TREESHREWS: AN ACCOUNT OF THE MAMMALIANFAMILY TUPAIID^.
By Marcus Ward Lyon, Jr.,Formerly of the Division of Mammals, United States National Museum.
INTRODUCTION.This -review of the treeshrews, constituting the mammahan familyTupaiidse, was originally contemplated in 1904 by Mr. Gerrit S.Miller, jr., curator of mammals, United States National Museum, butowing to pressure of other work he was unable to carry it out. In1910, shortly after I severed my active connections with the Divisionof Mammals, United States National Museum, Mr. Miller suggestedto me the desirability of making a study of the treeshrews. I took uphis suggestion and the present paper is the result. At that time heturned over to me some preliminary notes on the group he had madeduring a visit to European museums when he was primarily engagedin other lines of research. The increase of new material, both in theUnited States National Museum and in other museums, made itimperative that the entire field be gone over again. The collectionsin Washington were first studied, and during the summer of 1911 Ivisited most of the museums which Mr. Miller's previous workshowed contained material valuable for this revision.Specifically, the material examined consists of about 800 speci-mens, all of which are listed in the tables of measurements and dis-tributed as follows:British Museum, 355 specimens, 27 types.United States National Museum, 324 specimens, 29 types.Civic Museum of Natural History, Genoa, 37 specimens, no types.Royal Zoological Museum, BerHn, 29 specimens, 1 type.Museum of Natural History, Paris, 20 specimens, 1 type.American Museum of Natural History, New York, 14 specimens, 1type.Natural History Museum of Geneva, 3 specimens, no types.Natural History Museum of Turin, 1 specimen, no types.In addition to the specimens mentioned above, in most museums,particularly the older ones, there are a number of specimens of veryuncertain or generalized localities, which are unsuitable for systematicwork, and they are not included in the above figures.Proceedings U. S. National Museum, Vol. 45--No. 1976.80459??Proc.N.M.vol.45?13 1
2 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.In addition to examining specimens in these museums, I have alsohad for study specimens sent to Washington from the followinginstitutions:Museum, Philippine Bureau of Science, 12 specimens, 1 type.Selangor Museum, Selangor, Straits Settlements, 8 specimens, notypes.
^
Academy of Natural Sciences, Philadelphia, 6 specimens, no types.I was unfortunate in my time of visiting the Natural HistoryMuseum in Leyden. The director. Dr. F. A. Jentink, who has directcharge of the mammals, was on his vacation and I was unable toexamine the specimens of Tupaiidse in that museum. The materialcontained there as listed by Jentink ^ does not appear vitally impor-tant for a systematic review of the group, yet it contains some veryinteresting and historical specimens, which I regret not having seen.Among them are the unique type of Dendrogale murina and theonly skeleton of the genus Ptilocercus that I know of existing inmuseums and the cotypes of Tana dorsalis. Dendrogale murina isthe only species of treeshrew of which I have not seen examples.I take pleasure in here expressing my thanks to the directors of themuseums which I personally visited for giving me the privilege ofstudying the available material in their institutions, or from whichmaterial was borrowed.The importance of the explorations of Dr. W. L. Abbott in ourknowledge of the treeshrews can not be lost sight of. With theexception of less than a dozen specimens in the United StatesNational Museum the entire series of treeshrews there was collectedthrough his untiring efforts. This means that more than a thirdof the specimens of treeshrews in all the museums of America andEurope have been personally collected by Doctor Abbott. Amongthem are 29 types. Indirectly he is also responsible for the tree-shrews collected by Messrs. Kloss and Robinson on the MalayPeninsula, or adjacent islands.The text figures of the skulls and teeth of the various generawere made bj^ Mr. A. J. Engel Terzi.Measurements.?-All the measurements are in millimeters. Withthe exception of those of the head and body and of tail of skins, theyhave all been made by the writer, including those of the hind foot,which includes the claws. In most cases the measurements of thehead and body and tail were made by the collector in the flesh. Inthe tables of measurements where the head and body and tail measure-ments are followed by ?, those measurements were made by thewriter from the dried skin or mounted specimenf Head and bodyand tail measurenlents of specimens preserved in alcohol were also
1 There are, however, in the Selangor Museum, 4 types, none of which I have seen.2 Cat. Ost6ol. Manun. Mus. Hist. Nat. Pays-Bas., vol. 9, 1887, and Cat. Syst. Mamm. Mus. Hist. Nat.Pays-Bas, vol. 12, 1888.
NO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 3made by the writer. Measurements of the skull followed by ?indicate that the measurement is onl}?" approximate owing to damageto the skull. Hindfoot measurements followed by ? are also onlyapproxunatc. In the tables of measurements the specimens in theUnited States National Museum will be recognized by simple cata-logue numbers, as 104362, without qualifying initials or footnotes;those in the British Museum by the well-known separation of thesenumbers into sections by means of periods, as 99. 6. 12.3; specimensin other museums will be found designated by appropriate and self-explanatory initial letters or by footnotes,HISTORICAL ACCOUNT.The earliest published account of treeslirews is that of Ellis,^ oneof the surgeons of Captain Cook's expedition. On Tuesday orWednesday, 25th or 26th of January, 1780, *Ellis remarks: ''Oursportsmen * * * having seen only a few monkies, squirrels,and a cock and hen, the latter of which they shot. According toLinnaeus this island is their native place." The island referred to isPulo Condore, off the coast of Cochin China. The squirrels men-tioned in the account are not squirrels, but Tupaias. One of themwas evidently shot. A rough but very accurate sketch of the animalwas made by EUis and a Latin diagnosis of it written in his journal.This description of the animal was published by Gray in 1860.^Through the courtesy of the officials of the British Museum a repro-duction of a photograph of Ellis's drawing is here printed. Therecan be no doubt from Ellis's picture or description that his squirrelswere Tupaias (pi. 1).Tupaias as such were first brought to the attention of the worldby M. Diard, a French naturalist, at one time an assistant of SirThomas Stamford Raffles, in November, 1820, under the designationof Sorex glis}Six months later. May, 1821, the genus Twpaia was first proposedby Sir Thomas Stamford Raffles,* and the species ferruginea andtana described, the latter in the present paper being made the typeof a nev/ genus.Specimens of Tupaias had been seen by Europeans several yearsearlier, and one even sent to Europe. Geoffroy ^ remarks:The discovery of this remarkable group of Insectivores has been attributed to bothM. Diard and Sir Raffles. The fact is that it belongs to neither of these celebratedtravelers, but to Leschenault de la Tour, who had sent in 1807 to the Museum ofParis an individual of the species which has since been called Tupaiajavanica. Never-theless it is only since 1820 that the attention of naturalists has been called to Tupaias,and that these animals have really entered the domain of science.
1 Voyage by Capt. Cook and Capt. Gierke in sMps Resolution and Discovery, 1776, 1777, 1778, 1779, and1780, vol. 2, 1782, p. 340.
- 2 Ann. Mag. Nat. Hist., ser. 3, vol. 5, 1860, p. 71.8 Asiatic Journ. Month. Reg., vol. 10, p. 478, November, 1820.* Trans. Linn. See. London, vol. 13, p. 256, May, 1821.* Belanger, Voyage aux Indes-Orientales, Zoologie, p. 104, 1835.
PROCEEDINGS OP THE NATIONAL MUSEUM. vol. 45.Geoffrey was naturally quite unaware of the existence of Ellis'smanuscript notes and drawings. Since Diard's and Raffles's timethe group has become better and better known and its geographicrange widely extended. The most important discoveries in regardto the group since 1821 have been the announcement of the genusPtilocercus by Gray/ 1848, of the group now called Dendrogale bySchlegel and Miiller/ the discovery of treeshrews in the PhilippineIslands by Whitehead, about 1879,^ now formmg the genus Urogale,and the discovery of treeshrews in India by W. Elliot, about 1849.*DEFINITION AND RELATIONSHIPS.The Tupaiidse are diurnal insectivorous mammals characterizedby a general squuTel-like aspect, more or less arboreal habits, orbitscompletely encircled by bone, alisphenoid canal present, malar bonewith a more or less enlarged perforation, separate radius and ulna,and separate tibia and fibula, dental formula / ? C }, Pm f M. f
,
upper molars with typical W pattern. The family is composed oftwo very distinct groups for a long time regarded as genera, thetypical members of the family, Tupaia and the aberrant Ptilocercus.The old genus Tupaia has gradually been seen to be a compositegenus, and up to the present time has been divided into three sepa-rate genera: Tupaia, Dendrogale, Urogale. In the present papertwo more genera are recognized. These genera are now for the firsttime grouped to form the subfamily Tupaiinse. The single genusPtilocercus is here regarded as forming the subfamily Ptilocercinse.Eylomys of the Erinaceidse was formerly associated with the tree-shrews, but was removed in 1874 by Anderson.^TuPAIINiE.Tail bushy or close-haired tliroughoutits entire extent.Ears small and cartilaginous.Footpads of moderate development.Supraorbital foramen well developed.Foramen rotundum entirely distinctfrom sphenoidal fissure.Second upper incisor unicuspid.Upper molars with well-developed bifur-cated mesostyles.Upper molar teeth without a distinctcingulum.Lower molar teeth without a cingulum.
PTILOCERCIN.a;.Tail with terminal portion distichouslytufted, naked, and scaly basally.Ears large and membranaceous.Footpads relatively large and soft.Supraorbital foramen absent.Foramen rotundum confluent with sphe-noidal fissure.Second upper incisor with a distinctposterior cusp.Upper molars without mesostyles.A distinct cingulum encircles the uppermolar teeth.Lower molar teeth with a cingulum onouter surface.
1 Proc. Zoo!. Soc. London, 1848, p. 23.2 Verb. Nat. Gesch. Nederl. Overz. Bezitt., p. 167, 1839-44.8 Thomas, Ann. Mag. Nat. Hist., ser. 6, vol. 9, p. 250, March, 1S92.
* Waterhousc, Proc. Zool. Soc. London, 1849, p. 107.6 Trans. Zool. Soc. Londan, vol. 8, 1874, pp. 453-467.
NO. 1976. TREE8HREW8: FAMILY TUPAIID^?LYON.Genera and their types. Genus and its type.Tupaia Raffles, 1821, Tupaia/erruginea. Ptilocercus Gray, 1848, Ptilocercus lowii.Anaihana, new, Tupaia ellioti.Dendrogale Gray , 1848, Tupaia murina.Tana, new, Tupaia tana.Urogale Mearns, 1905, U. cylindrura ( =T. everetti.)The nearest relatives of the Tiipaiidse are the Macroscelididae,terrestial Insectivores of Africa. Many authors ^ place the twofamilies in a siiperfamily or subordinal gi'oup, the Menotyphla orTupaioidea as distinguished from all the other living Insectivores theLipotyphla.This grouping appears to me to be a natural one, and the differencesthat we now find between the Tupaiidee and the Macroscelididse arein large measure due to the very different modes of life of the twofamilies, the Tupaiidse bemg quite arboreal in their habits, and theMacroscelididae, terrestrial and saltatorial. The geographic distri-bution of the two families taken together show many resemblancesto the present day distribution of the Tragulidse, rhinoceroses,elephants, anthropoid apes, Cercopithecidge, and Megachiroptera,a circumstance lending some weight to their probable common origin.In spite of their great difference there is scarcely an osteologicalstructure in the Macroscelididie that does not have some counterpartm the Tupaiidse, and the opposite, the most conspicuous differencebeing the absence of the alisphenoid canal in the former and itspresence in the latter, and the complete bony orbit of the Tupaiidaeabsent in the African family. The skull of the Macroscalididse bearsmost general resemblance to that of Ptilocercus, and it is interestingto note that a supraorbital foramen is lacking in both, but is a con-spicuous feature of the Tupaiin^e. The main differential pointsbetween the two families are seen in the following table:TUPAIID^.Alisphenoid canal present.Supraorbital foramen present (except inPtilocercus)
.
Orbit completely surrounded by bone.Radius and ulna separate bones.Tibia and fibula separate bones.Metatarsals not unusually elongated.Premolars, 3 above and 3 below.Molars, 3 above and 3 below.
Macroscelidid^.Alisphenoid canal absent.Supraorbital foramen absent.Orbit not completely surrounded by bone,even postorbital processes lacking.Radius and iilna fused.Tibia and fibula fused.Metatarsals unusually elongated.Premolars, 4 above and 4 below.Molars, 2 above and usually 2 below(sometimes 3 below).
^
> Weber, Die Saugetiero, 1904, p. 377. Gregory, Bull. Amer. Mus. Nat. Hist., vol. 27, 1910, p. 268. Gill,
'
Bull. Geol. Geogr. Surv. Terr., No. 2, ser. 2, May 14, 1875, p. 20. Osborn, Age of Mammals, 1910, p. 522.? See Gregory, Bull. Amer. Mus. Nat. Hist., vol. 27, 1910, pp. 2S0-285; also Thomas (Proc. Zool. Soc.London, 1890, pp. 445, 446) who remarks on dentition of Petrodromus and the other genera.
6 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.Although the general appearance of the molarifoim teeth of theMacroscelididse is quite different from the typical W-patternedteeth of the Tupaiidse, yet it is easy to see how the teeth of the formermay have been derived from those of the latter. The teeth of theMacroscelididse show a greater departure from the more typicaltritubercular teeth of the Tupaiidse, just as the limb bones haveshown a greater departure from the normal.OSTEOLOGY.The skeleton of the Tupaiidas, as represented by the genera Tupaiaand Tana, has been rather carefully studied by Blainville/ Mivart,^Anderson,^ and Gregory.^ In the British Museum is most of theskeleton of the type of Urogale everetti, and in the Leyden Museum isa skeleton of Ptilocercus. I have not seen the latter, but Jentink^has published a few notes on it. It is the only skeleton of that genusthat I know of existmg m museums. I have not seen skeletons of thegenera Anaihana or Deridrogale, and know of none in collections.Skeletons of Tupaia are found in most of the larger museums, and inthe United States National Museum are the following:Cat. No. 124317, Tupaia glis ferruginea, Singapore.Cat. No. 174609, Tupaia demissa, Sumatra.Cat. No. 49468, Tupaia lacernata wilhinsoni, middle of Malay Peninsula.Cat. No. 111782, Tupaia nicoharica nicobarica, Great Nicobar Island.Cat. No. 154593, Tupaia javanica, western Java.Cat. No. 174611, Tana tana tana, Sumatra.The observations on the skeleton which follow are based upon anexamination of these skeletons of the genera Tupaia and Tana andskulls of the other genera. I have also made free use of the observa-tions of Mivart, Anderson, and Gregory.Slcull.?The skull of the genus Tupaia is characterized by its rathergeneralized structure; it is widest just posterior to the middle, andtapers toward either extremity both laterally and supero-inferiorly,the tapering being much more pronounced anteriorly, especially so inthe genera Tana and Urogale; posteriorly the skull is gently roundedoff. The brain case is relatively large and inflated and widest at thezygomatic roots. The orbit is completely surrounded by bone, islarge, directed mainly laterally but at the same time slightly inclinedupward and forward. Posterior to the orbit is a temporal fossa ofmoderate size. The temporal ridges are rather prominent and dis-tinct except for a short distance in front of the lambdoid crest, wherethey unite to form a short sagittal crest. In Ptilocercus the temporal
1 Ost&g. Mamm. Insect., 1840, pp. 31-35.2 Journ. Anat. Physiol., vol. 1, 18G7, pp. 292-295, and vol. 2, 1868, pp. 145-146.3 Zool. Res. West. Yunnan, 1879, pp. 108-123.
* Orders of Mammals, Bull. Amer. Mus. Nat. Hist., vol. 27, 1910, pp. 269-280.? Notes Leyden Museum, vol. 7, 1885, p. 7.
