" in the equatorial band are characteristic of Ja'^far's work.^^ Of the four early astrolabes in the collection inscribed with maghribi script, two (Nos. 144 and 4001) were certainly made in Moslem Spain. The plates of a third (No. 2572) also suggest a Moorish origin. The same may be said for the fourth example (No. 3643), engraved with a now oblit? erated signature and date. The umm of each of these instruments contains the stereographic pro- 12 NUMBER 45 13 jection of horizon coordinates for a specific lati? tude (i.e., the network usually found on a plate). The latitudes are: 37?30', 66?, 90?, and 90?. All four of these Moorish astrolabes illustrate the fact that such instruments, inscribed with Kufic letters and made in the western portion of the Moslem world, carry no astrological tables. Instead each is engraved with scales (zodiacal calenders) that correlate solar position during the year with a date in the Christian calendar.^^ Instructions for dividing these scales are given in surviving editions of Masha"'allah's treatise but not in al-Biruni's. The variety apparent in the kursis and '^ankabHts of Persian instruments is also apparent in the analogous elements of these Moorish astrolabes. Simple kursis combine with relatively elaborate ^ankabHts. Ahmad ben Husain's "^ankabUt, on an instrument dated A.H. 704 [A.D. 1304] (No. 144), is remarkable in that, while it differs noticeably from other early examples, it contains elements that reappear in the '^ankabUts of later instruments (see, for example. No. 2568, dated A.H. 1103 [A.D. 1691] and No. 2571 dated A.H. 910 [A.D. 1504]).^^ The basic design of Muhammad ben as-Sahli's '^ankabUt is similar to that of the early Persian examples in the collection. The "ankabut of Mu? hammad ben Fattuh, on an instrument dated A.H. 621 [A.D. 1224] (No. 4001), does not survive. FIFTEENTH CENTURY Remaining evidence suggests that manuscripts on the construction and use of the astrolabe were available in Europe at least as early as A.D. 1054. Chaucer undoubtedly relied on one of these when he composed in A.D. 1387 the first English treatise on the subject, "Brede and Milke for Childeren."^^ Published works in Latin appeared in various European centers of learning through? out the sixteenth century.^^ These representatives of the "construction and use" genre usually dif? fered only in the epoch^^ of the table of fixed stars recommended for inclusion in a rete. The identi? fication of instruments that are engraved with Latin characters and are contemporary with this early Latin literature has been complicated by the fact that, until the sixteenth century, few European astrolabe-makers bothered to date their work. Dates for instruments from this period have been assigned, however, based on epigraphic clues and on the calendric correlate of the vernal equi? nox as revealed by scales engraved on the back of each instrument (see note 20). The two very early European instruments in the collection have been dated by means of their zodiacal calendars.^^ No obvious signature ap? pears on either astrolabe, although the earlier of the two (No. 304) carries a curious configuration of letters stippled on the upper right and left quadrants of its back (see Figure 15). The zodiacal calender and the rete of the astro? labe with the stippled lettering (No. 304) indicate that it was made to be used in the mid-fifteenth century. It was designed to be used at a single latitude (52?), a characteristic it shares with some other European instruments.^^ The second prob? able fifteenth-century European astrolabe in the collection (No. 186) includes six plates, about as many as one would expect to find in con- FIGURE 15.?Astrolabe No. 304. The stippled letters may be a maker's signature, or, more probably, an owner's mark. 14 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY temporary examples from Persia or North Africa. Differences between European astrolabes and astrolabes made in the maghrib are even more subtle than differences between the latter and astrolabes originating in the mashriq.^^ Astrolabes used in Moorish Spain can hardly be distin? guished from those used in Lancastrian England or Medician Italy. However, the evidence on preserved instruments does suggest that Euro? peans had no need for the circular cotangent scales found on most maghribi instruments. Cer? tainly this scale is not found on any of the Euro? pean instruments in the collection.^^ A second subtle difference between European and Moorish tradition can be seen in the inclusion of a scale of equal hours on the rim of an instru? ment made in Europe. As a rule, no analogous scale appears on a Moorish instrument. A sim? ilar distinction between European and Moorish preferences for timekeeping scales can be made with respect to the scale of seasonal (or unequal) hours engraved on the back of astrolabes from each area. The scale is likely to fill both upper quadrants of the back only on European instru? ments. Astrolabes in the collection show that Euro? pean makers are as likely to distinguish them? selves through the design of the rete as Persian or Moorish makers are through the design of the "ankabUt. The unsigned member of the pair of early European astrolabes in the collection (No. 304) has been assigned to an English source based primarily on the recognition of similarities be? tween the rete pattern of this instrument and the rete pattern chosen by Chaucer to illustrate his "Brede and Milke. "^ ^ Although it is clear that the maker borrowed something from this pattern type, he has clearly modified it to suit his taste. The early Italian instrument in the collection (No. 186)^^ incorporates a rete with a functional FIGURE 16.?Astrolabe No. 2006. Lines for a sundial and the probable dial maker's signature can be seen on the back of the rete (at left). NUMBER 45 15 FIGURE 17.?The rete of Chaucer's astrolabe, as illustrated in his treatise, "Brede and Milke for Childeren." design that was adopted by makers of various European nationalities.^^ It is distinctively simple, much like the early Persian ^ankabUt, but with the addition of an upper equinoctial arc. It is possible that a third astrolabe in the collection (No. 2006) may be added to the group of fifteenth-century instruments. Although it is unsigned, its inscriptions contain a number of clues to its origin. Some of the clues are mislead? ing. The majority of the inscriptions, including the zodiacal calendar, suggest a fifteenth-century origin. The diversionary clues, all inscribed on the back of the rete, include a maker's name (see Figure 16); they appear to derive from the sev? enteenth century. The rete's resemblance to one used to illustrate an edition of Chaucer's work on the astrolabe (Figure 17) is noteworthy. SIXTEENTH CENTURY All but one of the sixteenth-century astrolabes in the collection are the work of European makers. This distribution reflects the distribution of Per? sian and Moorish makers vs. European makers in the whole corpus of extant sixteenth-century as? trolabes. There exists an overwhelming majority of European examples. The one sixteenth-cen? tury astrolabe in the collection inscribed with 16 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY Kufic characters of maghribi origin (No. 2571) is dated A.H. 910 [A.D. 1504]. Except for the addition of lines marking the times of Muslim prayer on each of its plates, there is little to suggest that use of this instrument differed from use of contem? porary European examples. The pattern of the "^ankabut is distinctive, identifying a maker ("Ali ben Muhammad ben Abdallah ben Faraj) influ? enced by Moorish decorative style (circular sup? port of the equinoctial arc, Moorish arches at the ends of the equinoctial arc, studded balls as bases for star pointers). European fascination with the astrolabe in the sixteenth century is suggested not only by the number of instruments that survive, but also by the number of published works on the topic that appeared during the century. A list of some of the most popular publications (i.e., works that ap? peared in several editions) would include works by: Gregor Reisch (quoting Masha^allah) (Fri- burg, 1503), Johannes Stoffler (Oppenheym, 1512), Reiner Gemma Frisius (Antwerp, 1556), Egnazio Danti (Florence, 1569), John Blagrave (London, 1585), and Robert Tanner (London, 1587). The publications are evidence of an inter? national European interest. Instruments in the collection support this evidence. Four sixteenth-century European instruments in the collection are signed and dated as follows: Georg Hartman, 1537 (No. 262); M.P., 1542 (No. 221); I Galois, 1548 (No. 204); Philip Danfrie, 1584 (No. 2007). The one anonymous sixteenth- century instrument (No. 2005) is quite similar to the astrolabe by Egnazio Danti illustrated in Gunther.^? If the instruments in the collection are any indication, it would seem that European makers were far more likely to introduce variety into the contents of the back of an astrolabe than were their Persian or Moorish counterparts. One astro? labe includes a quadrant (No. 2006). One in? cludes a De Roias projection (No. 2005). One has a table of stars, along with the longitude, decli? nation, and magnitude of each star (No. 204). One has a set of perpetual calendar wheels (No. 2007). One has a scale of unequal hours (No. 262). One includes both a scale of unequal hours and a scale of equal hours (No. 221). Only two scales seem to have universal appeal: a scale correlating the Christian calendar with solar po? sition (zodiacal calendar), and the ubiquitous shadow square. Examples in the collection suggest that, in addition to zodiacal calendar and shadow square, at least one other common element characterizes sixteenth-century European astrolabes. The divi? sions of the astrological houses have been in? cluded on each plate. These divisions appear on the plates of only one Persian instrument in the collection (No. 65, attributed to an eighteenth- century maker). Both Masha^allah and al-Biruni advocate using the unequal hour lines as house divisions. On the whole, the Persian astrolabes document the popularity of this practice. It is just possible that European inclusion of house lines distinct from unequal hour lines in the sixteenth century is evidence of the impact of Regiomontanus's contribution to astrology. The inscribed houses consistently conform to the method of division suggested by Regiomontanus as an alternative to methods attributed to Alca- bitus or Campanus.^^ The youngest European instrument in the col? lection (No. 2007) includes evidence that Euro? pean makers distinguished themselves not only by means of a preferred rete design but also by their choice of materials. Philip Danfrie's astro? labe with its elaborate baroque rete is made of heavy paper that has been glued to a wooden form and is fitted with a brass alidade. With over 20 of his astrolabes surviving to the present, Georg Hartman'*^ appears to be the most productive of European astrolabe makers. The example of Hartman's work included in the col? lection (No. 262) documents his obviously suc? cessful method of production. The telling evi? dence appears in the form of a very small number "12" engraved in unobtrusive places on each of the separable parts: in the depression on the front of the body, at the lowest point of each plate, on the back of the rete, the alidade, and the index arm. Hartman's use of such a numbering system suggests that he managed a workshop in which several astrolabes were being finished simultane- NUMBER 45 17 ously. There would be no need to number the component parts of astrolabes completed one at a time. SEVENTEENTH CENTURY Those astrolabes in the collection that date from the seventeenth century are all inscribed with Arabic characters.'*^ These astrolabes are typical of the larger corpus of seventeenth-century astrolabes by Moorish or Persian makers in that they exhibit stylistic variations associated with three distinct manufacturing locales, and they are products of families of instrument makers. Henri Michel has, with typical perspecuity, discussed time- and space-dependent elements of astrolabe design in Chapter 20 of his Traite.'*^ These elements comprise a few basic "styles" subject to elaboration by individual makers. Of the styles identified by Michel, three are associ? ated with instruments inscribed in Arabic char? acters: the Hispano-Moorish, the Persian, and the Indo-Persian. The collection includes seven? teenth-century examples of all three styles. The Hispano-Moorish example, made by Hasan ben Ahmad al-Battuti (No. 2568), is dated A.H. 1103 [A.D. 1691]. Characteristic elements of its design include a simple throne, undecorated except for a single engraved marginal line, and a rete fitted with four turning knobs, incorporating star pointers shaped like thin flames attached to pierced trefoils and mounted on rectangular bases. Two instruments in the collection (Nos. 44 and 25) made in the Persian style are associated with the name Muhammad Mahdi (al-Yazdi), who is known to have been involved in the production of some 20 other astrolabes during the period between A.D. 1640 and 1670. According to the signature, Mahdi "decorated" one of these instru? ments (No. 25) in A.H. 1078 [A.D. 1667]; he signed the other (No. 44) without dating it and without designating his contribution.'*^ The recognizably Persian aspects of Mahdi's astrolabes include a throne shaped like a scalloped equilateral triangle and decorated with embossed vines, flowers, and inscription on a stippled background. Like other Persian retes, Mahdi's are foliated; star pointers are leaf-shaped. Following Persian tradition, he has placed his signature in a cartouche below the shadow squares on the back of the astrolabe. In an attempt to characterize the Indo-Persian style, Michel has remarked that it is to the Persian style "as German rococo is to regency: too much ostentation; too many flourishes; an awkward and artificial grace."^^ In the collection, the works of Muhammad Muqim (No. 86) and piya"'al-Din Muhammad (No. 87), both of Lahore,"*' serve as specific examples. Their Indo-Persian origin is evident in pierced thrones and in the extensive use of dashed division lines.^^ Neither of these two specific examples includes a functioning rete. Other instruments by these same makers are equipped with quite delicate retes. If the retes of the two signed Indo-Persian astrolabes in the collection were like these, it is easy to see why they did not survive. Indo-Persian elements can be recognized in each of the remaining four seventeenth-century astrolabes in the collection, although there is no explicit evidence that any originated in Lahore. All have pierced thrones. One (No. 2569) also includes extensive use of dashed lines and a well- preserved delicate rete. It is unsigned. One (No. 85) is signed by a maker from Qayen (in Eastern Persia).^ The fact that its rete is missing suggests that it may have been delicate. Certainly the maker's extensive use of dashed division lines may be interpreted as Indo-Persian influence. Al? though Gunther described as "Indian" the A.D. 1611 work of "Ali ben "Awad (No. 70), there is no evidence other than the pierced throne that it really is. The stylistic origins of the instrument signed by Khalil and dated A.H. 1067 [A.D. 1656] (No. 2567) are equally uncertain. At all times in its history, the skills necessary for designing the astrolabe, in Hispano-Moorish, Persian, or Indo-Persian style, very often were cultivated by more than one member of an in? strument-making family. The classic example in the Muslim world is the family of al-Hadad of Lahore.^^ Astrolabes signed by al-Hadad and then by his son "^ Isa remain from the sixteenth and early seventeenth centuries.^^ Muhammad 18 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y Muqim, who in A.H. 1053 [A.D. 1643] signed an astrolabe (No. 86), was al-Hadad's grandson. He had a brother, Qa^im Muhammad,^^ whose son Diya" al-Din Muhammad, a great-grandson of al-Hadad, made in A.H. 1070 [A.D. 1659] an astro? labe (No. 87) now also in the collection. This remarkably well-established family interest in in? strument making apparently died with Diya^ al- Din. At least, no known subsequent makers claim a lineage through Diya". In the inscription on his instruments the prolific Persian astrolabe maker Muhammad Mahdi identified himself as the son of Muhammad Amin. This Amin may be the maker of two seventeenth century astrolabes bearing that sig? nature, or he may be Muhammad Amin ben Mirza Khan who signed an astrolabe in A.H. 996 [A.D. 1587].^^ Hasan ben Ahmad's membership in a North African instrument-making family can be firmly established. His (probably younger) brother Muhammad ben Ahmad al-Battuti made both quadrants and astrolabes in the early eighteenth century. EIGHTEENTH CENTURY By the eighteenth century, the astrolabe had all but disappeared from the workshops of Euro? pean instrument makers.^^ However, astrolabes inscribed with Arabic characters surviving from the period are no less numerous than those sur? viving from preceding centuries. Thirteen eight? eenth-century astrolabes can be identified in the collection (almost 25 percent). Many of these originated in Isfahan, marking that Persian city as a manufacturing center of the importance of Lahore or Yazd. Isfahan's most prominent astro? labe maker and decorator was "^Abd al-A^imma. His name appears on roughly 50 instruments. A careful comparison of many of these recently undertaken by Gingerich, King, and Saliba, re? vealed several to be deliberate forgeries. These forgeries contribute as much to the affirmation of '^Abd al-A^imma's reputation as do the authentic instruments which carry his signature. It is not easy to determine when the forgeries were made; but their existence does suggest that, at some point in its history, the value of a Persian astro? labe could derive more from the reputation of its supposed maker than from the precision of its component parts. All four of the astrolabes signed by "Abd al- A^imma in the collection (Nos. 31, 37, 39, 40) appear to be authentic examples of his work. Characteristically,^' none of the four astrolabes is dated. On three (Nos. 37, 39, 40) ""Abd al- A^imma's signature appears alone. On one (No. 31) he is identified as the decorator of an instru? ment made by Muhammad Amin ben Muham? mad Tahir.^^ By comparing these four astrolabes it is possible to distinguish "Abd al-A^imma's role as "maker" from his role as "decorator." As maker he apparently fashioned each of the component parts of the astrolabe: he pierced the sighting plates of the alidade with two holes; he cut the rete in a distinctive style; he habitually paired certain projections when he engraved the plates; and he selected the tables to be included on the back of the instruments. Equivalent component parts show the hand of Muhammad Amin in the instrument "made" by him. It would seem that when "Abd al-A^imma functioned as decorator, he worked with partially engraved parts supplied to him by a maker. He was responsible only for the engraved details. The remaining eighteenth-century astrolabes in the collection (those not signed by ^Abd al- A^imma) fall into three categories: well-made Persian instruments; a well-made Indo-Persian instrument; and poorly made instruments of un? certain origin. Two of the fine Persian astrolabes (Nos. 42 and 65) exhibit clear affinities with and yet are not identical to the work of "^Abd al-A^imma and his collaborators. One (No. 42) was made by Hajji "Ali; the other is unsigned. At least two, and perhaps a third astrolabe in the collection, serve as evidence that "Abd al-A^imma's style did not entirely dominate the eighteenth century. "Abd al-Ghafur (ben Muhammad Sa^id) made one of these (No. 55) in A.H. 1198 [A.D. 1783], and is almost certainly the maker of a second (No. 54), dated A.H. 1196 [A.D. 1781], but unfinished and unsigned. The two instruments exhibit similari? ties in the pattern of their retes, the selection of NUMBER 45 19 scales on the back of each instrument, and the inscription of the scale on the rim. This latter inscription identifies the rim scale as four 90? arcs, a practice associated with seventeenth cen? tury Indo-Persian instrument makers.?^ Yet an? other unidentified Persian maker whose undated work is included in the collection (No. 47) in? scribed the rim scale in this manner. This anon? ymous maker also resembles "Abd al-Ghafur in the shaping of the throne (exhibiting Indo-Per? sian tendencies) and in the dividing of the upper left quadrant of the back; he otherwise exhibits a unique style. A single undated astrolabe (No. 88) of definite Indo-Persian origin appears, judging on the basis of the placement of stars in its rete,^^ to belong to the group of eighteenth-century astrolabes in the collection. No maker's name is engraved on the instrument, but a probable owner (Sahibuhu Maghfur al-Husayni al-Jilani) identified his property along the outer edge near the throne. The relative uncertainty of its date of origin does not diminish the value of this astrolabe in defin? ing the enduring aspects of the Indo-Persian style. These include the high pierced throne and elab? orate delicate rete familiar from seventeenth-cen? tury examples. They also include a tendency on the part of the maker to used dashed lines in his engraving. Further confirmation of its Indo-Per? sian origin is provided by the two prayer lines graphed in the upper right quadrant of the astro? labe's back. A comparison of "^Abd al-Ghafur's works with this example emphasizes the former's Indo-Persian tendencies. An astrolabe signed by Muhammad Sadiq (No. 52), dated A.H. 1189 [A.D. 1775], and two anony? mous instruments (Nos. 57 and 66), also believed to be eighteenth century, together document the existence of a class of technically correct but aesthetically unremarkable products of the astro- labist's workshop. The number and placement of divisions on these instruments suggest that their makers understood the astrolabe's function but were not inclined toward precision or elaborate decorative detail. Two of these astrolabes (Nos. 57 and 66) have a single feature in common with Indo-Persian instruments; that is, the normal po? sitions of "feet" and "finger" divisions of tangent and cotangent scales (feet-left, fingers-right) have been reversed (fingers-left, feet-right). These in? struments, though not of fine quality, should in no way be confused with fake or pseudo-astro? labes. Examples of this latter type in the collection are discussed later (see p. 20). NINETEENTH CENTURY By no means confined to the late eighteenth century, astrolabes of marginal quality more of? ten than not date from the nineteenth century. One candidate for this category, dated A.H. 1216 [1801], bears the signature of Sadiq (No. 53). It so closely resembles the work of Muhammad Sadiq described above that the former is surely a copy of the latter, executed by Sadiq or perhaps by "^Abd al-Karim.^^ The two instruments may be distinguished on the basis of the units com? prising each rim scale. On Muhammad Sadiq's astrolabe this scale is divided into 5? intervals; on Sadiq's astrolabe, this scale is divided into 6? intervals. A pair of nineteenth-century instruments in the collection serves to illustrate how a well-made, if simple, astrolabe can serve as a model for a less exacting maker. Unfortunately, nothing is known about the place and time of origin of the finer member of this pair (No. 3811). Its gazetteer is distinctive in that the reported longitudes have apparently been measured from the observatory at Greenwich rather than from the Fortunate Isles. This same distinctive gazetteer fills the umm of the astrolabe signed by Mirza Jahan Bakhsh (No. 63), also in the collection. This and other similarities between the two instruments suggest that the latter is a fairly faithful "copy" of the former. There is no indication that these two astrolabes are the products of one maker. Bakhsh's workmanship has an uneven quality not apparent in the anonymous effort. By inscribing his 5? rim scale so that it begins one division to the left of the meridian, Bakhsh has revealed the limits of his understanding of both the construc? tion and the use of the astrolabe. The collection contains two astrolabes dated A.H. 1281 [1864]. One of these (No. 61) has been 20 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y signed by Sahib "Ali Kabir Khan; the other (No. 62) carries the signature of ^Ala" al-Din. The two instruments are remarkably similar in all respects except quality of workmanship. "Ala" al-Din's astrolabe may be a copy of, but is by no means equivalent to, the well-made astrolabe signed by Kabir Khan. Kabir Khan's instrument is itself at least reminiscent of the eighteenth-century astro? labes signed by "Abd al-A"imma. The influence is most apparent in his choice of rete pattern, but it extends to decorative details on the back and plates. Yet another pair of astrolabes can, on the basis of quality of workmanship and placement of stars, be traced to a nineteenth-century origin. The instruments are virtually identical from rete, through plates, to back. Neither one carries a maker's name. Because it is complete and care? fully (if not finely) made, one instrument (No. 2566) appears to be the model member of the pair. The unfinished, carelessly worked state of the other (No. 64) identifies it as the "copy." The gazetteer of each instrument begins with Isfahan, suggesting that this may have been the site of manufacture. A comparison between these twins and the works of "^ Abd al-A "imma and his Isfahan school reveals a definite difference in attention to detail and perhaps supplies subtle commentary on changes in the Muslim world. Like Kabir Khan's astrolabe, an example in the collection signed Hamza (No. 59) includes a rete patterned after those included on instruments signed by "^ Abd al-A"imma. Hamza's astrolabe is dated A.H. 1268 [A.D. 1851]. His workmanship is creditable. An anonymous and undated astrolabe in the collection (No. 49) also incorporates one of "Abd al-A"imma's rete designs. The relatively inferior workmanship on this example suggests that it probably should be attributed to a nine? teenth-century maker. None of the other nine? teenth-century instruments in the collection ex? hibit the independent quality of workmanship evident in the effort of Muhammad Akbar, dated A.H. 1234 [A.D. 1818] (No. 58).^^ His inscriptions are clear; his graphs and scales accurate; his decorative details beautifully executed. There is no evidence that he borrowed his rete pattern from an earlier maker. Muhammad Akbar's work identifies him as one of the last of the true "Asturlabi," exhibiting the qualities of both sci? entist and artisan. One last nineteenth-century instrument in the collection remains to be described (No. 4000). Except for a brief Persian inscription on the kursi, it is inscribed in Sanskrit characters.^^ A Sanskrit inscription in the umm identifies the maker of this astrolabe as Vitrabhadra, the son of Rama; the date is 1805. The Persian inscription identifies an owner, Rama of Jodhpur, who acquired the in? strument in 1815. The accuracy of the division and inscription of the scales of this astrolabe indicate that the maker was a knowledgeable astrolabist. Some of these scales are familiar, some distinctive. The shadow squares are examples of the former. On Indo-Persian instruments made in Lahore, the left square is divided into fingers; the right into feet. Curiously, the placement is reversed on this Sanskrit example just as it is on Persian astrolabes. A distinctive scale relating solar declination to degrees of right ascension occupies the lower half of the back of the Sanskrit astrolabe. No other instrument in the collection carries such a scale. PsEUDO-ASTROLABES There exists no more convincing evidence of the diminished scientific role of the nineteenth- century astrolabist than the number of astrolabe? like instruments characterized by incoherent if not entirely nonsensical inscription. The National Museum of American History preserves four ex? amples of these pseudo-astrolabes (Nos. 2570, 4002, 4003, 4004). Objects like them may be purchased any day in the markets of the Middle East. They are in no way to be confused with functional astrolabes. It is, however, easy to see how they could be. Pseudo-astrolabes have all of the separate pieces of their functional counterparts: body, plates, rete, alidade. However, any resemblance between the two classes of objects ends here. The body of a pseudo-astrolabe is characterized by a lack of plate holder (no tooth, notch, or cylinder NUMBER 45 21 in the umm). Each plate, in turn, lacks the means to be held (no notch, tooth, or hole). The rete is characterized by a symmetry, which also extends to the placement of the star names. There are invariably no sighting holes in the "sighting plates" of the pseudo-alidade. Inscription on a pseudo-astrolabe is, at the most, neat. Engraved lines may be carefully, if incorrectly, placed. "Pseudo" does not necessarily mean sloppy. Respectable appearances can be deceiving. A neatly engraved rim scale with di? visions marked in order "80 5 85 5 70 5 75" (as on No. 2570) is no rim scale at all. Ten neatly engraved lines do not constitute a projection of seasonal hours. Pseudo-plates that carry projec? tions forming geometric designs have only that to recommend them. They have no astronomical significance, and are of no use to serious astrolo? gers. In their clumsy, superfluous way, pseudo- astrolabes serve to illuminate the valued qualities of a scientific instrument: accuracy, precision, and utility. Comparative Analysis This comparative analysis of the collection fo? cuses on each of the functional elements of the instruments: suspensory apparatus, throne, rim, front central area, back, plates, rete, alidade, and index. Whenever a scale is discussed, both its position and its division units are reported before any general remarks are made. Conventional trig? onometric coordinates are used to specify the positions of scales or other features on the astro? labes. Thus, a scale that extends "from the 0? point counterclockwise around the rim to the 90? point" falls in the quadrant illustrated in Figure 18. A "tooth at 90?" specifies the feature shown in Figure 19. Discussions of the use of parts of the astrolabe generally rely upon statements made by Masha^allah, al-Biruni, or Chaucer. Size Astrolabes in the collection range in size from 75 mm in diameter (No. 15) to 248 mm in diameter (No. 2005). This is fairly representative of extant portable astrolabes, except for the very large instruments (284-590 mm) made in Europe in the sixteenth century by the Arsenius brothers. Few of the makers represented in the collection can be said to prefer a particular size. Neither is it possible to identify a given period of time in which astrolabes of uniform dimension were con? structed. Suspensory Apparatus Astrolabes in the collection fully illustrate the variety one is likely to encounter in the means by which this instrument is suspended. In general the apparatus consists of a stirrup-shaped handle with an axle passing through the throne from front to back (see Figure 7) and an attached ring which, as Chaucer points out, can be "put on the thumb of thy right hand when taking the height of things." On most instruments inscribed in Arabic characters the ring passes through an opening in the handle. Masha^allah emphasizes the importance of beveling the edges of the con? tiguous parts "lest their movement be impeded." Several instruments show that makers paid atten? tion to this detail: good examples in the collection are Nos. 37, 57, 85, 144, 2569. For the most part, the handles of instruments inscribed in Arabic characters are quite plain, but on some examples (e.g., Nos. 54, 55, 64, and 87) it has the shape of a Moorish arch. On some European instruments (e.g., Nos. 221, 2006, 2007) the top of the handle has been pierced and fitted with a rotating pin to which the ring is attached (see Figure 20). Throne T H E MASHRIQ Thrones on those astrolabes in the collection originating in the mashriq seem to fall naturally into three categories: high and pierced, high and inscribed, of medium height and simple decor. The high, pierced variety tends to be associated vvith Indo-Persian makers, the Lahore school of the seventeenth century. These include especially Nos. 70, 85, 86 (in this example, plates cover the original pierced throne), 87, 88, 2567, and 2569. At least one high, pierced example (No. 4) was made in Isfahan in the twelfth century, suggesting an early source for the Indo-Persian style. Astrolabes in the collection suggest that at least as early as the seventeenth century, Persian mak? ers considered the throne to be an appropriate location for inscriptions. The most popular is a brief quotation from the Koran which incorpo? rates the word kursi. It appears on Nos. 39, 55, and 66, all made in the eighteenth and nineteenth centuries. Placement of this passage on the kursi of an astrolabe amounts to a religious dedication. Occasionally, when it is not the appropriate pas? sage from the Koran, the throne inscription will identify either the person who commissioned the astrolabe (No. 40), or the maker of the instrument (No. 63). On at least one example (No. 59) the 22 NUMBER 45 23 180? 0? 180? 270? FIGURE 18.?Scale extending from the 0? point counterclock? wise around the rim to the 90? point. 270? FIGURE 19.?Plate with tooth at 90 ' FIGURE 20.?Astrolabe No. 2006. Handle pierced and fitted with a rotating pin. inscription on the throne records a transfer of ownership. One of the most interesting throne inscriptions appears on the astrolabe constructed by Muham? mad Mahdi in A.H. 1078 [A.D. 1667] (No. 25). It is a list of towns encircling a hole that once must have held a compass. The towns are arranged so that they proceed clockwise in order of increasing longitude. About half of the listed towns have longitudes less than the longitude of Mecca and about half have longitudes greater than that of Mecca. The third category into which thrones from the mashriq may be grouped includes examples from all Persian schools of design. The early examples evidence the simplest taste (see Nos. 15 and 47). Decoration involving vines and flowers embossed on a stippled background characterizes instru? ments from the eighteenth century (see, e.g., Nos. 37 and 42.) T H E MAGHRIB The thrones of astrolabes in the collection made in the maghrib add little bulk or decor. They are 24 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y small, generally less than half as high as they are wide. The earliest designs in the collection, those made before the fifteenth century (Nos. 144, 2572, 3643, 4001) are pierced with holes. Later exam? ples (e.g.. No. 2568) are solid and may carry inscriptions identifying the maker. EUROPE With a few notable exceptions (two in the collection, i.e., Nos. 2005 and 2007), the thrones of European astrolabes are unpretentious both in size and decoration. The latter generally consists of rosettes or scroll work. Inscription is rare. It occurs on only one instrument in the collection. On the back of the throne of astrolabe number 204, near the body, the word "Midi" has been engraved. The two exceptionally elaborate thrones among the European instruments in the collec? tion were designed in Italy (No. 2005) and France (No. 2007). The Italian example (made ca. A.D. 1580) incorporates a covered compass hole flanked by scroll-work. The French example (made ca. A.D. 1584) is decorated front and back with grotesque masks. Front Rim T H E MASHRIQ AND MAGHRIB Scales of Degrees There are two types of degree scales. The first (1) is continuous (i.e., a single 360? scale). It appears on instruments from both regions. The second (2), known only from the mashriq, is really a group of four 90? scales. Position: (1) The 360? scale extends from the 90? point clockwise around the rim back to the 90? point. (2) The 90? scales extend from (a) the 90? point clockwise around the rim to the 0? point (b), from there to the 270? point, (c) then from the 270? point to the 180? mark, and (d) finally from there back to the 90? point. Units: The scales may be marked on the outer band every 5? with inscribed numerals and on the inner band at 1? intervals (uninscribed). Or they may be marked at 6? intervals with inscribed numerals and at 1? intervals (uninscribed). Use: (1) The 360? scales are used with the muri (a marker on the ecliptic circle of the rete at 0? Capricorn) to measure the right ascension of the rising point of the celestial equator. (2) The 90? scales can be used with the alidade to measure either zenith distance (on arcs [a] and [c]) or altitude (on arcs [b] and [d]). Remarks: (1) The instrument in the collection that carries the scale consisting of units of 6? apparently dates from the early twelfth century (No. 3643). It is inscribed with Kufic characters of the maghribi type. On all other Kufic instru? ments, this scale consists of units of 5?. In inscrib? ing the 5? or 6? scales, some makers were quite explicit and wrote out all the digits of each des? ignation from " 5 " (No. 2572) or " 6 " (No. 3643) to "360"; some, on the other hand, streamlined the process by implying the hundreds digit except in the designation of "100," "200," or "300" (No. 144). Further modification of this last method, whenever scales of 5? intervals were being in? scribed, involved writing a " 5 " in the first and every other interval; thus: "5 10 5 20 5 . . . 300 . . . 5 60" (No. 4). A maker's decision to write out "360," or to drop the first digit, or to write "55" or " 5 " apparently depended on nothing more than individual taste. No national or time-based traditions reveal themselves. (2) The perception of this scale as a combination of four arcs of 90? is characteristic of seventeenth-century Indo-Per? sian astrolabes (Nos. 86, 87, 2569). It is also found on three eighteenth-century astrolabes (Nos. 47, 54, and 55) and on one astrolabe completed early in the nineteenth century (No. 53). All astrolabes incorporating this type of scale are engraved with Arabic Naskhi characters. The scale is com? prised of 6? divisions on three instruments (Nos. 53, 87, and 2569). Altitude Scales These 90? scales are found primarily on instru? ments made in the mashriq. Position: These scales extend from the 0? point counterclockwise around the rim to the 90? point; from the 180? point clockwise to the 90? point; NUMBER 45 25 from the 0? point clockwise to the 270? point; and from the 180? point counterclockwise to the 270? point. Units: The scales are marked in units of 5?, inscribed with numerals (outer band), and 1? uninscribed (inner band), or 6?, inscribed with numerals, and 1? uninscribed. Use: When used with the muri, these altitude scales measure the position of the vernal equinox relative to the zenith or nadir. (For the use with the alidade arm, see description of altitude scales on the back of the astrolabe, p. 30). Remarks: Astrolabes in the collection with rim scales arranged in this way date from the late eighteenth and early nineteenth centuries (Nos. 58, 61, 62, 66). In every case of their appearance on the rim, the altitude scales duplicate similar scales on the back of the astrolabe. None of these instruments is equipped with a second alidade arm. The scale is divided into units of 6? on only one instrument, dating from the mid-eighteenth century (No. 66). EUROPE Winds Position: This scale extends from the 90? point clockwise around the rim back to the 90? point. Units: Marked in units equal to one-twelfth of a circle, inscribed with the names of the winds. Use: This scale identifies direction (e.g., north, northeast) with the name of the wind associated with that direction (Septentrio, Aquilo).^^ Remarks: Wind scales appear on two astro? labes in the collection, these being the instrument signed "Galois" (No. 204) and the instrument signed by Georg Hartman (No. 262). The named winds compare as follows: Galois MERIDIES LIBONOTVS APHRICVS OCCIDENS CHORVS CIRCIVS SEPTENTRIO BOREAS Hartman AVSTER LIBONOTVS APHRICVS FAVONIVS CHORVS CIRCIVS SEPTENTRIO AQUILO VVLTVRNVS CECIAS ORIENS SVBSOLANVS EVRVS VVLTVRNVS EVROAVSTER EVROAVSTER Similar wind scales are incorporated into the design of horizontal sundials dating from the first and second centuries A.D. Equal Hours Position: This scale extends from the 90? point clockwise around the rim to the 270? point and from the 270? point back to the 90? point; or it extends from the 90? point clockwise around the rim back to the 90? point. Units: 1 hour or 15? inscribed with numerals, and 12 minutes or 3? uninscribed, or 4 minutes or 1? uninscribed. Use: A scale of equal hours is used with the index arm (or alidade) and the rete to indicate the time of day associated with a given configu? ration of the stars or the sun relative to the horizon. Remarks: On two instruments in the collec? tion, one signed " M P " (No. 221) and another apparently made in Italy (No. 186), the 24 divi? sions of the hour scale are inscribed with numbers from 1 to 24. On all other instruments, this scale is inscribed as two successive scales of 12 divisions. The scales inscribed by Hartman (No. 262) and Danfrie (No. 2007) use Roman numerals. All other astrolabists employed Hindu-Arabic nu? merals. All of the European instruments in the collection are fitted with an index arm or with an alidade, attached so that it can be used with the hour scale. Scales of Degrees Position: This 90? scale is marked in one of three ways. It extends from the 90? point clock? wise around the rim to the 0? point; from the 90? point counterclockwise around the rim to the 180? point; from the 270? point clockwise around the rim to the 180? point; and from the 270? point counterclockwise around the rim to the 0? point. Or it extends from the 0? point counter? clockwise around the rim to the 90? point; from 26 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y the 0? point clockwise around the rim to the 270? point; from the 180? point counterclockwise to the 270? point; and from the 180? point clockwise to the 90? point. Or it extends from the 90? point clockwise around the rim back to the 90? point. Units: This scale is marked in units of 5? inscribed with numerals (outer band), or 10? inscribed with numerals, and 2? uninscribed. Use: Degree scales in the first position were apparently designed to be used with an alidade to measure zenith distance. Degree scales in the second position were apparently designed to be used with an alidade to measure altitude. Degree scales in the third position were apparently de? signed to be used with the fixed index incorpo? rated into the rete to measure the right ascension of the rising point of the celestial equator. Remarks: The various methods of inscribing the scale of degrees are distributed as follows: scales in the first position appear on all European instruments in the collection, with the exception of that made by Georg Hartman (No. 262), who used the second position, and the early Italian astrolabe (No. 186), which includes a degree scale in the third position. Evidence in the collection suggests that the content of the rim was not as firmly established among European makers as it was among makers in the Muslim world. Front Central Area T H E MASHRIQ Over half of the astrolabes in the collection have been inscribed with a list of cities along with their latitudes, longitudes, and one or more other geographic parameters (such as distance from Mecca). This list or gazetteer can invariably be found in the circular depression that constitutes the front of the body of a Persian or Indo-Persian instrument. Both very early (e.g.. No. 15) and very late (No. 62) Persian astrolabes in the collec? tion incorporate such a list. For a number of reasons, an astrolabe's gazet? teer can be considered to be one of its most interesting elements. First of all, its presence ties the astrolabe to an instrument-making tradition whose earliest surviving products include some portable sundials from antiquity. Both the so- called "Tischendorf Dial," dating from the fourth century, and the Aphrodisias Dial Plate, probably from the same period, are engraved with circular tables of places and their latitudes.^? There is, then, at least a Byzantine precedent for Ja^far ben ^Umar's inclusion (in A.D. 1372) of a circular table identifying each of a series of places with a longitude and latitude on a portable scientific instrument. The tables on the Byzantine dials proceed in order of increasing latitude; Ja^far's table proceeds in order of increasing longitude. While the form of an astrolabe's gazetteer re? lates it to portable sundials, its content relates it to a body of astronomical literature, the earliest example of which being "one of the most impor? tant astronomical documents of antiquity," Clau? dius Ptolemy's 'irpoxeipojv Kavovuiv (Handy Tables). One of these tables is a list of important cities (the 'TToXeia ETnarjiioL), with geographical latitudes and longitudes reckoned from the For? tunate Isles. ^ Variations on Ptolemy's theme were produced as astronomical zijes by Arab authors, beginning in the early ninth century with the work of Yahya ben Abi Mansur. These zijes typically contain (among other things) lists of cities and their geographical coordinates, tables of ecliptic coordinates of selections of "fixed" stars, and astrological tables of various kinds. Persian and Indo-Persian astrolabe makers al? most certainly consulted zijcs like Yahya's for the information necessary to complete umm, '^ankabUt, and zahr. Publication of any one well-known zij could easily lead to the identification of astrolabes dependent upon it.^^ In the meantime, it will have to suffice to discuss the gazetteers of the instruments in the collection without knowing the exact source of their contents. Ja^far ben "^Umar's gazetteer (No. 15), in? scribed in the fourteenth century, has been de? scribed above. The majority of cities included are located in what is now Iran (formerly Persia). Longitudes (atwal), reported in degrees and min? utes of arc, have been measured from the Fortun- NUMBER 45 27 ate Isles. Latitudes {"^urud), reported in degrees and minutes of arc, have been measured from the equator. The third recorded geographical param? eter (inhiraf), the inclination of the azimuth of the Qibla, is the measure of the arc of the horizon, reported in degrees and minutes, intercepted be? tween the meridian of the listed city and a vertical circle passing through the zenith of Mecca. The data relating to each city seem to be arranged in order of increasing longitude beginning with Baghdad (80?00') and ending with Samarqand (98?00'). Ibn Ash-Shatir published ca. A.D. 1350 a zij that included a list of 290 cities related to just the three geographical parameters included in Ja'^far's gazetteer. It is possible that Ibn Ash Shatir was J a "far's source of geographic data. The gazetteer inscribed by "^ Ali ben "^Awad al- Mahmudi in the early seventeenth century (No. 70) is uniquely brief. It consists only of a list of 25 cities and their latitudes, grouped according to some geographic criteria that happen to preserve roughly modern national boundaries. The list places Mecca and Medina in successive places of honor,^^ a practice duplicated in almost all of the gazetteers containing these cities. One notable exception to the tendency to so distinguish these cities is the gazetteer inscribed in A.D. 1643 by the Indo-Persian maker Muhammad Muqim (No. 86). The city of Lahore is given the place usually reserved for Mecca. Indo-Persian astrolabe makers (at least in the seventeenth century), including Muhammad Muqim, tended to limit the content of gazetteers to longitude and latitude. The practice, no doubt, reflects their limited appreciation of the impor? tance of the Muslim religious center. Certainly they did not consider it to be an important geo? graphic reference point. The geographic param? eters which they do include apparently do not derive from a single zij. The collection contains only one example (No. 44, the work of Muhammad Mahdi) of a gazet? teer inscribed with geographic parameters rounded to the nearest degree. A second astrolabe in the collection, dating from the late seventeenth century and inscribed by Qasim "^Ali, is notewor? thy because the information contained in its gaz? etteer includes at least the beginnings of a table of length of daylight.