NO. 1976. TREESHREW8: FAMILY TUPAIIDJ3?LYON. 7
ridges remain separated and in TJrogdle they unite early to form a muchmore conspicuous sagittal crest than they do in TujMia. The lamb-doid crest is well marked and gently arched. It begins fauitly oneither side near the external auditory meatus and becomes welldeveloped along the upper border of the supraoccipital. The palateis long but neither specially wide nor specially narrow. In front arewell marked anterior palatine foramina; posteriorly the palate isslightly concave, and ends in a slightly thickened ridge, and a verysmall blunt median spine. The most anterior part of the posterioredge is about on a line with the posterior edge of the last molars. Inthe posterior half of the palate in the genera Twpaia and Tana areusually irregular vacuities. The other genera, Urogale, Anathana,Dendrogale, and Ptilocercus, are usually without defects of ossifica-tion in the palate. The external pterygoid foss?e are large, short, andwide, formed by the well marked, pointed, and slightly directedinward pterygoid bones, and the pterygoid plate, rather short andtriangular, of the alisphenoid. The choansB are rather wide, andnarrower between the pterygoids than anteriorly. The bullae are ofmoderate size and formed of the endotympanic. The outer edge ofthe bulla is produced outward so as to cover up or enclose the tym-panic ring or ectotympanic. The small foramen ovale is almostcovered over by the antero-external edge of the bulla. The glenoidfossa is rather wide and shallow and limited in front and behind byshort and inconspicuous anterior and posterior glenoid processes.Only the minutest trace of a paroccipital process is present. Thealisphenoid is pierced by an alisphenoid canal. The foramen magnumis directed downward and backward.The external opening of the infraorbital canal is situated abovethe second premolar. In Ptilocercus the canal is much shorter andits external opening is over the last premolar. The internal openingof the canal lies shortly inside the orbit. The lachrymal canal hasits opening in a distinct notch except in Ptilocercus and is rathermore outside of the orbit than inside of it. Except in Ptilocercusthere is a conspicuous supraorbital foramen at the upper outer angleof the orbit, continuous with a groove under the edge of the roof ofthe orbit. Except in Ptilocercus the optic foramen is separatedfrom the sphenoid fissure by a narrow spicule of bone, and the foramenrotundum lies at the base of the external pterygoid plate. InPtilocercus the optic foramen is separated from the sphenoid fissureby a broad bridge of bone and the foramen rotundum is blended withthe sphenoid fissure. The foramen ovale is situated almost underthe antero-outer edge of the bulla except in Ptilocercus where theopening of the foramen is plainly visible in front of the bulla. InTupaia and' Tana the malar is pierced by a large fenestra, in theother genera by a small foramen. The external auditory meatus is
8 PROCEEDINGS OF THE NATIONAL MV8EVM. VOL. 45.situated under the posterior root of the zygomatic process of thesquamosal, and is moderately large. Just within its orifice may beseen the tympanic ring.The mandible shows no points of special interest. In comparisonwith the Macroscelididse the coronoid process is well developed andstands about as high above the condyle, as the condyle does abovethe angular process. In the Macroscelididse the condyle is drawnupward as high as the coronoid. Ptilocercus has a relatively widerand larger coronoid than the other genera of the Tupaiidse. A wellmarked mental foramen is found under the first or second premolars,
/lOchUfrnQ/ Jar.
i^emexilli
-Jin/erfiarie/af
JnftrflCFig. 1.?Diagram showing the forms and relationships of theindividual bones of the skull of tupaia as determined byexamination of young indmduals.but in Ptilocercus the foramen is usually not single and situatedslightly more posteriorly.The relative position and shape of the more important bones of theskull as shown by examination of immature specimens may be seenin figure 1.Auditory ossicles.?"In the Tupaiidse the malleus assumes to anextreme degree ths neckless and nonlaminated type common in mostCebidas and some lemurs, as well as in Sciurus; but the manubriumis rather of the form prevailing in the above-named primates than
NO. 1&76. TREE8HREWS: FAMILY TUPAIID^?LYON. 9that seen in the squirrels; and so in every respect is the incus. Allthe ossicula differ from the varied forms in other families of theinsectivora, except that the incus somewhat approaches in type thatof Galeopithecus. Taken as a whole, the ear bones of Tupaia arehigher in type than in any other family belonging to this order."^Vertebral column?Five skeletons in the United States NationalMuseum have the vertebral formulae:Cat. No. 124317, Twpaia glis fenuginea, C, 7; D., 13; L., 6; S., 3; C, 24.Cat. No. 49468, Twpaia lacernata vAlhinsGni, C, 7; D., 13; L., 6; S., 3; C, 27.Cat. No. 174609, Tupaia demissa, C, 7; D., 13; L., 6; S., 3; C, ?
.
Cat. No. 111782, Tupaia nicobarica, C, 7; D., 13; L., 6; S., 3; C, 28.Cat. No. 174611, Tana tana, C, 7; D., 13; L., 6; S., 3; C, 22.Certain authors give the lumbar series as 5, 6, or 7.^ Blainville ^in the text of Osteography of Mammals says that the number oflumbar vertebrae is seven, but on studying the first plate I can countonly six lumbars. By the same author ^ the sacral vertebrae aresaid to be two, and the caudal 22-23, figures which are confirmedby the plate.The vertebral formula of Ptilocercus is given by Jentink* as D. 14,L. 5, S. 3, Cd. 31.The individual vertebra3 are well developed, and with well-definedprocesses, that is relatively better developed and with better developedprocesses than in the case of the Macroscelididge. The first sixcervical vertebree are pierced by vertebral foramina; the atlas islarge and relatively heavy, and so is the axis which bears a conspicu-ous posteriorly directed dorsal spine. On all the other cervical ver-tebrae dorsal spines are absent or represented by minute projections.As is usual, the under and lateral surface of the sixth cervical is welldeveloped with a prominent costal or pleurapophysial plate. Thedorsal vertebrae show no noteworth}^ peculiarities. The dorsal proc-ess on them is rather low and directed backward till the ninth isreached, which has the dorsal process directed upward; the dorsalprocesses on the three remaming dorsal vertebree are low and directedforward. The lumbar vertebrae taken as a whole have well-developedanteriorly directed dorsal processes, and still better developed, wideanteriorly directed transverse processes, the processes being smalleron the anterior vertebrae of the series. In Alacroscelididae, repre-sented by a skeleton ElepTiantulus roseti, the transverse lumbarprocesses are shorter and wider. The sacrum m Tupaia and Tana iscomposed of three firmly fused vertebrae, that is, no large foraminaexist between the transverse processes of the first and second, andsecond and third sacrals, as is the case in Macroscelididas. The first
1 Doran, Trans. Linn. Soc. London, ser. 2, 1S79, vol. 1, p. 444.> See Gregory, Bull. Amer. Mus. Nat. Hist., vol. 27, 1910, p. 275.s Ost6ographie des Mamniiferes, vol. 1, p. 33, pi. 3, lower figure.
* Notes, Leyden Museum, vol. 7, 1885, p. 38.
10 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.and second sacrals in Twpaia are practically devoid of dorsal spines,but the thii'd has a conspicuous one. All three have distinct spinesin Macroscelididse, and that of the first is very large and conspicuous.The sacrum is attached to the ilia by means of the transverse proc-esses of the first sacral vertebra in Twpaia, by those of the first andsecond in Macroscelididse. The first five caudal vertebra m Twpaiahave a neural canal and well-developed transverse processes; inMacroscelididse it is only the first three.Rihs and sternum.?There are 13 pairs of ribs in Tupaia, of whichthe first 8 are attached along their ventral border directly to thesternum, the seventh and eighth reaching the sternum together,where the xiphoid segment of the sternum is attached to the fifthmesosternal segment. The ventral ends of the last two pairs of ribsare entirely free, while the ventral ends of the ninth, tenth, andeleventh ribs are attached to the costal cartilages of one another andto that of the eighth rib.The sternum consists of a large well developed manubrium, theanterior extremity of which is largely expanded, as is usual in mammalswith a well developed clavicle, and relatively better developed thanit is in Macroscelididse. The manubrium is followed by five narrowmesosternal segments, and these in turn by the narrow xiphisternumending in a rather distmct fiat rounded piece of cartilage. In Macro-scelididse there are only four distinct and very wide mesosternal seg-ments, and the posterior extremity of the xiphisternum is forked.Shoulder girdle.?^The clavicle is well developed in Tupaia, articu-lating at one extremity with manubrium of the sternum and at theother with the acromion process of the scapula. It appears to berelatively as well developed in Macroscelididse.The scapula presents no special peculiarities; its shape may beseen from an examination of plate 2. It possesses a flat wideacromion process and a short blunt metacromion. The coracoidprocess is well developed. The spine of the scapula is relativelymuch longer in Tupaiidse than in JMacroscelididse, and conversely theacromial process relatively shorter.Pelvis.?^All three bones of the pelvis are well developed inTupaiidse,the ilia are large and flaring and relatively wide, the tuberosity of theischium well developed, and the symphysis pubis very long, and theobturator foramen large and oval. In Macroscelidid^ the ilia aremuch narrower, the symphysis relatively shorter, and the obturatorforamen more oblique and elongated.Fore limh.?The humerus is long and weU developed and shows nospecial peculiarities. The deltoid ridge is prominent and beginsslightly above the middle of the shaft. A distinct supracondylarforamen is above the internal condyle relatively smaller in Tupaiidae
NO. 1976. TREE8HREWS: FAMILY TUPAIIDJE?LYON. 11than it is in the Macroscelididas. The radius and ulna are welldeveloped and present as distinct bones, and they and the humerusare siibequal in length.In Macroscelididae the ulna is intimately fused with the radius andappears but little more than a process at the upper end of the radius.The radius-ulna is much longer than the humerus.The carpus is composed of a scapho-lunar, cuneiform and pisiform(well developed) proximally, the usual trapezium, trapezoid, magnum,and unciform distally and a distinct os centrale between the two rows.Hind limb.?The femur is only slightly shorter than the tibia, inMacroscelididse much shorter. The bone has a well marked head andneck and promment greater, lesser, and third trochanters.The tibia is well developed, with a conspicuous ridge in front. Thefibula is slender, and perfectly distinct from the tibia, except at thetwo extremities, where the two bones are in contact, the lower end ofthe fibula having a distinct surface for articulation with the astra-gulus. In Macroscelididas the fibula becomes fused with the tibiaslightly above the middle of the bone, and the anterior spine of thetibia is much more pronounced than in Tupaiidse.The calcaneum is laterally compressed and narrow and relativelyshort posteriorly, broad and rather flattened anteriorly. Its poste-rior portion is relatively much shorter than m the Macroscelididae.The trochlear surface of the astragulus is relatively wider and muchshallower in Tupaiidse than in Macroscelididse. The remainingbones of the tarsus, cuboid, navicular, and the three cuneiform bonesdo not show any special peculiarities in Tupaiidse. In Macrosceli-didse they are all much elongated.The metatarsals are without special peculiarities. The second,third, fourth, and fifth are all essentially subequal, but among them-selves in order of length they stand third, fourth, second, fifth. Thefirst metatarsal is distinctly shorter than the others, but is by nomeans a short bone like the first metatarsal of Macroscelididse, inwhich family the entire first digit is shorter than the second, third,fourth, and fifth metatarsals,- which are laterally compressed andmuch elongated. In Tupaiidse at the base of the fifth metatarsal is alarge unciform process lacking in Macroscelididse.There does not seem to be cny essential differences between theskeletons of Twpaia, Tana, and Vrogale. Urogale has relativelyheavier and thicker bones than has Tana, especially seen in those oflegs and feet, and has a higher and much better developed spine onthe axis than has either Tana or Twpaia.It is not probable that the skeletons of Dendrogale and Anathanadiffer essentially from those of Twpaia or Tana. An examination ofa skeleton of Ptilocercus, however, would probably reveal differencesfrom the other genera, in correlation with the pronounced cranial and
12 PROCBEDTNOS OF THE NATIONAL MUSEUM. vol.45.dental characters. Jentink ^ gives the vertebral formula D. 14, L. 5,S. 3, Cd. 31. He further says: ''The ribs are peculiarly broad. Thecla-sdcle is Well developed ; the bones of the forearms and of the hindlegs are separate." TEETH.The dental formula of the Tupaiidse is /. ?, C. |^, Pm |, M. f.The teeth are typically insectivorous and nearly represent the fullpossible Eutherian dentition; one upper incisor is lacking, and oneupper and one lower premolar. As to which of the theoretical teethare lacking does not seem to me to be a matter of great importance.My own view of the formula is 7. y^ , C. jP- z^W2^ , M. -r^-g ? Thereason for considering that the third upper mcisor is wanting andnot the first is that the third lower incisor is a vanishing tooth andhas almost disappeared in Urogale. This opinion is the same as thatof Gregory.^ He thinks that in Ptilocercus i^ is lacking and isrepresented by a minute alveolus. The only reason for assumingP \ are wanting is that at present the most anterior premolar is thesmallest of the series, and it seems not unreasonable to assume thatat one time it may have had standing in front of it a still smallertooth; furthermore, all of the premolars are preceded by milk teethwhich is usually not the case with the first premolars.The canines in the family are mteresting in that they do not havethe form and functions of true canines, but are almost indistinguish-able from the premolars. It might with some degree of proprietybe said that the ideal number of premolars is present in the familyand that the canines are lacking, especially since the canine in Ptilo-cercus and occasionally in the other genera is tWo-rooted, not a charac-ter of canine teeth, and since the canine is situated considerablyposterior to the premaxillo-maxillary suture, rather than in or almostin that suture, as in most other mammals. The only exception tothis is in Urogale where the lower canine has the form and functionof a true canine, and although the second upper incisor functions asa canine, yet the canine itself is more caniniform and less premolari-form than is the case in the other genera. The canmes are alwayspreceded by milk canines. ?For the shape and arrangement of cusps, etc., of the teeth figureson pages 33, 121, 128, 135, 155, and 161 should be consulted. Differ-ences that are useful for purposes of classification will be discussedunder each genus.Eruption of teeth in Tupaiinse.?The manner and order of eruptionof teeth in the subfamily Tupaiinae shows nothing not commonly seen
1 Notes Leyden Museum, vol. 7, 1885, p. 38.2 Bull. Amer. Mus. Nat. Hist., vol. 27, 1910, p. 271.
NO. me. TReESHRE^VS: FAMILY TVPAHD^-LYOH. 13in other mammalian groups. The youngest skulls that I haveexamined have always shown the complete milk dentition but some-times the last milk premolar ha, not been entnely m place. Theadult dentition is
:
71 /2_(7i_p2 p3 pi ji/i j/2 3pThe youngest dentition seen is:DP DP ^ DC DP' DP'
-DT^m^DTTDcrDPTDPlThe next teeth to appear are g^, followed by M,, then M\ thenM M' M M\ All the permanent molars are in place before themilk teeth 'are shed. The first of those permanent teeth havingpredecessors to appear in the majority of cases is ^, the upper andlower seemmg to come in about the same time, jr appear at nearly thesame time as ^, sometimes just before. ^^ are the last of the per-manent premolars to come in, appearing just after p^^. The
canines appear at or about the same time as ^^or just a little later,
arriving with y. Of the incisors the lower appear slightly beforethe upper. /, appears about the same time as P, and before P,andlus't afte;/3''7. comes m just after 7. and 73. P comes m a terall the permanent upper premolars and camnes are m place and isfollowed by I\ which is thus the last of the permanent teeth to be mplace. By the time the last permanent tooth is in place the molarsalmost always show slight traces of wear. , -u a aThe teeth of Tupaiid^ in whole or part have been described andfigured by Horsfield/ Cuvier,^ Huschke,^ Owen,^ Giebel,^ and Gregory.VISCERAL ANATOMY.The visceral anatomy of Tupaiidse has been examined by A. H.Garrod in 1875/ based primarily on an exammation oi Tupamhelangeri which had just died after living in the S^^^ens of the Lon-don Zoological Society nearly two years, supplemented by an exami-nation of Tupaia splendidula and Tana tana from specimens preserved
1 zool Res. Java, unnumbered plate ( Tupaia, Tana), 1824.i Dents des Mammiferes, 1825, p. GO ( Tupaia, Tana).? Isis vol. 20, 1827, pp. 758-759, pi- 10.? Odontoeraphy vol. 1, p. 419; vol. 2, pi. Ill, fig. 3, 1840-1845 ( Tana).4 g pny, p , 15_18, copied from Horsfield and Owen.:r*?rM;fNLf. His!:: vi s m., pp. ^i ^n. n.. n.^^ t.,.,.. p.,?^).
? Proc. Zool. Soc. London, 1879, pp. 301-305, figs. 1-3, brain.
14 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.in alcohol. Nothing unusual or of strilving importance was foundexcept that the two specimens of Tupaia each possessed a distinctcecum, while the specimen of Tana did not.Cecum.?The presence or absence of a cecum in certain msectivoreshas been made use of in the superfamUy classification of Tupaia.Thus Peters ^ grouped the insectivores into those with and mthout acecum, in the former group being Galeopterus, Tupaia, and Macrosce-lides. Gill ^ removed Galeopterus as a distinct suborder, and groupedthe Tupaiidse and Macroscelididse, as the Tupaioidea mainly on thepresence of a ''large" cecum. Haeckel is said to have proposed thetermsMenotyphla for theTupaiidse and Macroscelididse and Lipot}q3hlafor all the other insectivora, terms which have been retained byWeber.^ Garrod's * dissected specimens of Tupaia helangeri andT. splendidula both showed cecums one-half to three-fourths of aninch in length; Tana tana showed no cecum. Chapman '^ statesthat a cecum is wanting in a, specimen oi " Tupaia ferruginea" fromBorneo, as well as in an example of T. picta.All the specunens of Tupaia that I have examined possess a smallbut distinct cecum. Unfortunately, I have not exammed the intesti-nal tract of Tana tana and am so unable to confirm Garrod'sobservations. The specimens in the United States National Museumthat I have examined, with length of cecum, are: mm.112660, Tupaia glis ferruginea 13105013, Tupaia glis ferruginea 13124083, Tupaia helangeri 12123989, Tupaia lacernata lacernata 10124698, Tupaia discolor 11121893, Tupaia chrysogaster 8. 5111783, Tupaia nicobarica 8144306, Tupaia siaca 8It is not to my mind a ''large cecum," and can scarcely have anydefinite function, bemg almost as relatively small as the humanvermiform appendix.The Indian genus Anathana is said by Anderson ? to possess a
"long and narrow" cecum 1.17 inches in length, that is about 30 mm.It would not appear that the presence or absence of a cecum is agood character for determining larger groups. The majority of ourspecimens are so preserved as not to show the soft parts, and theorgan being vestigial appears to be absent at times, though as a ruleit is present in the majority of the species of Tupaia.
1 Abh. kon. Akad. Wiss. Berlin, 1863, p. 20.2 Synopsis of insectivorous mammals, Bull. Geol. Geogr. Surv. Terr., No. 2, ser. 2, May 14, 1873'.3 Die Saugetiere, 1904, p. 377.