^^ About half of the Persian (distinct from the Indo-Persian) gazetteers in the collection include a second table (in addition to the inhiraf), which relates listed cities to Mecca. This second table, labeled jihat or al-jihat, gives the direction of the azimuth of the Qibla with respect to the four cardinal points. The entries in the table are "abbreviations," which can be interpreted as fol? lows: SHSH is Sharqi ash-Shamal or the northeast; SHJ is Sharqi al-Janub or the southeast; GHJ is Gharbi al-Janub or the southwest; J A Q is Janub al- Qibla, or south of the Qibla; S H A Q is Shamal al- Qibla, or north of the Qibla. The latter two terms appear on instruments in the collection that date from the eighteenth or the nineteenth century. These terms do not occur in the two jihat tables on astrolabes dating from the seventeenth cen? tury. Both of these seventeenth-century instru? ments (Nos. 25 and 44) are ascribed to the maker- decorator Muhammad Mahdi. Figure 21 illustrates the gazetteer of the astro? labe decorated by Muhammad Mahdi in A.D. 1667 (No. 25). Table 6 is a transcription of the contents. The selection and sequencing of cities apparent in this example can also be detected in the gazetteers of about 14 others in the collection (see, e.g., Nos. 31, 37, 39, 40, 42, 47, 49, 55, 57, 58, 59, 61, 62, 65). Undoubtedly all are derived from a single textual source, or zij. The geo? graphic parameters included in these gazetteers are similar enough to confirm a common heritage. Astrolabe gazetteers which do not follow the tradition exemplified by Muhammad Mahdi's instrument, at least in this collection, are those made most recently, i.e., in the nineteenth cen? tury. For example, the astrolabes signed by Mu? hammad Sadiq (No. 52) and Sadiq (No. 53) contain similarly sequenced gazetteers, quite un? like Mahdi's. Astrolabes No. 2566 and No. 64 contain twin gazetteers that cannot be tied to either of the two Sadiqs or to the Mahdi instru? ment. The sequence of the contents of the gazet? teer of the anonymous astrolabe No. 66 is unique 28 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 21.?Astrolabe No. 25. Gazetteer, with alidade, wedge, and pin. NUMBER 45 29 TABLE 6.?Transcription of gazetteer inscribed on CCA No. 25 Place name Geographic parameters OUTER BANDS Bilad Mecca Medina Mi?r Sana "^ a Lahsa Bait al-Maqdis Dimashq Halab Mo?ul Nakhijewan Maragheh Tabriz Ardabil Surra Man Ra^a Kufa Baghdad Basra Shustar Kazerun Shiraz Yazd Hamadan Qazvin Gerfadman Isfahan Kashan Qpm Taliqan Asterabad Semnan Damghan Bilad Meshhad Nishapur Tun Sabzawar Herat Qayen Balkh Hormoz Kerman Qandahar Lahore Kashmir Ganjah Tiflis Shirwan Atwal 77 10 75 20 63 20 77 00 83 30 66 30 70 00 72 10 77 00 81 15 82 00 82 00 82 30 79 00 79 30 80 00 84 00 84 30 87 00 88 00 89 00 83 00 85 00 85 30 86 40 86 00 85 40 85 45 89 35 88 00 88 15 Atwal 92 30 92 30 92 30 91 30 94 20 93 20 101 00 92 00 92 30 107 40 109 20 88 00 83 00 83 00 91 30 ^Ard 21 40 25 00 30 20 14 30 24 00 31 50 33 15 35 50 34 30 38 40 37 20 38 00 38 00 34 00 31 30 33 25 30 00 31 30 29 15 29 00 32 00 35 10 36 00 34 14 32 25 34 00 34 45 36 10 36 50 36 00 36 20 "Ard 37 00 36 21 34 30 36 00 34 30 33 40 36 41 25 00 29 50 23 00 31 15 31 00 41 20 43 00 36 00 INNER BANDS Inhiraf 00 00 37 10 18 38 01 15 69 30 45 16 30 31 18 29 07 12 12 15 16 17 15 00 17 33 07 16 12 31 12 15 37 19 35 24 11 17 33 38 48 28 22 16 27 34 33 41 40 18 34 31 31 14 29 13 18 48 36 17 28 00 Inhiraf 45 06 46 25 55 20 44 12 14 08 14 36 65 36 74 30 62 11 75 05 78 26 75 20 15 49 14 41 44 13 Al-jihat 00 SHJ SHJ SHSH GHJ SHJ SHJ SHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ Jihan GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ GHJ 30 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y in the collection. While the sequencing of all these gazetteers may differ from Mahdi's, the recorded geographic parameters are not significantly dif? ferent. In all cases longitude has been reckoned from the Fortunate Isles. Two nineteenth-century astrolabes in the col? lection (Nos. 63, 3811) include gazetteers that differ from Mahdi's not only in the sequencing of the selected cities but also in the reckoning of the geographic parameters. As the inscription on the throne of No. 63 explains, the longitudes have been reckoned from the Observatory of Green? wich, London. That these parameters were de? rived independently of the source for earlier pa? rameters is suggested by the fact that no constant difference exists between the longitudes on No. 63 or No. 3811 and the longitudes on Mahdi's gazetteer. In addition, considerable differences exist in the latitudes recorded by Mahdi and the latitudes on Nos. 63 and 3811. The gazetteer of Sadiq's astrolabe (No. 53) contains, in addition to longitude, latitude, and inhiraf, an as yet undescribed scale called al-mas- afat. This particular parameter is a measure of the arc of a great circle intercepted between the zenith of a listed city and the zenith of Mecca. On Sadiq's instrument this measure is given in farsakhs. Sadiq's is the only instrument in the collection to contain this scale. Appendix I includes a transcription of all gaz? etteers in the collection. T H E MAGHRIB Sterographic projections sometimes appear on the front central area of maghribi astrolabes. There are two types of such projections. The first (1) applies to the habitable latitudes. The second (2) applies to the Tropic of Capricorn (60 ?N) and the polar regions (90 ?N). Position: This projection fills all four quad? rants of the base plate. Units: (1) See the section on plates (p. 47). (2) The projection is marked in units of 3? inscribed with numerals, or 6? inscribed with numerals (altitude and azimuth). Use: (I) See the section on plates (p. 47). (2) A stereographic projection for the latitude of the Tropic of Capricorn 66?, when used with the "^ankabut, functions effectively as an indicator of celestial latitude and longitude. A stereographic projection for the North Pole, when used with the '^ankabut, functions effectively as an indicator of the celestial coordinates of right ascension and declination. Remarks: (1) The two instruments in the col? lection that incorporate a plate projection into the umm were made in A.D. 1090 (No. 2572) and A.D. 1504 (No. 2571). The projections correspond to 37?30' and 30?, respectively. (2) The two astrolabes in the collection containing polar pro? jections were made in A.D. 1101 [?] (No. 3643) and 1304 (No. 144). The astrolabe containing a projection for latitude 66?N was made in A.D. 1224 (No. 4001). EUROPE Stereographic projections of habitable latitudes sometimes appear on European instruments. Position: The projection of habitable latitudes fills all four quadrants of the base plate. Units: See the section on plates (p. 47). Use: See the section on plates. Remarks: Most of the front central areas of European instruments in the collection are either blank or contain only the skeleton (tropics, equa? tor, and solstitial and equinoctial colures) of a stereographic projection. Three instruments con? tain complete projections, however. For all three, this projection constitutes each instrument's only plate. The latitudes represented are 52? (Nos. 304 and 2006) and 48? (No. 204). Back T H E MASHRIQ Altitude Scales Position: These central circular scales extend from the 0? point counterclockwise around the N U M B E R 45 31 outer margin to the 90? point, and from the 180? point clockwise to the 90? point. Units: The scale is marked every 5? inscribed with numerals (outer band) and 1? uninscribed (inner band), or 6? inscribed with numerals, and 2? uninscribed. Use: The scale is used in conjunction with the alidade (see p. 6) to measure the height of the sun or of a well-known fixed star above the horizon. Remarks: Two astrolabes in the collection, both apparently from Lahore in northern India (Nos. 87, 2569), have altitude scales divided into units of 2? and 6?. Cotangent Scales Position: These central circular scales (see Fig? ure 1) extend from the 270? point clockwise around the outer margin to the 180? point, and from the 270? point counterclockwise to the 0? point. Units: The scale is divided into "feet," in? scribed with numerals and the identification zill aqdam (lower left), and "fingers," inscribed with numerals and the identification zill asdbi'^ (lower right).^? Use: The cotangent scales may be used as a nomogram (with the altitude scale and the ali? dade) to determine the altitude of the noon sun from the ratio between a gnomon of specific length (e.g., 7 feet or 12 fingers) and the length of its noon shadow (read on the cotangent scale). The scales are also useful for determining times of Muslim prayer defined by the ratio between a gnomon's length and the length of its shadow. The alidade is set at a shadow length appropriate for a given prayer, the astrolabe is hung plumb, and the prayer begun when a ray of sunlight passes between the pinholes of the alidade's sight? ing plates. Remarks: The units adopted for this scale (or, more correctly, these two scales) have their origin in the shadow table tradition.'^ The use of 7- and 12-unit gnomons (feet and fingers) to compute such tables is known from Greek times. Hart? ner ?? clearly illustrates the relationship between this scale and a second foot or finger scale, the box tangent/cotangent scale (or shadow square), found on the back of the astrolabe. The latter can in a certain sense be viewed as an aid to the construction of the former. While the two scales can function independently, they tend to appear together. This is, in fact, the case for all but five of the astrolabes in the collection that are in? scribed with Arabic characters. On only one of these exceptions (No. 4) does the circular scale stand alone. Hartner reports that scales divided into "feet" are most often found in the lower left quadrant of the astrolabe, while scales divided into "fingers" are most often found in the lower right. Indeed, most of the instruments in the collection (23 examples) follow Hartner's model. The 11 exceptions reverse the quadrants (feet lower right, fingers lower left). It is difficult to assign a reason for the difference, except to note that five of the nonconformists were made in western Persia or northern India. Arc Scales Position: The declination arcs are distributed in the upper right quadrant (see Figure 1). The arcs representing the azimuth of the Qibla are graphed on the declination arcs in the upper right quadrant between the 0? radius and the 45? radius. The arcs representing the altitude of the noon sun are graphed on the declination arcs in the upper right quadrant between the 45? radius and the 90? radius. The arcs representing equal or unequal hours are generally graphed onto the upper left quadrant, especially when other scales occupy the upper right. Prayer arcs are graphed onto the upper right quadrant when it contains the declination scale. Units: The number of declination arcs varies from 6 (or one division for each sign of the zodiac) to 60 (or one division for each 1/10 sign of the zodiac). On some instruments the number of divisions varies from one sign of the zodiac to another. On most instruments the zodiacal seg? ments are each divided by five arcs. Generally the 32 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y zodiacal segments are identified by name along the 0? radius and the 90? radius. Along the 0? radius the inscriptions indicate that the outer arc corresponds to the winter solstice and the inner arc corresponds to the summer solstice. The in? scription along the 90? radius identifies the outer arc with summer and the inner arc with winter solstice. The number of arcs representing the azimuths of the Qibla varies from two to seven, each corresponding to a Middle Eastern city and inscribed with the name of that city. The arcs representing the altitude of the noon sun corre? spond to latitudes of position ranging from 22? to 50? and are inscribed with numerals. The hour arcs generally represent seasonal hours or 1/12 daylight, and are inscribed with numerals. The prayer arcs represent prayer times, inscribed khatt akhir al- ^ asr (line of end of afternoon prayer), khatt awwal al-^asr (line of beginning of afternoon prayer), khatt az-zawal (line of the going down of the sun), khatt nisj an-nahar (line of midday), or awwal az-zuhr (beginning of a little time after midday). In some cases these inscriptions are modified by the designation of a specific latitude (e.g., '^ard 32). Use: The declination arcs serve primarily as the context for other graphed scales. A declination scale without accompanying graphs does occur on one instrument in the collection (No. 47). The quality of workmanship on this particular exam? ple, however, suggests that the maker did not finish his work on the upper right quadrant. It is difficult to imagine a use for the declination arcs other than as context for graphs of azimuth, altitudes, hours, or prayers. The lines representing azimuths of the Qibla have been graphed on the declination arcs so that when the alidade of the astrolabe is placed at the point where the line for a city crosses the arc corresponding to the solar declination on a given day of the year, it marks on the altitude scale the altitude of the afternoon sun as it passes through the azimuth of the direc? tion of Mecca from that city. The lines thus provide a means for determining the direction of Mecca from important centers of the Muslim world. The lines representing the altitude of the noon sun function nomographically in a manner analogous to the lines indicating the azimuth of the Qibla. The noon sun lines essentially aid in the determination of due south (the direction of the noon sun) for cities and towns at various latitudes. As Henri Michel has explained,^^ the hour arcs can be considered a nomogram to be interpreted using the alidade in combination with the central circular altitude scale. Given the altitude of the noon sun, one can set the alidade at this altitude on the altitude scale and mark the point on the alidade arm which crosses the sixth hour arc. Measures of solar altitude made at any time on the given day can be related to a seasonal hour by observing the hour line coinciding with the mark made on the alidade corresponding to solar altitude measured at noon. Prayer lines are used as a nomogram with the alidade and central circular altitude scale to as? sociate solar altitude with times of Muslim prayer. Since each prayer line functions for a unique latitude, one might expect to find (as one does on examples in the collection) a latitude included as part of each inscribed identification. Remarks: The declination arcs function equally well as a nomogram whether they result from stereographic projection or are equally spaced. Equally spaced arcs appear on four in? struments in the collection (Nos. 42, 63, 70, 4000). It seems that the decision to employ one distri? bution rather than the other was purely a matter of taste. On most of the instruments in the collec? tion the Qibla azimuth lines are arranged so that cities for which the angular distance between due South and the direction of Mecca is greatest appear lowest in the quadrant. In order to ensure that these Qibla lines will fall into the lower eighth of the quadrant, they are associated with declination circles identified by the scale along the 0? radius of the quadrant. In order to ensure that they will fall in the areas of the upper right quadrant between the 45? radius and the 90? radius, the altitude arcs are related to declination circles identified by the scale inscribed along the 90? radius. On no instrument in the collection NUMBER 45 33 are prayer lines combined with other graphed arcs. Prayers are graphed on the declination arcs of six of the instruments in the collection made in the mashriq (Nos. 4, 70, 85, 86, 87, 88). The majority of these are of Indo-Persian origin. Angle Scales Position: These scales of sines and cosines oc? cupy the upper left quadrant. Units: These are the sines (and cosines) of angles that are multiples of 1? or 5? (see Figure 22), or sines (and cosines) of angles having sines (and cosines) that are multiples of 1/60 of the radius of the quadrant (see Figure 23). Use: Sines (and cosines) distributed according to multiples of degrees are used with an alidade which is graduated so that it can measure off a given vertical (or horizontal) component of a sine (or cosine) to be related to an angle which can be read from the altitude scale. Sines (and cosines) distributed according to multiples of 1/60 of the quadrant radius require an alidade to measure an angle to be related to a sine or cosine, which can be read directly from the scale. The former distribution would be most useful in cases where the trigonomic function is known and its corre? sponding angle is to be found. The latter scale would be more useful when the angle is known and its sine or cosine is to be found. Remarks: Sines and cosines which divide the quadrant into 60 parts are found on astrolabes in the collection more often than sines and cosines related to multiples of degrees. Only seven instru? ments exhibit the latter configuration (Nos. 15, 25, 39, 47, 54, 55, and 70). On a few of the astrolabes in the collection, the quadrant contain? ing the cross-hatching of trigonometric parts also contains radii evenly dividing the quadrant (Nos. 58, 63, and 70) or concentric quarter circles cor? responding to the solar declination at various times of year (Nos. 47, 52, and 85). On six ex? amples, the unequal hours also divide the trigo? nometric quadrant (Nos. 25, 39, 54, 55, 57, and 59). On two other examples (Nos. 53 and 4000) only a noon line appears (a semi-circle with di? ameter equal to the quadrant radius). Box Scales These examples of tangent scales (a and d) and cotangent scales (b and c) can be seen in Fig? ure 1. Position: The tangent scales extend vertically from the radius at 180? to the radius at 225?, and from the radius at 0? to the radius at 315?. The cotangent scales extend horizontally from the radius at 270? to the radius at 225? and from the radius at 270? to the radius at 315?. Units: These scales are marked in "feet," in? scribed with numerals and identified as zill aqdam ma'^kHs (a in Figure 1) and zill aqdam mustawi (b in Figure 1); and "fingers," inscribed with numerals and identified as zill asabi" mustawi (c in Figure 1) and zill asabi"^ ma'^kus (d in Figure 1). Use: See section on central circular cotangent scales (p. 31), and note that the tangent scale has a use when the altitude of the noon sun is less than 45?; the cotangent scale should be used when the altitude of the noon sun is greater than 45? Remarks: Judging from the examples of their work in this collection, maker-decorator "Abd al-A^imma (Nos. 37, 39, and 40) and maker Muhammad Mahdi (Nos. 25 and 44) preferred to abbreviate the identification of each cotangent scale to mustawi. For related remarks see the sec? tion on central circular cotangent scales (p. 31). Circular Tables Position: The tables extend from the 180? point on the astrolabe counterclockwise to the 0? point (see Figure 24). Units: These tables are marked according to the following: astrological term or limit, 1/60 of a semicircle, inscribed with the name or the last letter of the Arabic name of one of the five visible planets that governs the term (generally the out? ermost band; see Table 7); astrological term or limit, 1/60 of a semicircle, inscribed with numer? als from 6 to 30, representing term boundaries within zodiacal signs (generally the band just inside the band of planetary governors of the terms; see Table 7); zodiacal sign, 1/12 of a 34 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 22.?Astrolabe No. 25. Upper left quadrant includes sines of angles, which are multiples of r NUMBER 45 35 FIGURE 23.?Astrolabe No. 87. Upper left quadrant includes sines, which are multiples of 1/60 of the radius of the quadrant . 36 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y -Iff it-'''.^ l^. m ^ . ' ? FIGURE 24.?Astrolabe No. 37. The lower quadrants contain a full set of astrological tables NUMBER 45 37 TABLE 7.?Planetary governors and degree divisions of terms (I = Mushtari, Jupiter ; H = Zuharah, Venus; D = ^ Utarid, Mercury; K H = Mirrikh, Mars; L = Zuhal, Saturn) TABLE 8.?Zodiacal signs Zodiacal S Planetary governors I H D K H L H D I L K H D I H K H L K H H D I L I H L D K H D H I K H L gns 0-5 Degree divisions 6 12 20 25 30 8 14 22 27 30 6 12 17 24 30 7 13 19 26 30 6 11 18 24 30 7 17 21 28 30 Zodiacal S Planetary governors L D I H K H K H H D I L I H D L K H D I H L K H D H I K H L H I D K H L gns 6-11 Degree divisions 6 14 21 28 30 7 11 19 24 30 12 17 21 26 30 7 14 22 26 30 7 13 20 24 30 12 16 19 28 30 semicircle, inscribed with the Arabic name of each sign of the zodiac (generally the third band from the outside; see Table 8); astrological face, 1/36 of a semicircle, inscribed with the Arabic name or the last letter of the name of the sun, moon, or one of the five visible planets that governs the face (fourth band from outside; see Table 9); and lunar mansion, 1/28 of a semicircle, inscribed with the Arabic name of the lunar mansion (innermost band; see the accompanying tabulation, Lunar Mansions). Use: The term tables associate the five visible Number 0 1 2 3 4 5 6 7 8 9 10 11 Name English Aries Taurus Gemini Cancer Leo Virgo Libra Scorpio Saggitarius Capricorn Aquarius Pisces Arabic al-Hamal al- Thawr alfawza as-Saralan al-Asad as-Sunbula al-Mizan al-'^Aqrab al-Qaws al-Jady al-Dalw al-Hul planets with unequal divisions of each zodiacal sign. The zodiacal signs are associated with the terms and the faces. The faces table associates a visible planet or the sun or moon with each 10? division of the zodiac. The table of lunar man? sions specifies yet another division of the heavens. Sharatan Butain Thurayya Dabaran Haq'^ah Han'^ah Dhira^ LUNAR MANSIONS Nathrah Tarfah Jabhah Zubrah Sarfah '^ Awwa Simak Ghafr Zubana Mil Qalb Shaula Na "^a^im Baldah Dhabih Bula^ Su'^Ud Akhbiyya Muqaddam Mu ^akhkhar Rishsha Remarks: All of the collection's astrolabes containing term tables indicate a preference for a system of association between planets and zo? diacal divisions that Ptolemy attributes to Egyp? tian astrologers.?^ On two of these (Nos. 4 and 2569) the numbers on the inner term table indi? cate the number of degrees in each division of a zodiacal sign rather than the position of each division point within the sign. Only one astrolabe in the collection contains in addition to a table of faces a semicircular table of thirds of a sign, called (by Hindus according to al-Biruni)?^ decanates. This relatively rare example was signed in West? ern Kufic characters by Ja^far ben '^Umar (No. 15). 38 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y TABLE 9.?Planetary governors (first row) and degree divisions (second row) of astrological faces (KH = Mirrikh, Mars; S = Shams, Sun; H = Zuharah, Venus; D = "^Utarid, Mercury; R = Qamar, Moon; L = Zuhal, Saturn; I = Mushtari, Jupiter) K H 10 S 20 H 30 D 10 R 20 L 30 I 10 K H 20 S 30 H D R 10 20 30 L 10 I 20 K H 30 S H D 10 20 30 R L I 10 20 30 K H 10 S H 20 30 D 10 R L 20 30 I 10 K H 20 S 30 H 10 D 20 R 30 L I 10 20 KH 30 Rectangular Tables Position: A table of triplicities fills the shadow square (if any) in the lower left and lower right quadrants. Generally it consists of five rows (the upper one of which contains the table name) and ten columns (see Table 10). Units: Signs of the zodiac inscribed with nu? merals or names and associated with planetary lords, each identified by the last letter of its Arabic name and each associated with one of the Aristotelian elements: earth, air, fire, or water. Use: The table associates those signs of the zodiac which are situated in the zodiac at the angles of equi-angled triangles and records the nature of each trigon thus formed. In addition, the table associates planets with similar "natures" with each trigon. Remarks: The triplicity scale on No. 2569 fills the lower left shadow square only. The inscrip? tions that specify the nature of each trigon (fiery, earthy, airy, watery) can be used to delineate subgroups of astrolabes carrying this table. The four words that can be recognized as "Arabic" terms for these natures appear consistently on astrolabes made in Isfahan. At least two instru? ments (Nos. 55 and 85) carry what Morley calls the "Persian" terms for these natures.?^ Some instruments (e.g.. No. 58) seem to carry a mixture of "Arabic" and "Persian" terminology. T H E MAGHRIB Altitude Scales Position: These circular scales are centered on the instrument (see the section on "The Mash- r iq ,"p . 30). Units: See section on "The Mashriq." Use: See section on "The Mashriq." Remarks: On No. 3643 the altitude scale is continued into the lower quadrants of the astro? labe back. These additions extend from the 180? point counterclockwise to the 270? point, and from the 0? point clockwise to the 270? point. Cotangent Scales Position: These circular scales are centered on the instrument. They extend from the 225? point clockwise approximately to the 190? point, and from the 225? point counterclockwise approxi? mately to the 253? point; also from the 315? point clockwise approximately to the 287? point, and from the 315? point counterclockwise ap? proximately to the 343? point. TABLE 10.?Planetary lords, zodiacal signs, and natures of the triplicities (L = Zuhal, Saturn; S = Shams, Sun; K H = Mirrikh, Mars, H = Zuharah, Venus; I = Mushtari, Jupiter, D = "Utarid, Mercury, R = Qamar, Moon) Planetary lords L K H I R S H L H I R D KH L K H I R I R D KH S H L H Zodiacal signs Saggitarius Capricorn Aquarius Pisces Leo Virgo Libra Scorpio Aries Taurus Gemini Cancer Natures Fiery Earthy Airy Watery NUMBER 45 39 Units: These scales are divided into "fingers" or arcs of a circle (with center at the center of the astrolabe) bounded by radii that mark off lengths along the extended horizontal scale of the shadow square. The arcs are inscribed with numerals specifying the length marked by the radii, meas? ured in units equal to 1/12 of the shadow square's gnomon (see Figure 2). Use: See section on "The Mashriq" (p. 31). Remarks: On No. 4001 this scale is identified with the inscription sah(?) asdbi" az-zillayn. If the astrolabes in this collection are representative, it appears that this scale is often omitted entirely from examples made in the maghrib. Zodiacal Tables Position: The table extends counterclockwise through 360?, starting at the 0? point. Units: This table is, inscribed with numerals every 5?, and inscribed with the names of the zodiacal signs at 30? intervals. Use: The table is combined with a circular calendar table and the alidade to read the longi? tude of the sun for a given calendar date. Remarks: All non-European astrolabes in the collection having a complete 360? central circular zodiac table originated in the maghrib. Calendar Tables Position: The table extends counterclockwise 360? from a beginning point (representing 1 Jan? uary), which varies in position from 286? to 292?. Units: The table consists of one-day units un? inscribed on the outer band, and five- to six-day units inscribed with numerals (central band), and one-month units, inscribed with month names (inner band). Use: Combined with a circular table of zodia? cal signs and the alidade, this table is used to read the longitude of the sun given the date. Remarks: All non-European astrolabes carry? ing a circular calendar originated in the maghrib and are associated with a 360? central circular zodiac table. Two relatively early non-European astrolabes (Nos. 144 and 304) contain eccentri? cally-centered calendar tables (see Figure 25). TABLE 11.?Solar cycle (1 = Sunday, 2 = Monday, etc.) Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Day of the week 3 4 6 7 1 2 4 5 6 7 2 3 4 5 Leap year indicator K K K K Year 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Day of the week 7 1 2 3 5 6 7 1 3 4 5 6 1 2 Leap year indicator K K K The remaining calendar tables on instruments originating in the maghrib (Nos. 2568, 2571, 3643, and 4001) are concentric with the zodiac table. On both maghribi instruments with eccen? trically centered calendar tables, the radial posi? tion of the apsidal line (the line connecting the centers of the zodiacal and calendar tables) is about 75?. On one maghribi astrolabe in the collection (No. 4001) a series of central circular scales relates a day in the week (represented by the numerals 1 to 7) to a date in the Julian calendar. The table (see Table 11) reflects the fact that the day of the week of the first of January advances one day every year, except following a leap year, when it advances two days (365 leaves a remainder of 1 upon division by 7). The pattern of possible first days repeats after 28 years (7 days X 4 years). The count of years is often called the solar cycle. Arc Scales The arc scales on maghribi instruments are usually confined to scales of hours. See the dis? cussion of mashriqi arc scales (p. 31). Box Scales Position: See description under "The Mash- riq," (p. 33). 40 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 25.?Astrolabe No. 144. The center of the calendar scale does not coincide with the center of the scale of zodiacal signs. NUMBER 45 41 Units: These scales are marked in fingers, in? scribed with numerals, and identified by the in? scription az-zill al-ma "kus and az-zill al-mabsUt. Use: See the section on "The Mashriq" (p. 33). Remarks: Two characteristics distinguish the box scale on the maghribi instruments in this collection from the examples originating in the mashriq: the scales are divided into fingers only, and the cotangent scale is identified with the term mabsut (extended) instead of mustawi (level). EUROPE Altitude Scales Position: These circular scales are centered on the instrument. They extend from the 0? point counterclockwise to the 90? point, and from the 180? point clockwise to the 90? point, and from the 180? point counterclockwise to the 270? point, and from the 0? point clockwise to the 270? point. Units: See section on "The Mashriq" (p. 31). Use: See section on "The Mashriq" (p. 31). Remarks: The altitude scale on one European instrument in the collection (No. 2007) can be read to within half a degree. This is the paper astrolabe by Danfrie. The units of the altitude scale on the "de Roias" instrument in the collec? tion (No. 2005) are 10? inscribed with numerals, and 2? uninscribed. Zodiac Tables Position: See section on "The Maghrib" (p. 39). Units: See section on "The Maghrib." Use: See section on "The Maghrib" (p. 39). Remarks: The 5? divisions on the astrolabe signed by Har tman (No. 262) are marked with Roman numerals. Calendar Tables Position: The scale extends counterclockwise 360? from a beginning point which varies from 279? to 291? (see Figure 26). Units: The scale is divided into units of one day, uninscribed, and 4-11 days, inscribed with numerals, and one month, inscribed with month names. Use: See section on "The Maghrib" (p. 39). Remarks: About half of the calendars on the cataloged European instruments have eccentric centers on apsidal lines at 90? The paper astro? labe in the collection (No. 2007) contains in addition to the calendar a set of tables designed to facilitate the determination of the date of Easter. The first of these is a table (see Table 12) of the solar cycle associated with a dominical letter. (On No. 2007, this table is a small circular scale in the upper left quadrant of the astrolabe back.) The dominical letter is a letter from A to G associated with the first Sunday of the year. Assignment of the letter A to the first Sunday FIGURE 26.?Astrolabe No. 186. The date of the vernal equinox (Aries 0?) is 10.5 March. 42 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y indicates that it is the first day of the year; B, that it is the second day, and so on. It should be apparent that the pattern of dominical letters repeats after 28 years in the same way that the pattern of days of the week associated with Jan? uary first repeats. Hence, the dominical letters are found in conjunction with the so-called solar cycle. (See the discussion of a related table in the section on central circular scales found on magh? ribi astrolabes, p. 39). Note that leap years in the cycle are indicated by two dominical letters. The letter on the right is that corresponding to the first Sunday of the year. That letter corre? sponds to subsequent Sundays of the year until the leap day is intercalated when the "dominical" letter becomes the one on the left. A second circular table (see Table 13) found on the paper astrolabe (No. 2007) relates the golden number, the epact, and the paschal index (see Figure 27). The golden number is nothing more than a reference to a year in the 19-year Metonic cycle connecting lunar and solar cycles (235 X 29.53 = 19 X 365.25). The epact is a statement of the moon's age on the first day of the year. (The table on the paper astrolabe is in keeping with the tradition of beginning the Me? tonic cycle [or golden numbers] at a point when the first day of the year coincides with the new TABLE 12.?Solar cycle and associated dominical letter (taken from CCA No. 2007; extrapolated dates in brackets) TABLE 13.?Paschal index and epact with related golden number Year A.D. [1568] [1569] [1570] [1571] [1572] [1573] [1574] [1575] [1576] [1577] [1578] [1579] [1580] [1581] Solar cycle 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Domin? ical letter F, G E D C A, B G F E C, D B A G E, F D Year A.D. [1582] 1583 1584 1585 [1586] [1587] [1588] [1589] [1590] [1591] [1592] [1593] [1594] [1595] Solar cycle 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Domin? ical letter C B G, A F E D B, C A G F D, E C B A Year A.D. [1577] [1578] [1579] [1580] [1581] [1582] 1583 1584 1585 [1586] [1587] [1588] [1589] [1590] [1591] [1592] [1593] [1594] [1595] Paschal index 12 Apr 1 Apr 21 Mar 9 Apr 29 Mar 17 Apr 6 Apr 26 Mar 14 Apr 3 Apr 23 Mar 11 Apr 31 Mar 18 Apr 8 Apr 28 Mar 16 Apr 5 Apr 25 Mar Epact I XII XXIII IIII X V XXVI VII XVIII X X I X X XXI II XIII XXIIII V XVI XXVII VIII X X I X Golden number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 moon.) Note that each epact is 11 days later than the one preceding, reflecting the fact that the lunar year (12 X 29.5 = 354 days) is 11 days shorter than the solar year. The paschal index is the date of the first full moon falling on or following after the vernal equinox (21 March in the Gregorian calendar). The third table in the series is a central circular table adjacent to the calendar (see Table 14). Although it is not identified on No. 2007, the table is commonly referred to as the "Gregorian Calendar of Epacts." It enables the user to predict the dates of the new moon for any year, given its epact. The epacts are distributed so that they fall next to the date that would correspond to the new moon if the moon was the age of the epact on 1 January. Sometimes two epacts fall on the same day in order to allow for the circumstances that some synodic months have 30 days and some have 29. (In a year with epact 25, new moons fall on the dates opposite 25 if the golden number is less than 11, and on dates opposite 25' if the golden number is greater than 11.) Note that the maker of astrolabe No. 2007 included the years A.D. 1583, 1584, and 1585 at NUMBER 45 43 ^M^-' .!u//rr}.m jiMiit,. ^'i^j aun'- fttflin j:r.: ?-?:r''' ? 7ceu(tc^UJ.ir*yf.'- ?'?: r '?'^if^^iii^'^., FIGURE 27.?Astrolabe No. 2007. Calendric wheels can be used to find the date of Easter. 44 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y TABLE 14.?Gregorian calendar of epacts Count of days 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Jan 29 28 27 26 25'25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Feb 29 28 27 25'26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Mar 29 28 27 26 25'25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Apr 29 28 27 25'26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 29 May 28 27 26 25'25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 29 28 Jun 27 25'26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 29 28 27 Jul 26 25'25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 29 28 27 25'26 Aug 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 29 28 27 26 25'25 24 Sep 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 29 28 27 25'26 25 24 23 Oct 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 29 28 27 26 25'25 24 23 22 Nov 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 29 28 27 25'26 25 24 23 22 21 Dec 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 29 28 27 26 25'25 24 23 22 21 19'20 appropriate places in the calendric tables. These are crucial years following the Gregorian reform. These specialized calendric tables were undoubt? edly used to compute the date of Easter, a rather involved procedure with or without the help of tables. For example, to find the date of Easter in A.D. 1584, one would consult Table 13 to find that year's golden number (8), epact (18), and paschal index (26 March). Knowing that Easter would be the first Sunday following 26 March, one would consult Table 12 to find the dominical letter. In 1584, Sunday was the first day of the year (day A). By counting by sevens from 1 January on the calendar scale (and remembering that February has 29 days in 1584), one would find the first Sunday following 26 March to be 1 April, Easter Sunday. Non-calendric tables are rarely to be found on astrolabes of European origin. Only one Euro? pean instrument in this cataloged collection (No. 204) contains such a table, and that is a table of star names, which also gives their longitude, mag? nitude, "mediation" (see page 56), and declina? tion. In this example, the stars named are exactly those included in the rete design. Arc Scales Position: The scales of unequal or seasonal hours (1) are generally distributed in the upper NUMBER 45 45 left and upper right quadrants. The scale of equal hours (2) occupies the upper left quadrant of No. 221. Units: The seasonal hours are inscribed with numerals. The equal hours are marked in units equal to 1/24 of the time between meridian pas? sages of the sun. Use: See the section on "The Mashriq" for a discussion of unequal hours, which may be ap? plied to both equal and unequal hours. Remarks: (1) On astrolabe No. 221 the scale of unequal hours is confined to the upper right quadrant, with a scale of equal hours graphed in the upper left quadrant. (2) Any equal hour scale is only relevant for one latitude.^^ This fact can be used to establish the place of intended use for an instrument of otherwise uncertain origin. Box Scales Position: The shadow square, the box scale that appears on almost all astrolabes, is discussed in the section on "The Mashriq" (p. 33). Units: This scale is marked in units of one finger, uninscribed, and two, three, or four fin? gers, inscribed with numerals and identified Um? bra Recta (horizontal scales) and Umbra Versa (ver? tical scales). It may also be marked at every sexagesimal unit, uninscribed, and five sexagesi? mal units, inscribed with numerals. Use: See section on "The Mashriq" (p. 33). Remarks: Whereas the box scales are identified by name on every instrument containing them originating in the mashriq or maghrib, such titles are omitted from box scales on two European instruments in the collection (Nos. 186 and 304). Both of the latter are attributed to fifteenth-cen? tury makers. On one other European example (No. 2006) the box scales are inscribed Corda Recta and Corda Versa. Special Projections T H E QUADRANT.?Position and Units: A special projection on the back of one of the European instruments in the collection (No. 2006) was first described by Profatius in the thirteenth century.^^ It consists of the important circles of an astrolabe plate and rete folded into a single quadrant (see Figure 28). The folded circles in this example include Tropics of Cancer and Capricorn, Equa? tor, ecliptic and horizons for six cities Qerusalem, Carthage, Rome, Paris, Oxford, Berwyk). In or? der to divide these projected circles, the curved boundary of the quadrant carries two scales. The outer one (at least on No. 2006) is characterized by equal divisions corresponding to degrees of right ascension (angular distance along the celes? tial equator). The unequal divisions of the inner one (again, at least on No. 2006) correspond to degrees of longitude (angular distance along the ecliptic). The radial boundaries of the quadrant also carry various scales. Astrolabe No. 2006 con? tains three scales: an unequally divided scale identifying the terrestrial latitude of projected horizons; an unequally divided scale identifying the celestial declination (angular distance north or south of the celestial equator) of projected parallel circles; and an equally divided scale of sines {sinus droit) and versed sines {sinus versus). As outlined below, the sine scales are used in conjunction with the only lines comprising the quadrant, which are normally found on the back of an astrolabe, i.e., the graph of unequal hours, including the semicircle corresponding to noon. These lines appear on the quadrant as they would in the upper right quarter of a familiar European or maghribi astrolabe. Use: The quadrant projection is useless unless it has been fitted with a thin string, fixed to its vertex and threaded with a moveable bead (No. 2006 has both). With string and bead in place, the quadrant can best be used to tell the seasonal hour from the measured solar altitude. Recall that for any day of the year the maximum (or noon) altitude of the sun is equal to 90? plus the solar declination (which may be a negative num? ber) minus the latitude of position. Using the stretched thread, this angle is marked off along the right ascension scale (starting at the 0 hour line), and the bead is moved to the point of intersection of the thread and the sixth hour line. Leaving the bead to mark this spot, the stretched thread is moved so that it marks off the angle of 46 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY ^"^^ FIGURE 28.?Astrolabe No. 2006. Quadrant . NUMBER 45 47 solar altitude as measured. The bead marks the appropriate seasonal hour.?^ The declination scale can be used to measure solar declination by first moving the bead on the stretched string so that it intersects the ecliptic at the given solar longitude. Keeping the bead in place, it is only necessary to move the string to the scale and read off the declination. D E ROIAS PROJECTION.?Position and Units: Whereas in the stereographic projection, utilized in the earliest astrolabes, the center of projection is the south celestial pole and the plane of projec? tion is the equatorial plane, in the projection named after its originator, Johannes de Roias, celestial circles and points are projected orthogo? nally onto the plane of the solstitial colure. Jo? hannes de Roias published a description of this projection in A.D. 1550.*^ It is distinguished by the straight lines parallel to the equator corre? sponding to parallel circles of declination and by the semi-ellipses that are the projections of merid? ians of right ascension. On the example in the collection (No. 2005), 37 parallels are projected (one every 5?) and 37 meridians are projected (one every 5?; see Figure 29). The parallels cor? responding to the beginning of every sign of the zodiac are marked with the appropriate zodiacal symbol. The meridians are identified as hours before or after noon. In this projection the ecliptic becomes a diagonal line, passing through the center and the points of intersection of the tropics with the solstitial colure. In a major departure from the stereographic tradition, de Roias also projected bright stars on his network. The exam? ple (No. 2005) contains a Cygni, a Aquilae, a Leonis, 8 Capricorni, a Canis Minoris, and a Tauri. Use: In order to function as a nomogram, the de Roias astrolabe must be fitted with a ruler that can pivot about the center of the projection. The fiducial edge of this ruler is the projection of a horizon. Often (as the example illustrates) this ruler is divided by the orthogonal projections of azimuth circles. The ruler is completed by the addition of a small perpendicular rule, which can move freely along the ruler. This perpendicular rule, preserved in other examples, is missing from No. 2005. The rule is divided by projections corresponding to parallels of altitude. With the complete ruler and rule in place, one can use the de Roias projection to solve all of the familiar astrolabe problems. As an example, con? sider the determination of the equal hour given a measurement of solar altitude. Set the ruler so that the angular distance be? tween it and the projection of the Earth's axis is equal to the latitude of position. Note the division on the rule corresponding to the measured solar altitude. Move the rule along the ruler so that the noted division intersects the projected parallel of solar declination corresponding to the date. The projected meridian of right ascension also inter? secting at this point is the desired hour curve. Plates Over half of the astrolabes here considered are equipped with four (13 examples) or five (15 examples) circular plates, each plate containing projected lines on both sides. If these plate at? tachments are part of a proper instrument, their edges are each interrupted by a rectangular "tooth" or "notch" (see Figure 30) designed to fit into or over a notch or tooth in the raised rim of the astrolabe body. With one exception, sets of toothed plates in the collection are part of the instruments of western origin (whether European or maghribi). The exceptional set belongs to an unsigned astrolabe, which has been attributed to Fadl "^ Ali (No. 47). Each tooth in the exceptional set is situated at the lowest point of a plate (where a notch is found on other examples originating in the mashriq) and not at the highest point, where, based on the western instruments, one would expect to find a tooth. CONTENTS Colures While other projected circles may vary from instrument to instrument, no functional plate ever lacks the basic set of projected tropics, equa- 48 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 29.?Astrolabe No. 2005. De Roias pro? jection. *-v"V. Q / ? ??' iiltll C ?? ?? J * M I >I0|t , ^ ; 'Jl i ! ? .:.., J illMJ j y NUMBER 45 FIGURE 30.?Astrolabe No. 49. A tooth projecting from the rim holds the notched plates. 49 \ m ^WKmwKKItmSS^^^^^K^^ 50 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY tor, and solstitial and equinoctial colures. The celestial tropics and equator are concentric cir? cles; the colures are orthogonal straight lines. No inscription identifies the circles on Islamic instru? ments. On all six cataloged instruments of magh? ribi origin, however, the lower half of the solstitial colure is identified as Zawal, Az-Zawdl, or Khatt Az-Zawdl. Inscriptions do identify the tropics and equator on two European instruments: on No. 262, Capricornus (outer). Equator (middle), and Can? cer (inner); and on No. 2007, Tropicus Capricorni (outer), Aequinoctialis (middle), and Tropicus Cancri (inner). On one set of European astrolabe plates (No. 2007) the equinoctial colure is marked Ho? rizon Rectus at left and right. The Horizon Line Aside from the stereographic projections of the solstitial and equinoctial colures (perpendicular diameters), the fundamental projected line on an astrolabe plate is the northern portion of the horizon circle for a given latitude between 23? and 64? (see Figure 31). On most instruments inscribed with Arabic characters (whether East? ern or Western) this projected line is marked with the word Al-Maghrib or Maghrib (west) at right and Al-Mashriq or Mashriq (east) at left. Analogous identification is rarer on European instruments. Only the horizons of the paper astrolabe in this collection (No. 2007) are marked Horizon Obliquus at left and right. Most Arabic or Persian instruments (30 out of 41 in this collection) include one plate face on which only horizons have been projected, from four to seven per quadrant, covering the latitudes from 4? to 66? at intervals of 2? or 3?. In general, only the northeastern segment of each horizon (half its usual extent) is projected on this plate. An exception is the horizon plate of a European instrument made in A.D. 1542 (No. 221) (see Figure 32). For four instruments in the collection (Nos. 59, 61, 62, and 65) over 50 horizons are projected in eight segments of a plate face (rather than in four quadrants). Three of these instru? ments are marked with nineteenth century dates. In the absence of compelling contradictory evi? dence, it seems likely that the fourth, undated, instrument (No. 65) is also of nineteenth-century origin. On almost half of the instruments in the col? lection inscribed with Arabic characters, the in- ?>??.*{? 'i?A''i%#v*f??