* Proc. Zool. Soc. London, 1879, pp. 301-305.6 Proc. Acad. Nat. Sci. Phila., vol. 56, 1904, p. 148.
? Zool. Res. West. Yunnan, 1879, p. 126.
NO. 1976. TBEESHREW8: FAMILY TUPAIID^?LTON. 15GEOGRAPHIC DISTRIBUTION.Tlie Tupaiidse as a whole range from India on the west to andincluding Mindanao of the Philippine Islands on the east, and fromsouthern China on the north southward to and including Java andthe chain of islands off the southwest coast of Sumatra. They arenot found eastward of Java, nor on the Celebes, Formosa, Ceylon, orthe Andaman Islands so far as known. I know of no specimens orrecords of the Tupaiidse on the island of Bali, off the east end of Javaand just west of Wallace's Line. It would not be surprising to findthem on Bali when the fauna of that island becomes better known.They are found on practically all the smaller islands of the MalayanAi'chipelago, within the limits just mentioned, and more frequentlythan not develop geographic races or species on them.Zoogeographically the distribution of the Tupaiidas coincidesalmost perfectly with what is termed the Oriental Region or Realmof Wallace and most zoogeographers, and serves perhaps better thanany other family of mammals to define that region. The only areasin this region where they do not occur so far as known are the islandsof Ceylon, Formosa, the members of the Philippine Islands, north ofMndanao, and the Andaman Islands.^ By Wallace, Ceylon is in-cluded in a separate subregion of the Oriental Region.No one genus of the family has a range coextensive with the rangeof the family.The genus Anathana occupies an area almost coextensive withWallace's Indian subregion, but so far as our records of specimensshow, does not extend quite so far to the north, or with the Sclaters'Indian subregion excepting Ceylon.The weU-marked genus Urogale is confined to Mindanao of thePhilippine Islands. This group of islands has not been made a sub-division of the Oriental Region, but the Philippine mammals for themost part are so different from their relatives of the rest of the OrientalRegion that it would seem advisable to have them constitute a dis-tinct subregion of the Oriental . Urogale is thus one of its characteris-tic genera.The genus Tupaia has the widest geographic distribution of any ofthe genera in the familj^, and if we recognize the Philippine Islandsas a distinct subregion, it is characteristic of Wallace's Indo-Chineseand Indo-Malayan subregion, or of the Sclaters' Burmo-Chineseand Malayan subregions. The northern of these two subregions ischaracterized by but a single species group, the helangeri-chinensis;while the southern, the Indo-Malayan or the Malayan subregion ischaracterized by several well-marked species groups. Of the islandsin this subregion Borneo is inhabited by the greatest number of
1 The absence on the Andamans of treeshrews is rather interesting, as they occur on Preparis Island tothe north, and on the Nicobars, or at least the southern islands of the Nicobars to the south.
16 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.distinct types, among them Tupaias of the discolor, picta, montana,gracilis style. Three of these occur on islands to the westward ofBorneo; on Banka, discolor and gracilis types, on Billiton, gracilistype, on Karimata, gracilis and montana types, thus showing theaffinities of these islands with Borneo, and not with Sumatra. Ofthe two species found on Java, a related form of one of them occurson some of the small islands of the chain off the southern coast ofSumatra, and of the other apparently the same form occurs on oneisland of the same chain, and in the mountainous region of thesouthern coast of Sumatra. The intimate relationship of the MalayPeninsula with the island of Sumatra is shown by the occurrence onboth of Tupaia glis ferruginea and T. minor malaccana. Only oneweU marked group, minor, occurs on both Borneo, Sumatra, and theMalay Peninsula. It is not found on Java. On Palawan theCalamianes, and Cuyos Islands, all politically part of the Philippines,but geographically part of Borneo, is found a rather distinct speciesgroup without decided affinities to Bornean forms. On the Nico-bars occurs one of the most distinct species in the genus, withoutany apparent relationship to other members of the genus. So faras known no Tupaias are found on the Andaman Islands.Dendrogale, with two distinct species groups is found on Borneo,with one of the groups occurring also in French Indo-China. Thisdistribution is so peculiar and not paralleled so far as I laiow byother forms of mammals, that it seems almost certain that the genuswill be found elsewhere in the Indo-Chinese and Indo-Malay sub-regions. Of the form occurring on Borneo and the Asiatic mainlandnot a dozen examples are in existence in collections, showing that itis a particularly rare animal. See page 131.The genus Tana parallels that of the Orangs in its distribution, beingconfined to Borneo and Sumatra, and some of the adjacent islands.It contains two well-marked species groups, the smaller of which isknown only from Borneo.To my mind Urogale and Tana are derived from the same stockform; but Urogale on Mindanao being more restricted in area andmore remote from the source of origin, probably Borneo, has becomethe more highly differentiated of the two. Urogale must havereached Mindanao from Borneo by way of the Sulu Archipelago.At present there are no records of treeshrews from the islands of theSulu Archipelago, but in view of the occurrence of Urogale on Minda-nao it seems likely that treeshrews occur on them and they ought tohe of a genus or genera the same as or similar to Urogale and Tanxi.The genus Ptilocercus with a single species group is found in Bor-neo, Sumatra, and southern Malay Peninsula and some of theadjacent islands, and parallels the distribution of Tupaia minor.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 17The following natural divisions or areas of the Oriental, based uponthe genera and species found in the family Tupaiidse, may be recog-nized. They are not of coordinate importance, nor are all of themmutually exclusive. Those divided up by water seem to be suffi-ciently distinct, however, to indicate that at some not very remoteperiod connections of some sort may have existed between them.Indian (excluding Ceylon), genus Anathana, and absence of other genera.Pliilippine (Mindanao only), genus Urogale, and absence of other genera.Indo-Chinese, the belangcri-chincnsis group of the genus Tupaia, absence of othergenera and species. (The distribution of Dendrogale is so irregular that I have disre-garded it.)Nicobaran, the well-marked species Tupaia nicobarica, absence of other generaand species.Palawan-Calamine, a fairly well-marked species group of the genus Tupaia, absenceof other species and genera.Bornean, a well-marked species group in each of the genera Dendrogale and Tana,and by four well-marked species groups of the genus Tupaia.Belonging to this subdivision but without all the characteristics are Banka, Billiton,and the Natuna and Karimata Islands.Sumatran-Peninsular, glis group of Tupaia and T. minor malaccana.Java-Borussan (apparently including high mountain region of southern Sumatra),two species groups of Tupaia, absence of other genera and species.Sumatra-Bornean, genus Tana.Sumatra-Bomeo-Peninsular, genus Ptilocerus and the minor group of Tupaia.So few Tupaias are known from the Rhio-Linga Archipelago thatlittle can be said regarding its affinities. It has both Peninsular andBornean elements. The rather isolated Tambelan Islands have asingle species of genus Tana, evidently of Bornean origin, and theisolated Anamba Islands, inhabited only by members of the splen-didula group of Tupaia, also appear to be Bornean in their relations.GEOGRAPHIC INDEX.The names of the coimtries and islands are arranged geographically and notalphabetically.India, south of the Ganges: Anathana ellioti, p. 122; Anathana wroughtoni, p. 123;Anathana pallida, p. 124.India, north of the Ganges: Tupaia chinensis, p. 63.Burma: Tupaia chinensis, p. 63; Tupaia belangeri, p. 59.Tenasserim: Tupaia belangeri, p. 59.China: Tupaia chinensis, p. 63.Hainan: Tupaia modesta, p. 69.Siam (upper): Tupaia chinensis, p. 63.Siam (lower): Tupaia belangeri, p. 59; Tupaia lacernafa wilhinsoni, p. 52.Anam and Cochin China: Tupaia concolor, p. 68; Dendrogale frenata, p. 128.Pulo Condore: Tupaia dissimilis, p. 67.Malay Peninsula: Tupaia belangeri, p. 59; Tupaia laccrnata ivilkinsoni, p. 52;Tupaia glis ferruginea, p. 41; Tupaia minor malaccana, p. 114; Ptilocercuslowiiconti-nentis, p. 165.80459??Proc.N.M.vol.45?13 2
18 PR0CEEDIN?^8 OF THE NATIONAL MUSEUM. vol. 45.Islands adjacent to Malay Peninsula:Mergui Archipelago: Tupaia belangeri, p. 59.Lankawi: Tupaia lacernata lacernata, p. 53.Terutau: Tupaia lacernata lacernata, p. 53.Butang Islands: Tupaia lacernata raviana, p. 54.Penang Island: Tupaia gli.s glis, p. 45.Perhentian Island: Tupaia lacernata obscura, p. 55.Redang Island: Tupaia lacernata longicauda, p. 56. *Tioman Island: Tupaia glis sordida, p. 48.Pemangil Island: Tupaia glis pemangilis, p. 48.Aor Island: Tupaia glis pulonis, p. 47.Singapore Island: Tupaia glis ferrnginea, p. 41Batam Island: Tupaia glis batamana, p. 46.Bintang Island: Tupaia castanea, p. 90.Sumatra: Tupaia glis ferruginea, p. 41; Tupaia demissa, p. 58; Tupaia siaca, p. 91;Tupaia minor malaccana, p. 114; Tupaia javanica, p. 106; Tana tana tana, p. 139.Ptilocercus lowii continentis, p. 165.Rhio-Linga Archipelago (between Sumatra and Malay Peninsula):Batam Island: Tupaia glis batamana, p. 46.Bintang Island: Tupaia castanea, p. 90.Linga Island: Tupaia minor malaccana, p. 114; Tana lingx, p. 145.Sinkep Island: Tupaia phaeura, p. 49; Tupaia minor sincipis, p. 115.Borus?an Islands, along southern coast of Sumatra:Banjak Islands: Tupaia tephrura, p. 50; Tana tana tuancus, p. 145.Nias Island: Tupaia javanica, p. 106.Batu Islands:Pinie: Ptilocercus lowii continentis, p. 165.Tana Bala: Tana cervicalis cervicalis, p. 147.Tana Massa: Tana cervicalis masae, p. 148.Sipora: Tupaia chrysogaster , p. 71.Pagi Islands: Tupaia chrysogaster , p. 71.Java: Tupaia javanica, p. 106; Tupaia hypochrysa, p. 70.Borneo: Tupaia longipcs longipes, p. 76; Tupaia longipes salatana, p. 77; Tupaiamontanamontana, p. 94; Tupaia montana baluensis, p. 95; Tupaia picta, p. 96; Tupaiagracilis gracilis, p. 117; Tupaia minor minor, p. 110; Tupaia splendidula, p. 83; Dendro-gale murina, p. 129; Dendrogale melanura melanura, p. 132; Dendrogale melanura balu-ensis, ^.1Z2; Tana tana tana,]). 139: Tana tana utara, p. HI; Tana tana besara, Tp. 141;Tana chrysura, p. 149; Tana paitana, p. 150; Tana dorsalis, p. 152; Ptilocercus lowiilowii, p. 164.Islands faunistically related to Borneo
:
Laut off southeast corner: Tupaia minor minor, p. 110.Karimata Island, off west coast: Tupaia carimatse, p. 98; Tupaia gracilis edarata,p. 118.Banguey: Tupaia minor minor, p. 110; Tana paitana, p. 150.Palawan: Tupaia palawanensis, p. 78.Balabac: Tupaia palawanensis, p. 78.Culion: Tupaia mollendorffi, p. 81.Cuyo: Tupaia cuyonis, p. 82.Natuna Islands:Sirhassen: Tana tana sirhassensis, p. 142; Ptilocercus lowii lowii, p. 164.Bunguran: Tupaia natunse, p. 85.Laut: Tupaia splendidula, p. 83.Banka: Tupaia discolor, p. 73; Tupaia gracilis injlata, p. 118.Billiton: Tupaia gracilis injlata, p. 118.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LTON. 19
Islands not clearly related to large land masses:Tambelan Islands: Tana tana hunom, p. 144.Anamba Islands:Siantan Island: Tupaia chrysomalla, p. 88.Jimaja Island: Tupaia anambx, p. 89.Riabu Island: Tupaia riahus, p. 88.Philippine Islands:Palawan: Tupaia palawanensis, p. 78.Balabac: Tupaia palawanensis, p. 78.Culion: Tupaia mollendorffi., p. 81.Cuyo: Tupaia cuyonis, p. 82.Miadanao: Urogale everetti, p. 157.On the maps showing the distribution of the various members ofthe family Tupaiidse I have endeavored, as far as possible, to indi-cate the localities mentioned in the text or the tables of measure-ments and lists of specimens. In a few cases, however, I have beenunable to find some of the localities. Most of the maps are some-what diagrammatic, but that on page 143 is carefully made andshows virtually most of the localities whence specimens of treeshrewshave been obtained. In cases where a form is known, but froma single locality the figures indicating the distribution have beenlimited to the area around that point, where known, from two orthree rather separated localities, the distribution figures have beenextended to cover the intervening area, the assumption being thatthe animal will be found there; when known from several scatteredareas, or a large land mass, or part of one, the distribution figureshave been liberally applied around the whole area. Future explora-tions will undoubtedly show much wider ranges for many of theforms shown on the maps. On the whole, I have been rather con-servative in indicating the distributions, leaving it to the reader toimagine a more extended range. Thus, we know that Ptilocercus isfound in the Deli-Langkat region, Sumatra, and it has been indicatedat that locality only, on the map. There can be but little doubt,however, that it is found elsewhere in Sumatra, but owing to lackof records I have not so indicated it,FOOD.Judging by the typically insectivorous nature of the cheek-teethin Tupaiidse, the diet of these animals must be largely insects. Manyobservers say they naturally eat fruit as weU. Ptilocercus havingteeth sHghtly less insectivorous than the Tupaiinae, may perhapshave a more varied diet. However, it is a very rare animal anddirect observations on living specimens still rarer. The only ones I
20 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.
recall are those of Schneider/ who had a pak alive for some hours.The only food he offered them was bananas, which they did not eat.Of Tupaia Cantor ^ says: "The natural food is mixed insectivorousand frugivorous. In confinement, indi%dduals may be fed exclu-sively on either, though preference is evinced for insects; and eggs,fish, and earthworms are equally rehshed."Of the Indian Tupaias Anderson^ says: "One stomach was fullof the imperfectly digested remams of a small yellow ladybird witha sprinkling of the elytra of small beetles. Tliere were also smallmasses of a jeUy-Hke substance with very fine fibers."Hardwicke * in an introduction to Diard and Duvaucel's accountof Tupaia glis, says: "A Hving one was brought to Bengal by amedical gentleman some months ago; it runs about the house tame>but will not allow itself to be caught for close inspection; thoughat liberty to run out of doors whenever it hkes, it shews no dispo-sition to leave its quarters, and evinces some attachment to thefamily; for whenever strangers enter the house, it shews disquietudeby a chattering like noise. It occasions no trouble in feeding, forit is always on the search after insects, and its favorite food seemsto be flies, crickets, grasshoppers, and cockroaches."Jerdon^ writes of Tupaia chinensis at Darjeehng: "It frequentsthe zone from 3,000 to 6,000 feet, and was said, by the natives, tokUl small birds, mice, &c."Robinson and Kloss," speaking of Tupaia glis ferruginea, say:"The diet is very mixed, consisting of ants and other insects, fruits,seeds, and buds." HABITS.Cantor ^ writes on the habits of Tupaia: ^^The young of this verynumerous species (T. ferruginea) in hilly jungle, is easily tamed,and becomes famihar with its feeder, though toward strangers itretains its original mistrust, which in mature age is scarcely reclaim-able. In a state of nature it lives singly or in pairs, fiercely attackingintruders of its own species. When several are confined together,they fight each other, or jointly attack and destroy the weakest. Ashort pecuHar tremulous whisthng sound, often heard by calls andanswers, in the Malayan jungle, marks their pleasurable emotions,as, for instance, on the appearance of food, while the contrary isexpressed by shrill protracted cries. Their disposition is very rest-less, and their great agihty enables them to perform the most extra-ordinary bounds in all directions, in which exercise they spend the
1 Zool. Jahrb., vol. 23, 1905, p. 84, pi. 1.
' Joiirn. Asiat. Soc. Bengal, vol. 15, 1846, p. 189.9 Zool. Res. West. Yunnan, 1S79, p, 120.< Asiat. Soc. Bengal, vol. 14, 1822, p. 471.6 Mammals of India, 18G7, p. C5.
? In Tbomaa and Wrougliton, Journ. Fed. Malay States Mus., vol. 4, No. 1, December, 1909, p. 112.