- FIGURE 31.?Astrolabe No. 4. Inscription on the horizon line of the plate marks its eastern and western extension. FIGURE 32.?Astrolabe No. 221. Plate for multiple latitudes. NUMBER 45 51 eluded "horizon plate" also contains scales of declination along the solstitial and equinoctial colures between the tropics. Al-Biruni does not comment on the use of this scale; but no doubt it was used to mark the declination of a point on the ecliptic. Altitudes The number of altitudes engraved on an astro? labe plate generally depends on the plate's di? mensions. In al-Biruni's words, "A complete as? trolabe is one that has ninety altitudes num? bered . . . 1 to 90 from horizon to zenith. If the instrument is too small to contain all these, then only every second altitude (half-size) or third, or sixth, or tenth is marked (but not fifth, although this form should be made)."^^ All but three of the instruments in this collection contain plates with altitudes projected at every 2?, 3? or 6?. Astro? labes containing altitudes at every 5? are seem? ingly as rare as al-Biruni indicates. Only one maghribi (No. 3643) and one European (No. 186) instrument in the collection are so engraved. The diameters of plates carrying altitudes every 3? range from 220 mm to 135 mm; diameters of plates containing altitudes every 3? range from 160 mm to 100 mm; diameters of plates contain? ing altitudes every 6? range from 139 mm to 64 mm. On Islamic instruments, altitudes are identified with abdjad^^ numbers, normally on both the left and right sides of the meridian' (see Figure 33). Hindu-Arabic numerals identify the altitude on European instruments, which, if they are marked at all, are marked both at left and right of the meridian line (see Figures 34 and 98). Twilight The moments of nightfall and dawn are occa? sions for Muslim prayer, and may be indicated on the plates of instruments inscribed with Arabic characters by the stereographic projection of the lower boundary of twilight (a circle parallel to and about 18? below the horizon). Only one Arabic instrument in the collection (No. 52) in? cludes such a twilight line (engraved shafaq al-fajr) on its plates. However, plates on three European astrolabes (Nos. 262, 2005, and 2007) carry the FIGURE 33.?Astrolabe No. 65. Plates for 36? latitude (left) and multiple latitudes (right). Compare with Figures 32 and 34. 52 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 34.?Astrolabe No. 221. Plates for 49? (top) and 36? (bottom). Compare the lower plate with the plate for 36? in Figure 33. the plates of all but two of the astrolabes in the collection. These are distributed, in the over? whelming majority of cases, at intervals of 10?. The few exceptions contain azimuths projected at 5? (Nos. 25, 64, 187, 2007, and 2566), 6? (Nos. 2569 and 3643), or 15? (Nos. 85, 304, 4000) intervals. Abdjad numbers identify azimuth lines on instruments inscribed in Arabic characters; Hindu-Arabic numerals, on European instru? ments. In every case the numbering begins at the intersection of prime vertical and horizon and continues to 90? at the intersection of solstitial colure and horizon. A set of astrolabe plates commonly includes a projection for latitude 66? (see Figure 35). The practice appears to be limited to instruments inscribed with Arabic characters (at least in this collection). That this face served to verify the celestial latitude and longitude of stars included in the "ankabUt is suggested by its typical identify? ing inscription: (safiha) mizan al- "ankabut (Nos. 85, 86, 87 and 2569). The content of this face is generally limited to altitudes and azimuths (with major divisions identified by names of zodiacal signs). The inclusion of an "^ankabut plate appar? ently is not confined to a single Islamic maker or locale. The instruments in the collection associ? ated with "Abd al-A^imma suggest that some makers preferred not to include a rete projection. Azimuth lines projected onto an astrolabe face generally occupy the area above the projected oblique horizon; but, as examples in this collec? tion illustrate, some makers preferred to project only the portion below the horizon on some plates (see Nos. 15, 44, 57, 65, and 86). Some astrolabists projected all the azimuths that fit between the projected zenith and Tropic of Capricorn (see Nos. 47, 59, and 62). On many examples of plates, the prime vertical alone was projected above and below the horizon (Nos. 61, 66, 85, and 2569). twilight projection. In one case (No. 2007), the projection is marked Linea Aurorae sive crepusculinae. Unequal Hours Azimuths Projections of azimuths, or great circles passing through zenith and nadir points, are found on On the plates of all but one astrolabe in the collection, the projections of tropic and equator circles below the horizon are divided into 12 equal parts by the so-called "seasonal" hour lines. NUMBER 45 53 -^--//o, FIGURE 35.?Astrolabe No. 88. Plate for 66? latitude, used to verify celestial latitude and longitude of stars on the '^ankabHi. The set of plates without seasonal hour lines belongs to an instrument of questionable func? tionality inscribed with Arabic characters (No. 2570). On European instruments, the unequal hours are generally marked by Hindu-Arabic numerals. On instruments in the collection in? scribed with Naskhi characters, abdjad numbers mark the unequal hours; on two Naskhi examples (Nos. 25 and 40) these numbers are replaced by the ordinals from "first" to "twelfth": Al-Awwal, Ath-Thdni, Ath-Thdlith, Ar-Rabi'^, Al-Khdmis, As- Sddis, As-Sdbi", Ath-Thdmin, At-Tdsi", Al-^Ashir, Al-Hddi, Ath-Thdni ^Ashar. The latter convention is more typical on maghribi instruments (specifi? cally, four out of six in the collection). Astrolabe No. 2572 incorporates seasonal hours, which are identified according to astro? logical practice as being either male or female, and either animal, vegetable, or mineral (see p. 175). Equal Hours Astrolabes in the collection that are of Euro? pean or maghribi origin do not carry plates en? graved with hours other than seasonal. Other 54 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y types of hours are projected on plates of the collection's instruments originating in the mash? riq, in all but three cases. These supplementary hours are 1/24 of a day and are of equal duration throughout the year. In most examples in the collection they are counted from the western ho? rizon {horae ab occasu or Italian hours); in one example. No. 2566, they are counted from the eastern horizon {horae ab ortu, or Babylonian hours); and in four examples (Nos. 54, 58, 87, and 88) they are counted from both eastern and western horizons. As Hartner^^ points out, an explanation for the preference for horae ab occasu can be found in the fact that the Muslim day begins at sunset. Except in cases where the same numbers identify both equal and unequal hours, the equal hours on instruments in the collection are identified by Hindu-Arabic numerals. On only one plate carrying hours counted from both horizons are those hours identified (No. 58); other double projections are uninscribed. Houses The astrological houses, a division of the eclip? tic into six parts above and six parts below the horizon, are rarely projected onto the plates of instruments inscribed with Arabic characters. No. 65 is the only one mashriqi instrument in the collection with such a projection (see Figure 33). On the other hand, four western instruments (Nos. 204, 221, 2005, 2007) include plates with house divisions. On the astrolabe from the mash? riq, the houses are marked with the ordinals (starting at the eastern horizon and proceeding counterclockwise): Tdli", Thdni, Thdlith, Rdbi", Khdmis, Sddis, Sdbi"", Thdmin, Tdsi", "Ashir, Hddi "Ashar, and Thdni "Ashar. The houses on the Eu? ropean instruments are marked either with Ro? man numerals (Nos. 204, 2005) or with Hindu- Arabic numerals (Nos. 221, 2007). The fact that houses are not normally engraved on instruments inscribed with Arabic characters is perhaps explained by the accepted Muslim procedure for determining house divisions as de? scribed by al-Biruni (using the rete and appro? priate plate): To equalize the twelve houses, place the degree of the ascendant [the point of the ecliptic rising at the moment of birth] on the east horizon, the point of the ecliptic on the west horizon is the cusp [beginning boundary point] of the seventh house. Then look at the meridian, what has arrived there is the sign and cusp of the tenth house To deter? mine the cusps of the other houses, turn the rete inversely so that the degree of the ascendant comes under the horizon, and is placed on the line indicating the beginning of the eleventh unequal hour, i.e., through two temporal hours . . . then look at the meridian line to see what sign is there and what degree; it will be the cusp of the ninth house and its degree, the nadir of which is the cusp of the third. Then move the rete . . . so that the degree of the ascendant rests on the beginning of the ninth hour. Such a systematic use of unequal hour lines makes the projection of house divisions unnecessary. Prayers Plates contained in each of the maghribi in? struments in the collection carry lines that mark four times of Muslim daily prayer. Although these are day times, the lines seldom interrupt the altitude circles. Generally they are distributed between the projected unequal hours, within the second, eighth, tenth, and eleventh. Placed as they are, it is thus the intersection of the nadir point of the sun with these lines that marks the prayer times in the morning, at midday, in the afternoon, and at evening. The lines (from right to left) are inscribed Khatt Ash-Shafaq (morning prayer), Khatt Az-Zuhr (midday prayer), Khatt Al- '^Asr (afternoon prayer), and Khatt Al-Fajr (eve? ning prayer). In the light of the evidence pre? sented by the collection, it seems that Hartner's statement^^ that "most Islamic astrolabes" have prayer lines should be revised to refer to "most maghribi astrolabes." Latitude and Length of Daylight The differences between networks of lines pro? jected on the faces of a set of astrolabe plates are important, to be sure, but subtle enough to re? quire distinguishing inscriptions. On every func? tional plate face, an inscription somewhere along the meridian, below the horizon, identifies the latitude for which the projection applies. On most instruments inscribed with Arabic characters this NUMBER 45 55 inscribed latitude {"Ard) is accompanied by an associated length of daylight {sd^dtuhu). These associations of latitude with length of daylight were presumably taken from tables that had been a part of the geographic literature since Ptolemy. Two distinguishable methods are represented by astrolabes in the collection (see Table 15). The first (A) is represented by a majority of the ex? amples, including the four instruments signed by ?^Abd al-A^Imma. It is based on the assumption that the obliquity of the ecliptic is 23.59?. The second (B) follows the convention of incrementing the length of daylight by four minutes of time for every increase of a degree of latitude. One very early (No. 4) and two relatively late (Nos. 52, 53) astrolabes in the collection follow the latter. Most Eastern instruments contain plates only for integral degrees of latitude. One exception (No. 88) has a plate face for latitude 21?40' (= Mecca) and a plate face for 31?50' (= Lahore). A second exception (No. 42) has a plate face for 29?26' (=^ Shiraz 29?36') and a plate face for 32?20' {^ Isfahan 32?35'). Both No. 37 and No. 64 have plate faces marked 38?05', corresponding most closely to the latitude of two towns on the gazetteers of these instruments: Tabriz (38?00') and Ardabil (38?00'). No. 64 also has a plate face marked latitude 42?08'; there is no equivalent latitude on its gazetteer. TABLE 15.?Length of daylight associated with latitude via two distinct traditions Latitude (?N) 30 31 32 33 34 35 36 37 38 39 40 Tradi Hours 13 14 14 14 14 14 Length of tion A Minutes 57 7 17 28 39 52 daylight Tradit Hours 14 14 14 14 14 14 ion M B inutes 8 16 24 32 40 Latitudes engraved on maghribi and European instruments are not generally associated with lengths of daylight. Maghribi instruments in the collection tend to include plate faces for latitudes 21?40 ' (o r2r30 ' ) ,31?30 ' , and33?40 ' (or33?30 ' ) . Plate faces on the European instruments in the collection are limited to integral degrees except for the plate engraved 43?30' (= southern France?) on No. 221. ENGRAVING STYLE The networks projected onto an astrolabe plate face are generally made of continuous engraved lines, blackened for emphasis. Notched, dotted, or jagged lines may appear on maghribi instru? ments (see Nos. 186, 2571, 3643, 4001). The latter technique is used to emphasize prayer lines, prime vertical, and certain altitudes (the third, fifth [or sixth], tenth, or twentieth). Dotted lines are oc? casionally incorporated into the networks en? graved on plates of European instruments (see Nos. 221 and 262). These emphasize every fifth altitude or (in one case) the equal hours. The unequal hours on plates belonging to instruments made in the mashriq are often marked with dot? ted lines (Nos. 4, 15, 55, 57, 58, 59, 62, 64, 85, 86, 2566, 2569, 3811, and 4000), very likely to distin? guish them from seasonal hours also projected below the horizon. On some plates of Eastern origin, dotted lines also distinguish some azimuths (at least the prime vertical) and altitudes (includ? ing the oblique horizon). On the basis of the evidence in the collection, it is not possible to attribute the use of dotted lines to any motivation other than the whim of the maker. None of the instruments made or decorated by "^Abd al- A^imma incorporate dotted or otherwise inter? rupted lines. Rete As instruments in the collection illustrate, func? tional astrolabe retes invariably incorporate a stereographic projection of the ecliptic circle. This element of the design is usually defined by the beveled outer edge of a circular band attached 56 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y by thin struts to a flat ring surrounding the projection of the celestial poles. The beveled edge of the ecliptic band is most often divided into degrees; the band itself usually is divided into 36 subdivisions of 10? of the 12 zodiacal signs. On astrolabes made in Europe and the mashriq, the zodiacal signs are invariably identified by name; the familiar symbols appear on some maghribi examples. Numerals identify each subdivision as being 10?, 20?, or 30? of a sign. When a properly divided ecliptic circle is ro? tated about the projection of the celestial poles, the degree of celestial longitude corresponding to solar position on any given day traces the appar? ent path of the sun. The implications of this fact become apparent when the uses of the astrolabe are detailed. Instructions for properly dividing the stereo? graphically projected ecliptic circle are supplied perhaps for the first time, in Ptolemy's essay on the planisphaerium.^^ Ptolemy uses radii corre? sponding to the right ascension of the end points of each zodiacal division to achieve his goal. Emmanuel Poulle describes five other popular division methods, each with its own history in the literature^^ (see Appendix III). On every astrolabe inscribed with Arabic char? acters and on two European astrolabes (Nos. 186 and 204) there is a projection on the ecliptic (called a muri in Arabic) at the beginning of Capricorn (see Figure 9). It is used to indicate positions on the rim scale when the rete is rotated. The bands that attach the ecliptic circle to the ring around the projection of the celestial pole generally follow the projection of the equinoctial colure. The best makers incorporated counter- changes into these bands so that the exact position of the projection of the equinoctial colure could be seen readily. A lack of equinoctial colure bands may, but need not, identify a non-functional rete. Certainly four suspect retes in the collection (Nos. 49, 64, 66, and 86), all of which lack these bands, support this suggestion. On the other hand, curving vines without equinoctial colure bands connect the ecliptic to the celestial poles on two well-made retes from the mashriq (Nos. 54 and 55), both dating from the eighteenth century. One could conclude on the basis of examples here studied that only very good or very bad rete makers eliminated the connecting equi? noctial colure bands. The astrolabe as a timekeeper is distinguished from its near relative, the sundial, by the fact that its usefulness can be extended beyond sunset. This advantage is achieved through the incorpo? ration of the stereographic projections of selected stars into the rete design. As these projected stars rotate about the projection of the celestial pole, they duplicate the apparent motion of their coun? terparts in the heavens. If they are sufficiently well distributed, the projected stars can thus serve as adequate markers of the passage of night time. Rete makers positioned star pointers using co? ordinates supplied by tables of fixed stars. The Arab astronomer as-Sufi (A.D. 903-936) effec? tively published such a list when he designated 37 stars in his "Treatise on the Fixed Stars" (A.D. 964) as "astrolabe stars." European works on the construction and use of the astrolabe tend to include a tabula stellarum fixarum.^^ Most of these tabulae give the name of the star, the coeli mediati- onem (or the point on the ecliptic having the same right ascension as the named star), and the dec? lination of the star. All editions of Johannes Stof- fler's Elucidatio Fabrica include, besides the typical table of mediations and declinations, a table of stars along with their celestial latitude and lon? gitude. A maker could plot these latter stars using the same procedure he employed in engraving an astrolabe plate for latitude 66?. Poulle^? describes how makers utilized various parameters (includ? ing latitude and longitude) to position stars. The list of stars projected onto the retes of most instruments (over 30 mashriqi and maghribi) in the collection would have to include (1) a Tauri, (2) ^ Ononis, (3) a Ononis, (4) a Canis Majoris, (5) a Canis Minoris, (6) a Hydrae, (7) a Virginis, (8) a Bootes, (9) a Corona Borealis, (10) a Ophi- uchi, (11) a Lyrae, (12) a Aquilae, and (13) a Cygni (see Figure 9). The first seven of these "most popular" stars lie south of (outside) the ecliptic between Aries and Libra, the last six lie north of (inside) the ecliptic between Libra and Pisces. All of the "popular" stars are included in NUMBER 45 57 al-Sufi's list. About 36 other recognizable stars appear with varying frequency on retes in the collection (see Appendix II). No obvious single quality characterizes all of them. Most are either quite bright or are located in well-defined posi? tions within a constellation figure (at the corners of a box, at the vertices of a triangle, at the beginning or end of a line). The point projections of stars appear on each rete as the tips of leaves or flames or daggers. Since none of these could spring from the outer edge of the ecliptic band without detracting from its usefulness as a timekeeper, additional support? ing strapwork had to be included in each rete design. Every rete in the collection contains an almost complete Capricorn circle in the support? ing strapwork. This circle is tied to the ecliptic circle by projected equinoctial colure bands (usu? ally counterchanged) and, generally, also by thin curved ties near Pisces 0? and Scorpio 0?. No other elements of rete strapwork are common to all examples in the collection. Inside the Capricorn circle, the pattern of sup? porting strapwork forming an astrolabe rete tends to identify its maker (see, for example, Nos. 40, 37, and 39). Certain components of the generally individualistic designs are, however, shared by more than one maker. In a majority of examples, makers have incorporated some portion of the stereographic projection of the celestial equator into the pattern of rete strapwork. A second strap- work component shared by a majority of exam? ples is some portion of the stereographic projec? tion of the solstitial colure or meridian circle. Retes that do not include either of these elements generally consist of arabesque or baroque open? work, or of combinations of non-astronomical circular arcs. On the back of some astrolabe retes, there can be seen the remains of engraved lines and circles. These undoubtedly evidence steps in the con? struction process. The guidelines usually consist of stereographically projected colures, tropics, and equinox. Occasionally the remains of a di? vided ecliptic may be detected (see, e.g.. No. 62 in Figure 36). The astrolabe rete is used in combination with one of the plates to represent a given configura? tion of the heavens with respect to the horizon. The representation may in turn be interpreted as an indication of the time of day or night or as an indication of an individual's horoscope. According to Masha^allah: If you wish to know the hour at night, take the altitude of any star marked on the alhanthabuth [rete] . . . ; place the cacumen [tip of the star pointer] of the star on . . . its altitude [on the plate], and the degree of the sun [on the ecliptic] will show you the hour of the night .^ According to al-Biruni: To determine the ascendant from [a measured solar] alti? tude, select the disc [plate] constructed for the place of observation (or as nearly as possible), place it uppermost, and fit the rete over it. Then find and mark the muqanlara [projected altitude circle] which has the same number (or nearly the same number) as the [measured solar] altitude. Thereafter ascertain from a [zodiacal] calendar the exact place [longitude] of the sun at that time, and mark that point on the corresponding sign of the zodiac [ecliptic] on the rete. Then rotate the rete until the degree in question is over the muqanlara already marked, and note what sign and what degree thereof coincides with the eastern horizon. This is the ascendant.^?? Accessories Alidade Scales Position: The scales extend along the left arm with beveled edge at top and along the right arm with beveled edge at top (see Figure 8). Units: The only divisions on the left arm gen? erally correspond to the major divisions of the declination scale engraved on the back of the astrolabe; i.e., the arcs representing solar decli? nation at the time it enters each zodiacal sign, inscribed with names of zodiacal signs (left arm at bottom in Figure 8). The beveled edge of the right arm normally contains a length equal to and coterminal with the radius of the upper left quadrant on the back of the astrolabe. This length is divided into 60 equal parts, with every fifth division extending onto the upper half of the flat surface of the alidade and inscribed with a nu? meral (Figure 8). The lower half of the right arm of the alidade in most cases contains five unequal divisions extending from the right sighting plate # r -^'Z I. C" 9 ^^;0:^.- / ? > - Tk - -i FIGURE 36.?Astrolabe No. 62. Traces of con? struction guidelines are visible on the back of the '^ankabul (upper left). NUMBER 45 59 to the center and corresponding to the length of the shadow of the sighting plate as it falls on the alidade at the seasonal hours of the day, inscribed with numerals (Figure 8). Use: The left arm of the alidade essentially functions as a moveable equivalent of the names of zodiacal signs inscribed along the 0? radius of the astrolabe back. The inscription identifies the declination arcs upon which prayer lines and Qibla azimuths are engraved. By repeating these identifiers on the alidade, the astrolabist has made easier the interpretation of the lines en? graved in the area between the 0? radius and the 45? radius. The equally divided scale along the upper right arm of the alidade can be used to measure sine lengths along the 90? radius of the back of the astrolabe (and cosine lengths along the 180? radius). For that matter, the scale can be used to measure the distance along any quad? rant radius. The scale of unequal divisions engraved on the lower right arm represents a time-keeping tech? nique known from Graeco-Roman antiquity. The presence of such a scale identifies the astrolabe as a portable sundial, equivalent in design and util? ity to five ancient portable dials found near Bra? tislava, near Cret-Chatelard, at Rome, Memphis, and Aphrodisias. In order for the dial to mark the seasonal hours for any day at any given latitude of position, the alidade must be sus? pended so that it lies in the meridian plane, so that it is inclined to the vertical by an angle equal to the altitude of the noon sun at the given latitude on the day in question, and so that the shadow of the right sighting plate falls on the sundial scale. The second contingency can easily be satisfied by making use of the plates of the astrolabe (see page 56). If there is no plate for a given latitude, the dial can still be used to ap? proximate the seasonal hours by setting the ali? dade so it is inclined to the vertical at an angle equal to the latitude. According to Masha^allah: If you wish to know the natural [seasonal] hour of the day by the horary alidade, set the alidade on the [sun's] altitude at midday on the back of the astrolabe suspended, turn the back to the sun until the shadow of each corner of the upper pinnule falls on the alidade anywhere in line with its side. Then the division on which it falls will be the desired hour.^ Remarks: Without exception, the alidades on European astrolabes in the collection are coun? terchanged about the center. Only one astrolabe of non-European origin in the collection (No. 55) has a counterchanged alidade. All other non-Eu? ropean alidades in the collection are not counter- changed. The scales described above appear on the ali? dades of about half the non-European astrolabes in the collection. The remaining non-European alidades in the collection are uninscribed. None of the scales described above appear on European astrolabes in the collection. Five of these Euro? pean alidades are blank. One (No. 2007) is in? scribed with a maker's name. On one (No. 204) the words Linea Fiducie and Alhidada have been inscribed. The alidade accompanying the astro? labe incorporating a de Roias projection (No. 2005) has been inscribed with two scales: one is a scale of equal divisions to be used with a plate to convert equal to unequal hours; the other represents the stereographic projection of the 5? and 10? points between the equator and the North Pole of a sphere of radius equal to the radius of the Capricorn circle on the plate. Index Scales Position: The scales extend along the beveled edge. Units: These scales are marked at twelfths of the radius of the front central area (numbered), or 5? (unnumbered) and 10? (numbered) units of declination (No. 262). Use: See the section on "Alidade Scales," the discussion of the upper right arm (see above). The divisions of this scale coincide with stereographi? cally projected parallels of declination. Thus, the scale may be used to measure the declination of the sun or any star on the astrolabe rete. Blank index arms may conveniently be used in combi? nation with rim scales to determine the equal hour. Remarks: Only nine astrolabes in the collec? tion are equipped with an index arm. Four of 60 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y these are radial. Five are complete. The index arms on five astrolabes (Nos. 31, 186, 221, 304, and 2007) are blank. The complete index arm on a non-European astrolabe made in A.D. 1783 (No. 55) carries no scale but is decorated with vines and flowers on a stippled background. The com? plete index arm on the European astrolabe signed "I. Galois" (No. 204) carries no scale but is in? scribed Linea Fiducie on each arm. The scale (see above) on the radial index arm of the Hartman astrolabe (No. 262) is identified with the inscrip? tion, Lati: Meridio : Latitvdo Septrionalis. The scale on the complete index arm of the astrolabe signed Sadiq (No. 52) is untitled. Catalog of the Collection Each of the 52 astrolabes described in the following catalog is identified by the number assigned to it in A Computerized Checklist of Astro? labes compiled by Sharon Gibbs, Janice Hender? son, and Derek Price and published by the De? partment of History of Science and Medicine at Yale University in New Haven, Connecticut, in 1973. The Checklist is referenced in the catalog as CCA. The numbers assigned by Gibbs, Hender? son, and Price preserve numbers assigned to al? most 400 astrolabes by R.T. Gunther (1932) in his Astrolabes of the World. These numbers are correlated with Catalog Numbers of the National Museum of American History and the S.V. Hoff? man collection in Table 16. The date given for each astrolabe, unless it is preceded by the qualifier "ca.," is exactly what appears inscribed on the instrument. When the inscribed date refers to a calendar other than the Julian or Gregorian (e.g., the Hegiral calendar), it is followed by brackets enclosing the Julian or Gregorian equivalent. Signed but undated astro? labes have been assigned to the middle of the maker's productive period. In the absence of any inscriptions, the astrolabe's date has been esti? mated from the celestial longitudes of the stars in its rete or from the date of the vernal equinox indicated on its calendar scale. The "Signature" section has been reserved for the name of the maker or decorator. Translations of inscriptions that were contrib? uted by George Saliba have been identified with the symbol "GS." The catalog entries include mention of all inscription occurring on each as? trolabe. The names of cities and geographic pa? rameters inscribed in the front body of an astro? labe are reported in tabular form in Appendix I. Similarly, star names appearing on an astrolabe star map or rete are tabulated in Appendix II. Unless otherwise indicated, the reader may assume that there are no engraved details or other TABLE 16.?Cross references to numbers identifying astrolabes (CCA = number assigned in A Computerized Checklist of Astrolabes, Gibbs, Henderson and Price [1973]; NMAH = catalog number assigned by the National Museum of American History; H = Number assigned by Samuel Verplanck Hoffman) CCA 4 15 25 31 37 39 40 42 44 47 49 52 53 54 55 57 58 59 61 62 63 64 65 66 70 85 NMAH 322458 336113 336112 316767 333589 336114 316761 322460 316763 316764 333586 316765 316766 333590 322461 333587 322459 336090 316768 316769 215454.34 336116 322327 316762 316755 316750 H 28 13 12 39 20 16 18 37 26 6 8 24 27 30 31 10 11 14 7 22 25 23 32 21 4 38 CCA 86 87 88 144 186 204 221 262 304 2005 2006 2007 2566 2567 2568 2569 2570 2571 2572 3643 3811 4000 4001 4002 4003 4004 NMAH 316754 336115 322464 322463 316757 316760 316759 336117 316758 316756 318198 322083 333588 316770 316751 94624-l-A 315064 316752 318178 316753 215454.33 330778 330779 333932 333933 79.852.01 H 29 19 5 2 33J 35 34 33 36 40 15 17 41 1 9 original marks on the edge of the astrolabe. The two sides of a given astrolabe plate have been designated side "a" and side " b " (e.g., 6a and 6b). All of the astrolabes described in this catalog, including those no longer in the National Mu? seum of American History (CCA Nos. 4, 42, 55, 58, 65, 88, and 144), have been examined by the author. 61 62 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 37.?Astrolabe No. 4. Front view. CCA No. 4 FIGURES 31, 37, 38 Date: A.H. 547 [1152]. Signature: Hamid ben Mahmud al-Isfahani. Components: Body with throne (rim with throne riveted to backplate with throne), handle, ring, three plates, rete, alidade, pin, ringlet, and wedge. Diameter: 134 mm. Material: Brass. Characteristics: The pierced throne is a Moor? ish arch formed by intertwining vines. The handle is ring-shaped. The ring itself has no visible join. Two concentric circular scales divide the rim into 1? and 5? intervals. Inscription on the latter identifies it as a 360? scale. The first subdivision is marked with an abdjad " 5 ; " the last, "60." Eight concentric bands engraved in the front central area suggest that a gazetteer was planned, but never completed. The back margin is divided into altitude scales in the upper sections and a cotan? gent scale in the lower right section. The lower left margin has been reserved for the maker's signature. The two upper quadrants contain a set of six equally spaced declination arcs crossed by four prayer lines, two {khatt awwal waqt az-zuhr and khatt awwal waqt al-^asj) each for latitudes 36? and 32?25' (?Isfahan). The two lower quad? rants contain five semicircular astrological scales: zodiacal signs, planetary governors of terms, terms, planetary governors of faces, and face de? grees. Five symbols identify the planetary gover? nors of the terms: a Roman Z with horizontal slash is Jupiter, a vertical arrow with a ball at its tip is Venus; a vertical line topped by a ball topped by a Roman U is Mercury; a vertical arrow intersecting a ball is Mars; an unmodified Roman Z is Saturn. Three notched plates fit into the recessed area in the front of the astrolabe. Of the six plate faces, only one (3a) carries the complete standard net? work (of colures, equator, tropics, horizon, alti? tudes, azimuths, and seasonal hours) and it is supplemented by a projection of equal Italian hours and a projection of Muslim prayer lines. No azimuth lines are included in the networks of four of the remaining plate faces; two faces that lack azimuth lines do have lines for the equal Italian hours. Altitudes have been projected at 3? intervals. The identifying inscription on each face specifies latitude and length of daylight as fol? lows: Plate lb 2a 2b 3a 3b Latitude '^ard 31 "ard 30 "ard 34 "^ard 32 '^ ard 36 Length of daylight sa^atuhu 14 03 sa'^atuhu 14 sa^atuhu 14 19 sa^atuhu 14 08 sd'^atuhu 14 30 The sixth plate face contains four obliquity scales and 16 quarter horizons for latitudes 15?, 16?, 17?, 18?, 25?, 26?, 27?, 28?, 35?, 36?, 37?, 38?, 45?, 46?, 47?, and 48?. The rete is inscribed with the names of 27 stars. Its design is severely simple, incorporating NUMBER 45 63 FIGURE 38.?Astrolabe No. 4. Back view. a non-counterchanged east-west bar and an equinoctial arc. The ecliptic band is subdivided into units of 6?. The star-pointers are dagger? like. The uninscribed alidade is not counter- changed. It includes a beveled fiducial edge and two sighting plates (separated by a distance of 77 mm), each pierced with a single large hole. The pin is a bolt with a flattened head and a doorknob-shaped tip. The wedge is a horse's head, and it includes appropriate engraved details. The lettering on this instrument is Kufic of the eastern type, and it is consistent throughout. Remarks: No other astrolabe by Hamid is known. Plate 3a is notably more elaborate than the other plate faces. Plates 1 and 3 have each been pierced with two holes near the perimeter at about 54? and 124? The obliquity scale on 64 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y plate face la is inconsistently inscribed. The segment at 90? is the only one inscribed in standard fashion. CCA No. 15 (FIGURE 39) Date: A.H. 774 [1372]. Signature: sana'^ahu Ja'^far ben "Umar ben Daulatshah al-Kirmani. Components: Body with throne, three plates, rete, alidade, pin, ringlet, and wedge. Diameter: 73 mm. Material: Brass. Characteristics: The uninscribed throne is in the shape of a Moorish arch. It is pierced by two small holes placed symmetrically on either side of the handle hole. The rim is divided into 360? with the first unit marked with an abdjad " 5 , " the last, "60." A gazetteer containing 21 place names, each associated with longitude, latitude, and distance from Mecca, is engraved in the front central area. Twenty of the place names and parameters are engraved on spiral arms; one name (Mecca) is engraved in the center of the table. The margins of the back of this instrument contain altitude scales in the upper portion and cotangent scales (fingers left, feet right) in the lower portion. The upper left quadrant is divided by sine lines for each degree. The upper right quadrant contains, in addition to the signature, a shadow square divided into fingers and a box scale of triplicities inscribed: J... I II jL^-:-Jl o l |-.o II This translates as follows: "[The Table of| Triplicities [with the Planetary Lords of] Day [and] Night" (GS). The two lower quadrants contain six semicircular astrological scales (signs, planetary governors of terms, terms, planetary governors of faces, planetary governors of decanates, degrees of decanates). Except for the horizon face, all of the plate faces are engraved with projections of horizon, FIGURE 39.?Astrolabe No. 15. Gazetteer. altitudes, and unequal hours. One plate face contains in addition projections of azimuths; two faces also contain equal Italian hours; and one face contains additional projections of both azimuth and equal Italian hours. The plate face which contains only horizon, altitudes, and unequal hours contains these projections for two distinct latitudes. The following engraved latitudes and lengths of daylight locate each projection: Plate la lb 2a 2b 3a Latitude ^ard 32 "ard 30 ^ard 36 ^ard 38 "ard 28 "ard 34 Length of daylight sd^atuhu 14 08 sa'^aluhu 13 58 sd^dtuhu 14 30 sd'^atuhu 14 48 sd^atuhu 13 08 sd'^atuhu 14 19 The horizon plate face contains 21 quarter horizons for every 6? of latitude from 12? to 84?. Each horizon is identified by inscription. In addi t ion , the curious phrase al-bum [.^] 146 NUMBER 45 ?65 appears in the lower left quadrant of this plate face, the phrase al-bum [?] 114 appears in the lower right quadrant, and al-bum [?] 72 in the upper right quadrant. The rete incorporates 19 dagger- or sickle- shaped star pointers, each inscribed with a star name. The ecliptic is divided into 5? units. The plain alidade has a beveled fiducial edge, diagonal ends, and a central Moorish arch. It is not counterchanged. The sighting plates are each pierced with two equal holes, one above the other. A plane ringlet fits over the pin. The latter has a hemispherical head and doorknob- shaped tip. The wedge is a horse's head, complete with engraved eye and mouth. The alphabet employed by the maker is Arabic, Kufic throughout. Remarks: Other astrolabes signed by Ja^far ben "^Umar include CCA Number 16, Indian Museum, Calcutta (dated 790 [1388]); CCA Number 1205, private collection (dated 755 [1354]); and CCA Number 3660, private collection (dated 751 [1350]). This instrument is incomplete. It has no handle or ring. The simple alidade may be a replacement. Ja^far ben "^Umar's gazetteer is arranged in order of increasing longitude beginning with Baghdad just to the left of 90? and reading counterclockwise. No other gazetteer in the collection is ordered in this way. Except for the table of triplicities, none of the astrological tables are titled. CCA No. 25 (FIGURES 21, 22,40) Date: A.H. 1078 [1667]. Signature: namaqahu Muhammad Mahdi ben Muhammad Amin al-Yazdi. Components: Body, handle, five plates, rete, alidade, pin, and wedge. Diameter: 114 mm. Material: Brass. Characteristics: The throne is shaped like a Moorish arch. It is decorated front and back with embossed vines and flowers. In addition, on the front, two embossed birds face a central cartouche in which is inscribed the opening words of the quotation from the Koran which includes the word "throne." ( ^ . 2 I 9 Cl< I n a m 11 rt _ I ?" > ^ f^^"^ The translation is, "His throne holds the Heavens and the earth" (GS). Above this is a second inscription which reads: This translates, "He is the most exalted" (GS). At the base of the throne, on the right, the words "O! You who lifts" (GS) appear; on the left is the word "the Heavens" (GS). There is considerable inscription on the back of the throne. It is framed at left, top, and right by the three words mashriq, hut, and maghrib, respectively. In addition to these words, there are inscribed the names of 23 places arranged as follows (beginning at the lower left and proceeding in a clockwise direction): Bejeh, Shahr{?), Habashah, Mahdiyyah, Darnalar, Belzar{?), Tabariyyah, Tardblus, Halab, Amid, Arz Al Rum, Damdam{?), Shamakh, Tdliqdn, Bistdm, Tun, Hims, Qpm, Istdneh, Ajmer{?), Sumiyan{?), and Najd{?). A shallow circular depression 15.5 mm in diameter has been cut into the center of this back face of the throne. The rim of the astrolabe is divided by a scale of 360?. Subdivisions of 5? are identified by abdjad numbers from 5 to 360. In the front central area has been engraved a gazetteer listing 46 places, along with the longitude, latitude, azimuth of the Qibla, and direction of the azimuth of the Qibla of each. On the back of the astrolabe, the upper margin has been divided into two altitude scales. The lower margin contains cotangent scales divided into feet (lower left) and fingers (lower right). At the lowest point in the margin, the following phrase has been engraved: 66 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 40.?Astrolabe No. 25. Front view. C--?^ J-i^J J'^J^ Translated this reads: "The familiar year you seek is manifest, 1078 is its date" (GS). Stereographically-projected declination arcs divide the upper right quadrant. Some of these arcs are defined by dotted lines. Lines representing the altitude of the noon sun have been graphed onto the upper left portion of these declination arcs. The graphed lines are identified with numbers that represent every 3? of latitude from 24? to 48?. The inscription relating to this graph outside the outermost declination arc is transcribed and translated in the description of CCA No. 59. In the lower right portion of this quadrant, the altitudes of the sun when it crosses the azimuth of the Qibla at various times of the year have been graphed. There is one graphed line identified with Tus (lowest line), Yazd, Isfahan, Hamadan, NUMBER 45 67 Baghdad, and Shiraz. The inscription relating to the graph inside the innermost declination arc is transcribed and translated in the description of CCA No. 49. Sine lines for every degree have been engraved in the upper right quadrant. In this same quadrant, six identified equal hours have been graphed. Shadow squares divided into feet (left) and fingers (right) occupy the lower quadrants of the back of this astrolabe. The area framed by these squares has been divided into a rectangular table of triplicities. The natures are identified with Arabic terminology. The maker's signature ap? pears in a cartouche below the shadow square. The area between the shadow squares and the marginal cotangent scales contains five semicir? cular astrological scales: 1) the planetary gover? nors of the terms, 2) the terms, 3) the zodiacal signs, 4) the planetary governors of the faces, and 5) the lunar mansions. Each of the five plates is notched at 270?. One plate face contains 29 stereographically projected quarter horizons, one for every two degrees of latitude from 10? to 64?. There are, in addition, four obliquity scales at the ends of the stereo? graphically projected colures. All the remaining plate faces contain stereographically projected horizon coordinates for individual latitudes: alti? tudes (every 3?) and azimuths (every 10?). Eight plate faces contain, in addition to these coordi? nates, projections of seasonal hour lines. (These are omitted from plate face 3a). On plate faces 3b and 4b, the seasonal hours are identified with the ordinal numbers in Persian. These two faces contain, in addition to seasonal hours, the equal Italian hours. One plate face (4a) contains a projected twilight line. This last is identified with Persian phrases at left and right. Except for the horizon face (5b) and plate face 3a, the plate faces are each identified with a latitude and length of daylight as follows: Plate la lb 2a 2b 3b Latitude "ari 32 00 ^ard 22 00 ^ard 30 00 ^ard 35 00 "ard 36 00 Length of daylight sa'^atuhu 14 06 sa'^atuhu 13 22 sd'^dtuhu 13 58 sd^dtuhu 14 24 sd^dtuhu 14 28 4a 4b 5a "ard "ard "ard 34 00 37 00 40 00 sd"atuhu sd"dtuhu sd"atuhu 14 17 14 32 14 52 Five lines on plate 4b are inscribed as follows: Samt Qiblat Lar (at 135?), Massisjyah (at 115?), Lahijan (at 80?), Qpm (at 75?), and Barfurush (at 70?). The rete of this astrolabe is quite simple. It consists only of a counterchanged equinoctial bar and a section of the stereographically-projected equator, in addition to the usual ecliptic and Capricorn circles. The latter is decorated with Persian inscriptions. * L ll ^ J j ^ k v>? ^J>^ ^Sjj ^ &s AJ* ,L?. -J u ^ J ?"'????. ..J ^ I j i > l iL, . This may be translated as follows: "O! You Who set the zodiac in the heaven. The count includes the throne and the zodiac, that according to some is in nine layers. The fixed stars are one thousand and twenty-four, and each has its own place in the throne" (GS). The ecliptic is subdivided into units of 3?. There are 22 inscribed star pointers. Engraved lines divide the ecliptic on the back of the rete. The alidade is not counterchanged. It is deco? rated with vines embossed on a stippled back? ground. The left arm is divided by a sundial scale. The right arm is divided by two scales: one a scale of 60 equal units; the other composed of arcs of stereographically projected declination cir? cles. Two lines of Persian inscription decorate the center of the alidade. Two holes pierce the sighting plates. The pin is a bolt with a hemi? spherical cap and a doorknob tip. The wedge is in the shape of and is decorated with the details of a horse's head. The maker employs Arabic Naskhi script throughout. Remarks: This is one of the most elaborate of the Persian astrolabes in the collection. The dec? oration on the front face of the throne is most 68 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y unusual. At first glance it appears to consist entirely of embossed vines and flowers, but closer study reveals the two facing birds. In addition, above these birds, two "eyes" can be seen to look out under a furrowed "brow." The inscription which appears on the back of the throne is not typical of that found on most Persian astrolabes. Its geographic emphasis supports the notion that the shallow depression once held a magnetic com? pass. The word hut is probably a mistake for janubi (or "south"). In general, abdjad numbers are used to identify projected or dividing lines, but Hindu-Arabic numbers identify the unequal hours scale on the back of the instrument and the unequal hours on some plates. Some lines on the back of the astro? labe are dotted for emphasis. The places in the gazetteer are arranged in a pattern similar to that adopted by makers of other astrolabes in the collection (see, e.g.. No. 65.) The collection includes another, less elaborate astrolabe signed by Muhammad Mahdi, No. 44. CCA No. 31 (FIGURE 41) Date: Ca. A.D. 1715. Signature: sana' 'ahu Muhammad Amin ben Muhammad Tahir (maker) namaqahu "^Abd al- A^imma (decorator). Components: Body with throne, handle, ring, five plates, rete, alidade, pin, wedge, and index arm. Diameter: 74 mm. Material: Brass. Characteristics: The throne is a broad, sym? metric Moorish arch pierced with two holes. It is decorated front and back with embossed vines and a central flower. The stirrup-shaped handle includes a decorative design above the axle. The ring has a circular cross-section. The rim contains a scale of 360? subdivided into units of 5? iden? tified with abdjad numerals. Identifying inscrip? tion is embossed on a stippled background. Di? visions at 15? intervals (starting at the top) are marked by an embossed "X." The front central ->:... PS-' FIGURE 41.?Astrolabe No. 31. Front view. area contains a gazetteer consisting of 23 place names along with longitude, latitude, azimuth of the Qibla, and direction of the azimuth of the Qibla for each. The margin of the back of the astrolabe con? tains in its upper half two altitude scales; the lower left quadrant contains a cotangent scale divided into feet; the lower right quadrant con? tains a cotangent scale divided into fingers. The identifying inscription is embossed on a stippled background. The upper right quadrant contains stereographically projected declination arcs upon which are graphed lines marking the altitude of the noon sun at latitudes 32? (leftmost line), 36?, 40?, and 44?, and lines marking the azimuth of the Qib la at Tus (lowest l ine), Isfahan, and Qazvin (see CCA Nos. 49 and 59 for transcrip? tions and translations of the titles of these graphs). NUMBER 45 69 The upper left quadrant is divided by 30 equally spaced sine lines and by 30 equally spaced cosine lines. The chord of this quadrant is marked by a dotted line. Two shadow squares divided into feet (left) and fingers (right) occupy the lower quad? rants. There are also four semicircular scales in these two quadrants. One lists the planetary gov? ernors of terms; one lists the term divisions; one lists zodiacal signs; and one lists the planetary governors of the faces. Each of the five plates is notched at 270?. On one plate face 17 quarter-horizons (for every 3? of latitude from 18? to 66?) have been stereo? graphically projected. There are four scales meas? uring declination, along the ends of the projected colures between the projections of the solstitial circles. All other plate faces contain other stereo? graphically projected arcs in addition to an ho? rizon: altitudes (every 6?), azimuths (every 10?), and unequal hours. On all but one face (2b), the equal Italian hours have also been projected. The faces are identified with various latitudes as fol? lows: Plate lb 2a 2b 3a 3b 4a 4b 5a 5b Latitude "ard "ard "ard "ard "ard "ard "ard "ard "ard 24 25 34 27 36 30 33 32 38 Length of daylight sd"atuhu sa"atuhu sd"dtuhu sa"atuhu sa"atuhu sd"dtuhu sa"atuhu sa"atuhu sd"dtuhu 13 30 13 .34 14 16 13 52 14 28 13 56 14 12 14 07 14 39 Abdjad numerals are used to identify all projected lines on plates except the Italian hours, for which Hindu-Arabic numerals are used. The rete design includes a counterchanged equinoctial colure, on which diagonal lines have been engraved, and an M-shaped support be? tween ecliptic and lower half of Capricorn circle. A chain of leaves decorates both the Capricorn circle and the support. Twenty-four pointers have been inscribed with star names. The ecliptic has been subdivided into units of 6?. There are some engraved lines including ecliptic divisions still visible on the back of the rete. The alidade is not counterchanged. Divisions on the right half correspond to the positions of stereographically projected declination arcs; the left half of the alidade contains a partially effaced scale of equal divisions and a partially effaced sundial scale. There is a notch above a hole in each sighting plate. The pin is a screw with a hemispherical head. The wedge is a wing nut. A radial index arm has been attached to the front of this astrolabe. It is thin and narrow, being slightly wider at its outer end than it is at the point where it connects with its attachment ring. The alphabet employed in inscription is Arabic Naskhi. Remarks: Other astrolabes, similarly signed, include CCA No. 32, Indian Museum, Calcutta (undated) and CCA No. 1171, City Art Museum, St. Louis (undated). The outer surface of this instrument is quite worn; the plates are thin and bent; the alidade shows signs of having been repaired. The index arm, pin, and wedge are almost certainly not the work of Muhammad Amin. The towns listed in the gazetteer are a subset of those included in the gazetteer of No. 39. CCA No. 37 (FIGURES 24, 42, and 43) Date: Ca. A.D. 1715. Signature: ""Abd al-A^imma Components: Body with throne, handle, ring, four plates, rete, alidade, pin, ring, and wedge. Diameter: 93.5 mm. Material: Brass. Characteristics: The throne is shaped like a Moorish arch. Both front and back are decorated with an embossed design consisting of a vine with slender leaves and flowers. On the front this design surrounds a central cartouche. The handle is stirrup-shaped with diamond cross-section. The ring also has a diamond cross-section. The rim is divided into a scale of 360?. The last 5? subdivi? sion is marked "60" in abdjad numerals. The front central area contains a gazetteer listing 34 place names along with longitude, latitude, azimuth of the Qibla, and direction of the azimuth of the Qibla for each. 70 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y > ^ ^ ' ^ ' " FIGURE 42.?Astrolabe No. 37. Front view. On the back of the astrolabe, the upper margin contains an altitude scale. The lower margin contains cotangent scales divided into feet (left) and fingers (right). The upper right margin con? tains stereographically projected declination arcs. On these are graphed lines representing the alti? tudes of the noon sun for latitudes 28? (left), 30?, 32?, 34?, 36?, 38?, 40?, and lines representing the azimuth of the Qibla for four cities: Tus (bottom), Isfahan, Baghdad, and Yazd (for a transcription and translation of the inscription on the periphery of the declination arcs, see CCA No. 59; for inscription inside the declination arcs see CCA No. 49). The upper left quadrant contains 60 equally spaced sine lines and 60 equally spaced cosine lines. There are shadow squares divided into feet (left) and fingers (right) in the lower quadrants. There is a scale of triplicities in the rectangle formed by the shadow-square bound? ary. The scale is entitled: ^ L ^ . lL-> L..