NO. 1976. TREESHREWS: FAMILY TUPAIIDJE?LYON. 21day, till night sends them to sleep in then* rudely constructed lairsin the liighest branches of trees. At times they will sit on theirhaunches, holding their food between the forelegs, and after feeding,they smooth the head and face with both forepaws, and Hck thelips and palms. They are also fond of water, both to drink and tobathe in."Raffles says of Tupaiaferruginea : ^ "These animals are as tame andsprightly as squirrels. The tame one above mentioned was sufferedto go about at perfect liberty, ranged m freedom over the whole house,and never failed to present himself on the breakfast and dmner table,where he partook of fruit and mUk." It is also described beingdiurnal and arboreal. Tana is mentioned as bemg "always found onor near the ground."Mr. C. BodenlGoss^ thinks Tupaias are less arboreal than generallyaccredited. He says: "Of the numerous species of Tupaia whichI have collected personally T. longicauda with T. nicobarica, Zelebor,and its subspecies, T. (N.) surda Miller, alone are truly arboreal inhabit. As a rule the so-called ' treeshrews ' are seen and trapped onthe ground, where they live and feed, or, at most, climb occasionallyinto low bushes; in them the tail is shorter than the head and bodylength. The above-named animals, which are met with in high treesand have the habits of squirrels, all possess a taU that is considerablylonger than the length of head and body."The collector of Tupaia chinensis, reg. '^os. 97.11.2.10, 97.11.2.11,97.11.2.12, and 97.11.2.13, British Museum, says: "The four weretaken from one nest in a hollow bamboo," one of the few observationson their nests that I know of.Of Tupaia glis ferruginea, Robinson and Kloss ^ remark: "Thepopular name of ' treeshrew ' for these animals is hardly descriptive
? of their habits, as, in the majority of species, at any rate, it is quiteexceptional to see one anywhere than on the ground, among theroots of trees or on low bushes. The jungle near Changi, Smgapore,was an exceedingly good trappmg ground, and out of 70 or 80 trapsset every night hardly one was found unsprung or without an occu-pant next morning. Six or seven of these shrews were usually thuscaptured and many more were shot every day. The nest is found inholes, often m fallen timber."Regarding the food and habits of Tupaia glisferruginea, as observedon the Malay Peninsula and Singapore, Mr. H. N. Ridley^ writes:"The common species is very destructive in gardens, as it is almostif not entirely frugivorous. It bites holes in the chocolate pods to
1 Trans. Linn. Soc. London, voL 13, 1822, p. 257, May, 1S2L2 Journ. Fed. Malay States Mus., voL 4, p. 191, October, 191L8 In Thomas and Wroughton, Journ. Fed. Malay States Mus., vol. 4, No. 1, p. Ill, December, 1909,
* Natural Science, vol. 6, 1895, p. 28.
22 PROCEEDINGS OF TEE NATIONAL MUSEUM. vol. 45.eat the pith which mcloses the seeds, strewing the latter all over theground, and even digs up the seeds planted in flower boxes. * * *Twpaia ferruginea is more terrestrial in its habits than a squirrel.When alarmed it darts up a tree, but never very high, and turningits head downward utters a series of little scolding grunts, whichsound like some one talking at a considerable distance. When astone is thrown near it, it usually immediately jumps to the ground.It is evidently as yet only half accustomed to an arboreal life. Inconfinement it is very nervous, dashing about the cage whenapproached, and it never lives long in captivity." ^BREEDING.Tupaias are evidently able to bear young at practically all timesof the year. An examination of the collector's remarks on the labelsof specimens shows tliat 3 individuals were pregnant m January, 1in February, 1 m March, 3 in June, 1 in July, 2 in September, and1 in October. Specimens showing distinct signs of nursing or aboutwhich the collector remarks "milk in mammae" are distributedthrough the year as foUows: February, 2; March, 1; April, 1; May,1; June, 1; August, 1; September, 2; November, 1. December isthe only month without a record of pregnancy or breeding. Whilethe number of records is too small to justify any generalizationsit would appear that the beginning and the middle of the calendaryear are the periods of greatest productivity. (See table, p. 23.)NUMBER OP YOUNG.The number of offspring produced at one time by Tupaias probablyvaries with the species and directly with the number of mammaecommon to that species. Two is apparently the usual number, butit is sometimes one {Twpaia nicoharica surda, Cat. No. 111785), oras many as four (T. chinensisB. M. 97.11.2.10-13). (See table, p. 23.)Cantor 2 says: "The female usually produces one young; she hasfour mammae, the anterior pair of which is situated on the lowerlateral part of the chest, the posterior on the side of the abdomen,"Robinson and Kloss ^ note that two young are produced at a birthin Twpaia glis ferruginea. MAMM.^.The number of mammae in Tupaia varies from one pair in certainspecies to three pairs in others. The number is of some impor-tance as a character for certain species or groups of species. Wherethe mammae are six they have been designated by Mr. Oldfield
? But see specimen dissected by Garrod, living for two years, in London Zoological Society, page 13.? Journ. Asiat. Soc. Bengal, vol. 15, 1846.? In Thomas and Wroughton, Joum. Fed. Malay States Mus., vol. 4, No. 1, p. 112 December, 1909.
NO. 1976. TBEE8HREWS: FAMILY TUPAIID^?LYON. 23Thomas ^ as postaxillary, lateral, and preinguinal. When there areonly jiwo pairs of mammae, the preinguinal pau' seems to have disap-peared, and when only one pair is present it would appear to corre-spond with the lateral pair. The number seems to be very constant.The only exceptions to constancy that I have observed are : Twpaiapemangilis, Cat. No. 112499, U.S.N.M., where the mammae are 2-3 = 5,in a group where 2-2 = 4 is normal; T. chinensis, No. 26841, Amer.Mus. Nat. Hist., with 4 mammae instead of 6. Here the postaxillarypair is wanting; both of the remaining pairs are more posteriorlyplaced than usual, so that the preinguinal pair is really inguinal andthe lateral pair almost preinguinal. An alcoholic specimen of Twpaiahelangeri in Genoa from Mount Mooleyit, Tenasserim, with only twopairs of mammae, belongs to a group that normally has three pairs.
Dates of pregnant, and of nursing Tupaias.
Cat.No.
24 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.the color of the pelage in the two sexes is apparently the same.Change of pelage proceeds from the head downward in a more orless definite line across the body. The dates of specimens whichshow a distinct changing of the pelage are: February 25, June 4, 7,29, 30, July 1 (2 specimens), 26, September 4 (2 specimens), 18, 20,21, 23, October 27, November 5, 12, 14, 21, 26, December 4, 19, 22.The changes accordmgly take place mainly during the last half ofthe calendar year. There is only one specimen undergoing a pelagechange during the first five months. The dates of pelage changes arethus not so uniformly distributed throughout the year as are thosefor the production of young. Marked seasonal changes m color donot appear to exist. There is very little difference in color betweenthe new and the old pelages. Where two pelages exist in the sameanimal, the newer of them is of course brighter and fresher in color,but, strange to say, nearly always lighter. The lighter color is notdue to an excess of light-colored rings on the hairs which may later beworn off, but to a real difference m color or shade between the lightrings. One of the most striking examples of this difference in color isseen in the tails of the specimens of Twpaia injiata from BUliton.Cat. No. 124985 of that series has a tail with every appearance of anold pelage, and the fight areas of the hau's are ochraceous. Cat. No.124947 of the same series has the light areas of the hairs light buffyin the distal two-thirds, which is in old pelage, and almost whitishin the proximal third, which is distinctly new.Number of specimens showing changing pelage and evidences of breeding.
NO. 1976. TBEE8HREW8: FAMILY TUPAIID^?LYON. 25I have seen no specimens of Tupaia that suggest anything likemelanism, so that the condition probably occurs very rarely, if at all.FAMILY OR SUBFAMILY NAMES OF TREESHREWS.Cladobatse Fitzinger (Sitz. Akad. Wiss. Math. Nat. Wien, vol. 60,1869, pt. 1, 263). Genera: Hylomys, Cladohates, Dendrogale, Ptilo-cercus.Cladobatida Haeckel, 1895 (Syst. Phylog. Wirbelth., 1895, p. 593).Cladohates or Tupaja only genus mentioned.Cladobatidina Bonaparte, 1838 (Syn. Vert. Syst. in Nuov. Ann. Sci.Nat. Bologna, vol. 2, 1838, p. 111). Used as a division of the Sori-cida8, no genera mentioned. ^Cladobatina Bonaparte, 1845 (Cat Met. Mamm. Europ., 1845, p. 5).Used as a subfamily of the Soricidse, no genera mentioned.Glisoricina Pomel (Arch. Sci. Phys. Nat. Bibl. Univ. Geneve, vol.9, p. 250, November, 1848).Glisoricinse Murray (Geog. Hist. Mamm., 1866, p. 319). UnderPomeFs arrangement of the insectivora he uses Glisoricinge as a groupname embracing Hylogale, Sorexglis, and OxygompMus (fossil).Tupaiadse Bell. (Todd's Cyclop. Anat. Physiol., vol. 2, 1839, p.994). Work not seen.Tupaiidse Mivart., 1868 (Journ. Anat. Physiol., vol. 2, 1868, p.145). Comprising Tupaia, Ptilocercus, Hylomys.Tupaina Gray, 1825 (Thomson's Ann. Philos., vol. 26, November,1825, p. 339). Used as a probable fifth group of the Talipidae, withthe one genus Tupaia.Tupajidse Schlosser (Beitr. Palaort. Oester.-Ungarns, vol. 6, 1887,pp. 91, 114). Work not seen.Tupayse Peters, 1863 (Abhandl. kon. Akad Wissensch. Berlin,1863, p. 20). As a group name for Cladohates, Ptilocercus, and Hylo-gale (probably intended for Hylomys)
.
Tupayidse Gill (Arrang. Fam. Mamm. Smiths. Misc. Coll., No. 230,p. 19, 1872). Another spelling of Tupaiidce Mivart.NAMES THAT HAVE BEEN PROPOSED FOR TREESHREWS USED IN AGENERIC SENSE, OR NAMES OF ESTABLISHED GENERA TO WHICHTREESHREWS HAVE BEEN ERRONEOUSLY ASSIGNED.Chladobates.?A typograpliical error for Cladohates. It occurs inSchinz, Naturgeschichte und Abbildungen der Siiugethiere, 1824, p.87. Included species javanicus, vittatus, tana, ferrugineus.Cladobates.?This term was first proposed by Cuvier (Dents desMammiferes, p. 60, pi. 17) in 1-825. It is used synonymously withSorex-glis, and was probably considered by him to be more euphoni-ous or more descriptive of the animals. It contamed three species
?
tana, ferruginea, javanica. It is thus seen to be a pure synonym
26 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.
of Tupaia. The term was adopted by many subsequent writers asthe generic designation of the treeshrews, among them Lesson,1827; Wagner, 1841; Giebel, 1855; Peters, 1864; Zelebor, 1869;Fitzinger, 1870.Dendrogale.?Proposed by Gray (Proc. ZooL Soc. London, 1848,pi. 16, p. 23) in 1848 as a genus for the species Hylogalea murinaSchlegel and Miiller, Twelve years later Gray apparently repudi-ated the name when he described another species as Tupaia frenata.Until recent years most authors did not consider Dendrogale to begenerically distinct from Tupaia. Fitzinger, however, used it in1870, and in 1879 it was employed by Anderson as a full genus.Flower and Lydekker in 1891 did not recognize it, and Trouessart in1898 gave it only subgeneric rank. In the present paper it is em-ployed as th,e generic term for a small but well defined group ofsmall treeshrews which have gone under the specific names murina,frenata, and melanura.Erinaceus.?Blainville (Osteogiaphie des Mammiferes Insectivores,1839-1864, p. 112, pi. 6, fig. 1) uses the combination Erinaceus(Glisorex) tana. On page 31 he uses Glisorex as a full genus in thecombination Glisorex ferrugineus. This is the third instance that Iknow of where treeshrews have been referred to a genus that has notbeen specially set aside for them. The others are Diard and Du-vaucel's reference of them to Sorex in 1822, and EUis's Sciurus pub-lished in Gray. Blainville's error is so evident that one wondershow he made it.Griadobates.?A typographical error for Cladobates. It occurs inSchinz, Naturgeschichte und Abbildungen der Menschen und derSaiigethiere, p. 54. Included species ferrugineus, javanicus.Glipora.?This was origmally a manuscript name of Diard, andwas published by Jentink in 1888 (Cat. Syst. Mus. Hist. Nat. Pays-Bas, vol. 12, Mammiferes, p. 118). The species included in it areG. leucogaster ( = Tupaia ininorf), G. rufescens = (Tupaia splendidula?)
,
and G. murina = {Dendrogale murina). It is not probable Jentinkhad any intention of establishing the name of the genus or of theincluded species, rufescens or leucogaster. It is thus an accidentalsynonym of Tupaia. Glipora does not occur in Palmer's IndexGenerum Mammalium, 1904.Glirisorex.?Used by Scudder (Nomenclator Zoolgicus, pt. 2,p. 131) in 1882 probably as an etymologic improvement over Desma-rest's Glisorex. No species are mentioned. It is of course a puresynonym of Tupaia used in a broad sense.Glisorex.?This name was propose4 by Desmarest in a footnote onpage 536 of his Mammalogie, 1822, as more euphonious than Sorexglis,which he and other authors seemed to think was Diard and Duvaucel'sgeneric designation of the treeshrews. Desmarest, however, does
NO. 1976. TREESRREWS: FAMILY TUPAIID^?LYON. 27not use the term in preference to Tupaia. Glisorex was adopted byBlainville and Owen.Glisosorex.?Used by Giebel in Odontographie, 1855, page 18. Heprobably intended to copy the term Glisorex from Owen's Odontog-raphy. On the same page he uses the term Cladobates, evidentlythinking it a different genus from Glisorex and assigning differentdental formulas to the two animals. His knowledge of Cladobatesappears to be based upon Horsfield's account and figures in ZoologicalResearches m Java. No species are mentioned under Glisosorex,and the term is simply a variant of Glisorex.Herpestes.?Anderson (Zool. Res. West. Yunnan, 1879, p. 132) says:
'
' Diard and Duvaucel's figure [of Sorex glis] in the Asiatic Researches(Asiatic Researches, vol. 14, 1822, pi. 9) appears to have been copiedin a slightly reduced form into the Calcutta Journal of Natural His-tory (Cal. Journ. Nat. Hist., vol. 2, 1842, p. 456, pi. 13*, fig. 1),where it is regarded as a Herpestes!" I have been unable to consultthe volume of the Calcutta Journal referred to. In his introductionto Diard and Duvaucel's account Major General Hardwicke says:
''It bears most resemblance I think to the genus Viverra, particularlyto V. Ichneumon."Hylogale.?Proposed by Temminck (Monographies de Mammalogie,p. xix) in 1827 as a substitute for Tupaia, which being derivedfrom the native name tupai he considered a "nom tres-vicieux."He further remarks, ' ' Ce changement est dans Vinteret de la science;il sera sans doute adopte." According to Temminck the genus con-tained three species, which, however, are not mentioned by name.The term as originally written by Temminck never seems to havebeen adopted by other authors.Hylogalea.?An emendation of Hylogale Temminck, used by Schlegeland Miiller (Verb. Nat. Gesch. Nederl. Overs, Bezitt, 1839-44,p. 159) as the proper designation of the treeshrews. It is a puresynonym of Tupaia, but included two species which were unlmownto Raffles. The forms included by Schlegel and Miiller are tana,ferruginea, javanica, and murina, the last since made the type of thegenus Dendrogale Gray. Aside from Schlegel and Miiller the namedoes not seem to appear in the literature. In subsequent publica-tions Schlegel adopts the term Tupaja.Ptilocercus.?Proposed by Gray (Proc. Zool. Soc. London, 1848,p. 23) for the very curious animal since then known as Ptilocercuslowii. So far as I am aware no other term has ever been proposedfor Ptilocercus, neither has the animal ever been placed in any otherestablished genus.Ptilocerus.?A misspelling of Ptilocercus found in Wallace's Geo-graphical Distribution of Animals, 1876, vol. 1, p. 337; vol. 2, p. 187;and in Island Life, 1881, p. 345, and in Brehm's Thierleben, 1864,vol. 1, p. 664.
28 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Sciurus.?Used in the manuscript of W. Ellis, 1780, on the thirdvoyage of Captain Cook. Drawing and manuscript in the library ofthe British Museum, Natural History. This manuscript name waspublished as Sciurus dissimilis in 18Q0, in the Annals and Magazineof Natural History, third series, vol. 5, p. 71, in an article by Gray.Sorex.?Employed by Diard in the account of the first describedtreeshrew, Sorex glis, in the Asiatic Journal and Monthly Register,vol. 10, November, 1820, p. 478, also used again by Diard andDuvaucel. (Asiat. Res., vol. 14, 1822, p. 472, pi. 9. This volumewas received in London, at the Geological Society, January 10, 1823,and hence was probably published in the third quarter of 1822.Personal communication from C. D. Sherborn.) A careful examina-tion of these works shows that glis was described as a new species ofthe genus Sorex. It was not anyone's intention to make a new genuscalled Sorexglis, as certain writers have thought. Palmer in NorthAmerican Fauna, No. 23, page 636, is of the same opinion that Sorexglis was used as a genus and a species, and not as a single name,and so was Horsfield.^ Desmarest, however, considered it as ageneric term and published it as such, Sorexglis, and also emended itto Glisorex, which was still further emended by Scudder Glirisorex andby Giebel as Glisosorex.Sorexglis.?First used by Geoffroy and Cuvier (Hist. Nat. Manun.,vol. 3, liv, 35, December, 1821, p. 1) as a compound word Sorex-Glisas a generic designation of treeshrews. No citation of either Rafflesor Diard is given, but both are mentioned, and the name is evidentlytaken from the latter's Sorex glis. They discuss the inappropriate-ness of using barbarous names like Twpaia. It is interesting to notethat Twpaia appeared in May and Sorexglis in December of the sameyear, 1821. Geoffroy and Cuvier included in their genus the speciesnow known as javanica srnd ferimginea.Desmarest was next to use the term, in his Mammalogie, in 1822,in a footnote only, preferring in the text to use the term Twpaia.Sorex-glis is also mentioned by Cuvier in Dents des Mammiferes in1825, although Cladohates is adopted as the generic term.Tapaia.?An accidental renaming of Tupaia by J. E. Gray (Ann.Mag. Nat. Hist., ser. 3, vol. 5, 1860, p. 71). The speUing occurs asTapaia in the title, in the body of the article, and in the index.Tupaia does not appear in Gray's article, or elsewhere in the volume.Tupaia.?The earliest generic name for the treeshrews as such pro-posed by Raffles in May, 1821 (Trans. linn. Sec. London, vol. 13,1822, p. 256, May, 182P). It contained two species, ferruginea andtana. In the present paper tana is made the type of a new genus.Tupaia is the name adopted by most authors, although it was rejected
1 Catalogue of the Mammalia in the Museum of the Hon. East-India Company, 1851, p. 130.*See Horsfleld, Zool. Res. Java, 1824, p. 2 of text of Tapiriis malayanus,
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 29by some because of its barbarous origin from tupai the Malay wordfor any squirrel-like animal, in favor of Eylogale or Eylogalea, andCladobates.Tupaia is often written Tupaja by German and Dutch authors, andoccurs as Tupaya in Geoffroy and Cuvier.^Urogale.?Proposed by Mearns (Proc. U. S. Nat. Mus., vol. 28, May13, 1905, p. 43^ for the very peculiar treeshrew from Mindanao, Philip-pines. The type is U. cylindrura Mearns, which in the present paperis considered a synonym of Tupaia everetti Thomas. It is one of themost characteristic of all the genera of the Tupaiidse excepting.Ptilocercus. KET TO GENERA BASED ON EXTERNAL CHARACTERS.Tail naked for its basal portions, with a distichous tuft of haii-s at end.Ptilocercin^, Ptilocercus, p. 159.Tail haired throughout its whole extent, without a distinct naked basal portion.TUPAIIN^.Tail rounded and close haired for its whole extent.Hind foot about 30 mm.; snout not unusually long and slender, head usuallyconspicuously marked Dendrogale, p. 126.Hind foot about 50 mm.; snout long and slender, head without conspicuousmarkings Urogale, p. 154.Tail clothed with longer hairs, squin-el-like in character.Snout long and slender, with a naked area on top of nose encroaching backwardin middle line into hau-ed area Tana, p. 134.Snout not unusually long and slender, with naked area on top of nose not en-croaching on haired area, but cut straight across.Lowerlobe of ear presenting a surface greater than upper half of ear, inner sideof ear fairly well haired, reticulations on naked area of nose coarser.Anathana, p. 120.Lower lobe of ear presenting a smaller surface than upper portion of ear, innerside of ear scantly haired, reticulations on naked area of nose finer.Tupaia, p. 30.KEY TO GENERA BASED ON CRANIAL CHARACTERS.Supraorbital foramen absent, temporal fossa about equal to orbit in size,Ptilocercin^, Ptilocercus, p. 159.Supraorbital foramen present, temporal fossa conspicuously smaller than orbit,TUPAIINyE.Fenestra in zygoma small and inconspicuous.Rostrum much elongated; distance from lachrymal notch to end of premaxillaabout equal to distance from notch to occipital condyles Urogale, p. 154.Rostnmi not conspicuously elongated; distance from lachrymal notch to end ofpremaxilla much less than distance from notch to occipital condyles.Distance from lachrymal notch to end of premaxilla equal to distance fromnotch to auditory meatus; rostrum slender Dendrogale, p. 126.Distance from lachrymal notch to end of premaxilla equal to distance fromnotch to outer pterygoid plate; rostrum heavy Anathana, p. 120.
I Hist. Nat. Mamm., vol. 3, liv, 35, December, 1821, p. 1.
30 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Fenestra in zygoma, conspicuous, elongated, oval in shape.Rostrum long and slender; premaxillee elongated; distance from lachrymal notchto end of premaxilla equal to distance from notch to condyles. . . Tana, p. 134.Rostrum not conspicuously elongated and slender, nor premaxillfe unusuallylengthened; distance from lachrymal notch to end of premaxilla equal to dis-tance from notch to auditory meatus Tupaia, p. 30.KEY TO GENERA BASED ON DENTAL CHARACTERS.Upper molars without mesostyle; upper canine (third tooth in upper jaw) with 2 rootsand premolariform in shape; i2 about twice as large as ii; middle lower premolarsmallest of lower premolar series Ptilocercin.e, Ptilocercus, p. 159.Upper molars with distinct mesostyle; upper canine (third tooth in upper jaw) usuallywith a single root, small but somewhat caniniform, i^ and i., almost subequal; firatlower premolar smallest of lower premolar series Tupaiin^.i^ much larger than i^, and canirdfoiTn Urogale, p. 154.i^ slightly smaller than i^, not caniniform.Hypocones on upper molars much reduced; not present as distinct cusps; lowercanines and premolars more trenchant Dendrogale, p . 126
.
Hypocones on upper molars not conspicuously reduced; present as more or lessevident cusps; lower canines and premolars less trenchant.Hypocones very strongly developed, and pm* distinctly wider than long, andCj not noticeably higher than adjacent i^ and pmj Anathana, p. 120.Hypocones moderately developed; pm* not conspicuously wider than long; Cistanding noticeably higher than adjacent i^ and pwj.Lower canine nearer % than to pmj Tana, p. 134.Lower canine equally spaced between i^ and pm2 Tupaia, p. 30.DESCRIPTIONS OF GENERA AND SPECIES.Genus TUPAIA Raffles.1820. Sorex Diard, Asiat. Joum. Monthly Register, vol. 10, p. 478, November,1820. Not of Linnaeus 1758.1821. Tupaia Raffles, Trans. Linn. Soc. London, vol. 13, 1822, p. 256, May,182-1.1821. Sorex-glis Geoffroy and Cuvier, Hist. Nat. Mamm., vol. 3, liv. 35, p. 1,December, 1821, perhaps early in 1822.1822. Glisorex Desmarest, Mammalogie, footnote, p. 536, 1822.1825. Cladobates Cuvier, Dents des Mammifferes, p. 60, pi. 17, 1825.1827. Hylogale Temminck, Monogr. Mamm., p. xix, 1827.1840. Erinaceus, Blainville, Ost6og. Mamm. Insect., p. 112, pi. 6, fig. 1.1843. Eylogalea Schlegel and Mijller, Verh. Nat. Gesch. Nederl. Overz.Bezitt., p. 159.1855. Glisosorex Giebel, Odontographie, p. 18, 1855.1860. Sciurus Ellis, in Gray, Ann. Mag. Nat. Hist., ser. 3, 1860, vol. 5, p. 71.1860. Tapaia (sic) Gray. Ann. Mag. Nat. Hist., ser. 3, vol. 5, 1860, p. 71.1882. Glirisorex Scudder, Nomenclator Zool., pt. 2, p. 131, 1882.1888. Glipora JE-NTiNK, Cat. Syst. Mus. Hist. Nat. Pays-Bas, vol. 12, Mammif^res,p. 118. Publication of manuscript name of Diard.Type.? Tupaia ferruginea Raffles. Two species were included inthe genus by Raffles
?
T ferruginea and T. tana. The latter is takenas the type of the new genus Tana, page 134^ and T. ferruginea isselected the type of the genus Tupaia,
NO, 1976. TREE8HREWS: FAMILY TUPAIID^?LYON. 31Diagnostic characters.?The most generalized member of the familyTupaiidse, mainly characterized by the absence of the specializationsof the other members, such as rostrum not excessively elongated, tailnot tufted nor close-haired, teeth without unusual development ofhypocones or other peculiarities, head without conspicuous markings.External cliaracters.?The external form of Twpaia is exceedinglysquirrel-like. The natives make no distinction in name between tu-paias and squirrels, calling both tupai. Collectors observe it is some-times impossible to say whether a squirrel or tupaia has been shot untilthe specimen is picked up. Even experienced workers in museumsnow and then confuse a tupai skin with a squirrel skin. Aside fromthe shrew-like snout of Tupaia, a character which is not usuallyobvious in the average skin, the genus is quickly distinguished fromSciurus hj the absence of the long black whiskers or vibrissse. Fromthe other members of the family Tupaia is distinguished by its rela-tively small ears, finelyreticulated naked areaof nose, not encroach-ing on haired area ofnose, well-haired tail,lack of face markings.A more or less dis-tinctly defined oblicjiieshoulder stripe present.Mammse vary from oneto three pairs. For thenumber and arrange-ment of the footpadssee fig. 3, page 32. Fig. 2.?Diagram illustrating the ehinarium (1) in the ge-nera Tana and Urog.vle and (2) in the genera Tupaia,Anathana, and Dendrogale. About natural size.Cranial characters.?Rostrum moderately short and heavy; dis-tance from the lachr3^mal notch to end of premaxilla is about equalto the distance from the notch to the region of the external auditorymeatus ; in some species the posterior leg of the dividers falling behindthe opening, and in others in front of it. In the group containing thespecies Tupaia minor and gracilis, the rostrum is particularly shortand stubby and the posterior leg of the dividers reaches to about theend of the pterygoid processes. In the species T. chrysogaster therostrum is ver}^ slender, but not so much elongated, and the skull as awhole strongly suggests that of the genus Tana. The width of thebraincase usually approximately equals the length of maxillary tooth-row, or exceeds it a trifle in most species ; in some of the smaller mem-bers of the genus it exceeds it by as much as 3 or 4 mm. Fenestra inzygoma, large and conspicuous, elongated oval in shape; a distinctsupraorbital foramen present; temporal fossa distinctly smaller thanorbit. (Fig. 4, page 33.)
32 PROCEEDINGS OF THE NATIONAL MUSEUM. TOL. 45.Dental characters.?The teeth of Tupaia are the most generalized ofany genus in the family. The two pairs of upper incisors are usuallysubequal, although the first is nearly always a little the larger. Incertain species T. nicoharica, T. javavica, T. minor, T. gracilis, how-ever, there is a marked difference in size, but it is always the first pairwhich is the larger of the two. This difference in size is fairly usefulas a group character, but can not be relied upon as a hard and fastone, as specimens showing intermediate degrees of development arefound. There are three pairs of lower incisors, all well developed
ftf>.
<.d?.
hy,ihn:
i.dP. i.d*
- i.dP. prx.e
? then. prx.vn.
Fig. 3.?Palmar surface of eight forefoot and plantar subface ofeight hindfoot of tup^ua glis ferruginea. cat. no. 124319, u. s. n. m.,Singapore. Enlarged about t-wice. After Gregory, Bulletln Ameri-can Museum of Natural History, vol. 27, 1910, p. 270. Hy. thn, hypo-thenar pad; i. d', i. d^, i.d^, i. d*, first, second, third and fourthinterdigital pads; prx. c, proximal external pad; prx. in, proximalINTERNAL PAD; thCTl, THENAR PAD.and functional, but the third pair is much smaller than the others;the second pair is somewhat larger than the first pair. In some casesthe third pair is relativel}^ more reduced in size than in others, and inthese cases the tooth is barely functional. This is particularly so inthose species which have the greatest development of the central upperincisors and is also correlated with a greater development of the lowercanine. In this connection it is interesting to note that in the genusUrogale where the third lower incisors are quite rudimentary we finda very unusual development of the lower canine, and of the second
NO. 1976. TREESHREW8: FAMILY TUPAIID^E?LYON. 33pair of upper incisors and not of the first pair, as seems to be thetendency in the genus Tupaia. The upper canine in Tupaia ismoderately well developed,and usually with only oneroot, except in certain anom-alous cases. Its crownalways projects distinctlybeyond the level of the ad-jacent premolar. The lowercanine is relatively betterdeveloped than the upperand always projects con-spicuously above the levelsof the adjacent premolar andincisor. In the group ofspecies with the enlargedcentral incisors and reducedthird lower incisor, the lowercanine is relatively verymuch enlarged and standsvery high above the man-dibular toothrow. At first
Fig. 4.?Tupaia glis feekuginea, X 1, Cat. No. 114548.U.S.N.M. Tapanuli Bat, Sumatra.sight -this development of the lower ca-nine appears to be a valuable group char-acter, but one encounters specimens orspecies where it is quite impossible todecide whether the lower canine is of theenormously developed type or not. Itsgreatest absolute and relative develop-ment occurs in Tupaia nicoharica. Inthe case of T. cuyonis it is difficult to saywhether the canine is unduly enlarged.There are three pairs each of upper andlower premolars. The first of each aresmall and almost functionless ; the othersincrease in size and complexity from be-fore backward. The last deciduous pre-molars have the same form as the typicalmolars. Upper and lower molars, threepairs of each; the first and second upper molars with the hypoconesalways present, but rather poorly developed. See fig. 5 above.80459??Proc.N.M.vol.45?13 3
Fig. 5.?Tupaia glis feeruginea up-per AND LOWER TOOTHEOWS X 2|,Cat. No. 114548, U.S.N.M. Tapa-nuli Bat, Sumatra.
34 PROCEEDINGS OF THE NATIONAL MUSEUM. tol. 45.A cecum about 10 mm. long is probably one of the generic cliar-acters. (See p. 14.)Geographic distribution.?The genus Tupaia has a wider distributionthan any of the other genera of the family, ranging on the west fromnortheastern India, Burma, and Nicobar Islands eastward to membersof the Philippine Islands which geographically are extensions of Borneo
;
on the north from central China south to Sumatra, including islandson the southwest coast, Java, and Borneo. It does not occur onCelebes, nor islands to the east of Java, with the possible exception ofBali, whose fauna is not well known.Number of forms.?Tupaia, the most generalized member of thefamily contains the largest number of specific and subspecific forms,48 named forms being recognized in this revision. As a matter ofconvenience, the genus may be divided into four fairly well-markedsections and each of these mto secondary groups and subgroups. I feelsure, however, that the divisions are not natural ones, and it is quitepossible in most cases to start with any given subdivision and bymeans of diverging forms pass into or at least toward some other sub-division. The most strikingly differentiated of the forms are Tupaianicobarica and T. picta. The sections, groups, and subgroups, intowhich the genus may be divided are:I. Members with the tail longer than head and body, central upperincisors large in comparison with lateral pair, lower canines welldeveloped, and third lower incisor reduced in size, general shape ofskull normal. Contains two groups.1. Nicobarica group. Size large, general color brown, black area onlower back; mammas, 1-1=2.Tupaia nicobarica nicobarica, Great Nicobar Island, page 103.Tupaia nicobarica surda, Little Nicobar Island, pi. 3, fig. 1; pi. 10,fig. 7, page 104.2. Javanica group, size small, general color olivaceous above anddark below; mammae 2-2 = 4.Tupaia javanica, Sumatra, Nias Island, Java, pi. 3, fig. 2; pi. 10,fig. 1, page 106.II. Members also -with, the tail longer than head and body, andlargely developed central upper incisors and lo;wer canines, but ros-trum of skull very short and stubby and brain case relatively inflated,the general shape of the skull being quite atypical. General colora-tion oHvaceous above, light below, mammse, 2-2 = 4. Contains onlyone group.1. Minor group. Characters as above.Tupaia minor minor, Borneo, pi. 3, fig. 3; pi. 10, fig. 3, page 110.Tupaia minor malaccana, Sumatra, southern Malay Peninsula,and Linga Island, pi. 10, fig. 4, page 114.Tupaia minor sincipis, Sinkep Island, page 115.
NO. 1976. TREESHBEW8: FAMILY TUPAIID^?LYON. 35III. Members combining the characters of Section II mth thoseof the next or typical section. The tail is long, the skull has a shortstubby rostrum and inflated brain case, like that of Section II, but theteeth do not have unusually developed central upper incisors andlower canines. General coloration olivaceous above, light below.Number of mammae unlaiown. Contains but one group.1. Gracilis group. Characters as above.Tupaia gracilis gracilis, Borneo, page 117.Tupaia gracilis inflata, Banka Billiton, pi. 10, fig. 2, page 118.Tupaia gracilis edarata, Karimata Island, page 118.IV. This section contains the great majority of the members of thegenus, the tail is not unusually elongated, the skull has the typicalform as in Section I, but the teeth do not show the peculiarity ofenlarged central upper incisors and lower canines. Size large ormediiun, never small. Coloration various. Mammae, 1-3 pairs.Contains eight fairly well-defhied groups.1. Cliinensis group. Color generally grayish or olivaceous, with-out characteristic marks or bright colors; mammae, 3-3 = 6.Tupaia cliinensis, Southern China, Siam, pi. 8, fig. 1, page 63.Tupaia concolor, Anam, page 68.Tupaia modesta, Hainan, page 69.Tupaia dissimilis, Pulo Condore, pi. 1, page 67.Tupaia belangeri. Northern Malay Peninsula and southern Burma,pi. 8, fig. 2, page 59.2. Mollendorfi group. General coloration grayish or ochraceouswithout characteristic marks or bright colors, teeth approaching thoseof the javanica group, section I; mammae, 2-2 = 4.Tupaia cuyonis, Cuyo Island, Philippines, pi. 9, fig. 1, page 82.Tupaia mbllendorfii, Culion Island, Philippines, page 81.Tupaia palawanensis, Palawan Island, Philippines, page 78.3. Ferruginea or Glis group. General coloration dark ferruginous,tail various but never clear below except in T. demissa; mammae,2-2 = 4. For convenience this group may be divided into 2 sub-groups, the extremes of which are different enough, but the 2 sub-groups practically merge into one another.a. Ferruginea subgroup, typically ferruginous
:
Tupaia glis ferruginea, Sumatra, southern ]\Ialay Peninsula,pi. 3, fig. 4; pi. 8 fig. 6, page 41.Tupaia glis glis, Penang Island, pi. 8, fig. 8, page 45.Tupaia glis batamana, Batam Island, pi. 8, fig. 9, page 46.Tupaia glis sordida, Tioman Island, pi. 9, fig. 8, page 48.Tupaia glis pemangilis, Pemangil Island, page 48.Tupaia glis pulonis, Aor Island, page 47.Tupaia phseura, Sinkep Island, pi. 9, fig. 7, page 49.Tupaia tephrura, Batu Islands, pi. 8, fig. 7, page 50.Tupaia demissa, Sumatra, page 58.