^_-l_.^l o^i-- . I .^ II ^ 1 meaning, "Natures, Triplicities, Their Lords in the Day, Their Lords in the Night." Arabic terms have been used to identify the natures. A car? touche below the shadow square contains the signature. Also in the lower quadrants are five semicircular astrological scales: planetary gover? nors of terms, terms, zodiacal signs, planetary governors of faces, and lunar mansions. There are four plates. Each has a notch at 270?. One plate face contains only scales of dec? lination and stereographically projected semi-ho? rizons for every 3? of latitude from 15? to 66?. Each of the remaining plate faces contains a single stereographically projected horizon and re? lated coordinates: altitudes (every 6?), azimuths (every 10?), unequal hours, and equal Italian hours. The projected networks are identified: Plate lb 2a 2b 3a 3b 4a 4b Latitude li-''ard li- "ard li-''ard li-"ard li-''ard li-^ard li- "ard 29 34 37 30 36 32 38 Length of daylight sd"atuhu sd"atuhu sa"dtuhu sd"aluhu sa "dtuhu sa"atuhu sa" dtuhu 13 52 14 16 14 38 13 56 14 28 14 07 14 36 The rete pattern incorporates an abbreviated counterchanged equinoctial colure (inside the ecliptic). The area between the ecliptic circle and the Capricorn circle contains a vine outlining the shape of an inverted Moorish arch. There are 26 pointers inscribed with star names. The ecliptic is subdivided by units of 6?. The circle of a stereographically projected equator and the di? visions of the ecliptic circle are visibly engraved on the back of the rete. The alidade (no counterchanges) is divided on one arm by arcs corresponding to the stereograph? ically projected declination arcs in the upper right quadrant. There are "sundial" divisions on the other arm, as well as 60 equally spaced divisions NUMBER 45 71 FIGURE 43.?Astrolabe No. 37. Gazetteer. along the beveled edge. Two holes pierce the sighting plates; the upper is larger. The pin is a bolt with a hemispherical head. The ring is a washer. The wedge is a spiral-decorated bar. The inscription is in Arabic Naskhi characters. Except for identification of the equal Italian hours using Hindu-Arabic numerals, the scales are identified with abdjad numerals. Remarks: This is a well-made instrument. The maker has used dotted lines for emphasis on some scales on the back. Most lettering is embossed on a stippled background. The four plates are iden? tical in content to four of six included in No. 40. For other works by the same maker, see No. 39. 72 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y CCA No. 39 (FIGURE 44) Date: Ca. A.D. 1715. Signature: sana^ahu "Abd al-A^imma. Components: Body with throne, handle, ring, cord, three plates, rete, alidade, pin, ringlet, and wedge. Diameter: 92.5 mm. Material: Brass. Characteristics: The throne, a broarl, symmet? ric Moorish arch, has been embossed with inscrip? tion on both sides. The front reads, j j P ? S I Q C ^ I g O III I 11 rt. ? "ly^ ^*"^ J "His throne holds the Heavens and the earth" (GS). The back contains a lengthy inscription. f J x C ^ I , j J j - o u J l ^ ? - - O t , - l ?? v j > i l A ^ a J ? ? rtJLJl r e a d ^^LasnAJ) ^ ^ La^iU) ?_,J;\.i> il I SI ll ?_^L-. , I ifta^oif 11 A i? 1 ? J? 9-1' ?'' ' ? This may be translated as follows: "[In the name of God.] It has been completed for the sake of the namesake of the person sacrificed [by Abraham] the most excellent master, the most distinguished of the men of pen, the leading pole of the men of eloquence [read/M.yaAa"^], the zodiac \falak\ of the ministry, Mirza Isma'^il may God preserve him" (GS). The stirrup-shaped handle has a rectangular cross-section. The ring has a circular cross-section. A relatively well-preserved cord, consisting of faded blue, yellow, and lavender threads, is at? tached to the ring. A worn gold-colored ornament decorates its loose end. The rim of this instrument is divided by a scale of 360?. Complete abdjad numerals identify each 5? division. The front central area contains a gazetteer listing the azimuth of the Qibla and its direction for each of 46 named places, along with their respective longitudes and latitudes. On the back of this astrolabe, the upper half of the margin is divided into two altitude scales. On them, every fifth degree is marked with a dot. The lower half of the margin contains cotangent scales divided into feet (left) and fingers (right). The lowest part of the margin contains the follow? ing phrase inside brackets: ^ ^ J L ^ - v ^ * ^ "The intention is to have an image remain after us" (GS). The upper right quadrant contains stereographically projected declination arcs. Two sets of lines have been graphed on these arcs: lines marking the altitude of the noon sun at latitudes 30? (at left), 33?, 36?, 39?, and 42?, and lines marking the azimuth of the Qibla sun at Tus (lowest), Yazd, Isfahan, Baghdad, and Kufa. For a transcription and translation of the titles of these graphs, see Nos. 49 and 59. Sine lines for every degree divide the upper left quadrant. Six lines marking the unequal hours have also been engraved in this quadrant. Shadow squares di? vided into feet (left) and fingers (right) have been engraved in the lower quadrants.These quadrants also contain six astrological scales. One of these, the scale of triplicities (Arabic natures; for title see CCA No. 37), fills the area bounded by the shadow squares. The remaining five are semicir? cular and consist of a scale of planetary governors of terms, terms, zodiacal signs, planetary gover? nors of faces, and lunar mansions. Much of the identifying inscription on the body of the instru? ment has been embossed on a stippled back? ground. There are three plates. Two of these are notched at 270? One has four equally spaced notches. There are also 18 quarter horizons (one for every three degrees of latitude from 15? to 66?), which have been stereographically projected onto one face of the multi-notched plate. The remaining plate faces contain, in addition to a single stereographically projected horizon arc, al? titudes (every 6?), azimuths (every 10?), unequal hours, and equal Italian hours. The plate faces NUMBER 45 73 FIGURE 44.?Astrolabe No. 39. Front view 74 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y are identified as follows: Plate lb 2a 2b 3a 3b Latitude "ard 29 "ard 30 "ard 36 "ard 38 "ard 32 Length of Daylight sd"dtuhu 13 52 sd" dtuhu 13 56 sd"dtuhu 14 28 sd" dtuhu 14 39 sd"dtuhu 14 07 flowers and vines embossed on a stippled back? ground. The inscription on the front of the throne reads The rete design includes only the part of the equinoctial colure inside the ecliptic band. It is counterchanged two times. In the area between the ecliptic and Capricorn bands, curling vines form an inverted Moorish arch. Twenty-six star pointers have been inscribed with Arabic star names. The ecliptic is subdivided into units of 6?. There are no visible marks on the back of the rete. The alidade is not counterchanged. On one arm, six arcs that correspond to the stereograph? ically projected declination arcs in the upper right quadrant have been engraved. The other arm is divided by a scale of equal divisions and by a sundial scale. Each sighting plate has been pierced with two holes, the upper larger than the lower. The pin is a bolt with a hemispherical head and a doorknob tip. The ringlet has a rectangular cross-section. The wedge is a horse- head shape with engraved details. The alphabet employed by the maker is Arabic Naskhi. Remarks: The three plates included with this instrument are identical (with respect to content and arrangement) to three of the five plates in? cluded with astrolabe No. 40. CCA No. 40 (FIGURES 45, 46) Date: Ca. A.D. 1715. Signature: sana' 'ahu"Abd al-A^imma. Components: Body with throne, handle, ring, five plates, rete, alidade, pin, and wedge. Diameter: 93.5 mm. Material: Brass. Characteristics: The throne, shaped like a Moorish arch, is decorated on the back with and can be translated as follows: "According to the decree of his excellency Mahdi Qoli Bey Zangeneh this was done" (GS). The stirrup- shaped handle has a rectangular cross-section. The ring has a diamond-shaped cross-section. The rim is divided (at intervals of 5?) by a scale of 360?. The last unit is marked "60" in abdjad numerals. Every fifth degree division is marked by a dot. The front central area contains a gaz? etteer of 34 cities, along with the latitude, longi? tude, and the azimuth of the Qibla for each. The upper margin on the back of the astrolabe contains altitude scales. The lower left margin contains a cotangent scale divided into feet; the lower right margin, a cotangent scale divided into fingers. Declination arcs divide the upper right quadrant. Lines marking the altitude of the noon sun for every two degrees of latitude between 28? and 38? and for 42?, and lines marking the azimuth of the Qibla for the cities Tus (bottom), Isfahan, and Baghdad have been graphed on the declination circles. For a transcription of the in? scription outside the outermost declination arc, see CCA No. 59. For a transcription of the inscrip? tion inside the innermost declination arc, see No. 49. The upper left quadrant is cross-hatched by sexagesimal sines and cosines. Shadow squares divided into feet (lower left) and fingers (lower right) divide the lower quadrants. Outside the shadow squares there are semicircular astrological scales of planets, terms, signs, faces, and mansions. Inside the shadow squares is a rectangular scale of triplicities. The natures are identified with Arabic terms. The signature is enclosed in a cartouche below the shadow square. There are five plates, each notched at 270?. One plate face contains quarter horizons, one for every 3? of latitude between 12? and 66?, and a scale of obliquity divided into units of 6? The NUMBER 45 75 I FIGURE 45.?Astrolabe No. 40. Front view of separate parts. rv fi\-'\'i r^ .^ - ? ? ;3^ ^^^'i^l\%?<'^^^ .;^ 4.;.^ ; V-'V J - ? ?'''*%;i^'<^^A^'"*'*''"' ] ? ^ 76 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y /j-^ ^ ^ ^ ^ \' ; ^ N\ ' V \ \ \ \ ^ 'W; , ? ' . - , ? : i - ^ . ^^.i:^::^ FIGURE 46.?Astrolabe No. 40. Back view. NUMBER 45 77 remaining faces contain stereographically pro? jected horizon, altitudes (every 6?), azimuths (ev? ery 10?) unequal hours, and equal Italian hours for a single latitude. These latter plate faces are inscribed as follows: Plate l b 2a 2b 3a 3b 4a 4b 5a 5b Latitude "ard "ard "ard "ard "ard "ard "ard "ard "ard 29 30 36 32 38 34 37 35 40 Length of daylight sd" dtuhu sd"dtuhu sd"dtuhu sd" dtuhu sd"dtuhu sd"dtuhu sd"dtuhu sd" dtuhu sd"dtuhu 13 52 13 56 14 28 14 07 14 39 14 16 14 38 14 22 14 51 The very simple rete pattern consists of a bracket in the area between the ecliptic and the Tropic of Capricorn. There is a portion of coun? terchanged east-west bar inside the ecliptic. The ecliptic band is subdivided into units of 6?. There are 28 flame-shaped star pointers, each inscribed with a star name. Straight lines and ecliptic divisions have been engraved on the back of the rete. The alidade arm is not counterchanged. One arm is inscribed with sundial divisions.The other arm is divided by a scale of 60 equal divisions along the beveled edge and by lines coinciding with the stereographically projected declination arcs. The sighting plates are pierced with two holes, with the upper larger. The pin is a bolt with a doorknob-shaped head and tip. The wedge is a simple uninscribed trapezoid. The maker has employed Arabic Naskhi script with some Hindu-Arabic numerals (equal Italian hours). Remarks: Most lettering is embossed on a stip? pled background. The maker has used dotted lines only on the declination scale. Plates 1, 2, and 3 of this instrument are identical to the three plates of CCA No. 39, also by "Abd al-A^imma. The selection and arrangement of place names in the gazetteer follows the pattern adopted by the maker of CCA No. 25. The surface of the body of this example is quite worn, but there are no signs of repair. The engraving is excellent. Note the possibility of an error in the inscribed number of unequal hours on plate face 5b. CCA No. 42 (FIGURES 47, 48) Date: Ca. A.D. 1795. Signature: ?ana'^ahu Hajji ""Ali. Components: Body with throne, handle, ring, six plates, rete, alidade, pin, ringlet, and wedge. Diameter: 91 mm. Material: Brass. Characteristics: The throne is shaped like a Moorish arch. It is decorated on front and back with embossed leaves and vines surrounding a large central flower. The handle is a narrow stirrup. There is a scale of 360? on the rim. The first division is marked " 5 , " the last, "60" (in abdjad numerals). Inside the front central area there is a tooth at 270? In addition, there is a FIGURE 47.?Astrolabe No. 42. Alidade (top) and "ankabut. 78 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O C i FIGURE 48.?Astrolabe No. 42. Plates, including plate for multiple latitudes (top), for latitude 30?, and for latitude 32? gazetteer listing 34 places and including for each, longitude, latitude, and azimuth of the Qibla. The upper margin on the back of the astrolabe contains altitude scales. The lower margin con? tains cotangent scales divided into feet (left) and fingers (right). There is a stereographically pro? jected declination scale in the upper right quad? rant. Lines marking the azimuth of the Qibla have been engraved upon the declination arcs. The azimuth lines are identified with the follow? ing cities: Tus (lowest), Isfahan, Baghdad, and Shiraz. For a transcription and translation of th( inscription relating to this scale, see CCA No. 49 Lines marking the altitude of the noon sun fo: every 2? of latitude between 28? (at left) and 40' are also graphed onto the declination arcs. For ; transcription and translation of the inscriptioi relating to these lines, see CCA No. 59. In th( upper left quadrant, sines and cosines have beei engraved at sexagesimal intervals. The lowe quadrants contain shadow squares divided inti feet (left) and fingers (right). There is a scale c NUMBER 45 79 triplicities inside these. The title of the scale is transcribed and translated in the description of No. 37. Both Arabic and Persian terms are used to identify natures. The signature occupies a car? touche below the shadow square. Five semicir? cular astrological scales have been engraved out? side the shadow squares. These include the plan? etary governors of terms, the degree divisions of terms, the zodiacal signs, the planetary governors of faces, and the lunar mansions. The margin contains the following inscription at 270?: It reads: "Fifteenth work" (GS). Six plates fit into the front central area. All have a notch at 270?. One plate face is inscribed with 17 stereographically projected quarter hori? zons, representing every 3? of latitude between 15? and 66?. There are four declination scales on this face as well. All the other faces contain stereographically projected coordinates for a sin? gle latitude. These include altitudes (every 6?), azimuths (every 10?), unequal hours, and equal Italian hours. The faces are identified as follows: Plate l b ' 2a 2b 3a 3b 4a 4b 5a 5b 6a 6b Latitude "ard li-''ard li-''ard li-^ard "ard "ard "ard "ard "ard "ard ^ ard 29 25 00 32 00 29 26 32 00 30 36 32 20 36 00 32 38 Length of daylight sa" atuhu sd" dtuhu sd"dtuhu sd"dtuhu sd"dtuhu sd"dtuhu sa "atuhu sd"dtuhu sd"dtuhu sd" dtjj.hu sd" dtuhu 13 52 13 24 14 08 13 55 14 06 13 56 14 28 14 08 13 26 14 08 14 39 The simple rete pattern includes a counter- changed equinoctial colure inside the ecliptic band. There is a curling vine springing from the lowest point of the Capricorn circle. The Capri? corn circle is itself decorated with a zig-zag design. The ecliptic is subdivided into units of 6?. There are 29 inscribed star pointers. The alidade is not counterchanged. One arm is divided into six equal units. The other arm contains two scales: one (along the beveled edge) consisting of 12 equal divisions; the other, of 6 unequal divisions. Each sighting plate is pierced by two holes. The pin is a bolt with hemispherical head and knob- shaped tip. There is a washer-shaped ringlet. The wedge is shaped like a bird. The inscription is in Arabic Naskhi; Hindu- Arabic numerals identify the equal hours on the plates. Remarks: At least 12 known astrolabes have been signed by this maker. Three of these are in the United States: CCA No. 3509, private collec? tion (undated); CCA No. 3614, Hayden Plane? tarium, New York City (undated); and CCA No. 3678, private collection (undated). Dated astro? labes by ^Ali were made in A.H. 1203 [1788], 1205 [1790], 1206 [1791], 1207 [1792], and 1208 [1793].'^' The pattern of the rete of this astrolabe is remarkably similar to patterns adopted by several other makers at work in Isfahan, including "^Abd al-A^imma (Nos. 37, 39, and 40). Some of the plates of this instrument are less carefully en? graved than others (plates 2, 3, 5, and 6). The alidade is a very inaccurately engraved replace? ment. CCA No. 44 (FIGURE 49) Date: Ca. A.D. 1660. Signature: Muhammad Mahdi. Components: Body with throne, handle, ring, five plates, rete, alidade, pin, and wedge. Diameter: 88 mm. Material: Brass. Characteristics: The throne is shaped like a Moorish arch. A symmetric floral arrangement has been embossed on the front; a symmetric vine with bifid flowers has been embossed on the back. The rim contains a scale of 360? The first division is marked " 5 , " the last, "60," in abdjad numerals. The front central area contains a gazetteer listing 50 places, along with the longitude, latitude, the azimuth of the Qibla, and the direction of the azimuth of the Qibla for each. On the back of the astrolabe, the upper margin is divided by altitude scales. The lower margin contains cotangent scales divided into feet (left) 80 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 49.?Astrolabe No. 44. Back view. NUMBER 45 81 and fingers (right). The first two divisions of these latter scales contain the following inscription: ,L^ t^ ^ i > ^ Translated, it reads: "The intention is to have an image remain after us" (GS). The upper right quadrant contains a set of six stereographically projected declination arcs. Two different sets of lines have been graphed on these arcs. Those graphed in the upper portion of this quadrant are identified from left to right with the numbers 27, 30, 33, 36, 39, 42, and 45; they can be associated with an inscription outside the out? ermost declination band, which reads "The circles of mid-day" (see CCA No. 59 for transcription). The lower graphed lines are identified as follows (beginning with the lowest): Yazd, Tus, Isfahan, Basra, Baghdad, Shiraz. These lines can be asso? ciated with an inscription inside the innermost declination circle, which reads, "The lines of the azimuth of the Qibla at the cities." (See CCA No. 49 for transcription.) Sixty equally spaced sine lines divide the upper left quadrant. The shadow squares in the lower quadrants are divided into feet (left) and fingers (right). Inside these squares is a box scale of triplicities consisting of four rows and 10 columns. The scale is identified with the following inscrip? tion: J.:JJL^ ^ .1 "? I*. o J l e-^ ' In translation this is, "Quadrant of Triplicities [with the Planetary Lords of] Day [and] Night" (GS). Both Persian and Arabic terms are used to specify the natures, which are fiery, earthy, airy, and watery. ^ 1 1 ^ A cartouche below the shadow square contains the signature. One of the four plates is notched four times at the ends of the stereographically projected solsti? tial and equinoctial colures. The remaining four plates are notched only once at their lowest point. One face of the multinotched plate contains 24 stereographically projected horizon segments, one for every 3? of latitude between 3? and 66? The remaining plate faces each contain stereographi? cally projected coordinates for individual lati? tudes; the parameters include altitudes (every 6?), azimuths (every 10?), and unequal hours. The plate faces are identified with the following inscriptions: Plate 2a 2b 3a 3b 4a 4b 5a 5b Latitude li-''qrd "ard li-"ard li-"ard "ard "ard ^ard "ard 23 00 30 00 27 00 34 00 33 34 37 00 36 00 Length of daylight sd"dtuhu sd"dl sd" dtuhu sd" dtuhu sd" dtuhu s a" dtuhu sd" dtuhu sd" dtuhu 14 20 [? 14 25 The projection on the back of the horizon face is unidentified. The pattern of the rete includes a counter- changed equinoctial bar inside the ecliptic and vines forming an inverted Moorish arch in the are between the ecliptic and the Tropic of Capri? corn. There are 29 inscribed star pointers. The ecliptic is subdivided into units of 6?. Some circles and arcs radiating from the center have been engraved on the back of the instrument. The alidade is neither counterchanged nor in? scribed. The pin is a bolt with hemispherical head and knob tip. The wedge is duck-shaped. The alphabet used in the inscriptions is Arabic Naskhi. Remarks: The workmanship on this instru? ment is of varying quality. For example, the gazetteer is beautifully engraved, with the places arranged (with a few exceptions) in order of increasing latitude. However, the lines on the 82 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y back of the astrolabe?especially the upper left quadrant?show signs of less careful work. The azimuth lines on all plate faces are of very poor quality. On the horizon plate, the outermost arc in each quadrant is identified as "66." The rete shows signs of having been cast. Its central hole is slightly off center. Other astrolabes in the col? lection with retes similar to this one are Nos. 37, 39, 40, 59, and 65. There is every reason to believe that the alidade, pin, and wedge are replace? ments. If this instrument is indeed the work of its signer, Muhammad Mahdi, it is not his best effort. Mahdi's signature appears on 21 other known astrolabes. In some cases he is specifically designated "decorator" or "engraver." Two of Mahdi's other works are preserved in the United States: one (signed and dated) is No. 25 in this catalog; the other, CCA No. 2545, (signed) is preserved at the Adler Planetarium in Chicago. CCA No. 47 (FIGURES 50, 51) Date: Ca. 1700. Signature: None. Components: Body with throne, handle, ring, four plates, rete, alidade, pin, and wedge. Diameter: 95 mm. Material: Brass. Characteristics: The throne, shaped like a Moorish arch, is undecorated and uninscribed. The handle is stirrup-shaped with half hexagonal cross-section. The ring has a circular cross-section. The rim is divided into four scales of 90? inscribed with abdjad numerals, each proceeding in a clock? wise direction starting at 90?, 0?, 270?, and 180?. A gazetteer has been engraved in the front central area. It consists of 24 place names with longitudes, latitudes, and azimuths of the Qibla. The follow? ing phrases have been engraved in the middle of the gazetteer, diametrically opposite each other: the first. On the back of the instrument, the upper margin contains altitude scales. The lower margin contains cotangent scales divided into feet (left) and fingers (right). One upper quadrant contains stereographically projected declination arcs. The following phrase is written inside these arcs in the upper right quadrant: "The beginning of the signs and lines of the Qibla" (GS). The upper left quadrant is divided by the sines of the angles at increments of 5? and by radii which divide the quadrant into five- degree angles. Shadow squares divided into feet (left) and fingers (right) have been engraved in the two lower quadrants. Also in the lower quad? rants there are six bands suitable for astrological scales. Only two, a scale of lunar mansions and a scale of zodiacal signs are complete. Each of the four plates has a tooth at 270?. One face on one of these (la) contains only stereographically projected semi-horizons, 37 in all, between 19? and 66?. Face lb contains only stereographically projected colures, tropics, and equator. All other plate faces contain a single stereographically projected horizon and altitudes (every six degrees). Three plate faces (2a, 2b, 3b) also contain azimuths (every ten degrees) and unequal hours. Five plate faces (all except la, lb, 2a) contain equal Italian hours. The plates are identified as follows: Plate la lb 2a 2b 3a 3b 4a 4b Latitude safihah dfdqi "ard tis"in "ard 30 "ard 42 "ard 34 "ard 36 "ard 37 "ard 38 Length of daylight sd" dtuhu 13 42 sd"dtuhu 15 05 sd" dtuhu 14 17 sd"dtuhu 14 27 sd" dtuhu 14 00 sd"dtuhu 14 40 means "southeast," the second, is "southwest." C J - 9 ^ The rete design includes a complete equinoctial colure and a complete celestial equator. Other supporting strapwork is in the shape of twining vines. There are 17 star pointers engraved with Arabic star names. The ecliptic circle is subdi? vided into units of 6?. Equator and Cancer circles have been engraved on the back of the rete. The alidade is not counterchanged. It is unin- NUMBER 45 83 FIGURE 50.?Astrolabe No. 47. Front view of separate parts. 84 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 51.?Astrolabe No. 47. Back view. scribed. Each sighting plate is pierced with two holes; the upper is larger. The pin is a bolt with hemispherical head. The wedge is a stylized horse's head. The alphabet used by this maker is Arabic Naskhi. Hindu-Arabic numbers identify altitudes on one plate face. Remarks: Gunther attributes this instrument to Fadl "Ali. One astrolabe signed by this man is preserved in the Museum of the History of Sci? ence, Oxford, England (CCA No. 46). On one plate of the instrument described here (plate 2) the tooth is broken. The body appears unfinished. The workmanship is not consistently of the finest quality. CCA No. 49 (FIGURES 30, 52, 53) Date: See "Remarks." Signature: None. Components: Body with throne, handle, ring, five plates, rete, alidade, pin, and wedge. Diameter: 116 mm. Material: Brass. Characteristics: The throne is shaped like a Moorish arch. It is relatively broad and symmet- NUMBER 45 85 .A-'- ?' FIGURE 52.?Astrolabe No. 49. Front view. FIGURE 53.?Astrolabe No. 49. Back view. rical. Embossed vines and flowers decorate front and back. The handle is stirrup-shaped with a double-lunate cross-section. The ring also has a double-lunate cross-section. The rim is divided by a scale of 360? The first unit is identified with an abdjad " 5 , " the last, with a "60." There is a tooth projecting into the front central area at 270?. A gazetteer has been engraved on this area. It lists 38 places along with the longitude, lati? tude, and azimuth of the Qibla for each. There are altitude scales in the upper margin on the back of the astrolabe. The lower margin contains cotangent scales divided into feet (left) and fingers (right). The declination scale in the upper right quadrant is stereographically pro? jected. Five lines marking the azimuth of the Qibla for Tus (bottom), Yazd, Isfahan, Baghdad, and Shiraz have been graphed on the declination arcs. The inscription relating to this graph (inside the innermost declination band) reads: o-i-c rt-ojJ^^i o ^ J I ^^ rt 1. iiJi ^z.^Ja.^ u^u.;. It can be translated: "Lines of Qibla directions in the cities marked on the edges in the western akitude" (GS). Also graphed on the declination arcs are lines 86 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y marking the altitude of the noon sun for every two degrees of latitude between 28? and 42?. For a transcription and translation of the inscription relating to this graph, see No. 59. Both sines and cosines have been engraved at sexagesimal inter? vals in the upper left quadrant. The shadow squares in the lower quadrants have been divided into feet (left) and fingers (right). A triplicities scale, identified as , L ^ ^ J l _ , i_^_,L_,^l CJ I -. L-, o II 2^-I .h II has been engraved inside the shadow squares. Its translation reads: "The natures of the triplicities [and the] Lords [of the] Day [and the] Lords [of the] Night" (GS). Arabic terms are used to specify the natures. The five semicircular astrological scales engraved outside the shadow squares are the planetary governors of the terms, the term divisions in degrees, the zodiacal signs, the plan? etary governors of faces, and the lunar mansions. The five plates belonging to this instrument are each notched at 270?. Twenty-one semi- horizons have been engraved on one face, repre? senting every three degrees of latitude between 9? and 66?. On this same face, there are four decli? nation scales. Other faces contain stereographi? cally projected coordinates for a single latitude. On six of these, the coordinates include altitudes (every 3?), azimuths (every 10?), and unequal hours. Faces 2b, 3b, and 4b contain these coor? dinates as well as the equal Italian hours. The single latitude faces are identified with the follow? ing inscription: Plate lb 2a 2b 3a 3b 4a 4b 5a 5b Latitude li-"ard li-"ard li-''ard li- "ard li-''ard li-"ard li- "ard li-''ard li- "ard 29 30 36 34 37 38 32 31 40 Length of daylight sd" dtuhu al-sd"dt sa" dtuhu sd" dtuhu sd" dtuhu sa "dtuhu sd" atuhu al-sd"dt sa" atuhu 13 52 13 56 14 28 14 16 14 33 14 39 14 07 14 01 14 51 The rete pattern is formed by curling vines. It includes a branching vine which springs from the lowest point on the Capricorn circle, and it in? cludes, inside the ecliptic, two vines which form a design in the shape of the number "88." The ecliptic is subdivided into units of three degrees. There are 29 inscribed star pointers. The ecliptic has been divided on the back of the rete, but these divisions do not correspond to the divisions on the front of the rete. The alidade is not coun? terchanged. There are, on the one arm, parts of s tereographical ly projected declination arcs. There are twelve equal divisions along the be? velled edge of the other arm. Five unequal, sun? dial divisions have been engraved along the flat edge of this arm. Each sighting plate is pierced with two holes of equal size. The pin is a bolt with a flat edge and a doorknob-shaped tip. The wedge has a trapezoidal shape. The inscription is Arabic Naskhi. However, Hindu-Arabic numerals identify the equal hours. Remarks: Clues to the date of this astrolabe are provided by the pattern of its rete. The "88" design was adopted by a number of makers at work in Isfahan in the early eighteenth century. Prominent among these are "^ Abd al-A ^imma (see CCA Nos. 11 and 38) and 'Abd al- "Ali (see CCA Nos. 10 and 33). That this instrument was made after these examples is evidenced by the slightly less exacting workmanship. The decorative details are typical of Persian astrolabes, including brack? eting of scale divisions and embossed lettering and decoration on stippled background, but the workmanship is unexceptional. There are errors in the inscription of the marginal scales on the back of the astrolabe; a correction occurs in the scale of zodiacal signs. The plates are character? ized by errors in the placement of altitudes and the inscription of azimuths. There are also some errors in the hour lines. There is an error in the division of the ecliptic band on the rete (Leo has too few and Sagittarius too many divisions). The maker has used some dotted lines in the upper quadrants on the back of the astrolabe. CCA No. 52 (FIGURES 54, 55) Date: A.H. 1189 [1775]. Signature: See "Remarks." Components: Backplate with throne laminated to rim with throne, handle, ring, five plates, rete. NUMBER 45 87 FIGURE 54.?Astrolabe No. 52. Front view. 88 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 55.?Astrolabe No. 52. Back view. NUMBER 45 89 alidade, pin, wedge, index. Diameter: 95.5 mm. Material: Brass. Characteristics: The throne is broad and shaped like a Moorish arch. A shallow, circular depression (10.5 mm in diameter) decorates the front face. The handle is stirrup-shaped. It has a loop for the ring. The ring has a circular cross- section. The rim is divided into a scale of 360? with the last division marked "360" in abdjad numerals. The gazetteer in the front central area is arranged in spiral bands. It includes 47 place names with longitude, latitude, and Qibla azi? muth for each. Two or three additional, barely legible places and parameters, including possibly Ka'^bah, are inscribed in the two innermost seg? ments of the spiral bands. On the back of the astrolabe, an altitude scale occupies the upper margin. Cotangent scales oc? cupy the lower margin with divisions of fingers at left and feet at right. The upper left quadrant is divided by 60 sine lines and 10 quarter circles. The upper right quadrant contains a cartouche inscribed as follows: >^ o'v.o (j'^ 11 ^ i 9 SI C (j_i I ( c 'I p) v3 I "' k n 11 rt g "iif) ?^ v3'^ L-*=' hours in addition to unequal hours, altitudes, and azimuths. The plates are inscribed as follows: .JJl , l _ J l . q .k Ul Translated, it reads: "Made by the one who yearns for the forgiveness of his rich Lord Mu? hammad Sadiq ibn "Alinaqi, rightness protects me?May the Gentle forgive them both, year 1189" (GS). The shadow squares in the lower quadrants are divided into fingers on the left and feet on the right. There are five plates, each with a notch at 270?. One plate face is inscribed with 24 quarter horizons, one for every 2? of latitude between 22? and 68?. The face also contains four declination scales. Two plate faces (5a, 5b) contain altitudes (every 6?) and azimuths (every 10?) for single latitudes. Two other plate faces (2a, 2b) contain unequal hours in addition to altitudes and azi? muths. Two plate faces contain equal Italian Plate lb 2a 2b 3a 3b 4a 4b 5a 5b Latitude "ard "ard "ard "ard "ard "ard "ard "ard "ard 34 30 39 32 33 36 37 42 45 Length of daylight sd" dtuhu sd" dtuhu sd"dtuhu sd"dluhu sd" dtuhu sd" dtuhu sd" dtuhu sd" dtuhu sd" dtuhu 14 14 14 14 36 14 08 14 12 14 20 14 28 14 48 15 The rete of this astrolabe is a very simple design. It incorporates a complete straight equi? noctial colure and a complete projected equator. An engraved chain of diamonds decorates the Capricorn circle. Engraved lines and dots deco? rate equator and equinoctial bar. A piece of metal connecting the ecliptic with the equator has been inscribed: It translates: "O! You Who is true to the promise" (GS). The ecliptic is subdivided into units of 6?. There are 18 inscribed star pointers. Projections of colures, tropics, and equator have been en? graved on the back of the rete. The alidade is not counterchanged. The one remaining sighting plate is pierced by a single hole. A sundial scale divides one arm. There are nine equal divisions on the other arm. The first division along the beveled edge is marked "6," the last, "54." The first division along the flat edge is marked "12;" the last, "60." The pin is a bolt with a flat head and a doorknob-shaped tip. The wedge is duck- shaped. The index is 83.5 mm long. Only one arm is engraved with twelve equal divisions. These are identified twice: once from center to edge from"5" to "60," and once from edge to center. The undivided arm is decorated with vines and flowers on a stippled background. The alphabet is Arabic Naskhi. Hindu-Arabic numerals identify lines projected on the horizon plate. Remarks: The phrasing of the inscription on the back of this astrolabe leaves open the possi? bility that Muhammad Sadiq is its owner rather 90 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y than its maker. Saliba has observed that the inscription on the rete is a play on a portion of the maker/owner's name. The circular depression on the throne may have once held a compass. One of the Hoffman photos shows a cord attached to this instrument. It has since been removed. While there is currently no plate holder in the front central area of this astrolabe, there are indications that there was once a tooth at 270? The inscription of the front central area is of poor quality. The rete, which includes examples of good lettering, is slightly bigger than the front central area. On most plates the word "mashriq" has been written in reverse to preserve a certain symmetry near the horizon. CCA No. 53 (FIGURES 56, 57) Date: A.H. 1216 [1801]. Signature: sana' 'ahu Sadiq (maker), namaq? ahu al-Haqir ""Abd al-Karim (decorator). Components: Rim with throne laminated to backplate with throne, handle, seven plates, rete, alidade, pin, ringlet, and wedge. Diameter: 96.5 mm. Material: Brass. Characteristics: The high, symmetric Moorish arch is embossed with vines and flowers. It is pierced by a central teardrop-shaped hole. The handle is decorated with a high, pierced crown. The axle of the handle pierces the throne from side to side rather than from front to back. The rim is divided into four 90? arcs. Each of these arcs is subdivided into units of 6? inscribed with abdjad numerals. The front central area contains a gazetteer of 23 cities along with their longitude, latitude, azimuth of Qibla, and distance from Mecca (in farsakhs). In the center of the table, the phrase ? ? ? Y*oS pL^ ^^^j^ ^J/^ .r^ rt_9 jji-o A . t,'^ has been inscribed. On the back of the astrolabe, the upper half of the margin contains two altitude scales. The lower half of the margin contains two cotangent scales. The upper left quadrant is divided by ten sine lines and ten cosine lines separated by equal intervals. Four arcs tangent to the first four lines of each scale intersect the sines and cosines. Semicircles have been engraved on the radii of this quadrant. The upper right quadrant contains a lengthy inscription embossed on a stippled background. I C J I ^ L u J I iJ.iLu ( g i c ) " - ^ ^ ^ ^ O-T" I q<^ Lt f ^ " '"^ ^^^^j-i\ rtJUl ^i_ \j L^ ^ l?...^ ? LjiS ^ ^ 3l Sll Ky-f' FIGURE 58.?Astrolabe No. 54. ^Ankabut (left) and plates for latitudes 34? and 37 ' 94 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y ^: 4 FIGURE 59.?Astrolabe No. 54. Back view. NUMBER 45 95 The verse translates as follows: "And with Him are the keys to the unseen?none knows them but He; and He knows what is in the land and the sea; and there falls not a leaf but he knows it, nor a grain in the darkness of the earth, nor anything green or dry but [it is all] in a clear book." This astrolabe has five plates. Each is notched at 270?. One plate face contains eighteen semi- horizons representing every 3? of latitude be? tween 12? and 63?, as well as one complete horizon for 66?. Eight plate faces contain stereo? graphically projected coordinates for single lati? tudes, including altitudes (every 6?), azimuths (every 10?), unequal hours, and equal Italian hours. One plate face (4b) contains, in addition to the coordinates engraved on the other faces, a set of equal Babylonian hours. On all single- latitude faces, the latitude is inscribed accompa? nied by the word sd'^dtuhu. However, no hours are given. The latitudes are: Plate lb 2a 2b 3a 3b 4a 4b 5a 5b Latitude li-"ard li-"ard li-" ard li-" ard li-" ard li-" ard li-"ard li-"ard li- "ard 34 29 35 32 33 36 39 37 38 The rete pattern is remarkably symmetric. It is formed by curling vines and incorporates 13 named stars. The ecliptic is subdivided into units of 6?. The Capricorn circle is decorated with an engraved chain. On the front of the rete there are traces of engraved colures and equator and Can? cer circles. On the back of the rete there are traces of colures, equator circle, and arc divisions of the equator. The alidade is not counterchanged. There are crude equal divisions along the beveled edge of one arm. The sighting plates are pierced with two holes above the fiducial line. The upper hole is larger. The pin is a tapered bolt with a hemispherical head. The wedge is shaped like a duck. The inscription is Arabic Naskhi. Remarks: The attribution of this astrolabe to "^Abd al-Ghafur by Gunther was apparently based on the obvious similarity between its rete and the rete of an astrolabe signed by that maker (see No. 55). There are several indications that this is an unfinished work: there are no entries in the gazetteer; the declination scale is incomplete; no hours of daylight have been engraved on the plates. In designing the rete this maker has se? lected stars that preserve symmetry. All are cor? rectly positioned. "Abd al-Ghafur projects these same stars on his instrument. CCA No. 55 (FIGURES 60, 61) Date: A.H. 1198 [1783]. Signature: ^amal "^Abd al-Ghafur. Components: Body with throne, handle, five plates, rete, alidade, pin, wedge, and index arm. Diameter: 133.5 mm. Material: Brass. Characteristics: The throne is a high Moorish arch, decorated with vines and flowers embossed on a stippled background. There is inscription on front and back. The inscription on the front reads: O ^ o . . ' " ' . ' O j ) ??'*' ??' '3 ? >?< o ?i g 11 p jLc L_. KZJ I O Olll I I rt, tMlyS fLtti a ? ' L-w I - ? ^ I A >. I g and can be translated, "O knower of mysteries, they encompass nothing of His knowledge save what He will. His throne holds the heavens and the earth" (GS). The inscription on the back gives the date of the instrument. The handle is basically stirrup-shaped, but it has a high, pierced crown. The rim is divided by four contiguous scales of 90?, each proceeding in a clockwise direction. There is a tooth in the front central area at 270?. In addition, a gazetteer has been inscribed there. It lists 46 places along with the longitude, lati? tude, and azimuth of Qibla for each. On the back of the astrolabe, the upper margin is divided by altitude scales. The lower margin contains cotangent scales divided into feet (left) and fingers (right). In the upper right quadrant 96 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 60.?Astrolabe No. 55. Front view NUMBER 45 97 FIGURE 61.?Astrolabe No. 55. Inscribed edge. there is a stereographically projected declination scale. For a transcription and translation of the inscription outside the outermost declination arc, see this catalog, CCA No. 59. In the upper left quadrant there are sine lines for every five de? grees. The unequal hours are also engraved in this quadrant, identified with the phrases: which translates, "Lines of unequal hours" (GS). There are shadow squares in the lower quadrants, divided into feet (left) and fingers (right). A scale of triplicities has been engraved inside the squares. Its title is transcribed and translated in the description of CCA No. 37. The natures are identified using Persian terms. Outside the squares there are semicircular lists of planetary governors of terms, terms, zodiacal signs, plane? tary governors of faces, and lunar mansions. The signature fills a cartouche below the shadow squares. On the outside edge of the body of this instrument, there is an inscription which reads: a _ o . ^ J I yj I J .^??^ ? 11 ^j-o A J U L-> ?^9-^ I i l l U . ? I "? ' ^ . I iV II 0 - 9 <-" I ^ o I ^J ^ L_a-o ^1 ^1 h ^^ Translated, it is: "I take refuge in God from the devil," followed by Chapter VI, verse 59 of the Koran (see p. 95). This astrolabe has five plates. Each of these has a notch at 270?. Nineteen semi-horizons have been engraved on one plate face; there is a hori? zon for every 3? of latitude from 12? to 66?. On this same face there are four scales useful for measuring declination. All other faces contain stereographically projected coordinates appropri? ate for a single latitude. The coordinates include altitudes (every 3?), azimuths (every 10?), un? equal hours, and equal Italian hours. The faces are inscribed as follows: ?late lb 2a 2b 3a 3b 4a 4b 5a 5b Latitude "ard "ard ^ard "ard "ard "ard "ard "ard "ard 32 29 40 33 35 34 36 37 38 Length of daylight sd"dtuhu sd" dtuhu sd" dtuhu sd"dtuhu sa"atuhu sd" dtuhu sd" dtuhu sd"dtuhu sd" dtuhu 14 06 13 14 14 51 14 12 14 22 14 17 14 28 14 37 14 43 The rete of this astrolabe exhibits remarkable symmetry. It is composed of curling vines, some of which form a Moorish arch inside the ecliptic. The arch is decorated with an embossed chain of leaves, as is the Capricorn circle. The ecliptic is subdivided into units of 3?. There are 22 inscribed star pointers. No construction guidelines are visi? ble either on the front or on the back of the rete. The alidade is counterchanged about the center. There is no decoration, but each arm is divided into 60 equal units. The sighting plates are each 98 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y pierced with two holes, a larger hole above a smaller one. The pin is a bolt with hemispherical head and doorknob-shaped tip. The wedge is shaped like a duck. This astrolabe is also fitted with a radial index arm decorated with embossed vines and flowers on a stippled background. The inscription is in Arabic Naskhi. Remarks: There is some evidence on the back of this carefully constructed astrolabe that the maker was interrupted in his work. The lengthy inscription in the upper right quadrant refers to lines marking the altitude of the noon sun at various latitudes, but the lines themselves have not been drawn. One other astrolabe bearing "Abd al-Ghafur's signature is in the Hermitage Museum, Lenin? grad, CCA No. 3669. CCA No. 57 (FIGURES 62, 63) Date: A.H. 1202 [1787]. Signature: None. Components: Body with throne, handle, ring. seven plates, rete, alidade, pin, ringlet, and wedge. Diameter: 89 mm. Material: Brass. Characteristics: The uninscribed, undecorated, and unpierced throne is shaped like a scalloped triangle. The stirrup-shaped handle has a dia? mond cross-section, as does the ring. The rim is divided by a scale of 360?. The last division is marked 60 in abdjad numeraXs. There is a gazetteer in the front central area which lists 30 places and gives for each the longitude, latitude, and azi? muth of the Qibla. On the back of the astrolabe the upper margin contains altitude scales. The lower margin con? tains cotangent scales divided into fingers (left) and feet (right). The upper right quadrant con? tains a rectangular multiple difference table en? titled: J Q O J I ,J-?-9 y^j 0-- Jl ^L>T which translates "Excess of Revolution / single years and tens of years" (GS). This quadrant also contains the following: in- i ^ FIGURE 62.?Astrolabe No. 57. "Ankabut (right) with plate for 36"^ NUMBER 45 99 Q;,';li,'- iV-^^-p'^ - .> / ( ! ' iK'T;^ FIGURE 63.?Astrolabe No. 57. Back view. scription: 'a?.:>.::i ll 9 O^LM-H rt ^ Ic 'L^^j^jb rt_:**M ^ T ? T which can be translated: "In the year 1202 of the Hejira may God's peace and salutations be upon him [the Prophet]" (GS). Twelve equally spaced sines and an equal number of cosines divide the upper left quadrant. Six arcs representing the unequal hours are graphed on this grid. Shadow squares divided into fingers (left) and feet (right) occupy the lower quadrants. Inside the shadow squares there is an untitled table of triplicities and a table of factors surmounted by the follow? ing inscription: J jj^ <^j^ -^ ' ^ J.y=' ^ ' - ^ oL-cL-- 100 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y Translated, it reads "66? 30' latitude, longest day 24 hours." The lower left quadrant contains the date "1153 Yazdijirdii.^^ The lower right quadrant contains the date "2098 Alexandra.'" rt-.^^ ?.'^ ... I Y ? HA The astrolabe is fitted with seven plates, each notched at 270?. The faces of one plate contain only straight lines and circles. Five faces (la, lb, 3a, 4a, 6a) contain a stereographically projected reference frame for a single latitude consisting of horizon, altitudes (every 6?), and unequal hours. One face (3b) contains a reference frame that consists of horizon, altitudes (every 6?), azimuths (every 10?), and unequal hours. Six faces (2a, 2b, 4b, 5a, 5b, 6b) contain a reference frame consist? ing of horizon, altitudes (every 6?), unequal hours, and equal Italian hours. All faces contain? ing the reference frame are engraved as follows: Plate la lb 2a 2b 3a 3b 4a 4b 5a 5b 6a 6b Latitude "ard "ard "ard "ard "ard "ard "ard "ard "ard "ard "ard "ard 26 22 29 27 34 30 36 37 34 36 40 19(?) Length of Daylight sd" dtuhu sd" dtuhu sd" dtuhu sd"atuhu sa" atuhu sd" dtuhu sd" atuhu sd"atuhu sd" dtuhu sd" atuhu 13 21 13 52 34 14 17 13 57 14 30 36 14 14 17 14 22 14 52 15 The rete exhibits a certain symmetry. In addi? tion to vines and leaves it also includes a straight equinoctial bar and the lower portion of the celestial equator. There are only nine inscribed and one uninscribed star pointers. The ecliptic band is subdivided into units of 6?. The alidade is not counterchanged. There are six equal divi? sions (representing declination arcs) visible on one arm. On the other arm there are eight equally spaced divisons corresponding with the sine scale and at least three sundial divisions. Each collaps? ible sighting plate is pierced with three holes, none above the fiducial line. The pin is a cylinder with a hemispherical head. The ringlet has a scalloped edge. The wedge is a horse with a duck on its back. Eyes of both are drilled. The inscription is Arabic Naskhi. Remarks: The workmanship evident in this in? strument is inferior. Several corrections in posi? tioning and numbering of lines can be seen. The inscription on the rete is in an amateur hand. The sighting plates cannot be used accurately. The maker has used dotted lines to emphasize portions of the projections on the plates. The tables inside the shadow square and in the upper right quadrant compare with ones inscribed on No. 2569. It looks as though the maker of this astrolabe intended to include the contents of the upper right quadrant in the shadow squares. For some reason he changed his mind and shifted them to their present position. CCA No. 58 (FIGURES 64, 65, 66) Date: A.H. 1234 [1818]. Signature: sana^^ahu Muhammad Akbar. Components: Rim with throne attached to backplate with throne, handle, ring, six plates, rete, alidade, pin, ringlet, and wedge. Diameter: 222 mm. Material: Brass. Characteristics: The throne is a broad Moorish arch, decorated (front and back) with an em? bossed dedicatory inscription. The handle is stirrup-shaped. The ring is quite thick. Four scales of 90? have been engraved on the rim. Two of these originate at 0? and two at 180? There is a tooth projecting into the front central area at 270?. Inscribed on the front central area is a gazetteer that lists 70 places along with their longitude, latitude, and azimuth of the Qibla. On the back of the astrolabe there are altitude scales in the upper margin. In the lower margin there are cotangent scales divided into feet (left) and fingers (right). The divisions and inscription of the cotangent scale interrupt a band of en? graved vines with leaves. The upper right quad- NUMBER 45 101 FIGURE 64.?Astrolabe No. 58. "Ankabut and alidade. rant contains a stereographically projected decli? nation scale. Four lines marking the azimuth of the Qibla have been engraved on these declina? tion arcs. They are marked Shiraz (bottom), Baghdad, Isfahan, Shustar. For a transcription and translation of the inscription relating to this scale, see CCA No. 49. There are also eight lines marking the altitude of the noon such for latitudes between 28? and 42?. For a transcription and translation of the inscription relating to this scale, see No. 59. In the upper left quadrant there are sines and cosines at sexagesimal intervals, as well as radii every 5?. There are shadow squares in the lower quadrants. These squares contain a scale of triplicities. The natures are identified using a combination of Arabic and Persian terms. Sig? nature and date are inscribed in a cartouche below the shadow square. Five semicircular as? trological scales occupy the lower quadrants be? tween the shadow squares and the margin. These 102 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 65.?Astrolabe No. 58. Gazetteer. NUMBER 45 103 FIGURE 66.?Astrolabe No. 58. Back view. 104 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y include a scale of planetary governors of terms, degree term divisions, zodiacal signs, planetary governors of faces, and lunar mansions. The astrolabe is fitted with six plates. Each of these has a notch at 270?. One plate is blank on one face except for three stereographically pro? jected circles; the other face is divided into squares. A second plate has 19 quarter-horizons and four declination scales engraved on one face. The other face of the second plate contains ster? eographically projected coordinates for a single latitude. These include altitudes (every 2?), azi? muths (every 10?), unequal hours, equal Italian hours, and equal Babylonian hours. Two plate faces (4b and 6b) contain these coordinates except for the equal Babylonian hours. Six plate faces contain only altitudes, azimuths, and unequal hours. The plate faces are inscribed with latitudes as follows: Plate 2b 3a 3b 4a 4b 5a 5b 6a 6b Latitude li-"ard li-"ard ll-'' ard li- "ard li- "ard li-''ard li-''ard li- "ard li- "ard 38 22 35 29 36 37 43 34 30 Length of Daylight al-sd"dt al-sd"dt al-sd"dt al-sa"at al-sd"dt al-sd"dt al-sd"dt al-sa"dt al-sd"dt 14 40 13 21 14 20 13 52 14 28 14 34 15 54 14 17 13 57 The rete is quite symmetrical. It includes a counterchanged equinoctial colure as well as thick twining vines. The ecliptic circle is subdi? vided into units of 6?. An engraved vine decorates both Capricorn circle and equinoctial bar. There are 13 inscribed star pointers. The alidade has no counterchanges. Six stereographically projected declination arcs have been inscribed on one arm. The other arm is divided into 12 equal units along the beveled edge. There are five unequal divisions along the top. Each sighting plate is pierced with two holes; the upper is larger. The pin is a bolt with a doorknob-shaped tip. The ringlet is ocatagonal. The wedge is shaped like a duck. The alphabet employed by this maker is Arabic Naskhi, with Hindu-Arabic numerals designating the unequal hours on the plates. Remarks: Gunther translates the inscription on the front of the throne as follows: "Made by the order of Mahmud Mirza Kajary Said." Accord? ing to Gunther the inscription on the back reads: "He is God, in accordance with the command of the son of the greatest Shah in the paradise-like city of Naharwand, the work of this slave Mu? hammad Akbar was accomplished." These read? ings are unconfirmed by Saliba. The gazetteer of this instrument is remarkable in that it includes all diacritical marks. The plates are quite heavy. The rete is much repaired. A second astrolabe signed by Muhammad Akbar (dated A.H. 1236) is in the Whipple Museum of the History of Science, Cambridge, England, CCA No. 1006. CCA No. 59 (FIGURES 67, 68) Date: A.H. 1268 [1851]. Signature: sana''ahu Hamza. Components: Body with throne, handle, ring, four plates, rete, alidade, pin, ringlet, and wedge. Diameter: 90 mm. Material: Brass. Characteristics: The throne is a simple Moorish arch inscribed (beginning on the back and con? tinuing to the front): ^jLM**ill ^1 a ^^ jib Q-O JJJLE A^^^-OAJI ^^^J O^L- -b.l-s^JU i/l J.a_x_, ^j p U II Jf t l ,j^ \ TIA ? t S ' ^ o-L: The inscription translates: "Donated by Hamza to whomever is in the charity house of the people of learning and not to be removed except with the permission of the teacher and under his su? pervision. May I be with them on the right path" (GS). The handle is shaped like a stirrup; it has a rectangular cross-section. The ring has a dia? mond cross-section. There is a scale of 360? on the rim. It is divided into 15? units, the first of these being marked "5 15" (in abdjad numbers) and the last "360." There is a tooth projecting from the rim at 270? A gazetteer has been en- NUMBER 45 105 I FIGURE 67.?Astrolabe No. 59. "Ankabut and gazetteer. 106 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURE 68.?Astrolabe No. 59. Back NUMBER 45 107 graved on the front central area which lists 46 place names with their longitudes, latitudes, azi? muth of Qibla, and distance from Mecca. On the back of the astrolabe, the outer margin contains altitude scales in the upper half and cotangent scales of feet (left) and fingers (right) in the lower half. There is a stereographically projected declination scale in the upper right quadrant. Four lines have been engraved on this scale representing the azimuth of the Qibla at Tus (bottom), Isfahan, Baghdad, and Shiraz. For a transcription and translation of the inscription relating to these lines (inside the innermost dec? lination arc), see No. 49. Also graphed on the declination arcs are fine lines representing the altitude of the noon sun at every 2? of latitude between 26? and 42?. An Arabic inscription out? side the outermost declination arc relates to this graph: "iuo-s j^oJl ^^Jys?i\ (j_9 jL^_;.^^ lS-Uft> ^ ? H - ^ J-zf-^' ^ ^ ^^^^^ CJ^ "-^^ - T * ^ This is a poetic code (explained at CCA No. 61) for the rulers of the triplicities. It translates: "At night the second letter is kept in mind." There are five semicircular astrological scales outside the shadow squares, which are the planetary gov? ernors of terms, the term divisions in degrees, the zodiacal signs, the planetary governors of faces, and the lunar mansions. Each of four plates is notched at 270?. On one face there are engraved 51 quarter-horizons rep? resenting every degree of latitude between 12? and 62?. All remaining faces contain stereograph? ically projected coordinates for a single latitude. On one plate (plate 2) these consist only of alti? tudes (every 6?) azimuths (every 10?), and un? equal hours. The equal Italian hours have been added to other faces. The projections are identi? fied with the following inscriptions: Plate Latitude lb li-"ard 2a li-"ard 2b li-"ard 3a li-"ard 3b li- "ard 4a as-safiha mizdn al-" ankabut 4b li-"ard Length of daylight 26 sd" dtuhu 13 12 33 sd"dtuhu 14 00 30 sd"dtuhu 13 52 35 sd"dtuhu 14 23 34 sd" dtuhu 14 18 36 sa"atuhu 14 27 The relativity simple rete pattern includes a counterchanged equinoctial colure inside the ecliptic. A branching vine springs up from the lowest part of the Capricorn circle. There are 30 inscribed star pointers. The ecliptic is subdivided into units of 6?. Lines making the colures, tropics, and equator are visible on the front of the rete. The alidade has no counterchanges. Both arms have been divided into 60 units along the beveled edge. On the flat of one arm there are portions of six stereographically projected declination arcs; on the other, there are five sundial divisions. Each sighting plate is pierced with two holes above the fiducial line. In each case the upper hole is larger. The pin is a bolt with hemispherical head. The ringlet has a rectangular cross-section. The wedge is shaped like a pistol. The Naskhi script is used to inscribe both Arabic and Persian phrases on this instrument. Remarks: Both Hamza's signature and the date appear in the lower margin of the back of the instrument. There are no other known astro? labes with this signature; however, the rete pat? tern adopted by Hamza is one used by makers in Isfahan (notably "^Abd al-A^imma) in the eight? eenth century. Some dotted lines are used in the upper quadrants on the back of Hamza's astro? labe. CCA No. 61 (FIGURES 69, 70) Date: A.H. 1281 [1864]. Signature: See "Remarks." Components: Body with throne (cast in one 108 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y ': wif.ji?.x-!i^,v>s->^, FIGURE 69.?Astrolabe No. 61. Front view. 109 NUMBER 45 FIGURE 70.?Astrolabe No. 61. Back view. no SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y (_j-i???v.o piece), handle, ring, five plates, rete, alidade, pin, and wedge. Diameter: 118.5 mm. Material: Brass. Characteristics: The throne is a high Moorish arch inscribed on both faces. The inscription on the front consists of a list of God, Muhammad, the 12 Imams, and a blessing, as follows: JI ^J'?^^>? (j i-o )l o - L f ijy-alarfuoJl I . ? _ , ? , ^J-Ls^ ,\ It ?? II I I.N n "v o O-s" n- \-^j-i\ cr-J-C C^L?^\ O^ The blessing may be translated, "May the prayers and peace of God be upon them all" (GS). The inscription on the back consists of the following quotation from the Koran: ^ ^ .JyJl ^ L ^ ^J\^J ^ ^ \ \ j . ^ . . j ^ o J l . H I I jSj^ i l l It translates: "And lo! those who disbelieve would fain disconcert thee with their eyes when they hear the Reminder, and they say: Lo! he is indeed mad; when it is naught else than a Re? minder to Creation" (GS). The handle is stirrup-shaped and has a dia? mond cross-section. The ring has no visible join; its cross-section is a flattened circle. The rim is divided into 360? The front central area contains a gazetteer consisting of 52 named places along with the longitude (measured from the Fortunate Isles), their latitude, the azimuth of the Qibla, and the direction of the azimuth of the Qibla for each. The back margin contains altitude scales divided into degrees in the upper quadrants and cotangent scales divided into feet (left) and fingers (right) in the lower quadrants. The upper right quadrant contains a set of stereographically projected declination arcs (separated by 10? or 6? intervals) intersected by graphs indicating (1) the altitude of the noon sun at 11 latitudes (beginning at 22? and separated by 3? intervals between 23? and 50?) and (2) the azimuth of the Qibla at Tus, Yazd, Isfahan, Basra, Baghdad, Shiraz, and Kufa (?). For transcriptions and translations of the inscriptions relating to these graphs, see CCA Nos. 49 and 59. The upper left quadrant is di? vided by both sines and cosines at sexagesimal intervals. Both lower quadrants contain shadow squares divided into feet (left) and fingers (right). A lengthy embossed inscription fills the area framed by the adjacent shadow squares. That the first line of this inscription, ^fiKi^ (5vi-lfi> ^ ^ H - - J-Tf"^ is a mnemonic for the astrological triplicities is suggested by the second line, which reads and may be roughly translated, "At night the second letter is kept in mind" (GS). The first letter in each "word" in the first line stands for one of the four natures: Ndri (fiery), Turdbi (earthy), Hawd^i (airy), Md^i (watery). The sec? ond letter represents the planetary lord of the day associated with that nature: Shams (sun), Zuharah (Venus), Zuhal (Saturn), and Zuharah (Venus). The third letter represents the appropriate plan? etary lord of the night: Mushtari (Jupiter), Qamar (moon) "^Utdrid (Mercury), and Mirrikh (Mars). The fourth letter represents those planets that, according to their nature, have dominion both day and night: Zuhal (Saturn), Mirrikh (Mars), Mushtari (Jupiter), and Qamar (moon). A name. Sahib "^ Ali Kabir Khan, and date, 1281, are inscribed in a cartouche below the shadow squares. The following semicircular tables fill the area between the shadow squares and the cotan? gent scales (beginning with the outermost and proceeding inward): planetary governors of terms, terms, zodiacal signs, planetary governors of faces, and lunar mansions. (An unidentifiable semicircular scale lies between the cotangent scale and the table of planets; it is divided into unequal sections but not inscribed.) Eight plate faces are engraved with the stan? dard network of lines including altitudes at 3? intervals and azimuths at 10? intervals. A latitude and length of daylight has been embossed below the horizon on each plate face as follows: NUMBER 45 Plate lb 2a 2b 3a 3b 4a 4b 5a Latitude "ard "ard "ard "ard ^ard "ard "ard "ard 45 22 25 28 34 30 38 32 Length of daylight sa" dtuhu sd" dtuhu sd" dtuhu sd"dtuhu sd" dtuhu sd"dtuhu sd"dtuhu sa" dtuhu 22 00 13 21 13 30 13 42 14 17 13 36 14 39 14 07 A ninth plate face (5b) contains the standard network (utilizing 3? and 10? intervals) supple? mented by a stereographic projection of the Ital? ian hours. The latitude and length of daylight embossed below the horizon are 36 and 14 18. The tenth plate (la) face contains stereographi? cally projected colures, tropics, equator, and quarter-horizons for every degree of latitude be? tween 14? and 66?; scales of declination (with 2? and 4? divisions, the latter inscribed) are also included. The rete is inscribed with the names of 27 stars. Its relatively simple pattern includes a counter- changed equinoctial colure and a modified Moor? ish arch in the area between the projection of the ecliptic circle and the Tropic of Capricorn. The ecliptic band is subdivided into units of 6?. Trop? ics, equator, and ecliptic divisions have been en? graved on the back of the rete. The alidade is not counterchanged, but it includes a beveled fiducial edge and sighting plates, each pierced with two holes of slightly different diameter. The plates are separated by a distance of 71 mm. Scales en? graved on the arms of the alidade include a set of declination arcs (distributed in the same way as the declination arcs on the back of the body), a scale of sexagesimal divisions, and a shadow scale. The pin is a bolt; the wedge is the head of an animal. The inscription is entirely Arabic Naskhi, ex? cept for the Hindu-Arabic numerals used to des? ignate latitudes on the horizon plate. Remarks: The degree scale on the rim of the instrument is engraved as if it were going to be used as an altitude scale; the 5? divisions in each quadrant are numbered from " 5 " to "90." In the selection and ordering of the series of cities in the inner circle of its gazetteer, the instrument closely 111 resembles Nos. 59 and 62. Scales and networks of lines are characterized by a lack of precision. The bolt, the wedge, and the sighting plate on the arm of the alidade engraved with declination arcs all appear to be modern replacements. The name in the cartouche below the shadow squares may be identified only tentatively as that of the instru? ment's maker. The possibility that he may be its owner is suggested by the fact that his signature differs noticeably from the other writing on the astrolabe. No other astrolabes have been attrib? uted to this "maker." CCA No. 62 (FIGURES 36, 71, 72) Date: A.H. 1281 [1864]. Signature: "amal "^Ala^ al-Din. Components: Body with throne, two rings, five plates, rete, alidade, pin, and wedge. Diameter: 120.5 mm. Material: Brass. Characteristics: The throne is broad and shaped like a bracket. It is decorated on front and back with two engraved parallel lines that follow the edges. The suspensory apparatus consists of two rings. The rim contains four scales of 90?. Two of these proceed clockwise starting at 0? and 180?. Two scales proceed in a counterclockwise direction starting at 0? and 180?. The abdjad numbers identifying subdivisions of this scale are bracketed in groups of three. A tooth projects from the rim into the front central area. The front central area contains an incomplete gazetteer that lists 23 places and associates each with a longi? tude, a latitude, an azimuth of the Qibla, and a direction of the azimuth of the Qibla. The upper half of the outer margin on the back of this astrolabe contains altitude scales. The lower half contains cotangent scales divided into feet (left) and fingers (right). Stereographically projected declination arcs fill the upper right quadrant. Several lines have been engraved on these arcs, six representing the altitude of the noon sun at various unspecified latitudes and one representing the azimuth of the Qibla at an un- 112 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY -^ "n> C . c - ?^-'?I'k ..? ir*" - ^ i ' . ? ? p :-^/ t ?^^?^ FIGURE 71.?Astrolabe No. 62. Front view of separate parts. NUMBER 45 113 r FIGURE 72.?Astrolabe No. 62. Back view. known city. (For a transcription and translation of the inscription relating to the altitude scale, see CCA No. 59. For a transcription and translation of the inscription relating to the azimuth arcs, see CCA No. 49.) Sines and cosines have been engraved at sexa? gesimal intervals in the upper left quadrant. The radii of this quadrant are the diameters of two intersecting semicircles. Shadow squares in the lower quadrants have been divided into feet (left) and fingers (right). The area inside these shadow squares has been divided into a rectangular scale of triplicities. The natures are identified using Arabic terminology. The area outside the squares 114 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y contains a cartouche inscribed with the signature and date and five semicircular astrological scales, including: the planetary governors of terms, the position of each term boundary within each zo? diacal sign, the zodiacal signs, the planetary gov? ernors of the faces, and the lunar mansions. There are five plates. Four of these are notched at 270?. One is notched at 0?, 90?, 180?, and 270?. One face of the plate with the four notches contains 53 quarter-horizons representing every degree of latitude between 14? and 66?. On this same face there are also four scales indicating declination. All other plate faces contain stereo? graphically projected coordinates for a single lat? itude. On seven faces the coordinates include altitudes (every 3?), azimuths (every 10?), and unequal hours. On both faces of plate 5 the coordinates include altitudes (every 3?), azimuths (every 10?), unequal hours, and equal Italian hours. The plate faces are identified with inscrip? tions as follows: Plate lb 2a 2b 3a 3b 4a 4b 5a 5b ll h h ll ll ll h h It Latitude -''ard 45 -^ard 22 -^ard 24 -^ard 28 -^ard 34 -"ard 32 -^ard 30 -"ard 36 -^ard 38 Length of daylight sd"dtuhu sd" dtuhu sd" dtuhu sd"dtuhu sd"aluhu sd"dtuhu s a" dtuhu sd" dtuhu sd" dtuhu 15 22 13 21 13 30 13 00 14 17 14 7 13 36 14 27 14 39 The rete pattern incorporates a complete, counterchanged equinoctial colure. Most of the support for star pointers comes from a branching vine, which springs from the lowest point on the inner edge of the Capricorn band. There are 27 inscribed star pointers. The ecliptic is subdivided into units of 6? There are a number of lines engraved on the back of the rete, including divi? sions of the ecliptic and Capricorn bands. The alidade is not counterchanged. One arm contains portions of six stereographically projected decli? nation arcs. The other arm is divided into 60 equal units along the beveled edge and into five sundial marks along the flat. The maker's signa? ture and date are also inscribed on the alidade. Of the two holes in each sighting plate, the upper is larger. The pin is a bolt with a flat head. The wedge is shaped like a horse and includes en? graved details. The inscription is in Arabic Naskhi. The equal Italian hours and the declination scale on the horizon face are marked with Hindu-Arabic nu? merals. Remarks: While this astrolabe appears to be functional, it is not impressively precise. Some divisions (as those in the upper left quadrant) seem crudely drawn. There are crude divisions on the horizon plate face. The alidade, pin, and wedge described in this catalog entry do not appear in photos taken by Hoffman. The latter two (pin and wedge) apparently replaced the screw and wing nut photographed by Hoffman. CCA No. 63 (FIGURES 73, 74) Date: See "Remarks." Signature: See "Remarks." Components: Body with throne, handle, eight plates, rete, alidade, pin, and wedge. Diameter: 128 mm. Material: Brass. Characteristics: The throne is quite high and triangular in shape, with scalloped edges. On the front of the throne, large embossed letters spell out the beginning of the throne inscription: "His throne includes the heavens and the earth" (see CCA No. 39). On the back of the throne there is a lengthy engraved inscription which reads: J ^ ? 6y>^ ^ ^ yjs.:^^ a^L uP^j J ^ Jai. -,1 , p ,^ 4 ? ^jySJ^ ^.?:...' 1^ ?^3^ '^?^^ ^ji^.^ I jib L ^ <\ on V n rtS ' ' " NUMBER 45 115 FIGURE 73.?Astrolabe No. 63. Gazetteer. the last, "360," in abdjad numerals. There is a tooth projecting into the front central area at 270?. Engraved on the front central area is a gazetteer with longitude, latitude, azimuth of the Qibla, and direction of Mecca for 42 named places. The inscription is incomplete. On the back of the astrolabe, altitude scales divide the upper margin. The lower margin con? tains cotangent scales divided into feet (left) and fingers (right). The upper right quadrant contains equally spaced declination arcs. Seven lines rep? resenting the azimuth of the Qibla have been graphed on these arcs. The lines are identified as follows: jLjiJLo or: Multan (lowest), Shustar, Qandahar, Tus, Isfahan, Basra, and Medina. The inscription in? side the innermost declination arc provides in? structions for finding the Qibla: J ,j?.a^ J aJaL 'LJ >>il.inc *^j)o (jj?- ,L_:iJI rtl-^ o . ?i<: FL-aJi^l vjL-j <^\ ".si rtUi ^j-j Jt , i> ?fOj J5 ^ I _? ? >-? a Jill <^ o,Xu> ^>-9l 9 i j l >?> n.jrtM) &S J ^ %^ fc^HI' ' In the upper left quadrant, there are 13 equally spaced sine lines. In addition, there are radii every 5?. The shadow squares in the lower quadrants are divided into feet (left) and fingers (right). There is a scale of triplicities inside the squares. The natures are identified using both Arabic and Persian terms. Outside the squares, there are three semicircular astrological scales: zodiacal signs, lunar mansions, and planetary governors of the zodiacal signs. The inscription in each division is contained within brackets. The astrolabe has been fitted with eight plates, each notched at 270?. Plate face la is illegible. Plate 2a contains altitudes, and 3a contains alti? tudes marked at every 6?, unequal hours, and an 116 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURE 74.?Astrolabe No. 63. Back view. NUMBER 45 117 arch of unknown significance. All remaining faces contain altitudes (every 6?) and unequal hours. There is only one inscription identifying the lati? tude of the stereographically projected coordi? nates. It is on plate 8a and reads "34 98 ^ard 34." The rete is relatively simple. It includes a com? plete equinoctial colure and part of the equator defined by a scalloped band. In all, there are 22 inscribed star pointers. There are no subdivisions on the ecliptic band. Engraved colures and Ca? pricorn circle are visible on the front of the rete. The alidade has no counterchanges and no mark? ings. Each sighting plate is pierced with two holes. The pin is a bolt with hemispherical head and doorknob tip. The wedge is shaped like a horse. The alphabet employed by the maker is Arabic Naskhi. Remarks: This astrolabe is quite similar, espe? cially in the content and arrangement of the gazetteer, to CCA No. 3811. According to Gunther, the entries in the gazetteer are explained in the inscription on the back of the throne: "Longitude and latitude of countries are recorded on the Hajrah. Longitude from the Observatory of Greenwich, London. Latitude from the Equa? tor as is well known, is taken from dictionaries (?), and the direction from Mecca extracted from the European sine of the shadow. All completed by Mirza Jahan Bakhsh." There are no other instruments with this signature. There are several indications on this instrument that the maker was a careless worker. For example, the sine scale does not originate at the center of the back of the astrolabe. The declination arcs, some of them dotted, are quite carelessly drawn. Altitudes and hour divisions are also carelessly projected. CCA No. 64 (FIGURE 75) Date: Ca. A.D. 1850. Signature: None. Components: Body with throne, handle, ring, four plates, rete, alidade, pin, and wedge. Diameter: 148 mm. Material: Brass. Characteristics: The throne is shaped like a Moorish arch and pierced with three large holes. The handle is basically shaped like a stirrup, but its top is shaped like a Moorish arch. The ring is a scalloped circle with rectangular cross-section. The rim contains a scale of 360?. The first sub? division is marked " 5 ; " the last, "360" in abdjad numerals. Inscribed in the front central area is a worn gazetteer which lists 23 legible places and associates each with a longitude, latitude, and azimuth of the Qibla. A small post interrupts this inscription near the rim at 270?. On the back of the astrolabe, altitude scales occupy the upper margin. In the lower margin, there are cotangent scales divided into feet (left) and fingers (right). The upper right quadrant is blank. Sines have been engraved in the upper left quadrant. The lower quadrants are occupied by shadow squares divided into feet (left) and fingers (right). There are four plates. Three of these are notched at 270?; one is notched at 0?, 90?, 180?, and 270?. One face of the multi-notched plate (3a) is inscribed with 28 quarter horizons repre? senting every 2? of latitude between 12? and 66? There are also four scales of declination along the outer ends of the projected colures. All other plate faces, including the reverse of the horizon plate, contain a sterographically projected network of coordinates for a single latitude. On plate faces la, 2a, 3b, and 4a these coordinates include alti? tudes (every 2?), unequal hours, and equal Italian hours. On plate faces lb and 2b these coordinates include altitude (every 2?), azimuths (every 5? or every 10?), and unequal hours. Plate face 4b contains altitudes, azimuths, unequal hours, and equal Italian hours. The plates are inscribed as follows: Length of daylight Plate la lb 2a 2b 3b 4a 4b Latitude "ard 30 00 "ard 28 00 "ard 38 05 "ard 36 00 "ard 42 08 "ard 34 00 "ard 32 00 sd dtuhu sd" dtuhu sd" dtuhu sd"dluhu sd" dtuhu sd" dtuhu sd" dtuhu 13 57 [?] 39 14 38 14 29 15 04 14 17 14 07 The rete pattern includes a straight equinoctial 118 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURE 75.?Astrolabe No. 64. "Ankabut. colure and a portion of celestial equator between ecliptic and Capricorn band. The central support of the celestial equator is shaped like foliage. There are 22 inscribed star pointers. The ecliptic is subdivided into units of 6?. The alidade is not counterchanged, inscribed, or decorated. There are two holes and a notch above the fiducial line on each sighting plate. The pin is a bolt with hemispherical head and doorknob tip. The wedge is shaped like a horse with engraved details. The alphabet employed by this maker is Arabic Naskhi. Remarks: This astrolabe is almost identical with CCA No. 2566. The rete of one is the image of the rete of the other, except that the rete of No. 2566 has one more star pointer than that of this astrolabe. Any identifying inscription that may once have been on the "extra" pointer has since been obliterated. The gazetteers are also alike in content and arrangement. The similarity NUMBER 45 119 is obvious even though the gazetteer of this in? strument is partially obliterated. The rim divi? sions appear carelessly executed. The same can be said of the sine lines in the upper left quadrant. There are numerous dents and scratches on the back of this astrolabe. Carelessness is also evident in the inscription on the plates. The rete of this astrolabe is less finely cut than the rete of No. 2566, and the back of this astrolabe seems incom? plete. Thus, the evidence suggests that the astro? labe being described in this entry may be a copy of the astrolabe described in No. 2566 or that both astrolabes derive from a third, as yet uni? dentified, instrument. The approximation of the date of this instrument is based on the placement of the stars on the rete, and, if one accepts the premise that workmanship has declined over time, the quality of the workmanship. CCA No. 65 (FIGURES 33, 76, 77) Date: Ca. A.D. 1750 Signature: None. Components: Body with throne, handle, ring, four plates, rete, alidade, pin, and wedge. Diameter: 114 mm. Material: Brass Characteristics: The throne is unsigned, unde? corated, and unpierced. Its shape is a broad, symmetric Moorish arch. The stirrup-shaped handle has a triangular cross-section. The ring has a circular cross-section. The rim is divided by a scale of 360?. The last unit is inscribed with the abdjad numeral "60." The front central area con? tains a gazetteer listing 46 place names along with the latitude, longitude, azimuth of the Qibla, and direction of azimuth for each place named. There is an altitude scale in the upper margin on the back of this instrument. A cotangent scale divided into feet (left) and fingers (right) fills the lower margin. A stereographically projected dec? lination scale divides the upper right quadrant. Upon it have been graphed lines representing the altitude of the noon sun for every 2? of latitude between 28? and 40?, and lines representing the FIGURE 76.?Astrolabe No. 65. "Ankabut and alidade. azimuth of the Qibla for Tus (bottom), Isfahan, Baghdad, and Shiraz. The inscription inside the innermost declination arc relating to the azimuth graph is transcribed and translated in the descrip? tion of CCA No. 49. The inscription outside the outermost declination arc relating to the altitude graph is transcribed and translated in the descrip? tion of CCA No. 59. Sixty sines and as many cosines divide the upper left quadrant. The two lower quadrants contain shadow squares divided into feet (left) and fingers (right). Outside these squares are semicircular astrological scales of planetary governors of terms, term divisions, zo? diacal signs, and planetary governors of faces. 120 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY 4a 4b "ard 36 00 "ard 38 00 sa" atuhu sd"dtuhu 14 27 14 39 FIGURE 77.?Astrolabe No. 65. Back view. Each of the instrument's four plates is notched at 270?. Plate face la contains only projected quarter-horizons, one for each degree of latitude between 14? and 66?. All other plate faces con? tain a stereographically projected reference frame for a single latitude, consisting of a horizon and altitudes (every 3?), azimuths (every 10?), un? equal hours, and astrological houses. The plate faces are inscribed as follows: Plate lb 2a 2b 3a 3b Latitude "ard 42 00 "ard 28 00 "ard 30 00 "ard 32 00 "ard 34 00 Length of daylight sd"dtuhu 15 07 sd"dtuhu 13 47 sd" dtuhu 13 56 sd"dtuhu 14 07 sd"dtuhu 14 n The rete's design incorporates an inverted Moorish arch between the ecliptic band and the Capricorn band. A counterchanged portion of the equinoctial band connects the ecliptic with the central circle. The ecliptic band is subdivided into units of 3? There are 23 inscribed star pointers. Straight lines and circles are engraved on the back of the rete. The alidade arm contains on one arm divisions coinciding with the decli? nation scale and on the other arm divisions co? inciding with the sine/cosine grid and sundial divisions. Each sighting plate is pierced with two holes, with the upper larger. The pin is a bolt with flattened hemispherical head and doorknob tip. The wedge is shaped like a duck. The alphabet employed by the maker is Arabic Naskhi. Remarks: Several aspects of its design distin? guish this instrument from other astrolabes with which it can be compared. The embossed and stippled decoration typical of other eighteenth- century astrolabes made in the Middle East is notably lacking on this example. Engraved vines and flowers surround only one astrological scale. There is no scale of triplicities inside the shadow squares. The stereographically projected horizon reference frames are arranged on the plates in order of increasing latitude. Such an arrangement of plate faces is rare among Eastern makers. Finally, the plate faces contain house divisions distinct from the unequal hours, another practice rarely found among Eastern makers. These distin? guishing elements combine with resembling ele? ments. The rete, for example, has affinities with the rete of CCA No. 40. The selection and ar? rangement of the gazetteer is similar to that found on CCA No. 25. A dot emphasizes every fifth degree subdivision of the rim scale; this is also the case on No. 40. On this instrument the numeral signifying "90" on the right-hand altitude scale forms a design with the numeral signifying "90" on the left-hand altitude scale. The maker uses dotted lines to emphasize every fifth sine/cosine. A few errors mar the astrological scales. The unequal hours are named rather than numbered. NUMBER 45 121 FIGURE 78.?Astrolabe No. 66. Front view of separate parts. 122 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y CCA No. 66 (FIGURES 78, 79) Date: Ca. A.D. 1750. Signature: None Components: Body with throne and handle, ring, five plates, rete, alidade, pin, wedge. Diameter: 89 mm. Material: Brass. Characteristics: The high, Moorish arch- shaped throne is decorated on the front with the following inscription: n ?? h g 11 (^ I g 11 3^ 3 It can be translated as follows: "His throne holds the heavens and earth and He is not fatigued by keeping them and He is the High, the Great" (GS). The handle, cast in one piece with the throne, is a pierced disk. The ring has a circular FIGURE 79.?Astrolabe No. 66. Back view. NUMBER 45 123 cross-section. The rim is divided by four scales of 90?, two commencing at 0? and proceeding clock? wise and counterclockwise, and two commencing at 180? and proceeding clockwise and counter? clockwise. The scales are subdivided into units of 6?, each inscribed with a^SK-o ^j-f t r ^ >Xo.-ai>.a diSJ L ^ >l ??? o , < a J l w-<3-C 9 I ..ii>. V - V \-> ( ^ ) rt>^?. ."^ 2-0J3 ? y^j^ It translates: "Its owner Muhammad Taqi son of Muhib [?] ^Ali Qa^'ini [may it ward off the wrath of God]" (GS). The back margin contains both altitude scales divided into degrees and cotangent scales (fingers left, feet right). Equally spaced arcs and a prayer line are engraved in the upper right quadrant; the upper left quadrant is divided by sines at sexagesimal intervals and by declination arcs con? tiguous with those in the upper right quadrant. Both lower quadrants contain shadow squares: tangent and cotangent scales divided sexagesi- mally on the left, and tangent and cotangent scales divided into feet on the right. A table of triplicities (Persian titles) has been engraved in? side the box formed by the shadow square scales, a cartouche below the shadow squares is inscribed with the signature. Six semicircular tables fill the remaining area of the two lower quadrants: zo? diacal signs, lunar mansions, planetary governors of terms, term divisions, planetary governors of faces, and face divisions. There are no engraved details or original markings on the edge of the instrument. Of the six plates, three are engraved on both faces with the stereographic projection of the horizon for a single latitude, altitudes (every de? gree on plates 3 and 5, every 2? on plate 6), azimuths (every 6? except on plate face 6a, which is marked every 3?), unequal hours, and equal Italian hours. Two other plates are similar to these three. Plate 2a is marked with altitudes every 5?, 2b every 3?, 4a and 4b every 2?, and 2a is marked with azimuths every 5?, 2b every de? gree, and on faces 4a and 4b every 6?, except that one face of each (2b and 4b) has no hour projec? tions, and one face (2a and 4a) has no equal Italian hours. (Plate faces 2b, 4a, and 4b, have coordinates for two latitudes.) Plate lb contains only horizon, altitudes (every 2?), and azimuths (every 6?), and plate la contains parts of stereo? graphically projected horizons for every degree of latitude. The horizon plate face also contains four declination scales subdivided into units of 6?. The plates are identified as follows: Plate lb 2b 3a 3b 4a 4b 5a 5b 6a 6b Latitude safiha mizdn al- "ankabdt "ard 18 "ard 72 li- "ard 22 li- "ard 25 ll- "ard la "ard tis "in "ard 24 "ard kan(?) khafi (?) ll- "ard 27 It- "ard 29 It- "ard 32 li- ^ard 36 Length oj sd"atuhu sd "atuhu sd"dtuhu sd"atuhu sd" dtuhu sd"dtuhu sd"dtuhu ' dayligl 13 13 13 13 13 14 07 14 28 The pin holding the plates in the body is a bolt with a flat head. The wedge is a pin with a rectangular head. Except for some Hindu-Arabic numerals designating equal hours on the plates, all inscription is Arabic Naskhi. Remarks: Both rete and alidade are missing from this instrument. The engraver has made use of dotted lines on plate faces and in a few places on the body of the astrolabe. Only four of the spaces in the gazetteer set aside for length of daylight have been completed. Workmanship of varying quality is evident in the plates. On plate NUMBER 45 129 6 the projections on the faces are displaced by 90? from the orientation of the projections on the other plates. Azimuths show signs of having been re-engraved on plates 2 and 4. A second astrolabe signed (and dated) by Qasim "Ali is in the History of Science Museum, Oxford, England, CCA No. 1219. CCA No. 86 (FIGURES 84, 85) Date: A.H. 1053 [1643]. Signature: sana' 'ahu Muhammad Muqim ben "Isa ben al-Hadad Asturlabi Hamayoni Lahori. Components: Body, throne, handle, ring, five plates, rete, alidade, pin, and wedge. Diameter: 144 mm. Material: Brass. Characteristics: The throne is basically trian? gular in shape, decorated with lobes and fillets. It is covered front and back with plates attached with rivets. The handle is stirrup-shaped; the ring is a corrugated circle. The rim is divided into four 90? scales. The front central area contains a gazetteer listing 77 places along with their geo? graphic latitude and longitude. Diacritical marks are used throughout. On the back of the astrolabe, the upper half of the margin contains two altitude scales. The lower half of the margin contains cotangent scales divided into feet (right) and fingers (left). Six equally spaced declination arcs divide the upper right quadrant. Two noon lines are graphed on these arcs. The upper is inscribed khatt nisJ an- nahdr li "^ard 27; the lower is inscribed khatt nisfan- nahdr li- "^ard 32. Thirty equally spaced sines and cosines divide the upper left quadrant. Shadow squares divided into feet (right) and fingers (left) have been engraved in the lower quadrants. In the area bounded by these squares are two con? centric semicircular tables: the outer is a table of zodiacal signs; the inner is a table of lunar man? sions. The signature is inscribed inside the astro? logical tables; the date is inscribed below the shadow square. Each of the five plates is notched at 270? Plate face la is inscribed with 32 stereographically projected horizons, one for every 2? between 4? and 66?. This face also carries scales which meas? ure the degrees of arc between the projected equator and the tropics. On the back of this face is one which is divided only by stereographically projected altitudes (every 6?) and azimuths (ev? ery 10?). Eight other plate faces contain stereo? graphically projected horizon, altitudes (every 6? or every 3?), azimuths (every 10? or every 6?), unequal hours, and equal Italian hours. Two of these (2b and 3b) also have equal Babylonian hours. The projections are identified as follows: ?'late lb 2a 2b 3a 3b 4a 4b 5a 5b Latitude safiha mizdn al- "anka[b]ul li-"ard 36 li-"ard li-"ard li-"ard li-"ard li-"ard li-"ard li-"ard 36 40 31 32 25 22 Length of daylight sd" dtuhu sd" dtuhu sd"dtuhu sd"dtuhu sd" dtuhu sd" dtuhu sd" atuhu sd"dtuhu 14 14 14 14 14 14 08 13 13 The rete pattern is comprised of trailing vines, trained to form a monogram inside the ecliptic. The ecliptic is rather crudely divided. Only one star name is legible; it is Qalb al-^Aqrab (at approximately 257? right ascension). The alidade is not counterchanged; it carries no inscription. Each sighting plate is pierced with a single hole. The pin is a bolt with hemispherical head and doorknob-shaped tip. the wedge is a small trapezoid. The alphabet used by this maker is Arabic Naskhi. Remarks: It is possible that the covers over the throne hide a delicate, pierced design characteristic of other astrolabes from Lahore. Certainly the use of dotted lines on the body and on the plates is consistent with this acknowledged origin. The plates on which these dotted lines are used to designate altitudes and azimuths (1, 3, 5) are notably more accurate than the two plates which use only continuous lines for these projections. These latter two plates, along with the rete and the alidade. 130 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURE 84.?Astrolabe No. 86. Front view of separate parts. ^^y-:^t--^--?-?: :^-fe:., l-p. It''' ' it^'':^^T'^M; ??rv>' '\\- \^^. v^:%-' NUMBER 45 131 f'ysswiagtt, FIGURE 85.?Astrolabe No. 86. Back view of separate parts. ' ' . ^ ^ w i J i j ^ '^y-t i ^ ^ - 132 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y probably are not the work of Muhammad Muqim. The rim scales measure both altitude (in the upper left and lower right quadrants) and zenith distance (in the upper right and lower left quadrants). The Computerized Checklist of Astrolabes lists 24 other instruments by Muhammad Muqim, including two on which he had collaborated with his brother Qa^im Muhammad (Nos. 69 and 3820). Examples of Muqim's work in the United States include an astrolabe in the Peabody Museum in Salem, Massachusetts (CCA No. 3655), and one at Yale University, New Haven, Connecticut (CCA No. 3807). CCA No. 87 (FIGURES 23, 86, 87) Date: A.H. 1070 [1659]. Signature: ["amal aqall al- ' ibad] Diya^ al-Din Muhammad ben Qa^im Muhammad ben Mulla "Isa ben Shaikh al-Hadad Asturlabi Hamayoni Lahori. Components: Body, handle, ring, five plates, rete, alidade, pin, and wedge. Diameter: 88 mm. Material: Brass. Characteristics: The throne is a pierced Moor? ish arch formed by intertwining vines. The stir? rup-shaped handle has a rectangular cross-sec? tion. The ring has a lens-shaped cross-section. Four 90? arcs, each divided into units of 6? occupy the rim. There is a gazetteer in the front central area that lists 38 places along with the geographic longitude and latitude of each. The band of place means is introduced by the phrase asmd" al-buldan. The upper half of the margin on the back of this astrolabe contains altitude scales divided into units of 6? and 2?. The lower half of the margin contains cotangent scales divided into feet (right) and fingers (left). The upper right quadrant con? tains a set of six equally spaced declination arcs. Two prayer lines have been graphed on these. The upper is identified khatt nisf an-nahdr li-'^ard 27; the lower is identified khatt nisf an-nahdr li-''ard 32. The upper left quadrant is divided by 30 equally spaced sine lines. There are shadow squares in the lower quadrants divided into feet (right) and fingers (left). The signature and date are each inscribed inside the shadow squares. Two semicircular tables of zodiacal signs and lunar mansions fill the area between the shadow squares a,nd the lower margin. Each of the five plates is notched at 270?. Plate face la contains the stereographic projections of 32 quarter-horizons, one for every two degrees of latitude from 4? to 66?. Opposite this face is a face containing only stereographically projected altitudes (every 6?) and azimuths (every 10?). All other plate faces contain, in addition to stereo? graphically projected altitudes and azimuths, the unequal hours and both the Italian and the Ba? bylonian equal hours. The projections are iden? tified as follows: ''late lb 2a 2b 3a 3b 4a 4b 5a 5b Latitude safiha mizan at-" ankabut It h It It h ll ll h -"ard -"ard -"ard -"ard -"ard -''ard -"ard -"ard 36 [?] 