38 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.
h. Wilkinsoni subgroup, approaching helangeri of the chinensisgroup
:
Tupaia lacernata wilkinsoni, middle portion of Malay Penin-sula, pi. 8, fig. 3, page 52.Tupaia lacernata lacernata, Lankawi and Terutau Islands, pi.8, fig. 5, page 53.Tupaia lacernata raviana, Butang Islands, pi. 8, fig. 4, page 54.Tupaia lacernata obscura. Great Redang Island, page 55.Tupaia lacernata longicauda, Perhentian Island, page 56.4. Splendidula group. Color various, usually solid, varying fromseal brown to bright ferruginous. Color of underside of tail alwaysclear along the central line at least; mammge, 2-2 = 4,Tupaia splendidula, Borneo, pi. 10, fig. 11, page 83.Tupaia natunse, Bunguran, Natuna Islands, pi. 10, fig. 12, page 85.Tupaia lucida, Pulo Laut, Natuna Islands, page 86.Tupaia chrysomalla, Siantan, Anamba Islands, pi. 10, fig. 10,page 88.Tupaia riabus, Riabu, Anamba Islands, page 88.Tupaia anambse, Jimaja, Anamba Islands, page 89.Tupaia carimatseS Karimata Island, pi. 10, fig. 6, page 98.Tupaia castanea^ Bintang Island, pi. 10, fig. 9, page 90.Tupaia siaca^ Sumatra, pi. 10, fig. 8, page 91.5. Discolor group. Underparts rather brightly colored, anteriorportion of upper parts ferruginous, posterior olivaceous; mammae,3-3 = 6.Tupaia discolor, Banka Island, pi. 9, fig. 4, page 73.Tupaia longipes longipes, northern Borneo, page 76.Tupaia longipes salatana, southern Borneo, pi. 9, fig. 5, page 77.6. Hypochrysa group. Underparts very brightly colored, upperparts darkly olivaceous or brownish; mammae, 1-1 =2.Tupaia hypochrysa, Java, pi. 9, fig. 6, page 70.Tupaia chrysogaster, Pagi and Nias Islands, pi. 9, fig. 9, page 71.7. Montana group. Entire animal a grizzled dark brownish, oftenwith a well-marked black area on back; mammae, 2-2 = 4.Tupaia montana montana, Mount Dulit, Borneo, pi. 9, fig. 2,page 94.Tupaia montana baluensis, Mount Kina Balu, Borneo, page 95.8. Pida group. Back \vith a narrow dorsal stripe; mammae,2-2 = 4.Tupaia picta, Borneo, pi. 9, fig. 3, page 96.Remarks.? Tupaia is the most generalized member of the Tupaiidaeand, as would be expected, shows the largest number of distinct formsand the widest geographic distribution. It is easy to see how theother members of the family with the exception of Ptilocercus havebeen derived by relatively slight modifications from Tupaia.
1 Suggests the montana group. ' Closer to the femiginea group than the others.
NO. 197C. TREE8HREWS: FAMILY TUPAIID^?LYON. 37KEY TO THE SPECIES AND SUBSPECIES OF TUPAIA.Tail distinctly longer than head and body, lower canine (except m T. gracilis) twicethe length of the incisor in front of it, and central upper incisors (except T.gracilis) very conspiciously larger than lateral pair.Size large, head and body 180-200 mm., much black on lower back. . T. nicoharica.Shoulders and neck bright grizzled ochraceous.Great Nicobar, T. n. nicoharica, p. 103.Shoulders and neck dull grizzled ochraceous... Little Nicobar, T. n. surda, p. 104.Size small, head and body, 140-150 mm., general color uniformly grizzled olivaceous.Skull with rostrum of normal shape, not stubby, distance between the two lachry-mal notches equals distance from that notch to space between first andsecond upper incisors, under parts rather dark and grizzled.Sumatra, Nias, Java, T. javanica, p. 106.Skull with a short stubby rostrum, distance between the two lachrymal notchesequals distance from that notch to front of first upper incisor, under partsusually clear whitish or buffy.Hind foot 30-35 mm., central upper incisors and lower canines well developed,grizzling of upper parts, not unusually fine T. rainor, p. 109.Tail not conspicuously darker than general coloration of upper parts.Sumatra, Linga, Malay Peninsula, T. m. malaccana, p. 114.Tail conspicuously darker than general coloration of upper parts.Rump not distinctly russet in contrast with upper parts.Borneo, T. in. minor, p. 110.Rump distinctly russet in contrast with upper parts.Sinkep, T. m. sincipis, p. 115.Hind foot 37-43 mm., central upper incisors and lower canine not unusuallydeveloped, grizzlmg of upper parts very fine T. gracilis, p. 116.Hind foot 40 mm. or over, bullae smaller Borneo, T. g. gracilis, p. 117.Hind foot not over 40 mm., bullae larger.Upper parts and legs grizzled olivaceous brownish.Billiton, Banka, T. g. infiata, p. 118.Upper parts and legs grizzled ochraceous brownish.Karimata, T. g. edarata, p. 118.Tail usually distinctly shorter than head and body, occasionally approximately equalto or slightly longer than head and body, but in that case, lower back without alarge black area; central upper incisors not conspicously larger than lateral pair,and lower canine not unusually developed, less than twice the length of theincisor in front of it.Middle of back with a conspicuous nan-ow black stripe Borneo, T. picta, p. 96.Middle of back without a conspicuous narrow black stripe.Tail not uniformly above and below grizzled blackish and gi-ayish or blackish andochraceous, if gidzzled on upper surface, always showing a clear ungrizzledreddish ochraceous area on either side of central line, when viewed frombelow.Tail uniformly above and below, dull cream color. .Sumatra, T. demissa, p. 58.Tail various, uniformly tawny above and below, or coarsely grizzled withblackish above, and tawny or ochraceous below, never cream color.General color of upper parts between seal and walnut brown.Borneo, T. splcndidula, p. 83.General color of upper parts varying from dark hazel or chestnut to brighttawny or ochraceous with or without distinct grizzlmg.Upper parts with a fine diffused giizzling of blackish and ochraceous witha tendency toward a dark patch on back.Karimata Island, T. carimatse, p. 98.
38 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Upper parts without fine diffused grizzling, grizzling absent or else coarse,and without tendency to dark patch on back.General coloration of head, body, and tail tawny.Tawny, light and bright Laut, Natuna Islands, T. hicida, p. 86.Tawny, dull and darker. . . . Riabu, Anamba Islands, T. riabus, p. 88.General coloration of head and body a mixture of dark ferruginous andblackish.Under parts ochraceous Bintang Island, T. castanea, p. 90.Under parts light or dark buffy, sometimes grizzled.Under parts bright buffy without grizzling.. Sumatra, T. siaca, p. 91.Under parts dull buffy more or less gi'izzled.Under side of tail tawny.Bunguran, Natuna Islands, T. natunse, p. 85.Under side of tail ochraceous.Upper parts inclining toward ferruginous.Siautan, Anamba Islands, T. carysomalla, p. 88.Upper parts inclining toward ochraceous.Jimaja, Anamba Islands, T. ajiambie, p. 89.Tail uniformly above and below grizzled blackish, and gi-ayish, buffy, orochraceous, never with a distinct clear area on either side of middle of tailwhen viewed from below.First and second upper molars rather quadrate in outline, with distincthypocones, shoulder stripe obsolete or nearly so, back often marked withwith a broad black band T. mo7itana, p. 93.Back usually marked with a broad black patch or band.Mount Dulit, Borneo, T. m. montana, p. 94.Back without a broad black patch.Mount Kina Balu, Borneo, T. m. baluensis, p. 95.First and second upper molars more triangular in outline and with hypo-cones poorly developed if at all, shoulder stripe usually fairly wellindicated; back never with a broad black band.Upper partg of body distinctly rusty or ferruginous in color, and if grayishareas are present they are posterior.Anterior portion of body ferruginous, posterior grayish; mammae, 3-3=6.Hind foot less than 50 mm.; difference in color between anterior andposterior portions of back very marked. . .Banka, T. discolor, p. 73.Hind foot usually 50 mm. or more; difference in color between anteriorand posterior portions of back less marked T. longipes, p. 74.General effect of lower back and tail bister.Northern Borneo, T. I. longipes, p. 76.General effect of lower back and tail clove-brown.Southern Borneo, T. I. salaiana, p. 77.Anterior and posterior portions of body of the same ferruginous color.Under parts dirty white to buffy, never orange rufous; mammee,2-2=4. Ferruginea gi'oup.Tail brownish, like rest of upper parts in color.Sinkep Island, T. phxura, p. 49.Tail darker or lighter than rest of upper parts.Tail black or blackish, seen above.Palawan Island, T. palawanensis, p. 78.Tail a grizzle of blackish and buffy, seen above.Hind feet darker than tail, nearly black, a light buffy color pre-dominating at end of tail. . Batu Islands, T. tephrura, p. 50.
tjo. 197G. TRUESIIREWS: FAMILY TUPAIIDJE?LYON. SOHind feet not darker than tail, and a light buffy color not pre-dominating at end of tail.General color of upper parts brighter, burnt sienna in generaleffect.Tail not noticeably grayish, size slightly smaller.Sumatra southern Malay Peninsula, T. glis/erruginea, p. 41.Tail rather grayish, size slightly larger.Batam Island, T. glis bataviana, p. 46.General color of upper parts duller, a dark mars brown ingeneral effect.Light colors of imder parts inclining toward ochraceous.Tioman Island, T. glis sordida, p. 48.Light colors of under parts inclining toward buffy or whitish.Size smaller, hind foot not over 40 mm. and condylobasallength not over 45 mm.Pemangil Island, T. glis pemangilis, p. 48.Size larger, hind foot over 40 mm. and condylobasal usuallyover 45 mm.Color of lower back sUghtly darker than rest of upperparts Aor Island, T. glis pidonis, p. 47.Color of lower back not noticeably darker than rest ofupper parts Penang Island, T. glis glis, p. 45.Under parts orange rufous; mammae, 1-1=2.Kias and Pagi Islands, T. chrysogaster, p. 71.Upper parts of body not distinctly ferruginous, a grizzle of blacldsh andvarious shades of buffy or ochraceous; if ferruginous tints appear theyare situated posteriorly.Under parts orange rufous Java, T. hypochrysa, p. 70.Under parts buffy or whitish, or ochraceous.Mammae, 3-3=6; ranging from middle of Malay Peninsula north-ward. Belangeri-chinensis group.Color over lower back more ochraceous or tawny than on upperback.Northern Malay Peninsula and southern Burma,T. belangeri, p. 59.Color over lower back not essentially different in color from restof upper parts.Condylobasal length 50 mm., maxillary, tooth row over 19 mm.Anam, T. concolor, p. 68.Condylobasal length less than 50 mm. and maxillary tooth rowless than 19 mm.Skull and rostrum narrower, width of brain case 18 mm.Pulo Condore, T. dissimilis, p. 67.Skull and rostrum not so slender, width of brain case more than18 mm.Color darker, especially underparts and legs.Hainan, T. viodesta, p. 69.Color not so dark, underparts whitish and legs often grayish,Southern China and adjacent territory, T. chinensis, p. 63.Mammae, 2-2=4; ranging from middle of Malay Peninsula southward,and occurring in Philippine Islands.Hind foot and condylobasal length usually less than 45 mm. andtail not darker than lower back.
40 PROCEEDINOS OF THE NATIONAL MUSEUM. vol. 45.Tail distinctly different in color from lower back.Tail more grayish than back.Culion Island, T. mollendorffi, p. 81.Tail more blackish than back.Palawan Island, T. palaivanensis, p. 78.Tail approximately of the same general color as lower back.Rostmm rather short and heavy with an abrupt origin fromrest of skull Cuyo Island, T. cuyonis, p. 82.Rostrum not imusually short and heavy nor with an abruptorigin from rest of skull.Great Redang Island, T. lacernata obscura, p. 55.Hind foot and condj'lobasal length of skull usually exceeding 45mm. and tail more or less darker than lower back.Tail conspicuously darker than rest of upper parts.Size slightly larger, condylobasal length about 47 mm., max-illary tooth row 19 mm. Middle third of Malay Penin-sula, T. lacernata ivilhinsoni, p. 52.Size slightly larger, condylobasal length short, 45 mm., max-illary tooth row 18 mm.Lankawi and Terutau Islands, T. lacernata lacernata, p. 53.Tail darker than rest of upper parts, but not conspicuously so.Skull relatively short and wide, general color more olivaceous.Butang Islands, T. lacernata raviana, p. 54.Skull relatively not so short and wide, general color moreochraceous,Perhentian Islands, T. lacernata longicauda, p. 5G.
TUPAIA GLIS (Dlard).(Synonymy, type specimens, etc., under the subspecies.)Geograjyliic distribution.?Southern third of the Malay Peninsula,and various adjacent islands, and Sumatra.Diagnosis.?Upper parts a grizzle of a color between ferruginousand chestnut, and blackish, tail a grizzle of buffy and blackish;mammse, 2-2 = 4.Color.?Upper parts of head, neck, and body, and outer side of legsa grizzle of ferrugmous or chestnut and blackish, tail both above andbelow a grizzle of buffy and blackish, the black in excess above andthe buff below, especially along the middle line ; underparts includinginner side of legs various shades of buff, often with dark bases of thehaii-s showing through; feet a fine grizzle of buffy and blackish, thelatter color in excess; shoulder stripe of average development, vary-ing from yellow to buff yellow. Often around base of tail and thighsthe ferruginous or chestnut color is replaced by an ochraceous tint.SlcuU and teeth.?These show no special distmguishing character-istics. The rostrum is relatively long, and the distance from thefront of the premaxilla to the lachrymal pit is generally greater than
NO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 41the distance from the pit to the posterior edge of the external auditorymeatus.MeasuHments.?Tupaia glis is a relatively large species. The usualmeasurements are : Head and body, 170-200 mm.; tail, 150-175; hindfoot, 45-50; condylobasal length, 45-50; zygomatic width, 25-27;width of brain case, 18-21; maxillary tooth row, 17-20.Subspecies.?Six geographic forms of Tupaia glis may be recog-nized. They are all insular, except ferruginea occurring on the MalayPeninsula and Sumatra, but appear so closely allied to one anotherthat it is often impossible to distinguish them one from the otherwith certainty.The forms are
:
Tupaia glis ferruginea, southern third of Mala}^ Peninsula,Sumatra, page 41.Tupaia glis glis, Penang Island, page 45.Tupaia glis hatamana, Batam Island, page 46.Tupaia glis sordida, Tioman Island, page 48.Tupaia glis pemangilis, Pemangil Island, page 48.Tupaia glis pulonis, Aor Island, page 47.RemarJcs.?Tupaia glis is quite a plastic species, as the number ofsubspecies shows. The next three species, T. pliaeura, T. tepJirura,and T. demissa, are all geographic forms of T. glis, but differentiationhas proceeded so much further that their specific distinctness can notbe questioned. It is perhaps a slight misfortune that the earliestname appHed to the species was given to one of the insular races andnot to the real parent form occurring on the large land masses. As amatter of convenience the mamland subspecies is here treated firstand most of the comparisons are made with it. The mainland formis the most common in collections and the most convenient to workwith. TUPAIA GLIS FERRUGINEA Raffles.1821. Tupaia ferruginea Raffles, Trans. Linn. Soc. London, voL 13, 1822, p. 256,published May, 1821.1821. Tupaya press Geoffroy and Cuvier, Hist. Nat. Mamm., voL 3, livr. 35,p. 1, and Sorex-Glis press on p. 2, December, 1821, perhaps early in 1822.1842. Cladobates ferruginea, Cuvier, Hist. Nat. Mamm., a-oI. 7, Tab. Gen. Meth.,p. 2.1843. Eylogalea ferruginea, Schlegel and Muller, Verh. Nat. Ges. Nederl.Overz. Bezitt., p. 163, pi. 26, fig. 3; pl. 27, figs. 7-10.1879. Tupaia ferruginea, Anderson, Zool. Rea. West. Yunnan, p. 130, pl. 7,figs. 4 and 5.Type-locality.?Bencoolen, Sumatra.Type-specimen.?British Museum Register No. 79.11.21.573, skinand skuU of adult male, marked "Indian Mus. Coll. Sumatra, Sir S.RaflEles." It is probably one of the specimens from ''the woodsnear Bencoolen." Raffles says: "First observed tame in the house
42 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.