1 32 22 24 18 20 Length of daylight sd "dtuhu sd" dtuhu sd" dtuhu sd" dtuhu sd" dtuhu sd" dtuhu sd"dtuhu sd" dtuhu 14 [?] 14 05 13 13 05 13 14[?; The rete pattern incorporates a decorative bar and a bird's head inside the ecliptic. A symmetric pattern of vines connects the Tropic of Capricorn to the center of the rete. No star names have been inscribed. The division line on the ecliptic corresponding to the first point of Aries corresponds with 110? when the muri is aligned with 90?. There are no counterchanges in the uninscribed alidade. A single hole above the fiducial line pierces each plate. The pin is a bolt with hemispherical head. The wedge is a rectangle. The alphabet employed by the maker is Arabic Naskhi. Remarks: The obvious difference in quality between the body of this astrolabe and the NUMBER 45 133 FIGURE 86.?Astrolabe No. 87. Gazetteer. 134 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y various separable parts suggests that the latter were not the works of Diya'' al-Din. The rete is unquestionably nonfunctional. The numbers identifying altitudes on the plate faces do not correspond to specific divisions. There are several stray arcs among the projected hour lines. The body is itself characteristic of the Lahore school. Many of its divisions are marked by dotted lines. The throne is delicately pierced. The scales on the rim can be used to measure both altitude (in the upper left and lower right quadrants) and zenith distance (in the upper right and lower left quadrants). The Computerized Checklist of Astrolabes lists 25 other astrolabes signed by Diya^. Five of these are in the United States, including three in the collection of the Adler Planetarium in Chicago: one dated A.H. 1047 [1637] (CCA No. 2558), one dated A.H. 1057 [1647] (CCA No. 1095), and one dated A.H. 1071 [1660] (CCA No. 2554). The example in the Brooklyn Museum in New York (CCA No. 3555) is undated; the example in a private collection (CCA No. 3809) is dated A.H. 1073 [1662]. CCA No. 88 (FIGURES 35, 88, 89) Date: Ca. A.D. 1750. Signature: See "Remarks." Components: Body with throne, handle, ring, five plates, rete, alidade, pin, and wedge. Diameter: 159.5 mm. Material: Brass. Characteristics: T h e high, Moorish arch- shaped throne is pierced in a symmetric pattern and decorated with an engraved vine and flowers and leaves. The stirrup-shaped handle has a rec? tangular cross-section. The ring has a double- lunate cross-section. The rim is divided into 360?. The last unit is marked with the abdjad numeral signifying "360." The front central area was pre? pared to accommodate a gazetteer of 70 place FIGURE 87.?Astrolabe No. 87. "AnkabUt with plates. NUMBER 45 135 FIGURE 88.?Astrolabe No. 88. Back view. 136 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y ML41JL sasssJSKWWW"**^ r ^ FIGURE 89.?Astrolabe No. 88. Inscribed edge. names. Only 47 places are listed, however, along with the latitude and longitude of each. On the back of this astrolabe, the upper margin is divided by altitude scales. The lower margin contains cotangent scales divided into fingers (left) and feet (right). The upper right quadrant contains an equally spaced declination sc^le. Lines representing the beginning and end of the prayer al-"asr and the sunset line are graphed onto these declination arcs. Sixty sine lines divide the upper left quadrant. There are shadow squares divided into fingers (left) and feet (right) in the lower quardrants. One of the five plates is notched at 0?, 90?, 180?, and 270?. One face of this plate contains the projections of quarter-horizons for every 2? of latitude between 4? and 66?. This same plate face contains four obliquity scales subdivided into units of 6? and ending at 23?30'. Remaining plates are notched only at 270?, and remaining plate faces each contain a stereographically pro? jected horizon reference frame for a single lati? tude. On plate face 4b this network consists only of a horizon, altitudes (every 6?), and azimuths (every 10?); the zodiacal signs are also named on this face. Plate face lb contains only horizon, altitudes (every 5?), azimuths (every 10?), and unequal hours. All other plate faces contain ho? rizon, altitudes (every 3?), azimuths (every 10?), unequal hours, equal Italian hours, and equal Babylonian hours, the inscription on the plate faces reads as follows: Plate lb 2a 2b 3a 3b 4a 4b 5a 5b Latitude 30 21 40 25 27 31 50 36 00 mizan al-"ankabut "ard 38 00 "ard 45 00 "ard "ard "ard "ard "ard "ard Length of daylight sd" dtuhu sd"atuhu sd"atuhu sd"atuhu sa "atuhu sd"dtuhu sd"atuhu 13 21 13 35 13 44 14 8 14 30 14 42 15 28 The relatively symmetric pattern of the rete of this astrolabe is formed by twining vines. There is a complete, counterchanged (once) equinoctial colure. The ecliptic band is subdivided into units of 6? There are 31 inscribed star pointers; on them a small dot marks the part of the flower which represents the star. Circles have been en? graved on the back of the rete. One arm of the alidade contains divisions corresponding to the placement of the declination arcs in the upper right quadrant. Divisions corresponding to the sine scale have been inscribed on the other arm. A single small hole pierces the sighting plates. The pin is a bolt with a hemispherical head. There is a washer functioning as a ringlet. The wedge is shaped like and decorated to resemble a horse's head. The inscription is Arabic Naskhi. Remarks: The pierced throne, relatively fragile rete, division of cotangent scale, and content of gazetteer all suggest that this instrument is of Indian origin. It is unusually thick (9.5 mm). No latitude is given for which the prayer lines in the NUMBER 45 137 upper right quadrant would be appropriate. The sine scale has been carelessly inscribed, but the other workmanship appears to be of good quality. The rete is in excellent condition. The alidade appears to have been cast in one piece. A name, Sahibuhu Maghfur al-Husayni al-Ji? lani, has been engraved on the edge of this astro? labe near the top. Since the hand is different from that of the other inscription on the astrolabe, it seems likely that the name identifies an owner rather than a maker. CCA No. 144 (FIGURES 25, 90) Date: A.H. 704 [1304]. Signature: Ahmad ben Husain ben Baso. Components: Rim with throne riveted to back? plate with throne, handle, ring, ten plates, rete, alidade, pin, and wedge. Diameter: 166.5 mm. Material: Brass. Characteristics: Both front and back faces of the throne are engraved with the details of trifid flowers. Three holes, in addition to a handle hole, pierce its Moorish arch-shaped form. The handle is a stirrup with a diamond cross-section. The ring also has a diamond cross-section. The rim is engraved with a scale of 360?. The engraved lines in the front central area essentially constitute a stereographic projection of altitudes and azi? muths for latitude 90?. Altitudes are projected every 6?; azimuths, every 10?. The outer margin of the back contains two altitude scales in the upper portion and two cotangent scales in the lower portion. A scale of the zodiacal signs follows the perimeter of the quadrant area just inside the outer margin. An eccentric calendar scale borders the quadrant area just inside the zodiac scale. The vernal equi? nox (Aries, 0?) on the latter scale coincides with the date 14.5 March on the former. A scale of unequal hours has been engraved in the upper left quadrant. The signature and date are in? scribed in the upper quadrants. Two shadow squares, both graduated into finger units, fill the lower quadrants. Each of the ten plates includes a tooth. On eight of the plates and on the body the inscription is in Kufic script of the maghribi type. Of the 16 faces on these plates, 9 (lb, 3a, 3b, 4a, 4b, 5a, 5b, 6a, and 6b) contain stereographically projected horizon, altitudes (every 6?), azimuths (every 10?), unequal hours, and prayers: plate 10b con? tains only horizon, altitudes, and azimuths (every 6?); five (8a, 8b, 9a, 9b, and 10a) contain horizon and houses; la contains only horizons and alti? tudes (every 6?). The plates are identified with the following inscriptions: Plate Inscription l a li-jdmi" al-"urild l b "ard Ghamdta wa kull balad "arduhu 37 30 3a ^ard Mecca wa kull balad "arduhu 21 40 3b "ard Iskandaria wa kull balad "arduhu 31 4a "ard Mardkesh wa kull balad "arduhu 30 4b "ard Fds wa kull balad "arduhu 33 5a "ard Sinna [.^ ] wa kull balad "arduhu 35 20 5b "ard al-Jazirah wa kull balad "arduhu 34 30 6a "ard al-Ria[?\ wa kull balad "arduhu 36 30 6b ^ard Malagha wa kull balad "arduhu 31 8a "ard 36 30 8b "ard 33 9a madhhab al-Ghafdr "ard 37 30 9b madhhab Batlamyus[?\ "ard 37 30 10a madhhab Hermes "ard 37 30 10b [no identifying inscription] Two of the plates are inscribed in Arabic Nas? khi script. Of the four faces on these, one (2a) is blank and three contain stereographic projections of horizon, altitudes (every 6?), azimuths, un? equal hours, and prayers. The projections are located as follows: Plate Inscription 2b "ard al-Madina al-Munawwarah wa kull balad 24 "arduhu 7a "ard Qustantiniyah wa kull balad 41 "arduhu 7b "ard Adrina wa kull balad 42 "arduhu The rete of the astrolabe incorporates 27 star pointers, each consisting of a pierced ball or trefoil base and a hook pointer. An engraved star name 138 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURE 90.?Astrolabe No. 144. "Ankabut and alidade. can be identified with each pointer. The ecliptic is subdivided into 6? units. The alidade has a central semicircle and half brackets at the ends of its arms. Its beveled fiducial edge is not inter? rupted by counterchanges. One arm of the ali? dade is divided by 60 (along fiducial line) and 12 equally separated lines and by 5 lines separated by unequal distances. No numbers or letters ap? pear on these scales. The boltlike pin has a semi? circular head and a doorknob-shaped tip. The wedge is shaped like a bird with a long beak. Eyes and headfeathers have been engraved on the basic shape. Remarks: Two black depressions on either side NUMBER 45 139 of the handle hole may have been holes, which were filled in. The third month on the calendar scale is spelled Mars. On the shadow squares, a single designation az-zill al-mabsut extends below both tangent scales. On two plates, the line of morning prayer is designated qaws ash-shafaq. On one of these, both the afternoon and the evening prayer are also so designated. The metal of the two plates engraved in Naskhi script is not as highly polished as that of the other plates. It is unlikely that these plates are the work of Ahmad ben Husain. The Ahmad signature appears on three other astrolabes: CCA No. 132 dated A.H. 664 [1265] and now in the Reale Academia de Historia, Madrid; CCA No. 1080 dated A.H. 704 [1304], now in the Societa Ligure de Storia Patria; and CCA No. 1203 dated A.H. 709 [1309], now in a private collection in Europe. CCA No. 186 (FIGURES 11, 26, 91) Date: Ca. A.D. 1475 (see "Remarks"). Signature: None. Components: Body with throne, handle, ring, six plates, rete, alidade, pin, wedge, and index. Diameter: 105 mm. Material: Brass. Characteristics: The uninscribed armilla fixa has Moorish affinities. Engraved leaves decorate the front. The handle is stirrup-shaped with diamond cross-section. The ring has an oval cross-section. The outer circular band on the rim is divided by a scale of 24 hours. The inner circular band is divided by scales of 360? composed of units of 5? and 1?. Stereographically projected tropics and equator are the only engraved lines on the front central area. The outer margin of the back contains four altitude scales (one in each quadrant). The cir? cular band inside these scales is divided by a zodiacal scale. An eccentric calendar scale is en? graved just inside the zodiacal scale. The vernal FIGURE 91 .?Astrolabe No. 186. Plates for latitudes 37?, 40?, and 48? 140 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y equinox corresponds to 10.5 March. A scale of unequal hours has been engraved in the upper quadrants. Two shadow squares (divided into finger units) have been engraved in the lower quadrants. The inscription is Latin. Four of the six plates have a tooth at 90?. Plate 6 has a notch at 90?; plate 5 has neither tooth nor notch. Each plate face is engraved with the stereographic projection of horizon, altitudes (ev? ery 5?), azimuths (every 5?), and unequal hours. The altitudes and the unequal hours are identi? fied with inscribed numerals only on plate 5a. Each face is inscribed as follows: Plate la lb 2a 2b 3a 3b 4a 4b 5a 5b 6a 6b City Sicilie Emed Alexandrie Damascum Venetie Parisius Neapolis Tvnini Castelli Roma Prvgis Rotomagvs Latitude 37 35 31 33 45 48 40 38 43 42 52 50 The rete design incorporates a complete coun? terchanged east-west bar, a partial counter- changed meridian, and upper and lower equi? noctial arcs. Twenty flame-shaped star pointers are inscribed with the Latin names of stars. One uninscribed star pointer is shaped like an anthro? poid stick figure. The ecliptic circle is subdivided into units of 10? and 1?. Three of the star names are engraved on the backs of their respective pointers. The alidade is counterchanged about the cen? ter. It has a beveled fiducial edge and diagonal ends. Each sighting plate is pierced with two holes; the lower hole is larger. The pin is a cylinder with a hemispherical head. The wedge is shaped like a pin with a head. The index arm is radial with beveled edges and a pointer near the center. Remarks: The numerals identifying the un? equal hours 10, 11, and 12 on the scale of the back of this astrolabe are mirror images of the correct numerals. The anthropoid figure incor? porated into the rete appears to point to the star y Bootis. The right ascension of a Leonis is 141? The assumed date is based on information con? veyed by the zodiacal calendar and the rete. CCA No. 204 (FIGURES 92, 93, 94) Date: A.D. 1548. Signature: I Galois. Components: Body and throne plate, handle, ring, rete, alidade, pin, wedge, and index arm. Diameter: 150 mm. Material: Brass. Characteristics: The throne is shaped like a high crown. On the back it has been inscribed with the word "MIDI." The stirrup-shaped han? dle includes decorative disks at the ends of the crossbar. The margin of the front face of the body plate is divided by five concentric circular scales. The outermost of these is a wind-circle with 12 divisions inscribed, "*MERIDIES* / LIBO*NOTVS / APHRICVS* / *OCCI- DENS* / CHORVS* / CIRCIVS* / *SEPTEN- TRIO* / BOREAS* / VVLTVRNVS* / *OR- lENS* / EVRVS* / EVRO*AVSTER." Next innermost is a scale of 24 equal hours inscribed with Hindu-Arabic numerals as two scales of 12 hours. Next is an uninscribed scale of 360? marked at intervals of 1?. Next is an uninscribed scale of 72 5? units. The innermost marginal scale measures altitude; it consists of 72 inscribed 5? units. The front central area contains the stereo? graphically projected horizon coordinates for lat? itude 48?. Projected lines include horizon (in? scribed at left, "P PRIMA H O R A DIEI" and on the right, "PRIMA H O R A NOCTIS") , alti? tudes (every 3?), azimuths (every 10?), houses (12), and unequal hours. The marginal scales on the back of the astro? labe include (from outermost to innermost) an uninscribed scale of 360? marked in 1? units, a scale of zodiacal degrees marked in units of 5?, a zodiacal scale, an uninscribed scale of 366 days, an inscribed scale of day-groups, and a scale of months. The vernal equinox corresponds to 10.5 NUMBER 45 141 z FIGURE 92.?Astrolabe No. 204. Front view. March. A semicircular table (giving the sign of have been engraved in upper and lower quad- the zodiac, the mediation [see page 56], the name, rants. The signature and date are inscribed below the magnitude, and the declination of 20 stars) the shadow squares, and shadow squares (divided into units of fingers) The rete is a simple design, incorporating a 142 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 93.?Astrolabe No. 204. Back view. complete equator and complete solstitial and equinoctial colures. An " 8 " has been engraved at 270? on the Capricorn band; a " Q " has been engraved at 90?. There are 20 flame-shaped star FIGURE 94.?Stdffler's table of fixed stars, including param? eters identical to those inscribed on the back of Astrolabe No. 204. This rendition appears on folio 17 verso of Elucidatio Fabricae Ususque Astrolabii (Oppenheym, 1512). NUMBER 45 143 ^o? L. Et ttolc cqptons 8^ybi rcgula tangit diamcmim^ Ch* pm^^ tcpcdcmvniimciramalioinccmro.c.ffantcmhocpundum:e^aVc^^^^ to vcrtc pcdcm mobilcmmlmeam occultam Sdimprimcnotam que propofitc ftdic fi^oaageaimalnslkUistabuIeJmponcndocuilibctffgnodu^ ABVLA STELLAR FIXAR continens earum longitudinem Fm cell mediationem dechV nationcs,?partcs ftCmagiiitudmes earundcm* pgcm6 Orftopcie, Dcyrmm Uritg perfWl Ca.ptie<>ijf?i? ~ y6ciim0^Auri. Gimfjgr peg <6riowit Cdnie mM^t* CAni6 min^r* g> i^fufnX ?^fng? 2V)?eer i^uimrud <6rioni6 Caput ?femini antccs^tntie, Capiat Cctntfti fftt?mng Ccrui>^ynie? 144 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y pointers, 18 of which are identified with Latin inscriptions. The ecliptic is divided into 180 units of 2? and 36 units of 10?. The alidade is counter- changed about its center. The words "LINEA FIDVCIE" and "ALHIDADA" are engraved on the arms. The index is complete, counterchanged about the center, and inscribed with the words "LINEA FIDVCIE" on each arm. An " 8 " has been engraved on one arm of the index; a " Q " on the other. A rivet serves as pin. The wedge is in its simple form. Remarks: There are a number of clues embod? ied in this instrument which indicate that the maker consulted an edition of Johannes Stoffler's Elucidatio Fabricae Ususque Astrolabii. The scales and projection on the front of this instrument are similar to those Stoffler describes and illustrates. But the most convincing clue can be found on the back of the work marked "I. Galois." The param? eters in Galois's table of star names are exactly those given in Stoffler's "Tabulae stellarum fix- arum." Nineteen of the entries come from Stof? fler's table giving the mediation or mediationem for certain stars; one entry comes from Stoffler's table giving stellar longitude.^"^^ The 20 stars listed on the back of Galois's astrolabe are the same stars incorporated into Galois's rete. There are some errors in the calendar scale: six day divisions between 19 April and 26 April; six day divisions between May 19 and May 26; five day divisions between 24 December and 1 Janu? ary. The engraver has employed both a Gothic "4" and a sixteenth-century "4" in inscriptions on this instrument. The marks " 8 " and " Q " on the rete and the moveable index resemble the symbols used to indicate conjunction (Q) and opposition (8) in astrology and astronomy. No other instruments are known to have been signed by Galois. CCA No. 221 (FIGURES 14, 32, 34, 95) Date: A.D. 1542. Signature: M.P. Components: Rim, backplate, throne, handle, ring, four plates, rete, alidade, pin, wedge, and index arm. Diameter: 207 mm. Material: Brass. Characteristics: The triangular throne is formed by two symmetric scrolls topped by a disk. The stirrup-shaped handle has a revolving attachment for holding the ring. The outer band of the rim is divided by a scale of 24 equal hours (numbered 1 to 24 in Hindu- Arabic numerals). The inner band is divided by four altitude scales. In between these bands is a band divided into 360 equal parts. The front central area is blank. Marginal scales on the back include a band of four altitude scales, a band divided into 360 equal parts, a band containing the 5? subdivisions of a zodiacal scale, and a band containing 12 equal divisions, each identified with a zodiacal symbol. Inside the zodiacal scale is an eccentric calendar scale, the months inscribed in Latin. The vernal equinox coincides with 10.5 March. The upper right quadrant contains a scale of unequal hours. The upper left quadrant contains a scale of equal hours constructed for use at a latitude of approx? imately 48? 15' (Munich?). Shadow squares di? vided into finger units are engraved in the lower quadrants. The initials of the maker and the date have been inscribed below these squares. All four plates have a tooth at 90? Except for one (la) which has only a series of stereographi? cally projected horizons, all plate faces have ho? rizon, altitudes (every 2?), azimuths (every 10?), houses (seven to twelve), and unequal hours. These latter conventional plate faces are inscribed with the following latitudes: Plate Latitude lb 36 2a 41 2b 43 30 3a 45 3b 47 4a 49 4b 51 The rete has an east-west bar that is counter- changed twice, and the lower half of a meridian NUMBER 45 145 FIGURE 95.?Astrolabe No. 221. Back view. bar. The remaining strapwork consists of inter? secting arcs of circles. There are 21 flame-shaped star pointers. Each of these is inscribed with a Latin star name, sometimes in combination with a zodiacal symbol. The ecliptic is subdivided into units of 2? and 10?. There are no markings on the back of the rete. The sighting plates of the counterchanged ali? dade are each pierced by a slit above a hole, both in the plane of the fiducial edge. The pin is 146 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y cylindrical with a flat head. The wedge is a rectangle. The radial index arm has a beveled edge and a curved bracket design at the outer end, similar to that which decorates the ends of the alidade. Remarks: The initials "M.P." are characteris? tic of the Munich clock- and compass-maker Mar? kus Purmann, who marked some of his work with his initials. All known instruments with Pur- mann's initials are sundials, however, and all date from a period at least 40 years after the date on this astrolabe. CCA No. 262 (FIGURES 12, 96, 97, 98) Date: MDXXXVII [or A.D. 1537]. Signature: Georgivs Hartman Norenberge. Components: Body with throne, handle, ring, four plates, rete, alidade, pin, wedge, and index arm. Diameter: 160 mm. Material: Brass. Characteristics: The throne is a triangle, com? posed of three disks and decorated with the details of rosettes. The ends of the axle of the stirrup- shaped handle are decorated with rosettes. The ring has a circular cross-section. The relatively wide rim of this instrument con? tains three different circular scales. The 12 divi? sions of the outermost scale are inscribed with the Latin names of the winds. Roman numerals (identified as two groups of 12 hours) mark off 24 equal hours on the next band. Inside the hour scale are four altitude scales. Stereographically projected tropics, equator, and colures have been engraved on the front central area. In addition, the following inscribed phrase is barely visible between the equator and the Tropic of Cancer: "Appartenne a Galileo Galilei;" just outside the Tropic of Capricorn at 270? there has been en? graved a small "12." Three different circular scales divide the mar? gin of the back. These include a set of four altitude scales (outermost circle), a zodiacal scale with 5? subdivisions identified with Roman nu? merals (middle bands), and a concentric calendar scale (inner band). The vernal equinox coincides with 10 March. A scale of unequal hours fills the upper quadrants. A double-headed eagle deco? rates the circle defined by the sixth-hour lines. Two shadow squares, divided into fingers, have been engraved in the lower quadrants. The area inside the left square is decorated with an eagle with a human head. Half the area inside the right quadrant is decorated with an eagle; the other half is decorated with diagonal bars. Each of the four plates has a tooth at 90?. Every plate face (except 3b) contains at least the following stereographically projected lines: hori? zon and altitudes (every 2?). Five plate faces also contain stereographically projected azimuths (ev? ery 10?), unequal hours, and houses. On plate 3a, only equal Italian hours supplement the basic projection. On plate 4a, the basic projection is supplemented by azimuths (every 10?), houses, and by both unequal hours and equal Italian hours. Plate face 3b is badly scratched; it contains no projected lines. The projections are identified with the following inscriptions: Plate Latitude la LATITV GRA XL VIII lb LATITV GRA XLV 2a LATITV GRA LIIII 2b LATITV GRA LI 3a [no inscription] 4a LATITV GRA XLII 4b LATITV GRA XXXIX In addition to the identifying inscription along the meridian between the equator and the Capricorn circles, the degrees of latitude for each face have been engraved in Hindu-Arabic nu? merals on the tooth. A small "12" has been engraved just outside the Capricorn circle on faces lb, 2b, and 3b. The rete design incorporates a multiply coun? terchanged (six times) east-west bar. Other struc? tural supports are made up of strings of arcs. The star pointers are 27 pierced bulbar sickles. Each is identified with a Latin inscription. The ecliptic circle is subdivided into units of 6?. The ecliptic divisions and the Capricorn circle have been en? graved on the back of the rete; a small "12" has been engraved at 270?. NUMBER 45 147 FIGURE 96.?Astrolabe No. 262. View of body with inscription. 148 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 97.?Astrolabe No. 262. Back view. The alidade is counterchanged about the cen? ter. Its sighting plates collapse; each is pierced with two holes in the plane of the beveled fiducial edge. A small "12" has been engraved on the underside. The pin is a bolt, the head of which is decorated with a five-petalled rosette. The wedge is shaped like a pipe. The radial index arm is divided by a scale of celestial declination. A leaf NUMBER 45 149 = < ' ? / / 'W'iW / \ / 7 - / ? J : , . ? ? - \ 2 )*s^ --xr--^ '-. y^y-fji^ ^-aSb ?i.iL. ^ j _ - ^ ^ 'K I .. It can be translated: "The direction of these cities [from Mecca] is S.W. except for Damascus" (GS). There is a post projecting from a point near the rim at 270?. On the back of the astrolabe, the outer margin contains two altitude scales in the upper quad? rants. There are cotangent scales in the lower quadrants which are divided into feet (left) and fingers (right). The upper right quadrant contains s tereographical ly projected declination arcs. NUMBER 45 157 ^S #kjf '? . ^*. I f ^^ ^ jr*.'' j ^ . j r ' o > ^i?'^'^~-M/^ FIGURE 102.?Astrolabe No. 2566. Back view. Graphed on these arcs are seven lines representing arc has been transcribed and translated in the the altitude of the noon sun at latitudes 5?, 15?, description of CCA No. 59. Also graphed on these 20?, 25?, 30?, 35?, 40? The inscription relating arcs are four lines representing the azimuth of the to this graph outside the outermost declination Qibla for the following cities: Tus (bottom). FIGURE 103.?Astrolabe No. 2566. Front view of separate parts. .^^, - h . '4^ ., V\ ?^^ '?. " fir'- ? " , '' - / ' ? / P^' '-t''-' : . - / V _' NUMBER 45 159 Isfahan, Baghdad, Shiraz. For a transcription and translation of the inscription relating to this graph inside the innermost declination arc, see the de? scription of CCA No. 49. Sines and cosines divide the upper left quadrant . Shadow squares divided into feet (left) and fingers (right) occupy the lower quadrants. The area circumscribed by these squares is divided into 6 rows and 13 columns. The rows in the first column contain the title of the scale inscribed in the remaining rows and columns. The scale proceeds from top to bottom and from right to left, and gives the names of the planetary governors of terms, the number of de? grees in each term division, and the position of each term division in its zodiacal sign. According to Saliba, the poorly written inscription outside the shadow squares associates angular measures with measures of time, such as 1 ? with 4 minutes and 90? with 6 hours. He notes, however, that many of the associations are erroneous: 54? with 3 hours, 51? with 5 hours. The entire outer edge is inscribed. The astrolabe is fitted with four plates. There is a small hole in each near the edge at 270?. On one plate face there are 28 stereographically pro? jected quarter horizons, one for every 2? of lati? tude between 12? and 66?. Each of the other plate faces contains a network of stereographi? cally projected coordinates for a single latitude. On two of these faces (3b and 4b), the stereo? graphically projected coordinates include alti? tudes (every 2?), azimuths (every 5? or every 10?), and unequal hours. Lines representing the equal Italian hours have been projected onto the other faces. Identifying inscription on the plates reads: Face lb 2a 2b 3a 3b 4a 4b Latitude "ard 42 00 "ard 34 00 "ard 32 "ard 30 "ard 28 00 "ard 38 5 "ard 36 00 Length of daylight sd"dtuhu 15 4 sd"dtuhu 14 17 sd"dtuhu 14 7 sd" dtuhu 13 17 sd"dtuhu 13 39 sd"dtuhu 14 38 sd"dtuhu 14 29 The rete design includes a complete equinoctial colure without counterchanges. There is a portion of a celestial equator outside the ecliptic. A fol? iated central support of the equator contributes to the generally symmetric pattern. An inscrip? tion along the Capricorn band appears to associ? ate signs of the zodiac with masculine or feminine natures, with "changing," "stable," or "mutable" characters, with the elements of fire, earth, wind, or water, with a planet, and with numerals. It reads: '^ *^-'>* ->^ ? ^ ? ^ ty^^ A-1 , ^ o?^^' I HJJ.O %S^^^a )J.a"? O l _ - ^i.> Oli * I JJ^ d L j L > .J^ v ^ ' -^ <*| I in ytM ? W \i*^jjo ( j _ * l ^ ^ a ? ^ ^ O ^ L L . * A J / L ^ A I ? ????. ? Y ? T Urf^ ^^ JLt ? JY ^^?f '-? I *>_:? o jS^i^ *,j^J> 4-:*L> ^J\ j-f^ AJ/Li. o-J-* T?T v ' .j^ ^ j ^ A_jLi- j _ ^ L > t ?. I '4 '? o (Ji'-Jj A_l.Li- ( 5 * ^ ? ^ ? T , ^ ? ^ o?-^' "-^^J*^ ? ^ ? ? ^ fS^jo ^ L_< Oi_i L i ( J * ^ *-^ Li- j-i^ ? Y '?^9-? A tentative translation reads as follows: Aries, Male, Changing, Fire, House of Mars, 201; Taurus, Female, Stable, Earth, House of Venus, 41; Gemini, Mutable, Wind, Male, House of Mercury, 42; Cancer, Changing, Female, House of the Moon, Water, 202; Leo, House of the Sun, Stable, Fire, Male, 202; Virgo, House of Mercury, Mutable?, Earth, Female, 42; Libra, House of Venus, Male, Changing, Fire, 42; Scorpio, House of Mars, Female, Stable, Water, 202; Sagittarius, House of Jupiter, Mutable?, Fiery, Male, 402; Capricorn, House of Saturn, Changing, Earth, Female, 2; Aquarius, House of Saturn, Stable, Fire, Male, 4001; Pisces, House of Jupiter, Mutable?, Water, Female, 201 (GS). There are 23 named star pointers. The ecliptic band is subdivided into units of 6?. Tropics, equator, and ecliptic divisions have been en? graved on the back of the rete. The alidade is not counterchanged. One arm has been divided into the five unequal hours; the other arm contains 11 equal divisions. Each sighting plate is pierced by 160 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y two holes and notched at its upper edge. The pin is a bolt with a tapering end. The wedge is shaped like the heads of a horse, upon which eyes and mane have been engraved. The inscription is Arabic Naskhi. Remarks: The inscription on this astrolabe suggests that at least two different scribes were responsible. One hand is apparent in the scales on the back of the astrolabe; another in the tables in the lower quadrants and around the edge. There is further evidence of combined efforts on the plates, where the azimuths are numbered in a hand different from that used elsewhere on the plates. The maker(s) have employed some dotted lines in the upper quadrant on the back and on the plates. The upper left quadrant is only crudely divided. The rete shows evidence of re? pair. Its pattern is identical to that of No. 64. Neither holes nor notch of sighting plates lie above the alidade's fiducial line. CCA No. 2567 (FIGURES 104, 105) Date: A.H. 1067 [1656]. Signature: "^amal [?] Khalil. Components: Rim with throne attached to backplate with throne, handle, ring, two plates, rete, alidade, pin, ringlet, and wedge Diameter: 132 mm. Material: Brass. Characteristics: The throne of this astrolabe is quite high and pierced to form intertwined vines. The stirrup-shaped handle has a double lunate cross-section. The rim contains a scale of 360? divided by units of 1? (uninscribed) and 5? (in? scribed). The front central area is blank except for engraved projections of the solstitial and equinoctial colures. The upper half of the margin at the back of the astrolabe contains two altitude arcs. The lower half of this margin is blank. There are no scales in the upper half of the quadrant area. The lower half of the quadrant area contains two adjacent shadow squares. The two plates are each notched at 270?. All faces contain stereographically projected horizon (inscribed), altitudes (every 6?), and unequal hours. The projections are identified: Plate la lb 2a 2b Latitude "ard 20 ^ard 24 "ard 40 "ard 35 The simple rete pattern includes the equinoc? tial colure and the lower arc of the equator circle. The ecliptic is subdivided into 60 units of 6? The 16 star pointers are dagger-shaped. The signature and date and parts of the stereographically pro? jected equator and Tropic of Cancer are inscribed on the back of the rete. There are no counterchanges in the alidade. The beveled edge of one arm is divided into 60 equal parts; every fifth division line continues onto the top of the arm. The units of five are numbered. Five unequal divisions on the other arm are not identified. Each sighting plate is pierced with one large hole. The pin is a bolt with a rosette head. The ringlet is octagon-shaped. The wedge is a high trapezoid. The alphabet of the inscription is Arabic Nas? khi. Remarks: All of the engraving on this instru? ment is quite worn. The plates are particularly thin and crudely inscribed. Concerning the word "amal in the signature, Mayer^? has said, "With the exception of a Sultan, not a single man who signed his astrolabe in this way is known from literature." CCA No. 2568 (FIGURES 106, 107) Date: A.H. 1103 [1691]. Signature: sana^ahu al-Hasan ben Ahmad al- Battuti. Components: Rim with throne, backplate with throne, handle, rin, end of cord, three plates, rete, alidade, pin, and wedge. Diameter: 131.5 mm. Material: Brass. Characteristics: The throne is high, but un- NUMBER 45 161 FIGURE 104.?Astrolabe No. 2567. Front view of separate parts. , J yL K\^^^/-' I -'?J^^ I * ? - - ? ? 162 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURE 105.?Astrolabe No. 2567. Back view of separate parts. O 1 NUMBER 45 163 FIGURE 106.?Astrolabe No. 2568. Front view. 164 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURE 107.?Astrolabe No. 2568. Back view. ^ 1 ,!? ?H -.?=^!^!! :. :\ ' \ , i^ - '^ - ' - i ) ^ ?'"H ? FIGURE 108.?Astrolabe No. 2569. Gazetteer. (upper) and khatt nisf al-nahdr li-'^ard 32 (lower). Sixty equally spaced sine lines (some dotted) divide the upper right quadrant. The shadow square in the lower left quadrant is divided into 60 equal units; the shadow square in the lower right quadrant is divided into feet. The lower quadrants of this astrolabe also contain nine as? tronomical or astrological tables. Three of these are rectangular and divide the area inside the shadow square. The upper left one is a table of triplicities, labeled. ^^^-(r^J ij-Kf-i J j a * or, "Table of rulers of triplicities by night and day" (GS). The lower one is a table showing multiples of the difference (in degrees and min? utes) between the approximately correct length of the tropical year and 365 days, labeled, NUMBER 45 167 I FIGURE 109.?Astrolabe No. 2569. Back view. This translates as, "Table of excess of revolution" (GS). The upper right rectangular table gives the length of longest daylight in hours and the cor? responding latitude for upper and lower limits of the seven "climates" or bands of geographic lati? tude, titled: ^ ' { > " * -?V ? ? ? y_^Jj^ yjj,^ which translates as, "Table of extent of [?] hours at midday" (GS). The six circular tables include lunar mansions, zodiacal signs, planetary gover? nors of terms, terms, planetary governors of de- cans, and decans. Each of the five plates included in this instru- 168 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 110.?Astrolabe No. 2569. Full "ankabut (2569A). ment has a notch at 270?. One plate face contains 67 stereographically projected quarter-horizons? one for each degree of latitude between 4? and 65?, four for latitude 66?30', and one for latitude 70?. There are also four obliquity scales at the ends of the projected colures. The greatest obli? quity is 23?30'. All remaining plate faces contain stereographically projected horizon, altitudes (ev? ery 2?), and azimuths (every 6?). Eight of these also contain stereographically projected unequal hour lines and hour lines corresponding to the equal Italian hours. Some of the projected lines are dotted. The plates are identified with the following inscriptions: NUMBER 45 169 FIGURE 111. Plate la lb 2a 2b 3a 3b 4a 4b 5a 5b ?Astrolabe No Latitude safiha ^dfdqi safiha mizan al-" ankabut li-"ard li-"ard li- "ard li- "ard li- "ard li-"ard li-"ard li-"ard 2569. 25 27 29 32 30 38 36 40 Small "ankabut (2569 Length of daylight sd" dtuhu sd"atuhu sd"dtuhu sd"atuhu sd"dtuhu sd" dtuhu sd"dtuhu sd"dtuhu 13 25 13 43 13 52 14 08 13 53 14 42 14 32 14 52 This astrolabe is equipped with two retes. The design of the larger of the two is quite delicate and non-symmetric. It incorporates a counter- changed east-west bar. Tropic of Capricorn, and ecliptic. There are 43 inscribed star pointers, shaped like leaves. A dot on each pointer marks the position of the star. The ecliptic circle is subdivided into units of 6? and 2?. There are no markings on the back of this rete. The second rete is smaller and of simpler design, incorporating the arcs of three circles. The 10 leaf-shaped star pointers, all inside the ecliptic circle, are inscribed with the names of stars, most of which duplicate stars on the larger rete. The alidade is not counterchanged. It is in? scribed with three different scales: a stereograph? ically projected declination scale on one arm and a scale of 60 equal divisions and a sundial on the other arm. Each sighting plate is pierced by three holes, with the uppermost largest and the other two of equal size. The pin is a bolt with a flattened hemispherical head and a knob tip. The wedge is a small trapezoid. Remarks: This anonymous instrument exhibits many of the characteristics of astrolabes known to have been made in Lahore: dotted division lines, equally spaced declination arcs, graphed prayer lines, delicate rete. The rectangular "table of differences" on the back of this astrolabe is similar to tables included on astrolabes with the following CCA numbers: No. 78 (made by Mu? hammad Muqim ben "^ Isa in Lahore); No. 74 (made in India); and No. 94 (inscribed in San? skrit). The rim scale is divided so that the arcs in the first and third quadrants measure zenith dis? tance and the arcs in the second and fourth quadrants measure altitude. All five plates appear to have been engraved by the same astrolabist who was responsible for the inscription on the body. On the alidade, both the declination scales and the sundial are incorrectly lettered and the declination scale (at least) appears to have been re-engraved by someone other than the original maker. The errors could have been made by the person who made the obvious repairs on the alidade sighting plates. In fact, the present en? graved area appears to have been laminated onto another alidade. CCA No. 2570 (FIGURE 112) Diameter: 94 mm. Material: See "Remarks." Remarks: This instrument can easily be distin? guished from functional astrolabes. There is no holder for the four plates and there are no holes 170 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURE 112.?Astrolabe No. 2570. Front view of separate parts. NUMBER 45 171 in the sighting plates of the alidade. The rim of the body has been accurately divided into 72 units, but the numbering of these divisions pro? ceeds in nonsensical order. The front central area has been neatly divided by concentric bands and radial lines, as if to contain a gazetteer, but again the inscribed numbers are meaningless. Of all of the scales engraved on the back of the astrolabe, only the altitude scales in the upper margin and a circular scale of zodiacal signs have any recog? nizable validity. All lines on the plates are neatly engraved and arranged as if stereographically projected. Identifying numbers, however, are in? scribed virtually at random and the inscriptions purporting to give latitudes have no relation to reality. The rete is neatly patterned and includes a bird figure above center and a counterchanged equinoctial colure. The inscription of the ecliptic band, which should begin at 180? and proceed counterclockwise actually begins at 90? and pro? ceeds clockwise. Most of the inscriptions on the rete are not familiar star names, and of the four recognizable names, two are obviously misplaced. The beveled edges on the alidade do not follow the fidicual line. Clearly legible names of zodiacal signs have been engraved on both arms but the numbers have no meaning. There is much inter? nal evidence that the maker of this instrument was familiar with the language found on an astrolabe but that he knew nothing of the astron? omy embodied in such an object. Neither the identity of the maker nor the provenance of this example of his work is known. The script employed throughout is Arabic Kufic. The instrument is made of a silver-colored metal. CCA No. 2571 (FIGURES 113, 114) Date: A.H. 910 [1504]. Signature: "" Ali ben Muhammad ben Abdallah ben Faraj. Components: Rim with throne attached to backplate with throne, handle, three plates, rete, alidade, pin, and wedge. Diameter: 178 mm. Material: Brass. Characteristics: The throne is low and shaped like a bracket. It is undecorated and uninscribed. The handle is shaped like a stirrup. The ring is pear-shaped with a groove around the outer edge. A scale of 360?, subdivided into units at 6?, has been inscribed on the rim. The first unit is des? ignated " 6 ; " the last unit, "60." There is a notch in the rim at 90?. A crude stereographic projec? tion of colures, tropics, equator, altitudes, and horizon for approximately 30? has been engraved on the front central area. On the back of the astrolabe, the upper margin is divided by altitude scales. Inside the margin, there are six complete concentric circular scales consisting of 1? and 5? subdivisions of the zodia? cal signs, the zodiacal signs, one and five, six, or eight day divisions of the calendar year, and the Christian months. The date 13.5 March is oppo? site Aries 0?. The signature and date are inscribed in the upper quadrants. There are shadow squares in the lower quadrants. These are divided into fingers on the left and feet on the right. There are three plates, each with a tooth at 90?. On one plate face (la) there have been engraved stereographically projected straight lines, declination circles (19), horizons (15), and altitudes (15 ?N and 4?S of the horizon at latitude 0?). The face is identified as li-jdmi'' al-''urud. The other five plate faces contain stereographically projected coordinates for a single latitude. These coordinates include altitudes (every 3?), azimuths (every 10?), unequal hours, and lines representing times of prayer. The prayer lines are identified fajr (below the left horizon), al-^asj (within the tenth unequal hour), az-zuhr (within the eighth unequal hour), shafaq (below the right horizon), az-zawal (mid-heaven). On each plate face, an inscription identifies the latitude of the stereo? graphic projection: Plate Latitude lb 2a 2b 3a 3b li-''ard li-''ard ti-''ard li-''ard li-"ard 31 30 30 33 40 34 30 36 30 The rete is a relatively simple design. It in- 172 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y *v>' FIGURE 113.?Astrolabe No. 2571. Front view. eludes a counterchanged equinoctial colure and the lower portion of a stereographically projected celestial equator. This equator arc is supported by a central circle. There are 30 inscribed star pointers. Each of these extends from a base dec? orated with stud. The ecliptic is subdivided into units of 3? The tropics and equator circles have been engraved on the back of the rete. The plain alidade is not counterchanged. Each sighting plate is pierced with a single wide hole. The pin is a bolt with a flattened head. The wedge is an animal shape with engraved eyes, nose, and mouth. The alphabet employed by this maker is Arabic Kufic maghribi. Remarks: An astrolabist with the name "Ali NUMBER 45 173 FIGURE 114.?Astrolabe No. 2571. Back view of separate parts. ben Muhammad ben [?] ben Faraj is listed in Mayer's Islamic Astrolabists. ^ ^^ An uncertainly read date of A.H. 1010 [1601] is assigned to the one example of the work of "Ali described by Mayer. Except for the use of a Kufic mashriqi numeral to designate 300, the instrument described in this catalog entry appears to be inscribed with Kufic maghribi characters. The rete shows signs of re? pair. Alidade, pin, and wedge are apparently replacements. 174 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y CCA No. 2572 (FIGURES 115, 116, 117) Date: A.H. 483 [1090]. Signature: sana' 'ahu Muhammad ben as-Sahli. Components: Rim with throne attached to backplate with throne, handle, ring, five plates, rete, and pin. Diameter: 107 mm. Material: Brass. Characteristics: The low Moorish throne is pierced by four holes, two on each side of the handle hole. The handle is shaped like a flattened stirrup decorated with pierced trefoils at its sides. The ringlet has a diamond cross-section. The rim has been inscribed with a scale of 360? The first unit is marked " 5 ; " the last, "360." The following coordinates, appropriate for latitude 37?30', have been projected stereographically onto the front central area: horizon, altitudes (every 6?), azi? muths (every 10?), and unequal hours. In addi? tion to numerals identifying the various projec? tions, the following inscription appears: (below horizon) Ishbiliyah wa Malagha wa Gharnata '^ard 37 SOsd'^dtuhu 14 19; (along the solstitial colure below the horizon) khatt az-zawdl. Four altitude scales (one in each quadrant) divide the outer margin on the back of this astro? labe. The inner margin is divided by a zodiacal scale. An eccentric calendar scale divides the FIGURE 115.?Astrolabe No. 2572. Front view of "ankabut. NUMBER 45 175 FIGURE 116.?Astrolabe No. 2572. Back view of "ankabut with inscription. perimeter of the quadrants just inside the zodiacal scale. According to this calendar, the date of the vernal equinox is 15 March. The signature and date are engraved in the upper quadrants. Shadow squares divided into finger units fill the lower quadrants. All faces of all five plates have been engraved with stereographically projected horizon, alti? tudes (every 6?), azimuths (every 10?), and un? equal hours. In addition, prayer lines {ash-shafaq, fajr, zuhr, '^asj) have been included on both plate faces of one plate (number 4). The unequal hours are not only identified by number words and symbols, but are also associated with the Arabic terms for animal {hayawan, hours 1, 4, 7, 10), mineral {ma^dan, hours 2, 5, 8, 11), vegetable {nabat, hours 3, 6, 9, 12), and male {mudhakkarah, hours 1, 3, 5, 7, 9, 11) and female {mu^annathah, hours 2, 4, 6, 8, 10, 12). Inscription on each plate face associates it with a geographic location as follows: Plate Inscription la lb 2a 2b 3a 3b 4a 4b 5a 5b [No latitude at the equator] "ard li-Madinat as-sin "arduha li-Mecca, [May God Honor it] [Medina of the Prophet, Peace upon him] wa "arduha Akhmin wa " arduha Qurtubah wa "arduha Ishbiliyah "arduha Gharndta al-Miria min al-iqlim ar-rdbi" wa'' arduha "ard Saraqusta Hours 12 66 18 21 30 24 27 33 30 37 30 36 42 176 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y ^%Mx K i: 1.1 f - - -/-^^f- /??-J vy <^ t^rzT^T?f^rT:i^' fe^)^ c^V f^%* FIGURE 117.?Astrolabe No. 2572. Back view of disassembled astrolabe. NUMBER 45 177 Inscription on these plates and on the body of the astrolabe is Arabic Kufic of the western type. The rete of this astrolabe is of simple design, incorporating a complete, counterchanged east- west bar and a portion of equinoctial arc (outside the ecliptic). There are 22 star pointers. Twelve of these are inscribed with Hebrew star names; the remainder are blank. The ecliptic circle is divided by scratched lines. There is no inscription on it. Two star pointers are inscribed in Hebrew on the back of the rete. In addition, there is a single Hebrew word inscribed along the back of the ecliptic circle. The pin is a short bolt with a flat head. A thin wire serves as wedge. Remarks: This instrument is remarkable for a number of reasons. If the reading of the date is correct, it is the earliest astrolabe in the collection. The phrase bi-Balansiya follows the date. Saliba believes that it refers to the city of Valencia. One of the plates (number 4) is heavier than its fellows, and it exhibits a slightly different style of com? position. The designation of the unequal hours found on all plate faces is rather unusual. It is unlikely that Muhammad ben as-Sahli is respon? sible for the inscription currently found on the rete of this instrument. However, smoothed sur? faces on the rete suggest that an inscription may have been erased to make way for the Hebrew, so the rete pattern may indeed be due to the origin? ator. No alidade and no wedge are preserved with this instrument. The maker is the younger member of a father- son team of instrument makers working in Val- encia.^^^ His name appears together with that of his father, Ibrahim ben Sa^id, on a globe finished A.H. 478 [1085], and preserved in the Museo di Storia della Scienza in Florence. CCA No. 3643 (FIGURE 118) Date: Not legible. Signature: Not legible. Components: Body with throne, handle, ring, six plates, rete, alidade, pin, and wedge. Diameter: 118 mm. Material: Brass. Characteristics: The throne is shaped like a Moorish arch. It is pierced with four holes, placed symmetrically on either side of the handle hole. The handle is stirrup-shaped; the ring has no visible seam. A scale of 360? has been engraved on the rim; it has been subdivided into units of 6? and 1?. The front central area contains ster? eographically projected altitudes (every 6?, un? inscribed), azimuths (every 5?), and a prayer line {az-zawdl) for latitude 90?. The outer margin of the back contains four altitude scales. The inner margin contains a scale of zodiacal divisions and a concentric calendar scale. The date of the vernal equinox is 14 March. The two upper quadrants contain an illegible inscription in Naskhi script. A barely legible date and an illegible signature are inscribed in the lower quadrants below two shadow squares. Each of the six plates includes a tooth (at 90?). All but two plate faces contain stereographic projections of a horizon, altitudes (every 6?) and azimuths (every 10?), unequal hours, and prayers {shafaq, fajr, az-zawdl, zuhr, "asj). The projections are identified as follows: Plate Latitude 2a li-"ard Misr wa kull balad "arduhu 30 00 2b li-"ard Menerah wa kull balad "arduhu 39 00 3a li-"ard al-Iskandaria wa kull balad "arduhu 31 00 3b il-"ard "Asqaldn wa kull balad "arduhu 34 00 4a li-"ard Mardkesh wa kull balad "arduhu 31 30 4b li-"ard Qurtubah wa kull balad "arduhu 38 00 5a li-"ard Fas wa kull balad "arduhu 33 40 5b li-"ard Falastin wa kull balad "arduhu 32 30 6a li-"ard Ishbiliyah wa kull balad "arduhu 37 00 6b li-''ard Tulaitilah wa kull balad "arduhu 40 00 Some prayer lines on these faces are hatched. Of the two remaining plate faces, one contains stereographically projected horizon, altitudes (ev? ery 5?), and azimuths (every 6?) for latitude 66?; the other contains horizons and altitudes for lat? itude 0?; it is inscribed li-jdmi'' al-''urud. The inscription on all plates is Arabic Kufic maghribi. The rete includes 26 bent dagger-shaped star pointers, each inscribed with a star name. The ecliptic is subdivided into 6? units. The inscrip- 178 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 118.?Astrolabe No. 3643. Back view of body with separate parts. ' \ i \* Ix-r 1 V i. -7*? f ' ?^^^^ * ^ . J ^ ' f ^Sfi|k I ^SBL /**. ^Mt **? / \ , V ? .=.-'9 iU*j-?- NUMBER 45 179 tion on both star pointers and ecliptic is Arabic Naskhi. The rete's pattern incorporates a coun? terchanged east-west bar, an interrupted and counterchanged meridian, and upper and lower sections of equinoctial arc. The uninscribed ali? dade is not counterchanged. It has a central semicircle, beveled diagonal edges, and a beveled fiducial line. The sighting plates are each pierced with a single hole. The pin is a bolt with a flattened semicircular head and a doorknob tip. The wedge is a simple rectangle. Remarks: The titles of the shadow square scales on this instrument have been inscribed inside the squares. Both digits and words identify the divisions of the unequal hour scales on all plate faces inscribed with such a scale. Some altitudes and azimuths on all plates are empha? sized using hatched lines. The fact that the rete is inscribed in a hand different from that of the body places its origin in doubt. The simplicity of the alidade suggests that it may be a replacement. In his discussion of this instrument, Gunther translates the modern Arabic inscription on the back as follows: "Bestowed on the Greatest Mos? que the 6th of Muharram year 1308." CCA No. 3811 (FIGURES 119, 120) Date: Ca. A.D. 1850. Signature: None. Components: Body with throne, handle, ring, four plates, rete, alidade, pin, and wedge. Diameter: 111 mm. Material: Brass. Characteristics: The relatively broad, Moorish arch-shaped throne is uninscribed and undecor? ated, except for a geometric design etched on the back face. The handle is stirrup-shaped. There is a heavy ring. The rim is divided into a scale of 360? with the first unit marked " 5 " and the last unit marked "360" in abdjad numerals. Inside the front central area, there is a tooth at 270?. There is also a gazetteer listing 52 places with longitude, latitude, azimuth of the Qibla, and direction of azimuth of the Qibla for each. Altitude scales divide the upper margin on the back of the astrolabe. The lower margin contains cotangent scales divided into feet (left) and fingers (right). There is a scale of stereographically pro? jected declination arcs in the upper right quad? rant. Seven lines, marking the azimuth of the Qibla for various locations, have been graphed on these declination arcs. These locations include Medina (top), Basra, Isfahan, Tus, Qandahar, Shustar, and Multan. For a transcription of the inscription relating to this scale inside the inner? most declination arc, see the description of No. 63. There are sines at sexagesimal intervals in the upper left quadrant. Shadow squares divided into feet (left) and fingers (right) occupy the lower quadrants. There is a scale of triplicities inside the squares. The natures are identified using Ar? abic words. Outside the squares, there are two semicircular astrological scales: a scale of the zodiacal signs, and a scale of lunar mansions. Three other concentric bands have been engraved in these lower quadrants. They are divided but not inscribed. There is a geometric design along the edge of the instrument similar to that etched on the back of the throne. This astrolabe is fitted with four plates. Three of these are notched at 270?. One has notches at 0?, 90?, 180?, and 270?. The plate with four notches contains 30 quarter horizons on one face, represent ing every 2? of latitude from 10? through 66?. This face also contains four decli? nation scales, one at each notch. The other face of this plate contains only stereographically pro? jected tropics and equator. The remaining plate faces contain stereographically projected coordi? nates for a single latitude. One of these (4b) has only altitudes (every 6?) and azimuths (every 10?). One plate face (3a) has altitudes (every 3?), azimuths (every 10?), and unequal hours. One plate face (3b) has altitudes (every 3?), azimuths (every 10?), unequal hours, equal Italian hours, and a prayer line (marked samt Qiblat Tus). Three plate faces contain altitudes, azimuths, unequal hours, and equal Italian hours. The plates are inscribed with the following latitudes: 180 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURE 119.?Astrolabe No. 3811. Gazetteer. ^late 2a 2b 3a 3b 4a 4b "ard "ard "ard "ard "ard mizdn Latitude al-" ankabut 29 32 34 37 36 Length of daylight sd"dtuhu 13 42 sd"dtuhu 14 07 sd" dtuhu 14 17 sd"dtuhu 14 32 sd"dtuhu 14 28 The rete pattern consists largely of curling vines, which frame a bird-like figure just above the center. There is a complete equinoctial bar with one counterchange. There are 34 inscribed star pointers. The ecliptic is subdivided into units of 6?. Colures, tropics, and equator have been engraved on the back of the rete. The alidade is NUMBER 45 181 FIGURE 120.?Astrolabe No. 3811. Back view. not counterchanged. It is neither decorated nor inscribed. There are two holes in each sighting plate. The pin is a bolt with hemispherical head and tapered tip. The wedge is an amorphous shape pierced by a small hole. The alphabet employed by this maker is Arabic Naskhi. Remarks: The longitudes in the gazetteer have been measured from Greenwich. Similarly meas? ured and arranged longitudes can be found on No. 63. The maker uses dotted lines to emphasize some division lines in the upper quadrants on the back of the astrolabe. There are also some dotted lines on the plates. CCA No. 4000 (FIGURES 121, 122, 123) Date: 1861 [1805] (see "Remarks"). Signature: Virabha (Vitrabhadra). 182 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 121.?Astrolabe No. 4000. Inscribed edge. FIGURE 122.?Astrolabe No. 4000. Rete. Components: Body with throne, handle, ring, eight plates, rete, alidade with sighting tube, pin, and wedge. Diameter: 141 mm. Material: Brass. Characteristics: The throne has the shape of a high Indian arch. The signature and date are inscribed on the front in Arabic. The handle is stirrup-shaped, with a diamond-shaped cross-sec? tion. The ring has a circular cross-section. The rim is divided by a scale of 360? divided into units of 6?. The first unit is marked " 1 ; " the last NUMBER 45 183 ^:'V:''^M ?^^*^^iii FIGURE 123.?Astrolabe No. 4000. Back view. unit, "60." A tooth projects into the front central area at 90?. The front central area has been divided by concentric circles and radial lines. The maker may have intended to include a gazetteer. but has filled one concentric band only, with the following inscription in Sanskrit: "The headman Sri Rama, government of the town of Jodhpur, 1235 [1819]." 184 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y On the back of the astrolabe, the upper margin contains altitude scales. The upper right quad? rant contains equally spaced declination arcs. Engraved on these arcs are lines representing the unequal hours and two prayer lines. In addition, there are two engraved semicircles and two en? graved quarter circles. Sines and cosines have been engraved at sexagesimal intervals in the upper left quadrant. There are shadow squares in the lower quadrants divided into feet (left) and fingers (right). Outside these squares are four semicircular scales. Two of these give measures of solar declination; two record right ascension. To? gether the scales relate solar declination to every degree of right ascension. There are two scales engraved on the outer edge of this astrolabe, each divided into 46 units. The divisions of the first of these are inscribed with the even numbers from 2 to 90. The divisions of the second are inscribed as follows: 343 39 (adjacent to 2), 181 36, 114 08, 85 23, 68 03, 46 27, 48 48, 40 26, 36 42, 32 58, 30 00, 27 00, 24 36, 22 10, 20 30, 18 40, 17 34, 16 31, 15 21, 14 00, 13 19, 12 06, 11 36, 10 48, 10 07, 9 22, 8 43, 8 06, 7 30, 6 52, 6 23, 5 51, 5 25, 4 50, 4 22, 3 54, 3 26, 3 00, 2 32, 2 16, 1 41, 00 50, 00 38, 00 12, 00 00 (adjacent to 90), 12 (adjacent to a division inscribed with an illegible character). The instrument has been fitted with eight plates. Each of these is notched at 90?. Plate face la contains only quarter horizons (41), represent? ing every degree of latitude from 10? to 67?. Plate 2a contains, in each of four quadrants, stereo? graphically projected altitudes (every 6?) for a single latitude. All remaining plate faces contain the following stereographically projected coordi? nates: altitudes (every 6?), azimuths (at various intervals), unequal hours, and equal Italian hours. The following latitudes, locations, and pa? rameters are inscribed on the plate faces: Plate Inscription lb Latitude 26 Ajmer / Shadow 5 50 2a [UL] Latitude 48 / Shadow 13 50; [UR] Latitude 60 / Shadow 20 3; [LL] Latitude 42 / Shadow 11 10; [LR] Latitude 52 / Shadow 15 50 2b Kuruksetra [=Panipet] 30 / Shadow 06 55 3a Tilanqa 1 8 / S h a d o w 03 55 Longest Daylight 32 (ghatikas) 3b Daulatabad 20 / Shadow 04 20 4a Ahmadabad 19 / Shadow 04 18 4b Delhi 29 / Shadow 06 40 Longest Daylight 34 34 (ghatikas) 5a Latitude 21 Burhanpur / Shadow 04 30 5b Latitude 27 Gwalior / Shadow 05 35 6a Agra 27 / Shadow 06 06; Mecca 22 / Shadow 04 04 6b Agra 27 / Shadow 06 06 7a Nepalapura 31 / Shadow 06 20 7b Samarqand 40 / Shadow 11 23 8a Lahore 32 / Shadow 07 31 8b Kashmir 35 / Shadow 08 03 The rete pattern is symmetrical. It includes a bird figure just above center, a straight equinoc? tial colure, and a stereographically projected ce? lestial equator. There are 15 inscribed star pointers. The ecliptic is subdivided into units of 6? and 2?. Tropics and equator have been en? graved on the back of the rete. There are no counterchanges in the alidade. Neither are there markings of any kind. The sighting device in? cludes a tube which connects the sighting plates. Each of these latter is pierced with two holes. The pin is a bolt with bell-shaped head and doorknob- shaped tip. The wedge is rectangular. Remarks: Inscription of the divisions of the ecliptic consists of the first two letters of the Indian names of the zodiacal signs. The "shadow" parameter given on the plates is the length in digits of the noon equinoctial shadow of a 12- digit gnomon at the given latitude. The date inscribed on the instrument is a year in the Vikrama era which began in 57 B.C. Hence, the equivalent date in the Gregorian calendar is A.D. 1805. CCA No. 4001 (FIGURES 124, 125) Date: A.H. 621 [1224]. Signature: Muhammad ben Fattuh al- Khama^iri. Components: Rim with throne laminated to backplate with throne, handle, four plates, pin, and wedge. Diameter: 184 mm. Material: Brass. Characteristics: The throne is shaped like a Moorish arch pierced with two small holes. The NUMBER 45 185 w/ttzJm FIGURE 124.?Astrolabe No. 4001. Back view. Hi front is decorated with an engraved geometric design. The back is inscribed with the maker's signature and date. The stirrup-shaped handle is decorated with rosettes at the end of the cross? bar. A geometric design decorates its top. A scale of 360? divides the rim. A stereographic projec? tion of altitudes (every 3?) and azimuths (every 6?) for latitude 66? fills the front central area. On the back of the astrolabe, the margin is divided by two altitude scales (upper portion) and two cotangent scales in finger units (lower portion). A scale of zodiacal divisions (in units of 1?, 5?, and 30?) and a concentric calendar scale (divided into units of days, groups of five, six, or eight days, 186 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 125.?Astrolabe No. 4001. Plates for latitudes 38?30' (upper left), 21?40' (upper right), 31?30'(lower left), and 30? (lower right). and months) are inscribed just inside the outer margin. The date of the vernal equinox is 14.9 March. Just inside the calendar scale there have been engraved three scales designed to assist in calendric computation. These are a solar cycle (28 divisions), a scale of days of the week corre? sponding to the solar cycle (28 divisions), and a scale of leap year designations (28 divisions). The upper quadrants contain no inscription. The two lower quadrants contain shadow squares divided into finger units. This astrolabe is fitted with four plates. Each includes a tooth at 90?, and each is inscribed with stereographically projected horizon, altitudes (ev- NUMBER 45 187 ery 3?), azimuths (every 10?), unequal hours, and prayers (on plates la, 2, and 4a: ash-shafaq, az- zawdl, az-zuhr, al-^asr, al-fajr; on plates lb, 3, and 4b: as above, but no az-zuhr and al-^asr). The projections are identified by the following inscrip? tion: Plate Inscription la li-"ard Mecca 2 1 4 0 lb li-"ard Medina al-Musharrafah 25 30 2a li-"ard Misr wa li-kull mawdi" "arduhu 30 00 2b li-"ard Mikndsa az-zitun wa li-kull mawdi" "arduhu 34 3a li-kull mawdi" "arduhu 31 30 3b li-kull mawdi" "arduhu 36 30 33 30 38 30 4a li-kull mawdi " "arduhu 4b li-kull mawdi " "arduhu Most of the inscription is Arabic, Kufic magh? ribi; but on two of the plates (numbers 3 and 4) some Naskhi has been scratched near the merid? ian. The pin is a bolt with a flattened head. The wedge is a bird with a pierced eye. Remarks: The plates included with this instru? ment are remarkably varied. Plates 1 and 2 ap? pear to consist of cast metal, and plates 3 and 4 appear to have been rolled. Plate color varies as well: number 1 is a dark golden color, 2 and 4 are the color of yellow gold, and 3 is copper colored. On all of the plates the prayer lines are jagged. A few other jagged lines designate altitudes (about every sixth) and azimuths (prime meridian) on all plates but the third. There are obvious con? struction guidelines on all plates. The rete is missing. A second astrolabe made by Muhammad ben Fattuh in A.H. 621 (CCA No. 130) is quite similar to this example. It is preserved in the Museum of the History of Science, Oxford. Another of ben Fattuh's astrolabes, in the collections of the Adler Planetarium, Chicago (CCA No. 153), is dated A.H. 634. Ben Fattuh's signature on the Smithson? ian instrument identifies him with the city of Seville in Spain. CCA Nos. 4002, 4003 (FIGURES 126, 127) Diameter: 80 mm (4002), and 103 mm (4003). Remarks: These two objects are examples of fC.^^S^'^^fJ^n^,, FIGURE 126.?Astrolabe No. 4003. Front view. pseudo-astrolabes. They each have all of the com? ponents of authentic instruments (body with throne, handle, three plates, rete, alidade, pin, and wedge), but claims to authenticity do not extend beyond form. For example, divisions of the rim number 44 in the one case (No. 4002) and 56 in the other (4003). Inscription in the front central area of each looks like a gazetteer but hardly reads like one. No notch or tooth holds the plates fixed. There are, in each case, four circular scales on the back. Only one of these is recogniz? able, consisting of pictures representing the zodia? cal signs. There is a rectangular "scale" composed of four rows and four columns inside these circles. The maker may have seen the back of an astro? labe, but he almost surely did not understand it. The lines on the plates in some cases form a floral design. In no case do they correspond to stereo? graphically projected coordinates. The retes ex? hibit amateur workmanship. No recognizable star 188 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y FIGURE 127.?Astrolabe No. 4002. Back view. NUMBER 45 189 names appear. In the one case (No. 4002) the ecliptic is divided into 19 units: in the other (No. 4003) it is divided into 26 units. The markings on each alidade have no significance. There are no sighting holes in the "sighting plates." All the inscription employs Hebrew letters, perhaps to make it less accessible to prospective buyers. It appears likely that the same anonymous maker is responsible for both objects. CCA No. 4004 (FIGURE 128) Diameter: 140 mm. Remarks: This object, like entry Nos. 4002 and 4003, exhibits the characteristics of a pseudo-as? trolabe. An irregularly divided rim; a "gazetteer," which lists place names without geographic pa? rameters; loose, indistinguishable plates; a zodia? cal scale incorporating pictures; and a symmet? rical rete without recognizable star names?all preclude the object's identification as a function? ing scientific instrument. Although the alidade of this object includes sighting plates pierced with holes, there is no altitude scale against which "sightings" can be measured. To the casual ob? server, the scales and tables on the back of the instrument look functional; but closer examina? tion reveals that with the exception of the zodia? cal scale, no useful astrological information is conveyed. The maker has used the Arabic alpha? bet and Naskhi script to decorate the object with numbers, and recognizable names of places and ?"'^i^^t>'7r. ? <- ' FIGURE 128.?Astrolabe No. 4004. Back view. zodiacal signs. The placement of the inscription, however, combines with other clues to suggest that although the maker may have seen a func? tional astrolabe, he did not understand it. Appendix I Gazetteers One faces a number of problems in attempting to transcribe cities written in the Arabic alphabet into a European alphabet. One can adopt a "generally accepted spelling" (such as that found in the Times Atlas of the World) or one can choose a phonetic representation which somehow ac? counts for etymological quirks and native shifts of sounds that have no equivalents in the Euro? pean alphabet. The city names that appear in the tables in this Appendix represent an effort to preserve the intent of the native scribe without losing any geographic information. Whenever the spelling in the Times Atlas could be considered a faithful transcription of the Arabic city name, it was adopted as the standard spelling. The list that follows Table LA. identifies deviations from the standard spelling and probable spelling er? rors. (An asterisk preceding the place name in the table indicates that it is annotated in the list.) When adoption of a Times standard would mis? represent the Arabic or when no Times standard could be identified, a phonetic transcription of the Arabic was included in the table. An effort was made to identify puzzling place names in works of prominent Arabic geographers. The task of transcribing the geographic coor? dinates associated with a place named in the gazetteer of an astrolabe presents no fewer prob? lems than the task of transcribing the place name itself. Whether written in Kufic or Naskhi, certain Arabic numbers are easily misread. For example, only the presence of a dot or the length of a stroke distinguishes the numerals 15, 35, and 55. Similar problems arise in relation to the numerals 45, 85, and 105. In general, numbers between 11 and 19 can be confused with numbers between 31 and 39 and with numbers between 51 and 59. In addition, numbers between 41 and 49 can be confused with numbers between 81 and 89 and between 101 and 109. Moreover, it is often diffi? cult to distinguish numbers between 101 and 109 from numbers between 111 and 119. The conven? tion for representing zero can easily be read as the number 45. The information included in the gazetteers of the astrolabes from the National Museum of American History is presented in three tables. Table LA. summarizes all of the geographic information appearing on the instruments. The table is annotated to report Saliba's research on citations in the Arabic geographic literature and to indicate readings that vary from the listed parameters. Wherever the information can be ascertained from the instruments, the table includes geo? graphical longitude (in degrees and minutes), latitude (in degrees and minutes), azimuth of the Qibla (in degrees and minutes), distance from Mecca (or al-masdfat measured in farsakhs), and direction of the azimuth of the Qibla with respect to the four cardinal points {the jihat). The following symbolism has been adopted: A = Name and geographic parameters appear on the instrument as listed in the table. V = Geographic parameters on the instrument vary from those listed in the table. The reader is referred to the note associated with the place name for information about the nature of the variation. R = Geographic parameters rounded to the nearest degree (for the rounded values see Table LB.). 190 NUMBER 45 191 G = Geographic parameters measured from * = Annotation in the list following the table. Greenwich Observatory (for these param- Numbers in parentheses following the CCA num- eters see Table I.C.). bers indicate the number of geographic parame- ? = Problematic reading or identification. ters included in the gazetteer (e.g., (2) = longitude - = Tentative identification. and latitude). 192 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY (t)n8? (2)6952 (e)99e2 te)88 te)z.8 (2)98 (2)58 {\)0L if)99 (t)59 (e)t9 (t)e9 (t)29 (t)l9 (t)65 (e)85 (?)A5 (?)55 (t)?e (e)25 (?)6t (?)^t ( t ) t t {?)Zf (?)0t (t)6C (t)^e ( t ) l? (t)52 (e)5i uopoaJiQ aouBisiQ qjnuiizy apnjuBq apnjiSuoq o o o CO o CO ^ CO c n -a ri x: ,< o o ^ - 1 "" o o < LO ^^ CO eg o Tf< CO o rt hm ad ab < * < o o CO o o L O CO hw az < < > > > o o l ^ CM un o ? 1 jm ire h < * < > < CO Tj< CM o o L O -a kb ar ab a < * < o CM en CO o l O l O r^ kh la t < ? 1 < o i n ^ CM - f CO L O o lla ha ba c < * < < o o ^ Tfl o o CO CO 1- la n < * < o -+< cr> CO o o r^ r~. 1- R um < * < o o CO CO o ? * CO I ^ m id < * < 34 37 39 3 A A V R V A V V L O o ID CO O CM CO CO m o l < * < i n "?' < 17 13 SW V V A V A A R A A V A A V V G V A A A o o CO CO o CO CM CO rd ab il < * < o ' f CT) CO rd R um < * < o '*' CT^ CO O O r~ r-~ e rz a l- R u < < i n ?' ^ H CO O CO T f CO e rt ta rr sk er M ai < < m i n r^ CO o CO a> CO '-S sk u Q as ab at < ? O > < 38 48 41 1 SW V A A V A A R V V A A A A V A A A G A V V o m (n CO m CO CT) CO St er ab ad < * < < CM CM r^ CM o CO o w a dh < * < < i n ?' CO CO i n ? * o r-~ in oa NUMBER 45 193 > < o o CO CO o 1 o 1 in CO 1 XJ n h a l- A bw a PQ * < o m CT^ CO o CO CO CO bu lg ha r o ? * to o ? ? ' r-^ CO Tt< CM -+? o ?"? da kh sh an a m * < o o CO T^ l o CM in CO o CO ^ CO dh ag hi s n M * O r^ -^ yD CM o CM o ^ in in o^ CM dk ub a m CQ o > < < < 2 4 2 S W V V A A A A V R V V A A V A V V G - V V A V m ? * CM ^^ in CM CO CO o o o CO gh da d V A A SW V R A A A A CD CO O CD ^ -f (D CO O O .?1 o ?' Ik h n CQ * < > < in in tD CM o CM f~~ o "^ n a ra s n CQ ?X- < 21 4 SW A V A A r^ CM (D ?' O CO o ^ o o CO 00 u n -a IM n OQ * < c/2 r~- CM (D ^^ O CO o ^ o o CO CO x: n u n -a IM n PQ o A O A A V A M S i~-- ?< ?^ CO o in x> CO o in tD CO rf ur us h rt CQ * < > < > 22 9 SW V V V V A A R V V V V V V A V V V A CT) in t^ CO o o o CO o o Tt- CO sr a < A - O A A A V A A A A V A A A M S CO ? ? ' en CO o ^^ A A A o CM r~- ?^ O CO in o '?' a pu r PQ * < to o in rf O O m CO o ^^ CT) CO CQ * < o o ^ CM o o ? A A A V A O V V A A A M S CO ^ in Tf- o in CT) CO o CO t^ CT) kh ar a 3 CQ * < < < o CO CT^ ? * O O O CT) Ig ha r 3 CQ * > < > < ^ M CO o CM o o CT) o '-' rh an pu r 3 PQ * < o CO CO tD O O CO CO o o o o ?' OT 3 CQ > < o '?1 in CM o CO CM O ^^ ib ul rt Q * < o in CM i^ o CM o ^^ 2 h ib ul w a hu n Q < o o o CO o CM to CO o CM p- 00 m a v a n d (n Q * O SW V V A V V V R V V V V V V V V G V in O CO CO O CM to CO in in CO 00 m gh an n Q * > > < o Tt- Tt- CM o Tt- CO o CM CO 00 u la ta ba d :n Q * 194 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y .^ ;te r B < > ^ CO i n i n CO CO CT) CO 0 0 C^ J i n CO CO '?' IS V Q K- < o o I ^ l>l o o o CO " ? IU la ka B en ga .i- Q K- < > < < < 1 o > > < < < < c^ > < w c/: ^ CO o CO i n ?' CO CO o o o r~- m a sh q Q * < o o CM CM O o i n CO o o CO CO n a w a r Q * < o o CM CO o o en CO o o o en ra w a h < ^ O) o -^ 0 0 CM o CO 0 0 u z a ba d UH * o ^ CO CM ?" ^ M ^^ O Tt- m CO i n Tt- Tt- i n u z ku h IM b < < ^ O) to '-' T f CM o CM f ^ CO o Tt- Tt- co w m a n o fe * O > < cr: < < ^ C/2 CT) Tt- i n ^^ o CM - M T f O o CO 0 0 n jah n O * < CT) Tt- CT) CM o CO '?' ?"- 3 rh -M an ik p rt o ?X- o < > > < > < < a; > < > > > ^ IZJ ^ Tt- co CO Tt- ^^ ?* CO o CO i n CO rf ad m an OJ o * < < m CO CO CO o CM Tt- o ^^ a z n i x : O < m CO _< i n Tt- Tt- to CO o o CD en a z ya b x ; O * < o o CT) -^ o Tf to o ?' 3 O > > < Tt- O CO CM o CM Tt- ^ M "^ Ik on da o O * < i n Tt- to CM CO CO to . - M ^^ pa m au o O > < ^ CO o CM _ c Tt- i n i n r~ CO o CO CT) CO rg an o O * < > < CT) CM to CM O o Tt- .?1 -^' 'a lio r ? O * < < < < > < o > > > < < > tti < < < w C/D en CM 0 0 ^^ o i n i n CO o CM r^ la b rt X- * > > < o o Tt- CM CO CO T i - co i n i n CM o ?"' Iv an m X- * o < < < > < < a > > < > > > < < c^ > > < < < < > ^ Tt- CM CO to CM CM o i n CO o o CO 0 0 m a da n rt X * < i n ?* en CM i n CM CM ^ *OT G rt X < o o Tj - CO Tt- CM CM r-~ SM 3 00. c a G X o < > < > < > < > < > o > > > > > > Qi > < > > ^ CM CM CO o o Tt- i n o CO Tt- CO o CM Tt- CT^ IM U X * > < [XI c/3 O O CD CM O O Tt- CO i n o r~ SU II X * > < < < < < c^ < > ^ c/3 O CO Tt- r^ O o i n CM o o CM CT) rm o z o X * o < < > < < ^ c/3 O CM to CO O CO CO CO i n en 0 0 G H-H * NUMBER 45 195 0 > < < < > > < < > < 0 > > > < > > > > > > c< < < > < > > > CO 35 4 CT^ CM O Tj< m CM CM CO O Tt- t o CO Is fa ha n * < CO o CO m t o rt Is ka nd ar i < o CO CM o o 10 9 Sh as h Ja j w a hu * < o o i n o o Ja m ku t < o o "^' o o 15 0 Su ly ? Ja zi ra t a < o o r-~ O CO CO 0 0 D an ak el i? Ja zi ra t a < o o CM o o 0 0 CO r L . a n i K aw an Ja zi ra t B; * < o o i n CM o CO 0 0 0 0 Ja zi ra t M ah ad ha j? < o o T j - o o t ^ a la n? Ja zi ra t S; < < o o CM o o t D Ji dd ah < < t o CO t o CM t o o 11 9 Ju ni pe r * < 35 4 o Tt- CM M us ha ra fa x: rt X ! u rt i4 > > > > > < > > > > > > C/3 i n CO CD t o t o CM Tt- CO o Tt- Tt- o K ab ul * < i n CM CM CO o o i n CO o Tt- Tt- o G K ab ul is t; < i n CM t o CO o CO K aju r < LO i n t o CM i n Tt- t D K al pi < m CO t D CM o i n o K an na uj # O < > < < < < < < > o < < < < < < < < < < c^ < < < < < < > c/3 32 3 CO Tt- CO o o Tt- c o O o t o 0 0 K as ha n ? > < > < < t o CO 0 0 i n o o i n Tt- o CO t o o K as hg ar ?K- < < < < < < > C^ > c/3 en o r-- o o i n CO o o CO o K as hm ir * > > > > > > < Oi > < < < < c/3 r - ^^ i n cn CM o o CO K az er un * < o o Tt- co m Tt- T f CO K er ej * O O < CO CO CM o CO Tt- CO o o CO CO ha h K er m an s O > > > < o < < < > ai > > > c/3 i n CM t o o i n en CM o CO CM CT^ K er m an * 0 o c / ) t o Tt- o T j - CM CO m i n CO CM K er na l < CO Tt- co o o 10 0 K ha bi r < o CO CM o o i n t o H ab as h K ha rm a * < > < o CM t o CM o CM 10 9 ? K he np ay * < o o CT^ CM o o 10 3 ba d K ho ha na < < < < o o CM Tt- O o 10 7 K ho ta n < > < i n i n Tt- m CO i n CO K hu jan d * < o Tt- CO O T j - CT^ K hu w ai 196 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y H te rs B a ra D-. 6 < U ( f ' ) n 8 ? (2)6952 (?)9952 (2)88 (2)^8 (2)98 (2)58 (1)0/: (1^)99 (1^)59 (?)f'9 (f')?9 (f')29 (^)19 (f')65 (?)85 ( ? ) A 5 (?)55 (^)?5 (?)25 (?)6^ {?)Lf ( t )^ f ' (?)2f' (?)0f' (^)6? {f)L? ( f ' ) l? (1^)52 (?)51 uonoaj iQ 3DUBJSIQ q j n u i i z y 9 p n ; n B ' - i apmiSuoq V B n G Pl ac e 1 < ? * CM CO CO CO CO to o ^-' M hu w an ; t^ * < o o r^ CO o ? 1 i n CO G O ?a O ez - hu ze m )l < r~- i n o ? * o ?* CM ? * o CO CM CD F= hw ar iz ] ^ * < i n CM to CO o i n to CO IU ^ o jur w a Ui O < > < > < > > < o < < 22 3 SW A A R A A _^ CO CM ^^ o CO . ? 1 CO o CO en r^ 3 t > A V V M S o CM en CM o ^^ to CO i n Tt- Tt- co L ah ija n * < O < < < < < SW A A R A V o CO CD to O O Tt- CM o CO CO CO L ah sa * < < < < < < < < < > CA3 to 0-4 0 0 Tt- o i n ^ M CO o CM en o ?' L ah or e * < o CO to C J^ CO '-' Tt- "?' L uc kn ow > < c/3 O CO CM "" CM CO CO CO CO o CM r~- a da yi n s ? < o CO CM CO o O CM Tt- x: m a hd iy y 2 * < o o r^ CO O O ^ M r~- a le tia S < i n "?' CM CM i n CM _. '-^ a n du S O < < > > < o < < < < SW A V A V V R V A A r^ ^^ to -^ o CM r~~ CO o o CM CO x : a ra gh e S # < o m r-~ CO i n Tt- o 0 0 a ra n d :? < m ? I r~- co o o Tj - r^ a rd in s o o > < > A V V A V M S o CO CM ^^ o ? * ? r^ CO o O r~ en a rw S * < o CO CM i n o CO t o CO o o o CT) XI T3 3 O Pd a rw a l- S ? < i n CM to Tj - o o to CO o CO CM CD N 'x, OT SM ;3 H a sh ha d < Tt- ^^ CM CO O O r~ CO o CO o CD a z in an s s * 1 * O < < < < < o < < < < < 00 A A A A A R A A o o o o o Tt- ^ M CM o t ^ r^ e c c a S < < < o Tt- ^ M CM o -^ r-~ t ^ n rt e c c a M u ah S 6 * O < < < o < < > < < SE A A A A V R V A o '^ r-^ CO o o i n CM o CM i n r^ e di na S * NUMBER 45 197 < o o in CM o CM m r-~ :d in a D . < 78 59 V V R V V V A A A o Tt- cn CM in CO r-~ O ilt an * o 24 20 A G A o CM Tf CO in ^^ CO 00 ha w an d rt 2 < o o CM CM o CO CM o hr va ra a 2 * 12 34 A j w a se ru 46 25 34 2 S W V A A A A A R A A A A V A V A A G A A _ CM to CO o CO CM cn ;h ap ur N is * O < o o ??' CO CO o CM 00 r^ w sh eh r o 2 > < CM in CO CM o CM CO ' ' li ph at eh Cu # < o Tt- to CM in Tt- o CM rt C rt CLI < o o ?n CO o Tt- Tt- 00 K ab u Pe nj ah ir e z - * < < o o .-M CO in m in CO ? sh aw er IU CL, a < > 75 05 44 9 SW V V A A R V V V V A V V V A V V V A A A o o CO CO o Tt- t^ o n da ha r * < > o CM to CM o in Tt- ^ G * < in Tt- co CO o ? * - M CO sr Sh ir in ^1 0^ > 54 01 SW V V A R A A V A V A A o Tt- CO CO o CM CO CT^ ye n * > < < o Tt- .?1 CO o o .?1 Tt- yr aw an * < o CO CM CO o CO to to ys ar iy a * o > < 27 34 3 5 3 S W V A A A A A V R A A A V V V V V V G A A V A V A A o o to CO o o in 00 z v in * < o CO en CM o o Tt- to Iz um & * a < > 31 14 3 3 5 S W V A V V A V A R A V V V A A A A G V V V m Tt- Tt- CO o Tt- m CO s & * < o o in Tt- O m cn m st an ti n & * < o o in CO to CM CO Tt- rt u ba h & 198 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y (f ')n8? (2)6952 (?)9952 (2)88 (2)Z.8 (2)98 (2)58 il)OL (1^ )99 (^)59 (?)f'9 (^)?9 (f')29 (^)19 (f')65 (?)85 (?)Z.5 (?)55 (^)?5 (?)25 (?)6^ (?)Lf (f')f'f' (?)2^ (?)0i7 (i^ )6? if)L? (f')l? (1^)52 (?)5l U01JD9JIQ 3DUBJSIQ q j n u i i z y apmijBq apnj iSuoq > > > > O o > o > > > > > > o^ ci G < & cr rt rt Cd Oi a: SM 3 N rt XI a x: c/3 * B in x: c/3 NUMBER 45 199 < < o o o CO i n m to CO SM 3 CL rt x: c/3 * 0 A A A A A 27 9 SW V V V V V V V R V V V V V V V V V G V V V A A o CM CO i n to CO en CM O o CO 00 N rt SM X in * 32 3 SW V A V R A V A A A A V CT) O O CM o m o Tt- o CO Tt- 00 c rt SM X c/3 * OI > 32 2 SW A A A A A A R A A V V A A A A V A A T j - CM i n CO o CO ^ M CO o CO Tt- co st ai 3 XI C/3 * < < O o CO CO o o CD O 3 "i 'in * < o CO ^ M CO O O r~- co x; in OT in B TZt ?^ (fl * < o o to CO o o tD r- S M in ]? '(n < c/3 CO ^^ CO to O CO CM CO o o t ^ CT^ G rt .^ c/3 > < O T j - tD CM to CO to CO o CO CO 0 0 x: V >. 'B n *-" 0 c/3 * < O o r^ i n o o to o 'J '? o c/3 * < o cd (^ < > >. ^ OT Ci S^ rt H- N IM X cn H H O > > < > O c/3 > cd < < H- s: bo n X! H H H 200 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y _ te rs IU B ar a 3 d 2 < O U (1 )^118? (2)6952 (?)9952 (2)88 (2)/.8 (2)98 (2)58 (l)0? (f')99 (1^ )59 (?)f'9 (f')?9 (^)29 (^)I9 (f^ )65 (?)85 (?)Z.5 (?)55 (^)?5 (?)25 (?)6f' (?)^f' (?)2^ (?)0f' (f^ )6? (f)L? (^)l? (f')52 (?)5l UOUD3JIQ 33UBJS1Q q j n u i i z y apnj i jB '^ apnj iSuoq IU e c V u a CLH > > < < < < > < < 0 SW V A A R A A CM o i n o CO Tt- co o CO CM CD G 3 * O < > > < 1 41 1 SW A V i n CO o o i n CO o o cn N OT IM 3 * > > > > < > > > < < < 6 45 1 SW V A A V o i n Tt- o o r^ CO o CO CM CD OT 3 * < o CO CO CM o l O o ^ G 'n D" < o CM CO CM o i n o o _G 'c5 ^ < > Tt- O CM o CM CM CO o CO CO 'OT ci % * o < > < > < > > > > o < < < > > > 9 35 9 SW V V V V V A R V V V V CM CO o o CM CO o o CD CO -a N rt > * < o Tt- "^ o CM Tt- r- -0 's> < < u e CO CM < -Q < * < < < < < < < < < < < < < < < < < < < < < < < < < h a CO < ^^ 3 03 H ~n c >- .< < < < < > < < < < < < < <. < < < < < < < > ^ D < a o CM < cr 3 >- >- ^ * < < < < < < < < < < < < < < < < < < < < cn a n^ CT) Ui x: CIS >- c? X "c3 cr c P < < CO ? (T) CD ^ cr c < < < < < < < < < ai > a CT) CO < 15 N < < > < > 0- . > > > < < H s CO < c XI 03 Q < * < > U 3 cr> l O < x: a Q < < Q u i n i n < X I c? C c^ S J2 Q < < h w U Q ^ > > Q >-X s CT^ CM < TJ u PH < * > > < > < > > > > > > > > > > < Pi OH ?=a. ^ < "5 O < * > > < S u a t o ? ^ < ci t/i* < < X 0 a ^ H m < cd cd E- < * < < C/D a IX) CD ^ 03 cd X < < < < 0 *-^ CO CO < cr cd c < ? ri-. 0 - . x: cd N ^ < * < < > > u a CO < cd 6 x: -a cd Id 03 ^ < ? > < < < > < < C/3 < U "SCJ- CM < '5- cd x: M < * r^. 0- . cd I-, 3 -U- >H cr " =2 -o ^ > ^ < * > < > > O (X, ?51. CM CD < X2 cd C ii < * > > < S 0 a CM CM < X5 cd c S < * < < P < >- o < x: j d "cd Id 3 S < * NUMBER 45 209 n u e d ? nt i II .A .- -J T A B 6 2 < U s in C 4l '^ Oi use ?I^9? U 5 2 96952 V6952 8952 /.952 9952 I^^I 88 Oi 99 59 ^^ 9 ?9 29 19 65 85 /.5 55 ^5 ?5 25 Gf Lf i^i' 2^ Ot 6? Z.? I? 52 51 f u J v St ar n a rr < 0 0 PQ QQ. N aq ar < * ^. % < a ? * i n < N as r < * < < h ?5a. CO (7) '^ 'n 2 < < 0 u C/D a CO ^ < Q al b < * < < < < > > < < > > > > > < > > > > > > < < 0 0 PQ a _( Thi < R am ih < * < 2 h a n ed Pi < * < < > > > > < > > > < > > > < < > > > > > > 0 o a CO i n < R id f < * < < 0 a. CO CM < R ijl < < 2 0 00. CD < 5' < < ? D a CT^ < R uk ba t < > > > < > > 0 J QQ. i n CO < Sa rf ah < # < < < l-H u a m CM < Sh am iy ya h < > > > > > > < ? U a m CM < x: 1- Sh a'^ ra a l- Sh am iy ya < ? < i-L. ru. Cd u IS C/D < * > > > > > > > > > > > > > > > > > > > > > > > < Pi l-H > a CT) CO < "cd N u Si m ak a l- A < # < < > < < > < < > < ^ < a Tt^ i n < T ay ir < * < > > > > < > < > > > < > > < < Pi h J a CO i n < 'u" ed < * < > > > > < < u a CO CM < J T Y am an iy ya < * < < 2 < U a ^ t^ Z ub an ah < < l-H Pi < CO r^ M C_ td' P3 < < Q 2 < QO. r~. < tn a l- H ut 03 PQ < < < u A ^ CO < B. >^ a s. td' m < < < J w Q tu i n i n < 3 CC Id Q < < ^ < ?>^ X J c3 cr "cd X J 03 C 03 x: Q * < < U ?o CD i n < a n a b 2 X Q < < < U a CD i n < x: 03 't? a n a b a l- D aj x: Q < J W Q w i n i n < a n a b D al fin x: Q 210 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y T l n u e i c o O 1 < 03 < H d < u u u cd u6 fe r- w Cti U 8 ? ?t9? U 5 2 96952 V6952 8952 /.952 9952 i ' t l 88 0/. 99 59 1^ 9 ?9 29 19 65 85 LQ 55 1^5 ?5 25 et Lf t t 2 t Ot 6? L? 1? 52 51 t lU CJ IU 6 cd St ar n < J W Q w i n i n < ^ a b a l- H i c OS xi Q < < < < < < > < < CL, < U ?c CTl i n < >> -3 a b a l- Ja c ed XI Q * > < h a u QQ. CD < 3 0^ 'B a b Ja nu c OS x: Q * < > < > < < < < < < > A A A A h ? o QQ. CD < i/i 3 a b Qa yt c OS Q * > < < H M U - Tt< < h w o ?=3, CO < 'B 3 C s Ja 3 a b Qa yt c cd x: Q * < h w u - rh i < cfl 3 ha na b Qa yt Sh am al i Q < < ?>.^ u >- a b a l-T ? c ed Q < < < < < Pi CQ O a i n ?* < ik ka h IX, < < > < < < < < < < < < < < > < A A A A A O w OH u CO i n < a l- Fa ra s B ed fc * < < cr) CL, a , , CO < a l- H ut 6 cd PL, < < Oi u a CM CO < n ed "rt 6 cd fc < > > < < < < < A A A A h fc U ?>- CO < ^ a yt us B cd fc # < o fc fc <3Q. 0 4 CD < < V A A A A Q >^ ; i : a CD CM < u a l- Sh uja Fa rd * < O fc fc ? = 1 . CM CO < B dd a 03 a l- M uq x: Fa rg < H fc U a CO < J - X Qa yt u; 05 x: XJ cd < < < < < < < < < > < < > < A A A A A oi 0 U ^ CD CO < X2 cd 1 a l- G hu i x:-03 C cd ? < C^ ^ < -^ o CO ^ X I OS i^ * < O fc fc t - a l- Fa ra s XJ "o3 ^ * < cn < U <=a. CM < af f ^ < PQ J a CO r f < h Ja nu b ,"J cd i4 < < < < < < < < < < < < < cn < O =0 . CM < X ! a l- K ah di cd i4 < X cu 0 a i n < 03 ^ ; a l- H aw cd I H fc * < Pi D < ?=0. ^ IS c 03 S * < Oi fc fc a i n < cr CS I H o3 S * < Oi fc fc a i n < 03 >- q T hu ra (*3 I H 03 S * NUMBER 45 211 T l 3 in ti i < fc TA B d 2 C C A 5 OT ll. "> oi n 8 ? ? t9? U 5 2 96952 V6952 8952 Z.952 9952 t t l 88 OZ. 99 59 t 9 ?9 29 19 65 85 LQ 55 t 5 ?5 25 6 t Lf t t 2 t Ot 6? L? I? 52 51 t u u c L; St ar n a m e < n-. > > > < > > > > > 2 w ^ r-~ < ia fa t a l- N ah r > < < < < < < < < < > < < > < o fc fc QQ. CM CO < ik ab a l-F ar as * < Q > X t) m 'B' Xi cn I H Ci X ! ? < < > < o-. > < < > o <3Q. ^ 1^ iq ar a l- D aja jah * < > H H O <=Q. CO ? * 1^ x : rt >->- z a m a l- Sh am al i M ir * < < fc O a ? < ^ > < < < < < < < < < < ^ < a ^ i n < r T ay ir N as ? < < < < < < < < < < < < < < < < < Pi > fc a CO i n < r W aq i" N as < < U u hr Sa ha bi N at * < > < > > < < < > > > > < > > < > > < < < < < Pi CQ U a m +^1 < yi r Fa kk ah N ay # < < 0 p- o ? * < rs '>. N ay * < < 2 0 <5i. < n n n B cn TJ * < < < D p- o -^ < T n 0^ < < < < < < < < < < < < < oi O a CM CO < id at 0 / < < < > < Pi U a CM CO < id at a l- B at iy ah u * < < < < < < < < < < < < < < > < < < < < < < SC O a CO ?* < X I n l l cr < u "ed o Qa ll * < < < < < < < < < < < < < < < < < > < 0 fc fc a o CO < TJ 0 - . < H fc O .^ ? * < B. o( * < fc U QQ. CO < tu s Ja nu bi < < < < < Pi fc fc QQ. ? * < s a l- G hu l n rt Pi < > < < < < < < < < < < < < < > < < < < < < X fc 0 a ^^ i n < s a l- H aw w a "rt Oi * < < Pi fc X a CD ? * < s a l- Ja th i n rt Pi < < 2 H a Tt< CD ^ s a l- M ut ha ll at h n a < r t . o u a CO i n < X i R id I ? < < D : t D ub b oi * < < < > < > < 2 0 "^Q. CD < "a N a R ijl * 212 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y "O 3 C on ti n 1 I.A t - H 03 < H d 2 < U rt OT tL tii q j oi n8? ?t9? U52 96952 V6952 8952 /.952 9952 t t l 88 0/. 99 59 t9 ?9 29 19 65 85 /.5 55 t5 ?5 25 6t Lf t t 2t Ot 6? L? l? 52 51 t IU CJ C IU V B cn c i-. a cn < < 2 0 X a c a 3 > "rt n a N rt 1 ? "rt 2 * < < < < 2 0 ?aa. 0^ < a c/] 3 >< "rt n rt N rt "S 2 * < 2 <; U CQ. -*! i n ^ C o! u rt rt ?) : ^ Oi 1 > < H H 0 X rt C a 3 >- Pi * < < < < < < < < < < < < > < < > 0 CQ. CD < n I H m 3 >- 1 ? ? ? < n.. fc 0 ^^ CO CD 5 "rt < n "rt _CT '_r ' rt C/D # < > r - . < < < < < < < < fc U ?^ 3 > ? rt . o -c rt C/J * c^. 0 - . o.. 5 . x; rt (^ 'n cn \ * < < 60 CT rt cn 0 - . < < < < < < < < < < < 0^ < &0 o CO < c rt a > < n " CT rt cn * 1 < 0^ < ?o o CO < .n '.2 a cn c a a < n - CT rt cn * < 0/ < ?o o CO < ~n ' 9 "rt ? CT cn \ < < < n-. < < \ > < > > > > > < > > > < < > > A A A A V < U a CO CM < ' ra Y am an iv va h rt x: cn * > < < > < < < < > < < < V A A A A A S U a i n CM < 'a e^ x : cn * < < < < Q X a CT) CM < u .ii ' 3 x: cn < < < > < > < < < < < 0 0 PQ a ^ H Tf < X i 'a a Oi c^ a cn * < Q > X a CT^ CM < Tl IM rt fc "K Xi 3 cn \ * < < < Q 2 < a < c/3 rt l l rt fc u 3 cn < < fc fc a r-~ CM < x: ed C I - rt cn <.? u rt < < Pi u 0 - . x: rt > rt CQ "K S ^ * < < P 0 - . XJ X ! 3 Q Tl rt ? < fc U 0 - . c/3 3 > ? rt 0 T3 rt * < < < < > < < < 2 0 a CM CM < n rt N rt -k T l rt ? < < 2 0 a CM CM < cn c a 3 "rt n cn N rt T: rt > < 2 0 ?>- I-. c/3 a l- Ta w za " al Y u TJ rt > * < < < < < < < < V V A A A A oi 0 a CM CM < m c ' 3 Tl rt ? NUMBER 45 213 13 IU 3 C n ti . ? C o < a < H d 2 < U U cd OT u 4 J C ^ - ' C ?H o s ; ' use et9? U 5 2 96952 V6952 8952 L9QZ 9952 t t l 88 OL 99 59 t 9 ?9 29 19 65 85 ^5 55 t 5 ?5 25 6 t Lf t t 2 t Ot 6? L? I? 52 51 t u J V a a c Si rt < < 2 0 r^ [^ I H 3 TI rt ;^ < < < < u a CO CM < 'S rt a in > < r. . < < < < < < A A A < P a CO CO < XI X2 3 Q "ed rt * < fc u n-. rt a a u rt 2 "rt CO ? < N ah r "rt 0 - . < ? P a X3 XJ 3 Q "rt X2 3 Oi Ti a cr r... < ? o 0 - . Qa yt us n-. < < Oi Q >? CM i n < c 'c c "rt o-. < Ji C rt S < 0 Q < -0 'n 2 0 NOTES TO TABLE II.A "Abur: Read al-"Abur on No. 144. ?^Ayyuq: Read al-"Ayyuq on No. 4 and No. 144. Al-Dabaran.: Read al-Dabar on No. 144; read Dabardn on Nos. 15, 58, 63, and 2571. Al-Dubb: This name, appearing on No. 2568, apparently refers to the whole constellation of Ursa Major. Al-Fard: Read Fard on Nos. 52, 53, and 54. Al-Ghul: Read Ghdl on Nos. 25, 31, 37, 39, 42, 44, 59, 61, 62, 65, 2568, 2569B, and 2571. Al-Ghumay?a: Read Ghumaysd on Nos. 2568 and 2571. Al Ghurab: Identified with y Corvis, but could refer to the entire constellation. Al-Hani al-Nahr: Kunitzsch has inhindt an-Nahr. Al-Hawwa: Probably identical with Ra^s al-Hawwd. Read Hawwd on No. 63. Al-Inaq: See Allen, page 440, for identification. Al-Jizah: Difficult reading; unidentifiable. Al-Kaff al-Jadhma ^: Read Kaff al-Jadhmd on Nos. 44 and 47. Al-Khadib: Read Khadib on Nos. 44 and 2571. Al-Maqbid Yad al-Yusra: Difficult reading; unidentifia? ble. Al-Minkab: Read Minkab on entry Nos. 4, 65, and 3811. Al-Minkab: Read Minkab on entry Nos. 2568 and 2571. Al-Musalsalah: Kunitzsch (1959) has rijl al-Musalsalah. Al-Naqar: See Allen, page 103, for identification. Al-Nasr: Difficult reading. Al-Qalb: On No. 144, this star name is in the position of Qalb al-"Aqrab. Al-Ramih: Read Rdmih on Nos. 25, 37, 39, 40, 42, 44, 59, 61, 62, 63, 64, 2566, 2567. Al-Ra^s: This star appears to be identical with Ra^s al- Muthallath. Al-Ridf: Read Ridf on Nos. 4, 37, 39, 40, 42, 44, 59, 61, 62, 64, 65, 70, 2566, 2567, 2568, and 2571. Al-Sarfah: Read Sarfah on Nos. 53, 64, 88, 2566, and 2571. Al-Sha^ra al-Shamiyyah: Read Sha''ra Shdmiyyah on Nos. 52, 55, 58, 88, and 2569A; read Sha"ra Shdmion No. 3811. Al-Shira": Unidentifiable. The phrase means the "sail" and may refer to a part of Puppis. Al-Simak al-A'^zal: Read Simak A "zal for all entries except No. 4, which reads al-Simdk al-A "zal, and Nos. 65, 70, and 2569A, which read, Simak al-A"zal. The word A"zal is repeated on entry No. 47. Al-Tayir: Read Tdytr on Nos. 44, 63, and 2569B. Al-Waqi": Read IVdqi" on Nos. 42, 44, 63, 64, 70, 2566, 2568, 2569A and B, and 2571. Al-Yamaniyyah: Read Yamdniyyah on Nos. 15, 64, 2566, and 2567. Dhanab al-^Uqab: See Allen (1899, page 61) for identifi? cation. Dhanab al-Jadiy: Read Dhanab Jadiy on No. 31. Dhanab Janub i Qaytus: Read Dhanab fanubi on No. 66. 214 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y Dhanab Qaytus: Read Dhanab al-Qaytus on Nos. 44 and 88 and Dhanab al-Qaytus 3 on No. 2569A. Dhanab Qaytus: This pointer could refer to either t or jS Ceti on Nos. 4 and 42. A pointer marked Dhanab Qaytus 12 on No. 2569A may refer to A Ceti. Dhanab Qaytus Janubi : Read Dhanab al-Qaytus Janubi on No. 55. Fam al-Faras: Read Fam Faras on No. 44. Read Fam on No. 2569B. Fam Qaytus: Read Fam al-Qaytus on Nos. 88 and 2569A. Fard al-Shuja*^: Read Fard on No. 15 and Fard Shujd" on No. 44. Janah al-Ghurab: Kea.d Jandh on entry No. 53 a.ndjandh Ghurdb on entry No. 44. Ka"b: See Kunitzsch (1959, page 180, note 2). Ka'^b al-Faras: This should probably be read as Karab. See Kunitzsch, Untersuchungen zur Sternnomenklatur der Araber (1961, No. 145). Lira^s al-Hawwa: The maker probably intended Ra^'s al- Hawwd. Mankib: See Allen (1899, page 89) for this identification. Marfiq: See Kunitzsch (1959) main entry No. 121, for this identification. See also Allen, page 331. Marfiq Thurayya: Ibid. Masafat al-Nahr: The reading on No. 144 is problematic. Nos. 31, 37, 39, 40, 42, 59, 61, and 62 have Masdfat. Minkab al-Faras: Read Minkab Faras on Nos. 39, 44, and 2571. No. 2569b has f'araj. Minkhar Shuja": See Allen (1899, page 250) for identifi? cation. Minqar al-Dajajah: Read Minkdr on Nos. 64 and 2566. The reading on No. 66 is problematic. Mirzam al-Shamaliyyah: Read Mirzam on No. 63. Muqaddam: See Allen (1899 page 231) for problematic identification. Muqaddam Qa'^idat: Read Muqaddam for No. 2567. This star is also known as Mutaqaddtm-li-l-Qattdf Nasr Tayir: Read Nasr al-Tdyir on No. 55. Nathr Sahabi: See Kunitzsch (1959, page 63, No. 10) for identification. Nayyir Fakkah: Read Nayyir al-Fakkah on Nos. 42, 47, 52, 53, 59, 61, 62, 64, 2566, 2567, and 3643. Nayyir Qa^id: Possibly a combination of Qd^id and Nayyir Fakkah. Qadam al-Jawza": Qadam is equivalent to rijl. Qa'^idat al-Batiyah: The word al-Bdtiyah is repeated on No. 47. Qalb al-^Aqrab: Read Qalb "Aqrab on No. 44. Qalb al-"Asad: Read Qalb on No. 15. Qaytus: On No. 3811, the word Qaytus appears to be followed by the word Shamali, thus linking this pointer with t Ceti. The identification is problematic for No. 70. Ra^s al-Hawwa: Read Ra" s Haw wd on No. 44; read Ra^s on No. 2569B. Ridfah: Problematic reading on No. 31. Rijl Dubb: See Kunitzsch (1959, page 75, for identifica? tion. Rijl al-Jawza^: No. 70 has Rijl on two separate pointers. One of these undoubtedly represents yS Orionis; the iden? tity of the second is problematic. No. 63 has Rijljawza" Rijl al-Jawza" al-Yumna: See Allen (page 318) for identi? fication. Rijl Yumna: Ibid. Read RijI al-Yumnd on No. 58. Rijl al-Yusra: Read RijI on No. 15; read Rijl Yusrd on No. 39. Sabiq al-"Awwal: The reading on No. 2569A is problem? atic. Sadr Qaytus: Read Sadr al-Qaytus on No. 2569A. The read? ing on No. 65 is problematic. Saif al-Shara: Doubtful reading; unidentifiable. Saq "Ayman: Reading on No. 66 is problematic. Saq "Ayman Sakib: Possibly a combination of Saq '' Ayman and Sdkib al-Md , both references to the constellation Aquarius. Saq Sakib al-Ma": The reading on No. 52 is problematic. Sha' 'ra Yamaniyyah: Read Sha"ra Yamdni on all entries except Nos. 25, 52, 53, 58, and 88. Shami: Read Shdmiyyah on Nos. 15, 53, 64, and 2566. Simak Ramih: Read Simdk al-Rdmih on Nos. 55 and 65. Suhail al-Fard: See Allen (1899, page 249) for identifica? tion. Wasat al-Batiyah: Identification is problematic. Yad Dubb: Ibid. Yad Qaytus: Ibid. Yad al-Jawza": Read YadJawzd on No. 63. Yad al-Jawza' al Yusra: Read Yad al-Jawzd" on No. 2569A. Yad Yumna: Read Yad on Nos. 4 and 15. See Kunitzsch (1959, pages 63, 64) for distinction between Yad Yumnd and Yad Yusrd. Zahr al-Dubb: Reading on No. 88 is problematic. 