of a gentleman at Penang, and afterwards found wild at Singaporeand in the woods near Bencoolen. " The Penang animal was describedseveral months before ferruginea under the specific designation glis.No specimens collected at Singapore by Raffles are known, so that it isperfectly justifiable to take the above specimen as the type. It hasrecently been made into a modern study skin and is in fair condition.Some hair has slipped from about the head, legs, and base of tail.The color seems well preserved. The skull is nearly perfect; the oc-cipital region, zygomata, and bullae being slightly injured.Geographic distribution.?Malay Peninsula south of about latitude7^? N., including the island of Singapore; and most of island of Su-matra. For exact localities from which specimens have been exam-ined, see table, pages 43, 44. See No. 6 on map on page 75.Diagnostic characters.?A bright-colored member of the group, withupper parts of head and body distinctly ferruginous, tail a grizzle ofbuffy and blackish, rather distinctly different in color from body.Color.?Upper parts of head, neck, and body, and outerside of legsa grizzle of ferruginous or chestnut and blackish, tail both above andbelow a grizzle of buffy and blacldsh, the black in excess above andthe buff below, especially along the middle line; underparts includ-ing inner side of legs various shades of buff, often with dark basesof the hairs showing through; feet a fine grizzle of buffy and blackish,the latter color in excess; shoulder stripe moderately conspicuous,buffy in color.SlcuU and teeth.?These show no special distinguishing character-istics. See plate 8, fig. 6, and figures 4 and 5 on page 33.Measurements.?Type: Hindfoot, 45 mm; condylobasal length, 51;zygomatic width, 25 ?; width of braincase, 20; maxillary toothrow,20. Usual measurements of adults: Head and body, 175-200; tail,150-175; hindfoot, 42-47; condylobasal length, 47-49; zygomaticwidth 25-27; width of brain case, 19-20; maxillary toothrow, 18-20.For measurements of all specimens examined, see table, pages 43, 44.Remarlcs.? Twpaia glis ferruginea has one of the largest areas ofdistribution of any of the forms in the family, and appears very con-stant in its characters. I have been able to find no essential differ-ences between specimens from the Malay Peninsula and the island ofSumatra. This case is exactly paralleled by Tragulus napu.^An old specimen, British Museum, Register No. 79.6.28.15, collectedat Pajo, Sumatra, by Carl Bock, approaches very closely TupaiaphsRura.Specimens examined.?Sixty in the United States National Museum,37 in the British Museum, 1 in the Genoa Museum, 3 in the BerlinMuseum,
1 Miller, The Mouse Deer of the Rhio-Linga Archipelago: A Study of Specific Diflerentiation UnderUniform Environment, Proc. U. S. Nat. Mus., vol. 37, pp. 1-9, Sept. 1, 1909.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 43Measurements of Tupaia glis ferruginea.
Locality.
44 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Measurements of Tupaia glis ferruginea?Continued.
NO. 1976. ? TREESHREW8: FAMILY TUPAIID^?LYON. 45TUPAIA GLI3 GLIS (Diard).1820. Sorex glis Diard, Asiat. Jonm. Month. Reg., vol. 10, p. 478, November,1820.1821. Tupaiafcrruginea Raffles, Trans. Linn. Soc. London, voL 13, 1822, p. 256,published May, 1821.1822. Sorex glis, Diard and Duvaucel, Asiat. Res., vol. 14, p. 472, pL 9.1911. Twpaia ferruginea penangensis Robinson and Kloss, Journ. Fed. MalayStates Mus., vol. 4, p. 242, November, 1911. (Type No. 1445/11, SelangorMuseum, collected at Telok Behang, Penang Island, Apr. 2, 1911.)Type-locality.?Penang Island, off west coast of Malay Peninsula.Tliere is no type-specimen of Twpaia glis glis. Penang is the onlyexact locality mentioned in the original description, and consequentlyis regarded here as the type-locahty. No mention of its occurringon the Malay Peninsula is made. Of course Diard and RafHes hadthe same animal in mind in describing their respective species, butthe practical results are that glis is the name for the Penang animalMid ferruginea for the Sumatran and peninsular one.Geographic distribution.?Penang Island. See No. 21 on map onpage 75.Diagnosis.?A sHghtly differentiated geographic form of Tupaiaglis, differing in smaller size and a duller paler coloration of the upperparts.Color.?The general stylo of coloration of Tupaia glis glis is verysimilar to that of wide ranging T. glis ferruginea, hut the upper partsof body instead of being a fine grizzle of ferruginous or chestnut andblackish, is a grizzle of rather dark tawny or tawny ochraceous andblackish, while the head and neck are a grizzle of ochraceous buffand blackish. The tail is generally duller in T. glis glis. In otherrespects tlie two animals are essentially the same.STcull and teeth.?-The skull and teeth of Tupaia glis glis are dis-tinctly smaller than those of T. glis ferruginea, especially seen in thebrain case. (Plate 8, fig. 8.)Measurements.?The usual measurements of adults: Head andbody, 1G5 mm.; tail, 155; hindfoot, 42-43; condylo-basal length, 47;Z3^gomatic width, 25; width of braincase, 19; maxiUary tooth row,18.5; most of them agreeing with the minimum measurements ofTupaia glisferruginea. For measurements of all specimens examinedsee table, page 46.Remarlcs.? Tupaia glis glis is closely related to T. glis ferrugineaof the neighboring mainland of which it is zoologically a geographicform. Robinson and Kloss were entirely right in describing it as asubspecies of the mamland animal, and it is perhaps a misfortune thatglis was not originally used for the mainland form, or rather thatferruginea was not proposed first, as the latter name has been so longin use that it seems a pity to have it rank as a subspecies.Specimens examined.?Seven in the British Museum, and two loanedby the Selangor Museum. See table, page 46.
46 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.
NO. 1976. TREESEREWS: FAMILY TUPAIID^?LYON. 47Measurements of Tupaia glis batamana.
48 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.TUPAIA GLIS SORDIDA Miller.1900. Tiipaia sordida Miller, Proc. Wash. Acad. Sci., vol. 2, p. 231, Aug. 20, 1900.Type-locality.?Pulo Tioman, off east coast of Malay Peninsula.Type-specimen.?In United States National Museum, Cat. No.101747, Pulo Tioman, October 2, 1899; collected by Dr. W. L. Abbott;in good condition.Geographic distribution.?Pulo Tioman. See No. 28 on map onpage 75.Diagnosis.?Essentially the same in color as Tupaia glis pulonis,but under parts darker and more buffy; condylo-basal length slightlyless.Color.?Color of upper parts of head, neck, and body and outerside of legs is not distinguishable from that of Tupaia glis pulonis;shoulder stripe less prominent; under parts and inner side of legs dullbuff to dull ocliraceous buff with considerable of the dark bases of thehairs showing tlirough; underside of tail grayer and less buffy thanin the case of T. glis pulonis.Slcull and teeth.?These do not show any very tangible differencesfrom those of T. glis pulonis, but the condylo-basal length averagesabout 2 mm. less, and the skull as a whole appears narrower andslenderer, especially when compared with T. glis ferruginea. (Plate9, fig. 8.)Measurements.?Type: Head and body, 171 mm; tail, 165; hindfoot, 40; condylo-basal length, 45.5; zygomatic width, 25.5; widthof brain case, 19; maxillary tooth row, 18. For measurements of allthe specimens examined see table, page 51.Specimens examined.?Six in United States National Museum andtwo in British Museum.TUPAIA GLIS PEMANGILIS Lyon.1911. Tupaia pcmangilis Lyon, Proc. Biol. See. Wash., vol. 24, p. 168, June 16,1911.Type-locality.?Pulo Pemangil, off east coast of Malay Peninsula.Type-specimen.?In United States National Museum, Cat. No.112499, Pulo Pemangil, June 12, 1901; collected by Dr. W. L. Abbott;original number, 1064; in good condition.Geographic distribution.?Pulo Pemangil. See No. 29 on map onpage 75.
-, . r.Diagnosis.?Essentially the same color as Tupaia glis pulonis, buthind foot smaller and skull smaller and slenderer.Color.?The color of Tupaia glis pcmangilis does not differ fromthat of T. glis pulonis. *Slcull and teeth.?The skull and teeth are very similar to those ofTupaia glis pulonis and T. glis sordida, but the skull is slenderer,sUghtly shorter; m^ is smaller; bullae are slightly smaller.Measurements.?Tupaia glis pcmangilis is slightly smaller than theclosely related T. glis pulonis. External and cranial measurements
NO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 49
of the type: Head and body, 185 mm.; tail, 150; hind foot, 40; condylo-basal length, 45; zygomatic width, 23.5; width of brain case, 19;maxillary tooth row, 17.Specimen examined.?One, the type.Remarks.?-The three forms just enumerated, Tupaia glis pulonis,T. g. sordida, and T. g. pemangilis, are very closely related to one an-other. These three insular forms are not based on large series and itdoes not appear improbable that larger numbers might even show thatbut a single form inliabited the three geographically associatedislands of Tioman, Pemangil, and Aor, as the differences separatingthe three are not greater than are often found in the same formhaving a greater area of distribution.TUPAIA PHiEURA Miller.1902. Tupaia phasura Millee, Proc. Acad. Nat. Sci. Phila., 1902, p. 157, June 11,1902.Type-locality.?Sinkep Island, Rhio-Linga Archipelago.Type-specimen.?In United States National Museum, Cat. No.113148, skin and skull, adult male, Sinkep Island, September 4, 1901;collected by Dr. W. L. Abbott; original number, 1275; in good condi-tion.Geographic distribution.?Sinkep Island. See No. 19 on map onpage 75.Diagnosis.?Related to Tupaia glis ferruginea, but rather dark andreddish, and with the tail not essentially different in color from restof the animal. Mammse, 2-2=4.Color.?Tupaia phxura differs in color from T. glis ferruginea inhaving less black mixed in with the ferruginous-chestnut in the regionof the neck and shoulders and more black and less grizzling on thelower back and rump. The tail both above and below is a coarsegrizzle of black and tawny ochraceous, the black predominating onthe upper surface and the tawny ochraceous slightly in excess in themiddle line below. Wlien viewed from above the tail is concolorwith the rest of the upper parts. The under parts, legs, feet, shoulderstripe, etc., do not differ in color from the corresponding parts ofT. glis ferruginea.Skull and teeth.?There are -no characters by which the skulls andteeth of Tupaia phseura can be distinguished from those of T. gliserruginea. (Plate 9, fig. 7.)Measurements.?Tjrpe: Head and body, 195 mm.; tail, 140; Iiindfoot, 44; condylobasal length, 47.5; zygomatic width, 26.5; width ofbrain case, 19.5; maxillary tooth row, 19. For measurements of 2other specimens see table, page 51.Remarks.?Tupaia phseura is a very strongly characterized geo-graphic form of T. glis and has no close relatives. It is at once dis-80459??Proc.N.M.vol.45?13 4
50 PROCEEDINGS OF THE NATIONAL MUSEUM. tol. 45.tinguished from T. glis and its subspecies by the characters of itstail. It seems curious that with all the large collections made inthe Rhio-Linga Archipelago by Dr. W. L. Abbott that on only fourislands?Batam, Bintang, Linga, Sinkep?have treeshrews beencollected by him: a slightly differentiated form of T. glis on Batam,the very distinct Tupaia phseura or Sinkep, and on Bintang a mem-ber of the splendidula group. On Linga and Sinkep occur treeshrewsof a very different group (Tupaia minor) and genus (Tana)
.
Specimens examined.?Three. See table, page 51.TUPAIA TEPHRURA Miller.1903. Tupaia. tephrura Miller, Smiths. Misc. Coll., vol. 45, p. 57, Nov. 6, 1903.1909. Tupaia tephura (sic), Lyon and Osgood, Bull. U. S. Nat. Mus., 62,p. 250.Type-locality.?Tana Bala, Batu Islands, off west coast of Sumatra.Type-specimen.?In United States National Museum, Cat. No.121752, skin and skull of adult female, Tana Bala Island, February12, 1903; collected by Dr. W. L. Abbott; original number, 2276; ingood condition.GeograpJdc distribution.?Known only on Tana Bala Island, butprobably occurring on other islands of the Batu group. See No. 20on map on page 75.Diagnosis.?A dark-colored member of the Tupaia glis group oftreeshrews, darker than T. pliseura, but with a distinctly lighter tail.Mammse 2-2 = 4.Color.?Compared with Tupaia glis ferruginea, T. teplirura has agenerally darker color effect, the ferruginous tones being deeper andricher, and on the posterior half of the body there is a greater ad-mixture of black; the ferruginous extends further forward, distinctlycoloring the top and sides of head. The legs are darker in colorthan they are in T. glis ferruginea; the feet are almost black. Theunderparts are slightly darker than is usual in T. glis ferruginea.The tail is a mixture of black or blackish and buff or buffy; aboveat the base the black predominates, but for the posterior half thetwo colors are about equally mixed ; on the underside the buffy colorin excess, and on either side of the middle line is almost unmixedwith blackish, and rather closely approaches the style of tail of theT. splendidula group.SJcull and teeth.?The skull and teeth of Tiqmia teplwura are notessentially different from those of T. glis ferruginea. (Plate 8, fig. 7.)Measurements.?Type: Head and bod}^, 193 mm.; tail, 130 hindfoot, 43; condylo-basal length, 48.5; zygomatic width, 25.5; width ofbram case, 19.5; maxillary tooth row, 19. See table, page 51.RemarJcs.?Tupaia tephrura is a well-marked geographic form ofT. glis. Curiously enough it has many external resemblances to T.phxura of Sinkep on the opposite side of Sumatra, but it is dis-tinctly darker and more ferruginous anteriorly, and its tail is buffy
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 51where that of T. phxura is tawny ochraceoiis, and is distinctly dif-ferent in color from the rest of the upper parts.Specimens examined.?Two, from Tana Bala.Measurements ofJive insularforms of the Tupaia glis group of treeshreivs.
Name.
T. glis pulonisDoT.glispemangilis.Tupaia glis sor-dida.DoDoDoDoDoDoDo ,T. phxura ,DoDoT. tephruraDo.
Locality.
Pulo Aor
....doPulo Pemangil.Pulo Tioman...
....do
.do.
.do.
.do.
-do.doPulo SinkepdodoBatu Ids.; Tana]5ala,do
Number.
52 PROCEEDINGS OF THE NATIONAL MUSEUM. tol. 45.and body cream color to dull orange buff, with, more or less of the darkbases of tbe hairs showing through, inner side of legs colored similarlyto under parts, but hairs scantier and colors duller; shoulder stripemoderately pronounced, buffy in color.S'kull and teeth.?These do not differ materially from those ofTwpaia glis; the rostrum is if anything slightly relatively shorter in themore northern animal.Measurements.?The usual measurements are: Head and body,175-200 mm.; tail, 150-185; hind foot, 41-46 (less than ui T. glis);zygomatic width, 24-26; width of brain case, 18.5-20; maxillarytooth row, 17.5-19.5.Subspecies.?In addition to the mainland form, four msular formsare recognized
:
Twpaia lacernata wilkinsoni, middle third of Malay Penmsula,page 52.Tupaia lacernata lacernata, islands of Lankawi and Terutau,page 53.Tupaia lacernata raviana, Butang Islands, page 54.Tupaia lacernata obscura, Great Redang Island, page b5.Tupaia lacernata longicauda, Perhentian Island, page 56.Remarlcs.?As a matter of practical con\enience the mainland sub-species is first described, and comparison of the insular races madewith it. Tupaia lacernata is almost a perfect intermediate so far ascolor is concerned between T. glis and T. helangeri. Tupaia hclangeriis lighter in color and the contrast of its rather tawny rump withthe lighter anterior parts of the body more marked; its tail is not sodark. The point of most perfect differentiation, however, is thenumber of mammas. Twenty-four females of T. lacernata showingdeveloped mammce have the number 2-2=4, while 21 females of T.helangeri have the number 3-3 =6.TUPAIA LACERNATA WILKINSONI Robinson and Kloss.1911. Twpaia ferruginea wilkinsoni Robinson and Kloss, Journ. Fed. Malay-States Mus., vol. 4, No. 2, p. 173, April, 1911.Type-locality.?Ko-ldiau, Tarang (also spelled Trang and Trong),Lower Siam.Type-specimen.?In the Selangor Museum, No. 1138/10, skin andskull of adult male. Ko-ldiau, Tarang, January 12, 1910. I havenot seen this type.GeograpTiic distribution.?Tarang and extending northward towardsouthern Tenasserim. See No. 5 on map on page 75.Diagnosis.?The brightest colored member of the species; rump moreta^vny, larger than the insular forms except T. lacernata longicauda;tail not so blackish as in the forms on the west coast islands, butdarker than those on the east coast islands.Color.?As described above under the species, brighter than any ofthe subspecies, tail moderately blackish, rump rather tawny; shoulder
NO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 53Slcull and teeth.?Aside from their relatively larger size, the skulland teeth of T. lacernata wiTkinsoni do not differ essentially fromthose of the related subspecies. (Plate 2 and plate 8, fig. 3.)Measurements.?Collector's external measurements taken in theflesh (probably the type but not stated in the description) : Head andbody, 180 mm.; tail, 175; hind foot, 42; ear, 16. Cranial measure-ments: Probably the type; greatest length, 51.8; basal length, 44.9;zygomatic breadth, 25.9; cranial breadth, 20.9. Usual measure-ments of adults: Head and body, 180-195; taU, 150-170; hmd foot,43-45; condylo-basal length, 47-49; zygomatic width, 24-26; widthof brain case, 19-20; maxillary tooth row, 18-19. See table below.Specimens examined.?Nineteen, in the collections of the UnitedStates National, British, and Selangor Museums. See table below.Measurements of Tupaia lacernata wilkinsoni, longicanda, ohscura.