3 al-Na^amat: This name appears to refer to a group of stars, possibly TJ and 6 in Cetus. NUMBER 45 215 TABLE II .B.?Star names inscribed with Hebrew characters on CCA No. 2572 Star Name Al-A "zal Dabaran Fakka Ha-Ramih Lev ha-'^Agrav Lev ha-^Aryeh Me'"off ef Nofel Regel Zenev ha-Gedi Qayt us[?]* Ridf* Reference A39 A18 A45 A41 A48 A30 A54 A53 A19 A59 ?> A56 Star a YIR a T A U aCBR a BOO a SCO a LEO a AQU aLYR )8 0 R I 5 CAP CET aCYG * These two star names are engraved on the back of this "ankabdt. There is a third, as yet undeciphered word or phrase inscribed on the back of the ecliptic band of this "ankabut. TABLE II.C.?Star names inscribed with Sanskrit characters on CCA No. 4000 (two star names, including one possibly referring to a Cygni, were undecipherable from the photo? graph) Star Name Ardra Avijit Chitra Dhanuk Hasta Lubdhaka Magha Purvab Rohini Samudrapakshi Skavana Swati Star a O R I a L Y R a VIR a O P H y C O R a C M A a LEO / ?PEG a T A U L C E T a AQU a BOO 216 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y TABLE II .D.?Star names inscribed with Latin characters Star Name Agorab Al Alabor Alacrab Alchimech * Aldebaran Alfard Alfeta Alfra Algeber Algnib Algol * Algomeisa Alhaio Alhave Allot * Altaire Andromeda 3 Aquila 2 Aramech * Arcturus 1 Arturus Audicns C E q * Canis Maior 1 Canis Meridionalis * Canis Minor 1 Cap Med 2 Capra Caput Dragonis * Caput Herculis Caput Ophiuchi 3 Caput Serpen Cast * Cauda Capricor 3 Reference A36 A41 A23 A48 A39 A18 A29 A45 A62 A19 A15 A14 A25 A20 A51 A28 A54 A7 A54 A41 A41 A41 A41 A62 A23 A23 A25 A14 A20 A52 A49 A51 A51 A24 A59 Star y COR a BOO a C M A a SCO a VIR a T A U a HYD a C B O fi PEG /3 0R1 a PER ;8PER a C M I a AUR a O P H t U M A a AQI /SAND a AQI a BOO a BOO a B O O a BOO /SPEC a C M A a C M A a CMI /8PER a AUG y DRA a HER a O P H a O P H a 8 CAP 18 6 B B A A A A 20 4 V A V A CCA No. - H CM ' f CM CO O CM CM CO A A V A V A A V A A V A A V V A A 20 05 A A A V A A V 20 06 A A A A A A A A A A A 20 07 A A A A A A A A NUMBER 45 217 TABLE II .D.?Continued Star Name Cauda Ceti 3 Cauda Delphin 3 * Cauda Leonis * Cauda Signi 2 Caudaurse CoUarium Canis * Cor Leonis 1 Corona Cor Scorpii 2 Corvus * Crus Aquarii Crus Pegasi 2 Delphin Denebalgida * Dex Humerus Orio 1 Dex Lat Per 2 * Dor Ur Maio 2 Extre Cau Ur Ma 2 Fundus Vasis 4 * Gallina Gorgo * Hemerus Equi Hidra Hircus 1 Holor Idra Lanceator 1 Lancis Libre Lira 1 * Lucida Hidrae 2 Markeb Menchar Mirach Mus Pega 3 Mychar Reference Star A6 A55 A35 A56 A33 A23 A30 A45 A48 A36 A62 A55 A59 A22 A15 A33 A40 A32 A56 A14 A63 A29 A20 A56 A29 A41 A43 A53 A29 A27 A13 A58 / 3 C E T e DEL y8LEO a C Y G a UMA a C M A a LEO a C B O a SCO y C O R a AQU ySPEG e DEL a O R I a PER a U M A TJ U M A a C R A a C Y G ySPER a PEG a H Y D a AUR a C Y G a HYD a B O O a LIB a L I R a HYD p P U P a C E T /3UMA e PEG e B O O CO 00 A A A A B V A 20 4 A V V V V A A V 22 1 A V V V A A V A A V A CCA No. CM ? * CO o CM CO A A A A A A A V A A A 5002 A A A A V A A A 20 06 A A A A A A A B B 20 07 A A V A A A A A A A A A A 218 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y TABLE II .D.?Continued Star Name Naris Ceti 3 * Oculus Tauri 1 Os Ceti Palma Ophiuchi 3 Patera Pentecaitos Pec Cassiop Pegasus Perses * Pes Orion Pleiades Prima Caude Ur Ma 2 Procion Rasaden Raz Alawe * Rigil Rostrum Corvi 3 Septentri * Sini Hu Bootis 3 * Sinister Pes Orion 1 * Spica Virginis 1 Tibia Aquar Umbi Andro 3 * Ursa Maior Venter Ceti 3 Vertebrum Aquarii * Vultur Cadens 1 * Vultur Cadens Vultur Volans * Wega Yed Reference A8 A18 A13 A47 A32 A8 A5 A63 A15 A19 A16 A3 7 A25 A52 A51 A45 A42 A19 A39 A60 A7 A40 A8 A60 A53 ? A54 A53 A47 f C E T a T A U a C E T S O P H a C R A f C E T a CAS a PEG a PER / ? O R I T; T A U e U M A a CMI y DRA a O P H X O R I a C O R a C B O y BOO / ? O R I a VIR 8 AQU /3AND TJ UMA f CET 8 AQU a LYR p a AQU a LYR 5 O P H CO CO V A V V A A A A 20 4 V A A V A V 22 1 A V V A V A A A CCA No. 26 2 A V A A A A A V A A 30 4 V V 20 05 A A A A V A A A A A 20 06 A A A A A 20 07 A A A A A A A NUMBER 45 219 NOTES TO TABLE I I . D Aldebaran: Read Aldeb'an on No. 304. Algomeisa: Read Algomesa on No. 304. Altaire: Read Altair on No. 304. Arcturus 1: Read Arctur' on No. 262. Canis Maior 1: Read C. Maior on No. 204; read Caniz on No. 304. Canis Minor 1: Read Canis Min on No. 221; read Can" Min / on No. 2005. Caput Herculis: Read C. Herculis on No. 204. Cauda Capricor 3: Read Cauda Capricorni on No. 2005. Cauda Leonis: Read Cauda Q 2 on No. 2007. Read Cauda Signis in No. 204. Read Cor Q I on No. 221; read C*Leo on Cauda Signi 2: Cor Leonis 1: No. 204. Crus Aquarii: Read Crus ssr on No. 221. Dex Humerus Orio 1: Read Dex Hu Orion I on No. read Hu Orionis on No. 204. Dor Ur Maio 2: Read Dorson Urse on No. 204. Gallina: Read GalP on No. 2005. 221 Hemerus Equi: Read Hurne:Equi on No. 262; read Hu E Ma on No. 221; read H Eq on No. 186. Lucida Hidrae 2: Read Lucida Hy 2 on No. 221; read Lucida Hydre on No. 204. Oculus Tauri 1: Read Oculus on No. 262 and Oculus I on No. 221; read 0 Tauri on No. 204; read Oculus Taurus on No. 186. Pes Orion: Read Pes Sinis Orio I on No. 221. Read Pes Orionis on No. 2005. Rigil: Read Regel on No. 304. Sini Hu Bootis 3: The rete of No. 186 incorporates an unscribed anthropomorphic figure, which appears to point to this star. Sinister Pes Orion 1: Read Si: Pes Orionis on No. 262. Spica Virginis 1: Read Spica on No. 186; read Spica ^1 on No. 221 read Spica Virgini on No. 204. Ursa Maior: Read Ursa on No. 262. Vultur Cadens 1: Read V Cadens on No. 204. Vultur Cadens: Entry No. 2005 has a star pointer marked Lira in addition to this one marked Vultur Cadens. Wega: Read Vega on No. 304. Appendix III The Principle of Stereographic Projection The design of the astrolabe is based on a simple model of the universe which assumes that the sun moves on the surface of a vast celestial sphere centered on the earth. The principle of stereo? graphic projection is used to represent the appar? ent circular path of the sun (the ecliptic) on the planar surface that constitutes the astrolabe rete. Stereographic projection, as developed by Hip- parchus and applied by Claudius Ptolemy in his "Planisphaerium" (see his Opera), involves pro? jecting figures on a sphere representing the celes? tial sphere from one of its poles onto a plane parallel to its equator (see Figure 129). The im? portant characteristic of this projection is its pres- STEREOGRAPHIC PROJECTION OF T H E \ ECLIPTIC p ^ ECLIPTIC CELESTIAL EQUATOR S O U T H POLE OF T H E CELESTIAL SPHERE FIGURE 129.?The principle of stereographic projection. 220 NUMBER 45 221 ervation of circles and angles. The stereographic projection of a circle on the sphere is a circle in the plane of projection. One way of judging the skill of an astrolabist is to assess the precision with which he utilizes stereographic projection to produce the various representational lines and circles on the parts of his instrument. For example, an evaluation of the precision of an astrolabe can result from exami? nation of the ecliptic circle, which forms part of its rete. In his account of the utilization of ster? eographic projection, Ptolemy shows that the zo? diacal divisions of the ecliptic may be accurately positioned on the stereographic projection of the ecliptic by first locating the points of intersection of the celestial equator with great circles passing through the poles of the celestial sphere and the zodiacal boundary points. Points thus located on the projection of the celestial equator (the "right ascensions" of the zodiacal boundary points) con? nected to the center of the projection of the celestial equator define a radius that intersects the stereographic projection of the ecliptic at points identical with the stereographic projections of the zodiacal boundary points. Figure 130 shows the correct angular separation of zodiacal bound? ary points as located by Ptolemy's method. Emmanuel Poulle describes five alternative STEREOGRAPHIC PROJECTION OF T H E ECLIPTIC STEREOGRAPHIC PROJECTION OF T H E CELESTIAL EQUATOR FIGURE 130.?Methods of dividing the ecliptic. 222 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y methods of dividing the ecliptic band in his "La fabrication des astrolabes au Moyen Age." The least accurate of these, preferred by Hermann Contractus and many astrolabists, entails approx? imating the location of the zodiacal boundary points by considering them to be identical with the points of intersection between the stereo? graphic projection of the ecliptic and radii con? necting the center of the projection of the celestial equator with points defining 30? arcs on that projection. Figure 130 shows that this method, however lacking in its theoretical base, can result in a very close approximation of the position of the zodiacal boundaries on an astrolabe rete. Notes 1. The modifier, planispheric, limits this discussion to instruments that include in their design a plane projection of the celestial sphere. The most frequently employed means of projection is described in Appendix III. 2. Henri Michel's Traite de I'Astrolabe, published in Paris in 1947 and now available in a 1976 edition, remains the best modern contribution to this genre. W. Hartner 's "The Principle and Use of the Astrolabe,' ' published in Oriens- Occidens in 1968 is an excellent treatment of the subject in English. The recent appearance of several make-it-yourself astrolabe kits suggests the existence of an enduring interest in the instrument's structure and application. 3. Theon's work on the astrolabe is available only indi? rectly in the works of later authors. Otto Neugebauer has traced the transmission process and reconstructed the origi? nal Theon in an article on "The Early History of the Astrolabe." 4. The Arabic (first) and Latin (second) terms are given for each of the astrolabe's major components. Conventional transliteration has been used throughout. 5. In this discussion, the term "European" is used exclu? sively to refer to astrolabes inscribed with Latin alphabetic characters. Astrolabes made in Spain and inscribed with Arabic characters are identified as maghribi. 6. The history of the sea or mariner's astrolabe is ably presented by David Waters in Agrupamento de Estudos de Cartografia Antiga, Seccao de Coimbra. There is one Spanish example dating from the mid-sixteenth century in the col? lections of the National Museum of American History. 7. The seasonal (or " temporal" or "unequal") hours are those utilized by the ancient Greeks and Romans to divide each day's light or nighttime into twelve equal parts. Because the length of daylight (and nighttime) varies with the sea? sons, so too do the lengths of the seasonal hours. Equinoctial hours are of equal length on all days of the year and represent 1/24 of the time between successive noons. 8. The term Kufic refers to a calligraphic style consisting of stiff, angular, Arabic characters analogous to printed Roman characters. 9. The numbers in parentheses refer to catalog entries. 10. Neugebauer, "The Early History of the Astrolabe," page 240, quotes from Synesius (d. A.D. 412) to support the suggestion that Hipparchus invented the method of stereo? graphic projection. 11. See note 3. 12. Several editions of the compendium, Margarita Philo- sophica, are known. The earliest of these, edited by Gregor Reisch was published in 1503. 13. For example, Abu-1-Husayn '^Abd ar-Rahman ben Umar a|-$ufi (A.D. 903-986), Kitdb al-"amal bi-l-asturldb. 14. Al-Biruni's discussion of the construction and use of the astrolabe appears in his Book of Instruction in the Elements of the Art of Astrology, completed ca. A.H. 1020; especially pertinent are sections 324-346. 15. An effort has been made throughout this introductory essay to employ terminology familiar to the maker in describ? ing his work. 16. The term mashriq is used throughout this essay to mean that portion of the Muslim world now referred to as the "Middle East." 17. Hejira dates inscribed on astrolabes made in the mashriq or the maghrib have been converted to A.D. dates (given in brackets) using the method described in Augustus de Morgan's The Book of Almanacs. Before 1585, the conver? sion involves the Julian calendar; after 1585, the Gregorian calendar. 18. The "ankabut is shown opposite page 195b in Wright's translation of al-Biruni. 19. R.T. Gunther illustrates another example of Ja'^far's work in The Astrolabes of the World, page 130. 20. Henri Michel has pointed out in his Traite', (pages 135-141) that this convention makes possible the dating of astrolabes constructed before the period of calendar reform. The telling element is the date of the vernal equinox (Aries, 0?) which, in the Middle Ages, progressed backward through the month of March at the rate of about one day every 100 years. A comparison of vernal equinox days with inscribed dates on the pre-fifteenth century maghribi astrolabes in the collection leads to the following results: CCA No. 3643 144 4001 2572 Aries 0? 14 March 14.5 March 14.9 March 15 March Inscribed date [illegible] A.H. 704 A.H. 621 A.H. 483 21. These enduring elements include numerous counter- changes in the bar representing the equinoctial colure and a support for an equatorial arc flanked by Moorish arches. 22. This is the date of death of Hermann Contractus, author of what is believed to be the first European treatise on the astrolabe (see Gunther, Astrolabes). 23. The most accessible edition of Chaucer's treatise can 223 224 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y be found in Gunther's Chaucer and Messahalla. 24. The bibliography appended by Gunther to his Astro? labes of the World lists most known early works and reveals that many appeared in several editions. 25. The epoch of a table of astronomical parameters is the date for which it has been calculated. Since precession affects the values of these parameters in a predictable way, knowledge of a table's epoch enables its user to update it if necessary. 26. These are No. 304, with a vernal equinox coinciding with 11 March, and No. 186, with a vernal equinox at 10.5 March. 27. In Morley's "Description of a Planispheric Astrolabe Constructed for Shah Sultan Husain Safawi" (page 9), first published in 1856 (also reprinted by Gunther in his Astrolabes of the World), Morley observes that the designation of this single-latitude instrument is unique to Latin writers who call it "unius latitudinis aut elevationis polaris astrolabium." 28. The term maghrib is used throughout this essay to refer to Arab-occupied territory in North Africa and Spain. 29. The relevance of the cotangent scale in Arab lands is easily understood. It readily enables the user to determine the ratio between the height of a gnomon and the length of its shadow at any time of day. Such a ratio is a determinant of prayer-time in many Muslim sects. The scale would have no analogous European use. To a European maker or user the cotangent scale would appear to duplicate superfluously the scales of the shadow square. 30. Gunther describes an interesting maghribi exception to this "rule." On its rim, the 24 hours are marked by small dotted circles [Astrolabes of the World, page 290). Hartner suggests a reason for the absence of equal hour scales on Arab astrolabes when he explains that while the European way of counting equal hours from midday and midnight was known to Islamic astronomers it was never used in civil life ("Principle and Use of the Astrolabe," page 297). 31. The relevant figure (precisely, from MS Cambridge Dd. 3.53) is reproduced on page 6 of Gunther's Chaucer and Messahalla on the Astrolabe. 32. The Italian ascription is due to the evidence supplied by the several plates engraved for Italian cities: Venetie, Neapolis, Castelli, Roma. 33. See Gunther, No. 192, p. 341, and No. 247, p. 423. 34. CCA No. 2006 is not the only one that presents dating problems. In fact, the student of early scientific instruments must be alert constantly to the possibility that an object of study may have passed through a number of hands before being subjected to his scrutiny. Previous owners of scientific instruments may well be suspected of modifying an object to suit their needs. The separable parts of an astrolabe make it especially vulnerable to this type of modification. It is there? fore risky to make any assumptions about the origins of unsigned parts of an astrolabe. 35. Gibbs, Henderson, and Price, in A Computerized Check? list of Astrolabes list 19 authentic (not fake) Arab instruments and 197 authentic European instruments made between 1499 and 1600. 36. See the plates accompanying the description of astro? labe No. 180 in Gunther, The Astrolabes of the World. 37. A portion of the comparative analysis section of this essay has been devoted to elucidation of these various scales. 38. For details see the section on plates, p. 54. 39. In his Tabulae Directionum, completed in 1467 (avail? able in a 1972 edition), Regiomontanus proposed that the boundaries of the astrological houses be great circles running through the north and south points on the horizon and dividing the equator into 12 equal parts. 40. Regiomontanus mentions these two men as authors of systems of houses, which his own system is designed to replace. Alcabitus or al-Qabisi was an Iraqi astronomer who, about A.H. 950, composed an "Introduction to the Art of Astrology" in which he described houses bounded by great circles that coincided with the great circle approximations of the unequal hour lines. Regiomontanus' mention of Cam- panus of Novara (d. 1296) and his astrological houses (bounded by great circles that pass through the north and south points on the horizon and divide the prime vertical into 12 equal parts) established the latter astronomer as an authoritative astrologer. No specific astrological treatise can be attributed to him with certainty, however. 41. Har tman was born in Eggolsheim in 1489 and died in Nuremberg in 1564. 42. Relatively few astrolabes inscribed with Latin char? acters survive from this period. They include an instrument signed by Anthony Sneewins in 1661, now in the Musee des Arts et Metiers in Paris (Gunther, The Astrolabes of the World, page 402) and an instrument signed by P. Sevin in 1682, now in the Adler Planetarium in Chicago. 43. Michel has grouped the variable elements of astrolabe design into the following styles: Egyptian, Hispano-Moorish, Gothic, Renaissance, Persian, Indo-Persian, and Hindu. 44. The geographic qualifier al-Yazdi is not part of Mahdi's signature on either of these instruments. His asso? ciation with the Persian city of Yazd is documented by signatures on other instruments (Gunther, The Astrolabes of the World, page 121) and by his collaboration with Muham? mad Muqim al-Yazdi (see L.A. Mayer, Islamic Astrolabists and Their Works, pages 70, 71). 45. In discussing this second instrument, Mayer [Islamic Astrolabists and Their Works, page 71) notes that the "character of the script and of the decoration differs from all his [Mahdi's] other signatures." 46. See Michel, Traite de I'Astrolabe, page 157. 47. The signature on each of these two instruments ends with the phrase asturldbi humdyuni Ldhdri; hence, the associa? tion with Lahore. 48. The pierced throne of M u h a m m a d Muqim's astro? labe is presently covered by two plates attached with rivets. 49. The use of dashed lines in the rim scale seems espe? cially indicative of the Indo-Persian style. NUMBER 45 225 50. The signature on this instrument reads: s,ahib Qasim "Ali asturldbi Qd^ini. 51. The family was first recognized by Gaston Wiet, who reported his conclusions in 1936 [Bulletin de ITnstitut Francais d'Archeologie Orientale, Cairo, volume 36, pages 97-99). Wiet's work was extended in 1937 by Nabia Abbott and Syed Sulaiman Nadvi in articles in Islamic Culture (volume 11, pages 144-146; volume 11, pages 537-539). 52. An astrolabe signed by al-Hadad dated A.H. 975 [A.D. 1567], is reportedly preserved in the Library of Sir Salar Jung Bahadar of Hyderabad. An astrolabe bearing "Isa's signature, dated A.H. 1013 [A.D. 1604], is preserved in the instrument collection of the Adler Planetarium, Chicago. 53. The collection of Qazi Obeidu^l-Bari in Calcutta preserves one of the few remaining astrolabes signed by Qa^im Muhammad; the instrument is dated A.H. 1034 [A.D. 1624]. 54. Dated astrolabes by Muhammad Mahdi , none of them in the collection, indicate that he worked between A.H. 1059 [A.D. 1649] and A.H. 1070 [1659] (see Mayer, Islamic Astrolabists, page 70.) The signature Muhammad Amin ap? pears on two instruments dated A.H. 1086 [1675] and A.H. 1097 [1685]. The chronology undermines the possibility of a father-son relationship involving these two. The date of the work of Muhammad Amin ben Mirza Khan makes it prob? able that if such a relationship existed, he is the father. 55. Muhammad ben Ahmad's name appears on some 20 instruments made between A.D. 1715 and 1737. Two of these are preserved in the collection of instruments at the Adler Planetarium, Chicago. Only three of Hasan's instruments (all astrolabes created between A.D. 1685 and 1694) are preserved. 56. No European instruments inscribed with explicit eighteenth century dates survived to be included in Gibbs, Henderson, and Price's A Computerized Checklist of Astrolabes. 57. The study of Gingerich, King, and Saliba, "The '^Abd al-A^imma Astrolabe Forgeries" shows that dates only rarely appear on authentic '^Abd al-A^imma instruments and then only in dedications (see their table 1, page 189.) 58. "^Abd al-A^imma also decorated at least one instru? ment for Muhammad Amin's father, M u h a m m a d Tahir; Tahir's astrolabe is in the City Art Museum in St. Louis. He collaborated with M u h a m m a d Khalil on several instruments (four are preserved in the Museum of History of Science, Oxford). His name appears with that of '^Abd al-'^Ali on one astrolabe (also in the Museum of the History of Science, Oxford). These other astrolabists can be considered to be the most important members of '^Abd al-A^imma's "school." 59. "^All's name appears on 12 surviving instruments made between A.H. 1203 [1788] and A.H. 1208 [1793]. The example in the collection is numbered "the fifteenth." This practise of numbering each instrument calls to mind the similar practise of the European astrolabist, Georg Hart? mann. There is a major difference, however, in that only the body of "^All's instrument is numbered. 60. See, for example, CCA No. 87, signed by Diya^ al- Din Muhammad ben Qa^im Muhammad. 61. See Derek Price's article "An International Checklist of Astrolabes" for a description of the method used. 62. The one curious stylistic anomaly evident in this example is the rim scale. It neither incorporates dashed lines nor consists of four 90? arcs. 63. Sadiq's name, preceded by the term "?ana'^ahu," appears in a cartouche on the back of this instrument. The date, "1216," and the name ""^Abd al-Karim" appear within the shadow squares. 64. See the commentary on gazetteers, p. 26. 65. The only other known astrolabe signed by Muham? mad Akbar is preserved in the Whipple Museum, Cam? bridge. Its diameter is less than half that of the example of his work in this collection. 66. This instrument is not the only example in the collec? tion inscribed in two languages. Astrolabe No. 2572 consists of a body and plates inscribed in Arabic and a rete inscribed in Hebrew. The names of maker and owner of the Sanskrit astrolabe were suggested by David Pingree, who kindly provided a translation. 67. Translation, from Gunther, Chaucer and Messahalla on the Astrolabe, p. 5. 68. The term Naskhi refers to a calligraphic style consist? ing of rounded, cursive, Arabic characters as distinct from the angular Kufic script. 69. For a discussion of the ancient origins and use of these wind scales see A. Rehm's "Griechische Windrosen." 70. The dials are identified in Derek Price's article, "Portable Sundials in Antiquity, Including an Account of a New Example from Aphrodisias." 71. O. Neugebauer emphasizes the importance of Pto? lemy's tables in his History of Ancient Mathematical Astronomy, page 969. The earliest known edition of the tables was prepared by Theon in the fourth century A.D. The identity of the primary point of longitude in Ptolemy's geographical coordinate system remains uncertain. It could be any group of islands off the African coast, e.g., Canary or Madeira. 72. E.S. Kennedy describes a number of zijes in his "Survey of Islamic Astronomical Tables." 73. For example, access to Sedillot's published translation of Ulugh Beg's list of important cities led William H. Morley to identify this list as the source of the gazetteer inscribed on the astrolabe made for Shah Sultan Husain in A.D. 1712. Ulugh Beg's zij was written about A.D. 1440. It could well have been the standard reference for all later astrolabes. 74. Sectors containing Mecca and Medina are adjacent (in a counterclockwise direction) to the sector containing the titles of the various components of the circular table. 75. This characteristic can be considered another indi? cation of Qasim '^All's Indo-Persian affinities. His gazetteer does not contain geographic parameters relating to Mecca. 76. The sequence is not based on increasing latitude or longitude or on any progression of the other parameters in 226 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y the scale. It does, however, seem to follow well-traveled routes between cities in the Muslim world. 77. The equivalence between farsakh and degree of arc varies from geographer to geographer anywhere from 25 farsakhs to \8^ farsakhs per degree. 78. The use of the term "feet" in this context has origins in the Greek practise of measuring the length of the shadow cast by a human gnomon by setting one foot in front of another. The height of the average human gnomon was seven human "feet," leading to the practise of dividing any given gnomon into sevenths and using the term "feet" to refer to the divisions. The use of the term "fingers " to refer to the 12 divisions of a gnomon of given length has its origin in the Babylonian observation that a finger held at arm's length obscured 1/12 of a degree of the night sky. 79. O. Neugebauer provides a detailed description of this tradition in of A History of Ancient Mathematical Astronomy, pages 736-748. 80. See Hartner, "The Principle and Use of the Astro? labe," page 302, figure 850. 81. See Michel, Traite, page 90. 82. Ptolemy compares this Egyptian system of terms with a Babylonian system in the Tetrabiblos, volume 1, page 20. 83. See al-Biruni, Book of Instruction, page 263. 84. For a discussion of how each group was formed, see Ptolemy, Tetrabiblos, volume 1, page 18. 85. See Morley's "Description of a Planispheric Astro? labe," page 19. 86. Michel explains this relationship in his Traite\ pages 83, 84. 87. See Michel, Traite, page 22. 88. See Michel, Traite, pages 127, 128. 89. See de Roias, Commentarium in Astrolabium. 90. See al-Biruni, Book of Instruction, p. 197. 91. The term abdjad is used to refer to numerals formed from letters of the Arabic alphabet (e.g., " = 1, b = 2, j = 3, d = 4). 92. See Hartner, "Principle and Use," page 297. 93. See Wright's translation of al-Biruni, Book of Instruc? tion, pages 205, 206. 94. See Hartner, "Principle and Use," page 299. 95. See Ptolemy, "Planisphaerium" (in the Opera), sec? tions 8 and 9. 96. See Emmanuel Poulle, "La fabrication des astrolabes au Moyen Age," pages 117-128. 97. See the section on Hermannus Contractus in Gunther's Astrolabes, pages 406, 407. Also relevant are Reisch's Margarita Phtlosofica; Gunther 's Chaucer and Messa? halla, pages 1057, 1058; Stoffler, pages 33, 34, and 39-41; and Egnazio Danti, pages 162, 163. 98. See Poulle, "La fabrication des astrolabes au Moyen Age," pages 124, 125. 99. See Gunther, Chaucer and Messahalla on the Astrolabe, page 171. 100. See al-Biruni, Book of Instruction, page 200. 101. See Price on "Portable Sundials," pages 253-262. 102. See Gunther, Chaucer and Messahalla on the Astrolabe, page 174. 103. For a discussion of other works by "^Abd al-A^imma, see Gingerich, King, and Saliba's 1972 article on astrolabe forgeries, pages 188-198. 104. All dates are from Gibbs, Henderson, and Price, A Computerized Checklist of Astrolabes. 105. See L.A. Mayer, Islamic Astrolabists and their Work, plate 13. 106. See Stoffler, pages 33, 34, and 39-41. 107. See Mayer, Islamic Astrolabists, page 14. 108. See Mayer, Ibid., page 42. 109. See Mayer, Ibid., page 50. 110. See Gunther, Astrolabes of the World, page 302. Literature Cited Abbott, Nabia 1937. Indian Astrolabe Makers. Islamic Culture, 11 (1): 144-146. Abulfeda 1848-1883. Geographic d'Aboulfeda. Translated by M. Stanislas Guyard, 3 volumes. Paris: L'Imprimerie Nationale. [Volumes 1 and 2, 1848; volume 3, 1883.] Al-Biruni, Abu al-Raihan 1934. Book of Instruction in the Elements of the Art of Astrology. Translated by R. Ramsay Wright. London: Lu- zac. 1954-1956. Al-Qdnun al-Mas"udi Volumes 1-3. Hydera? bad, India: Da^irat al-Ma^arif. Allen, Richard Hinckley 1899. Star Names and Their Meanings. London: G.E. Stech- ert. [A 1963 corrected edition is available under the title. Star Names, Their Lore and Meaning. New York: Dover Publications.] Al-Idrisi, al-Sharif 1960. India arul the Neighboring Territories tn the Kitab nuzhat al-mushtaq fikhtiraq al-"afaq of al-Sharif al-Idrisi. Lei? den: Brill. [Translated and with commentary by S. Maqbul Ahmad.] Ali, Muhammad, translator 1934. Translation of the Holy Koran. Lahore, India: Ah- madiyya Anjuman-i-ishaat-i-Islam. Barbier de Meynard, Charles Adrien 1861. Dictionnaire geographique, historique et litteraire de la Perse et des contrees adjecentes. Paris: L'Imprimerie Nationale. [A translation of sections of Yaqutibh's Kitdb Mu"jam al-Bulddn.] Bartholomew, John , editor 1959. Southwest Asia and Russia. Volume 2 of The Times Atlas of the World. London: The Times Publishing Company, Ltd. Bartholomew, John George, editor 1893 Constable's Hand Atlas of India. Westminster: A. Con? stable and Co. Blagrave, John 1585. The Mathematical JeweL London: Walter Venge. [Reprinted in 1971. Amsterdam: Da Capo Press, Thea t rum Orbis Terrarum, Ltd.] Danti, Egnazio 1569. Trattato dell'uso e della fabbrica dell'astrolabio. Flor? ence. [Available in the Dibner Library, Smithson? ian Institution; Washington, D . C ] de Morgan, Augustus 1871. The Book of Almanacs. Second edition, London: J . Walton, de Morgan, Jacques Jean Marie 1894. Mission scientifique en Perse. Volume I of 5 Etudes Geographique. Paris: E. Leroux. de Roias, Johannes 1550. Commentariorum in Astrolabium. Lutetiae. [Available in the Dibner Library, Smithsonian Institution.] Dey, Nundo Lai 1927. The Geographical Dictionary of Ancient and Me? dieval India. London: Luzac. Forrest, Charles 1824. A Picturesque Tour along the Rivers Ganges and Jumna. London: Ackerman. Gahlot, Sukhvir Singh 1979. Historian's Calendar. Ratanada, Jodhpur: Hindi Sahitya Mandur. Gemma Frisius, Reiner 1556. Gemmae Frtsit Medici ac Mathematici de astrolabo cath- olico liber quo latissime patentis instrumenti multiplex usus explicatur. Antwerp. [Library of Congress call no. QB41.627.] Gibbs, Sharon, Janice Henderson, and Derek Price 1973. A Computerized Checklist of Astrolabes. New Haven. [Available from author.] Gingerich, Owen, David King, and George Saliba 1972. The "^Abd al-A^imma Astrolabe Forgeries. Journal for the History of Astronomy, 3:188-198. Goldstein, Bernard 1976. The Hebrew Astrolabe in the Adler Planetarium. Journal of Near Eastern Studies, 35 (4): 251-260. Gunther, R.T. 1929. Chaucer and Messahalla on the Astrolabe. Volume 5 of Early Science in Oxford. Oxford: Clarendon Press. [Printed for the Oxford Historical Society.] 1932. The Astrolabes of the World. 2 volumes. Oxford: The University Press. Hartner, W. 1968. The Principle and Use of the Astrolabe. In Oriens- Occidens, pages 287-311. Hildesheim: Georg Olms. Holloway, H. Maxon 1946. Checklist of the Samuel Verplanck Hoffman Col? lection of Astrolabes and Sundials. Annual Report of the New York Historical Society for the Year 1945, pages 62-71. New York: New York Historical Society. 227 228 SMITHSONIAN STUDIES IN HISTORY AND T E C H N O L O G Y Kennedy, E.S. 1956. Survey of Islamic Astronomical Tables. Transac? tions of the American Philosophical Society, 46(2): 121- 177. Khareghat, M.P. 1950. Memorial Volume II, Astrolabes. Bombay: Trustees of the Parsi Pumchayet Funds and Properties. [Li? brary of Congress call no. QB851 .K48.] Khwarizmi, M. 1926. Das Kitab surat al- "ard des Abu Ga"far Muhammad ibn Musa al-Khuwarizmi. Edited by H.V. Mzik. Leip? zig: Harrassowitz. Kunitzsch, P. 1959. Arabische Sternnamen in Europa. Wiesbaden: Harras? sowitz. 1961. Untersuchungen zur Sternnomenklatur der Araber. Wies? baden: Harrassowitz. Lane-Poole, Stanley 1894. The Mohammadan Dynasties. Westminster: A. Con? stable and Co. Lelewel, Joachim 1852-1857. Geographic du Moyen Age. Volumes 1-5. Brux- elles: Pilliet. Le Strange, G. 1966. The Lands of the Eastern Caliphate. London: Frank Cass. [First printed in 1905. Cambridge: Cam? bridge University Press.] Maddison, Francis 1966. Hugo Helt and the Rojas Astrolabe Projection. Agrupamento de Estudos de Cartografia Antiga, Seccao de Coimbra, number 12. Coimbra. Mayer, Leo Ary 1956. Islamic Astrolabists and Their Works. Geneva: Albert Kundig. Michel, Henri 1947. Traite de I'astrolabe. Paris: Gauthier-Villars. Meynard, C. Barbier de, see Barbier de Meynard, Charles Adrian Morgan, Augustus, de, see de Morgan Morgan, Jacques Jean Marie de, see de Morgan, Jacques Jean Morley, W.H. (see also Gunther, 1932) 1856. Description of a planispheric astrolabe constructed for Shah Sultan Husain Safawi. London: Williams and Norgate. [Reprinted in R.T. Gunther, Astro? labes of the World 1932.] Nadvi, Syed Sulaiman 1937. Indian Astrolabe Makers. Islamic Culture, ll(4):537-539. Nallino, C. 1899-1907. Al-Battani sive Albatenii Opus Astronomicum. Vol? umes 1-3. Medioliani: Insubrum, Prostat apud U. Hoepiium [Library of Congress call no. QB41 .M63, No. 40.] Neugebauer, O. 1949. The Early History of the Astrolabe. Isis, 40(121): 240-256. 1975. A History of Ancient Mathematical Astronomy. Volumes 1-3. New York: Springer-Verlag. Pingree, David 1978. History of Mathematical Astronomy in India. Vol? ume 15 in Dictionary of Scientific Biography. New York: Charles Schribner & Sons. Poulle, E. 1955. La fabrication des astrolabes au Moyen Age. Tech? niques et Civilisations, 4(4): 117-128. Price, Derek 1955. An International Checklist of Astrolabes. Archives Internationales d'Htstoire des Sciences, 32:243-263; 33:363-381. 1969. Portable Sundials in Antiquity, Including a New Example from Aphrodisias. Centaurus, 14:242-266. Ptolemy, Claudius 1907. Opera Astronomica Minora. Volume 2 in Opera, edited by J .H. Heiberg. Leipzig: Teubner. [In? cludes "Introduction to the Handy Tables"(pages 157-185) and the "Planisphaerium" (pages 225- 259).] 1964. Tetrabiblos. Edited and translated by F.E. Robbins for the Leob Classical Library. Cambridge: Har? vard University Press. Regiomontanus, Johannes 1972. Opera Collectanea. Osnabruck: Ot to Zeller. Rehm, A. 1916. Griechische Windrosen. Sitzungsberichte der Kdntglich Bayerischen Akademie der Wissenschaften (Phtloso- phisch-philologische und histonsche Klasse). Abhan- dlung 3. Reisch, Gregor 1503. Margarita Philosophica. Friburg. [Available in the Dibner Library, Smithsonian Institution.] de Roias, Johannes 1550. Commentariorum in Astrolabium. Lutetiae. [Available in the Dibner Library, Smithsonian Institution.] Saliba, George 1973. The Buffalo Astrolabe of M u h a m m a d Khalil. Al- Abhath,2&:\\-\8. Stoffler, Johannes 1512. Elucidatio Fabricae Ususque Astrolabii. Oppenheym. [Available in the Dibner Library, Smithsonian Institution.] a?-$ufi, Abu-1-Husayn 1954. Suwaru'l-Kawdkib. Hyderabad, India. Suhrab 1929. Das Kildb "agd^ib al-Akdlim as-Sab"a des Suhrab. Edited by H.V. Mzik, Leipzig: Harrassowitz. Tanner, Robert 1587. The Traveller's joy and felicitie, or a Mirror for Mathe? matics. London. [Available in the Dibner Library, Smithsonian Institution.] NUMBER 45 229 Ulugh Beg 1917. Ulugh Beg's Catalogue of Stars. Translated by E.B. Knobel. Washington: The Carnegie Institution. Waters, David 1966. The Sea- or Mariner's Astrolabe. Agrupamento de Estudos de Cartografia Antiga, Seccao de Coimbra, num? ber 15. Coimbra. Wiet, Gaston 1936. Une famille de fabricants d'Astrolabes. Bulletin de I'Institut Francais d'Archeologie Orientale (Cairo), 36:97-99. Yaqut ibn "^Abd allah al-Hamawi al-Rumi al-Bagdadi 1861-1906. Kitdb Mu"jam al-Bulddn. Volumes 1-10. Cairo. Zinner, Ernst 1931. Die Geschichte der Sternkunde. Berlin: J . Springer. 1956. Deutsche und Niederlandische Astronomische Instrumente des 1L-18. Jahrhunderts. Munich: Beck. Index "Abd al-A"imma, 18, 20, 33, 52, 55, 68, 69, 72, 74, 77, 79, 86, 104, 225 note 58, 226 note 103 "Abd al "Ali, 86, 225 note 58 "Abd al-Ghafur (ben Muhammad Sa ' id ) , 18, 19, 95, 98 "Abd al-Karim, 19, 90, 225 note 63 "Ala" al-Din, 20, 111 "Ali, 203 "Ali ben "Awad al-Mahmudi, 17, 27, 124 "Ali ben Muhammad ben Abdallah ben Faraj, 16, 171, 172, 173 "Isa ben al-Hadad, 17, 225 note 52 Abbott, Nabia, 225 note 51 Abulfeda, 201, 203, 204 Abu-1-Husayn "Abd ar-Rahman ben Umar a?-Sufi, 223 note 13 Abu Tahir al-Hasan ben Sa^id al-Jannani, 203 Ahmad ben Husain ben Baso, 13, 137, 139 Al-Biruni, Abu al-Raihan, 12, 13, 16, 22, 37, 51, 54, 56, 202, 203, 204, 205, 223 notes 14 and 18, 226 notes 83, 90, 93, and 100 Alcabitus (al-Qabisi), 16, 224 note 40 Al-Hadad, 17, 225 note 52 Al-Hasan ben Ahmad al-Battuti, 17, 18, 160, 165, 225 note 55 Al-Idrisi, al-Sharif, 201, 202, 203 Allen, Richard Hinckley, 207, 213, 214 As-Sufi, Abu-1-Husayn, 56, 57, 207 Bahadar, Sir Salar Jung , 225 note 52 Barbier de Meynard, Charles Adrien, 204, 205 Bartholomew, John George, 200, 201, 204 Blagrave, John , 16 Campanus, 16, 224 note 40 Chaucer, 13, 14, 15, 22, 223 note 23 Contractus, Hermann, 222, 223 note 22, 226 note 97 Danfrie, Philip, 16, 25, 41, 154, 156 Danti, Egnazio, 16, 153, 226 note 97 de Morgan, Augustus, 223 note 17 de Morgan, Jacques Jean Marie, 202, 204 de Roias, Johannes, 47, 226 note 89 Dey, Nundo Lai, 202 Diya" al-Din Muhammad ben Qa^im Muhammad ben Mulla "Isa ben Shaikh al-Hadad, 17, 18, 132, 134, 225 note 60 Fa(^l "Ali, 47, 84 Forrest, Charles, 201 Galandius, F ranc , 154, 156 Galois, I., 16, 25, 60, 140, 144 Gemma Frisius, Reiner, 16 Gibbs, Sharon, vi, 62, 224 note 35, 225 note 56, 226 note 104 Gingerich, Owen, 18, 225 note 57, 226 note 103 Goldstein, Bernard, 207 Grossman, N., v Gunther, R.T., vi, 16, 61, 84, 95, 104, 117, 179, 200, 223 notes 19 and 22, 224 notes 23, 24, 30, 33, 36, 42, and 44, 225 note 67, 226 notes 97, 99, 102, and 110 Hajji "Ali, 18, 77, 79 Hamid ben Mahmud al-Isfahani, 12, 62, 63 Hamza, 20, 104, 107 Hartman, Georg, 16, 25, 26, 41, 60, 146, 149, 150 Hartner, W., vi, 31, 54, 223 note 2, 224 note 30, 226 notes 80, 92, and 94 Heilbrunner, Raoul, v Henderson, Janice, 61, 224 note 35, 225 note 56, 226 note 104 Hipparchus, 12, 220, 223, note 10 Hoffman, Edgar, v Hoffman, Eugene, vi Hoffman, Margaret, v Hoffman, Samuel Verplanck, v, vi, 61, 90, 114 Ibn Ash-Shatir, 27 Ibrahim ben Sa^id, 177 Ja"far ben "Umar ben Daulatshah al-Kirmani, 12, 26, 37, 64, 65, 223 note 19 Kennedy, E. S., 225 note 72 Khalil, 17, 160 Khwarizmi, M., 202, 203, 204 King, David, 18, 225 note 57, 226 note 103 Knobel, E.B., 207 Kunitzsch, P., vi, 207, 213, 214 Lane-Poole, Stanley, 204 Langley, Samuel P., v Lelewel, Joachim, 201, 203, 204, 205 Le Strange, G., 200, 201, 202 Masha"allah, 12, 13, 16, 22, 57, 59 Mayer, Leo Ary, 160, 173, 224 notes 44 and 45, 225 note 54, 226 notes 105, 107, 108, and 109 230 NUMBER 45 231 Michel, Henri, vi, 17, 32, 223 notes 2 and 20, 224 notes 43 and 46, 226 notes 81, 86, 87, and 88 Mirza Isma"il, 72 Mirza J a h a n Bakhsh, 19, 117 Moreau, Johan , 154, 156 Morley, W.H., 38, 200, 201, 202, 203, 205, 224 note 27, 225 note 73, 226 note 85 M.P., 16, 25, 144, 146 M u h a m m a d Akbar, 20, 100, 104, 225 note 65 Muhammad Amin ben Mirza Khan, 18, 225 note 54 Muhammad Amin ben Muhammad Tahir, 18, 68 Muhammad ben Ahmad al-Baftuti, 18, 225 note 55 Muhammad ben as-Sahli, 12, 13, 174, 177 Muhammad ben Fattuh al-Khama"iri , 13, 184, 187 Muhammad Khalil, 225 note 58 Muhammad Mahdi ben Muhammad Amin al-Yazdi, 17, 18, 23, 27, 33, 65, 68, 69, 79, 82, 224 notes 44 and 45, 225 note 54 Muhammad Muqim ben "Isa ben al-Hadad, 17, 18, 27, 129, 132, 169, 224 note 48 Muhammad Sadiq, 19, 27, 89 Muhammad $affar, 124 Muhammad Tahir, 225 note 58 Muhammad Taqi ben Muhib(?) "Ali Qa^ini, 128 Nadvi, Syed Sulaiman, 225 note 51 Nagel, E., v Nallino, C , 200, 201, 203, 204 Neugebauer, O., 223 notes 3 and 10, 225 note 71, 226 note 79 Pingree, David, vi, 207 Poulle, E., 56, 221, 226 notes 96 and 98 Price, Derek J. de Solla, 61, 224 note 35, 225 notes 56, 61, 70, and 101, 226 note 104 Profatius, 45 Ptolemy, Claudius, 12, 26, 37, 55, 56, 220, 221, 226 notes 82, 84, and 95 Purmann, Markus, 146 Q a ' i m Muhammad ben "Isa ben al-Hadad, 18, 132, 225 note 53 Qasim "Ali Asturlabi Q a ' i n i , 27, 124, 129, 225 note 75 Qazi Obeidu=l-Bari, 225 note 53 Regiomontanus, Johannes, 16, 224 notes 39 and 40 Rehm, A., 225 note 69 Reisch, 16, 223 note 12, 226 note 97 Rosenwald, Lessing J., v Sadiq, 19, 27, 30, 60, 90, 225 note 63 $ahib "Ali Kabir Khan, 20, 110 Sahibuhu Maghfur al-Husayni al-Jilani, 19, 137 Saliba, George, vi, 18, 207, 225 note 57, 226 note 103 Sebokht, Severus, 12 Sedillot, 26 Sevin, P., 224 note 42 Shahjahan, 204 Shems al-Din, 205 Sneewins, Anthony, 224 note 42 Stoffler, Johannes, 16, 56, 144, 226 notes 97 and 106 Suhrab, 202, 203 Synesius, 223 note 10 Tanner, Robert, 16 Theon Alexandrinus, 1,12, 223 note 3 Ulugh Beg, 207, 225 note 73 Vitrabhadra, 20, 181 Waters, David, 223 note 6 Wiet, Gaston, 225 note 51 Wright, Thomas, 154 Yahya ben Abi Man?ur, 26 Yaqut ibn "Abd allah al-Hamawi al-Rumi al-Bagdadi, 200, 201,202, 203, 204 REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION Manuscripts intended for series publication receive substan? tive review within their originating Smithsonian museums or offices and are submitted to the Smithsonian Institution Press with Form SI-36, which must show the approval of the appropri? ate authority designated by the sponsoring organizational unit. Requests for special treatment?use of color, foldouts, case- bound covers, etc.?require, on the same form, the added approval of the sponsoring authority. Review of manuscripts and art by the Press for requirements of series format and style, completeness and clarity of copy, and arrangement of all material, as outlined below, will govern, within the judgment of the Press, acceptance or rejection of manu? scripts and art. Copy must be prepared on typewriter or word processor, double-spaced, on one side of standard white bond paper (not erasable), with IV4" margins, submitted as ribbon copy (not carbon or xerox), in loose sheets (not stapled or bound), and accompanied by original art. Minimum acceptable length is 30 pages. Front matter (preceding the text) should include: title page with only title and author and no other information; abstract page with author, title, series, etc., following the established format; table of contents with indents reflecting the hierarchy of heads in the paper; also, foreword and/or preface, if appropri? ate. First page of text should carry the title and author at the top of the page; second page should have only the author's name and professional mailing address, to be used as an unnumbered footnote on the first page of printed text. Center heads of whatever level should be typed with initial caps of major words, with extra space above and below the head, but with no other preparation (such as all caps or under? line, except for the underline necessary for generic and specific epithets). Run-in paragraph heads should use period/dashes or colons as necessary. Tabulations within text (lists of data, often in parallel columns) can be typed on the text page where they occur, but they should not contain rules or numbered table captions. Formal tables (numbered, with captions, boxheads, stubs, rules) should be submitted as carefully typed, double-spaced copy separate from the text; they will be typeset unless other? wise requested. If camera-copy use is anticipated, do not draw rules on manuscript copy. Taxonomic keys in natural history papers should use the aligned-couplet form for zoology and may use the multi-level indent form for botany. If cross referencing is required between key and text, do not include page references within the key, but number the keyed-out taxa, using the same numbers with their corresponding heads in the text. Synonymy in zoology must use the short form (taxon, author, year:page). with full reference at the end of the paper under 'Li terature Ci ted." For botany, the long form (taxon, author, abbreviated journal or book title, volume, page, year, with no reference in "Literature Cited ") is optional. Text-reference system (author, year:page used within the text, with full citation in "Literature Cited " at the end of the text) must be used in place of bibliographic footnotes in all Contri? butions Series and is strongly recommended in the Studies Series: "(Jones, 1910:122)" or " . . . J o n e s (1910:122)." If bibliographic footnotes are required, use the short form (author, brief title, page) with the full citation in the bibliography. Footnotes, when few in number, whether annotative or bib? liographic, should be typed on separate sheets and inserted immediately after the text pages on which the references occur. Extensive notes must be gathered together and placed at the end of the text in a notes section. Bibliography, depending upon use, is termed "Literature Cited," "References," or "Bibliography. " Spell out titles of books, articles, journals, and monographic series. For book and article titles use sentence-style capitalization according to the rules of the language employed (exception: capitalize all major words in English). For journal and series titles, capitalize the initial word and all subsequent words except articles, conjunc? tions, and prepositions. Transliterate languages that use a non- Roman alphabet according to the Library of Congress system. UndeHlne (for italics) titles of journals and series and titles of books that are not part of a series. Use the parentheses/colon system for volume(number):pagination: "10(2) :5-9." For align? ment and arrangement of elements, follow the format of recent publications in the series for which the manuscript is intended. Guidelines for preparing bibliography may be secured from Senes Section, SI Press. Legends for illustrations must be submitted at the end of the manuscript, with as many legends typed, double-spaced, to a page as convenient. Illustrations must be submitted as original art (not copies) accompanying, but separate from, the manuscript. Guidelines for preparing art may be secured from Series Section, SI Press. All types of illustrations (photographs, line drawings, maps, etc.) may be intermixed throughout the printed text. They should be termed Figures and should be numbered consecutively as they will appear in the monograph. If several illustrations are treated as components of a single composite figure, they should be designated by lowercase italic letters on the illustration; also, in the legend and in text references the italic letters (underlined in copy) should be used: "Figure 9b.' Illustrations that are in? tended to follow the printed text may be termed Plates, and any components should be similady lettered and referenced: "Plate 9b. " Keys to any symbols within an illustration should appear on the art rather than in the legend. Some points of style: Do not use periods after such abbre? viations as "mm, ft, USNM, NNE." Spell out numbers "one" through "n ine" in expository text, but use digits in all other cases if possible. Use of the metric system of measurement is preferable; where use of the English system is unavoidable, supply metric equivalents in parentheses. Use the decimal sys? tem for precise measurements and relationships, common frac? tions for approximations. Use day/month/year sequence for dates: " 9 Apnl 1976. " For months in tabular listings or data sections, use three-letter abbreviations with no periods: "Jan , Mar, Jun,' etc. Omit space between initials of a personal name: "J .B. Jones." Arrange and paginate sequentially every sheet of manu? script in the following order: (1) title page, (2) abstract, (3) contents, (4) foreword and/or preface, (5) text, (6) appendixes, (7) notes section, (8) glossary, (9) bibliography, (10) legends, (11) tables. Index copy may be submitted at page proof stage, but plans for an index should be indicated when manuscript is submitted.