Name.
54 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.
,Geographic distribution.?Islands of Lankavi and Terutau, includingadjacent islets, west coast of IMalay Peninsula. See No. 23 on mapon page 75.Diagnosis.?Differs from Tupaia lacernata wilkinsoni of the oppositemainland in its slightly smaller size and less tawny rump, darker tail,and less conspicuous shoulder stripe, and slightly smaller skull andteeth.Color.?As described above under the species, the tawny element ofthe rump less pronounced than in the mainland form, and mth thedark elements of the tail when viewed from above more conspicuousand contrasted with the color of the back.Slcull and teeth.?These are nearly identical m appearance withthose of the mainland animal, but on the whole appear slightlysmaller. (Plate 8, fig. 5.)Measurements.^Type: Head and body, 180 mm.; tail, 155; hindfoot, 44; condylo-basal length, 47.5; zygomatic width, 25; width ofbramcase, 20; maxillary tooth row, 19. Usual measurements : Headand body, 170-185; tail, 140-150; hmd foot, 42-44; condylobasal,45-47, zygomatic width, 24-25; %vidth of brain case, 19; maxillarytooth row, 18. For details of measurements see table, page 57.Specimens examined.?Seventeen from Pulo Terutau and 14 fromPulo Lankavi. TUPAIA LACERNATA RAVIANA Lyon.1911. Tupaia raviana Lyon, Proc. Biol. Soc. Wash., vol. 24, p. 167, June 16, 1911.Type-locality.?Pulo Rawi, Butang Islands, off west coast of MalayPeninsula.Type-specimen.?In United States National Museum, Cat. No.104355, Pulo Rawi, December 18, 1899; collected by Dr. W. L.Abbott; original number, 172; in good condition.Geographic distribution.?Butang Islands, west coast of MalayPeninsula. See No. 24 on map on page 75.Diagnosis.?Similar to Tupaia lacernata lacernata and T. lacernatawilkinsoni, but skuU generaUy wider, and rostrum especially thickerand wider, but somewhat lighter in color, and not so tawny on therump.Color.? Tupaia lacernata raviana differs in color from T. I. lacernatain having the light annulations of the hau-s less ochraceous and morebuffy, and with only a trace of tawny on the rump ; the black elementof tail is less in evidence, so that the tail as a whole is somewhat lighter;the feet are grayer and less ochi*aceous than they are m T. I. lacernata;the underparts and shoulder stripe are not different in the two forms.Slcull and teeth.?In general, the skuU and teeth of Tupaia lacernataraviana resemble those of T. I. wilkinsoni and T. I. lacernata, but theskull is noticeably wider, with a tliicker and wider rostrum, and morespreading zygomata. The skull differences whUe not appearing con-siderable in a description, are quite marked when skuUs of the two
NO. 1976. TREE8HREW8: FAMILY TUPAIIDAH?LYON. 55forms are viewed together, and are greater than the differences seenin the skms. (Plate 8, fig. 4.)Measurements.?Type: Head and body, 184 mm.; tail, 165; hmdfoot, 46; condylo-basal length, 46.5; zygomatic width, 25.5; width ofbrain case, 19; maxillary tooth row, 17.5; width of rostrum back ofincisors, 7. For measurements of a second specimen from PuloAdang, Butang Islands, see table, page 57.Specimens examined.?Two from the Butang Islands, the type fromPulo Rawi, and one from Pulo Adang.TUPAIA LACERNATA OBSCURA Kloss.1911. Tupaia obscura Kloss, Aun. Mag. Nat. Hist., ser. 8, vol. 7, p. 116, January,1911.Type-locality
.
?Great Redang Island, off east coast of MalayPeninsula, near Tringanu.Type-specimen.?In Selangor Museum, No. 2279/10. Skin andskull, adult male, Great Redang Island, September 2, 1910; col-lected by Mr. C. Boden lOoss, original number, 3708. I have not seenthe type-specimen.Geographic distrihution.?Gresit Redang Island. See No. 39 onmap on page 75.Diagnosis.?Differs from the mainland Tupaia lacemata wiTkinsoniin possessing a generally distinctly lighter coloration and smallersize.Color.?Based on two topotypes kindly loaned by the SelangorMuseum, Reg. Nos. 2282/10, and 2287/10. Neither of the two spec-imens are in uniformly the same pelage, both having a newer anddarker pelage anteriorly, and an older and lighter posteriorly. Gen-eral color of the upper part of head, neck, and body and outer side oflegs a fine grizzle of blackish and tawny ochraceous (darker pelageareas) and of blackish and ochraceous (lighter pelage areas), theochraceous or tawny ochraceous always in excess; tail, above, auniform grizzle of blackish and cream buff, the darker color slightlyin excess; tad, below, similar but the lighter color predominating,particularly in the middle Ime; under parts cream color to almostbuff yeUow; shoulder stripe well defined, whitish or cream color.Skull and teeth.?The skuU and teeth of Tupaia lacemata ohscuradiffer from those of the related forms only m their generally smallersize and somewhat relatively larger bullae.Measurements.?"Collector's external measurements of type:Head and body, 173; tail, 167; hind foot, 40; ear, 15. Cranial meas-urements : Greatest length, 48 ; * * * basal length, 42; * * *greatest cranial breadth, 19.1; zygomatic breath, 25.8." For meas-urements of two topotypes, see table page 53, and see also Kloss.,Jour. Fed. Malay States Museum, vol. 4, p. 192, October, 1911.Specimen examined.?Two loaned by the Selangor Museum.
56 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.TUPAIA LACERNATA LONGICAUDA Kloss.1911. Tupaia ferruginea longicauda, Kloss, Ann. Mag. Nat. Hist., ser. 8, vol. 7,p. 117, January, 1911.1911. Tupaia longicauda, Kloss, Journ. Fed. Malay States Mus., vol. 4, p. 190,October, 1911.Type-locality.?East Perlientian Island, off Tringanu, east coast ofMalay Peiiiiisula.Type-S2)ecimen.?In Selangor Museum No. 2295/10, skin and skulladult female, East Perlientian Island, September 8, 1910. Collectedby C. Boden lOoss; original number, 3517. I have not seen thetype-specimen.Geographic distribution.?East and West Perhentian Islands offeast coast of Malay Peninsula. See No. 38 on map on page 75.Diagnosis.?Differs from the other subspecies in its longer tail andlighter color.Color.?Based on two specimens loaned by the Selangor Museum,Nos. 2303/10 and 2315/10. The general color of Tupaia lacernatalongicauda is essentially the same as it is m the mainland T. lacernatawiTkinsoni, but the ochraceous and tawny ochraceous bands on thehairs are wider and more conspicuous, and the lighter color is slightlyin excess of the black; the grizzlmg is coarser, and there is less con-centration of the tawny color on the rump, and the anterior partshave no indication of the slight olivaceous tint seen m T. lacernatawilkinsoni; essentially the same colors are found in the tails of T.lacernata longicauda and wiTkinsoni, but the black is not m excess inT. lacernata longicauda; the under parts of the two forms are essen-tially alike; the hands and feet of T. I. longicauda are more ochraceousthan they are in T. I. wilkinsoni.Skull and teeth.?No appreciable differences exist between the skulland teeth of Tupaia lacernata longicauda and T. lacernata wilkinsoni.Measurements.?"Collector's external measurements of the type:Head and body, 178; tail, 192; hind foot, 44; ear, 16. Cranialmeasurements: Greatest length, 51.7; basal length, 44.7; * * *greatest cranial breadth, 19.5; zygomatic breadth, 26.2." For meas-urements of two topotypes see table, page 53, and for measurementsof additional specimens see Kloss, Journ. Fed. Malay States Mus.,vol. 4, p. 193, October, 1911. The average external measurementsof 24 adults given by Mr. EHoss is: Head and body, 177. 5; tail, 185.6;and hind foot, 43.5. A few specimens have the tail actually shorterthan the head and body, 2304/10, 179, 176; 2315/10, 190, 184, and inone case equal 2214/10, 178, 178.Specimens examined.?Two loaned by the Selangor Museum.Remarks.?The two treeshrews just described are closely related toone another as well as to the mainland form from the middle of theMalay Peninsula from which they have evidently been derived.Although somewhat more hke Tupaia helangeri in point of colorationthan the other subspecies, yet the mammae are 2-2=4. Typical
NO. 1976. TREE8HREW8: FAMILY TUPAIIDjE?LYON. 57Tupaia glis ferruginea is found on the nearby mainland of Tringanu.T. I. ohscura and longicauda are quite different from the tupaiasfound on Tioman, Pemangil, and Aor, which are distinctly of theferruginea type.Mr. Kloss ^ makes the following interesting observations on thehabits of this treeshrew: "Of the numerous species of Tupaia which Icollected personally, T. longicauda with T. mcoharica, Zelebor and itssubspecies T. (N.) surda, Miller, alone are truly arboreal in habit. Asa rule the so-called 'treeshrews' are seen and trapped on the groundwhere they Hve and feed, or, at most, climb occasionally into lowbushes. In them the tail is shorter than the head and body length.The above-named animals, which are met with in high trees and havethe habits of squirrels, all possess a tail that is considerably longerthan the length of head and body."Measurements of Tupaia lacernata lacernata and T. lacernata raviana.
Name.
58 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.TUPAIA DEMISSA Thomas.1888. Tupaja ferruginea var. chrysura Jentink (not of Giinther), Notes LeydenMuseum, vol. 11, p. 29, 1889. Type-locality, Deli, Sumatra.1904. Tupaia ferruginea dcmissa Thomas, Zool. Anz., vol. 27, p. 723, July 12,1904.1905. Tupaiaferruginea demissa, Schneider, Zool. Jahrb., vol. 20, p. 8C, 1905.Type-locality.?Tanjong Bringin, lower Langkat, northeasternSumatra.Type-specimen.?In British Museum, Reg. No. 4.6.9.1, skin andskull of adult female, collected at Tanjong Bringin, lower Langkat,Sumatra, February 8, 1898, by Gustav Schneider. Originally pre-served in alcohol, but made into a skin probably about 1904. Thecolors do not appear to have been affected by the alcohol. Typespecimen in good condition.Geographic distribution.?Deli-Langkat region, northeastern Suma-tra. It was not found by Dr. W. L. Abbott immediately north oflower Langkat, in the vicinity of Aru Bay, nor in the region of theSiak River a short distance to the southeast. See No. 7, on map onpage 75.Diagnosis.?In all respects like T. glis ferruginea except that thecolor of the tail is cream buff; mammae, 2-2 = 4.Color.?The color of Tupaia demissa, with the exception of thetail, is so like that of T. g. ferruginea that no detailed description isnecessary. The tail, except its base, which is like adjacent parts ofbody, is cream buff in color throughout its whole extent on bothsurfaces, although a few brownish hairs may mar the clearness of theupper view.STcull and teeth.?As in Tupaia glis ferruginea.
' Measurements.?Type: Hind foot, 44 mm.; condylo-basal length,49.5; zygomatic width, 27; width of brain case, 20.5; maxillarytoothrow, 19. For individual measurements, see table, page 59.RemarTiS.? Tupaia demissa is a very distinct member of the 'ferj^u-ginea" group. Although described as a subspecies, and occurring onthe same land mass with T. g. ferruginea, I have here called it a fullspecies because so far as I am aware there is no evidence of it inter-grading with the usual form. It appears to be a well-establishedcolor anomaly of T. g. ferruginea quite parallel with the case of Tanachrysura of Borneo. Because the tail has a uniform color and is nota decided mixture of blackish and some buffy or rufescent color,Tupaia demissa might with a certain degree of propriety be classedwith the splendidula group, but I am much more inclined to considerit a perpetuated case of partial albinism in the ferruginea group.Specimens examined.?Six, all from Deli or Langkat. In additionto these I have seen a specimen of this species on exhibition in theNatural History Museum at Strassburg, collected in "Sumatra" in1903. A specimen of this species is recorded by Jentink in the Ley-
NO. 1976. TREE8HREW8: FAMILY TUPAIID.^?LYON. 59den Museum as Tuimia ferrucjinea chrysura in 1888, collected at Deliby Doctor Hagen. Doctor Jentink considered it a mere color freak,and tentatively applied the name chrysura in analogy with Twpaiatana var. chrysura Giinther of Borneo.Measurements of Tu paia demissa.
Locality.
60 PROCEEDINOS OF THE NATIONAL MUSEUM. vol.45.Clara, James, St. Luke, and St. Matthew of the Mergui Archipelago,and probably other islands of the same group. See No. 2 on mapon page 75.Diagnosis.?BiiieYS from T. glis ferruginea m generally lighter andmore grajash coloration and from T. chinensis in havuig a distinctochraceous wash over the rump and lower back; mammae, 3-3 = 6.(7oZor.?Upper parts of head, neck, anterior half of back and outerside of forelegs a fine distinct grizzle of cream buff and blackish, thebuffy color slightly in excess; lower back, rump, base of tail, outer sideof hind legs a fine grizzle of ochraceous and blackish, the ochraceousin excess; lower back distmctly different in color from upper back,but the one imperceptibly mergmg into the other; tail above usuallyintermediate in color between anterior and posterior back, but thegrizzlmg less distuact and with an excess of the darker color; under-side of tail similar to upper, but lighter and more buffy, and thegrizzlmg less distmct; underparts, mcluding inner sides of legs, creamcolor to buff yellow.SJcull and teeth.?The skull and teeth of Tupaia helangeri show nopecularitics to distinguish them with certainty from those of relatedspecies; the rostrum is relatively shorter than it is in T. glisferruginea,and the skull as a whole slightly smaller than it is in T. glisferruginea,and slightly larger than in T. chinensis. (Plate 8, fig. 2.)Measurements.?Usual measurements of adults: Head and body,175 to 190 mm.; tail, 150 to 160; hind foot, 42 to 45; condylo-basallength, 46 to 48; zygomatic width, 24 to 25; width of braincase, 18to 19; maxillary tooth row, 18 to 19. For details of measurements,see table, pages 62 and 63.Specimens examined.?Tenasserim and Pegu, 30, Mergui Archi-pelago, 26, from the follo^ving islands: Bentink, 2; Domel, 4; Kis-sering, 3; Sullivan, 2; Clara, 1; James, 8; St. Luke, 1; St. Mat-thew, 5.Reinarlcs.? Tupaia helangeri as represented by specimens from theMergui Archipelago and the mainland opposite is a well marked form.From the type-locality, however, specimens are less differentiated.In the British Museum are two skins from Rangoon, one of which,Reg. No. 6.4.5.3, is quite typical of the species as here described; theother, Reg. No. 7.7.20.7, is scarcely ochraceous posteriorly and bearsconsiderable resemblance to T. chinensis. Both these skins were col-lected in February of different years, and both are adult. It wouldappear that T. helangeri is not as differentiated at the type-locality asit is further southward along the coast. As origuially describedTupaia helangeri does not appear to be different from T. chinensis.In the old accounts mention is not made of the distinct ochraceouswash over lower back, but as the species at that time was bemg sepa-rated from T. ferruginea, this is not surprismg. The old specimen,
NO. 197G. TREESHREWS: FA2IILY TUPAIIDM?LYOS. 61that may be taken as the type of the species, has been exposed to thelight too long to be of value in determining whether it was a typicalT. hclangeri or approached T. chinensis. The f::cts appear to be thatlelangeri is the oldest name for the continental Tupaias that are notferruginea. From the description and supposed type-specimen, thename helangeri might be applied to what I call in this paper helangerior chinensis, but at the type-locality of helangeri occur treeshrewscertainly belonging to what is here called helangeri; and chinensis hasbeen proposed by Anderson for uniformly grizzled grayish treeshrewsfarther northward, thus leaving helangeri perfectly available for theTenasserim animal.The relationship between Twpaia helangeri and T. chinensis seemsintimate, and it would not be surprising if the two forms were foundto intergrade, helangeri being confined to the coastal region andchinensis to the higher region of the interior. As it is, many of thespecimens examined are not typical of helangeri, among them BritishMuseum, Reg. No. 7.7.20.7 Rangoon; Reg. No. 85.8.1.82 Meetan;Reg. Nos. 82.11.18.1, 85.8.1.86 Thaungyeen Valley, and Reg. No.85.8.1.90 Bankasun, and the two Kokareet specimens in Genoa.Before determinmg this point it would be desirable to obtain goodseries of skins from the mouths of the Irawadi and Salwen Riversand at various points from along the river until the upper courses inor near Yunnan are reached.The specimens in the United States National Museum from thevarious islands of the Mergui Archipelago are fairly uniform in mostof their characters. A few differences in size or color are indicated insome of them, but it does not appear possible to divide them intogeographic races, or to separate the island forms as a whole from thoseof the adjacent mainland. On the mainland, Bentink, St. Luke, St.Matthew, the skms appear brighter on the lower back, and on Domel,Sullivan, Clara, James, and Kissering, the skms are slightly duller onthe lower back, and perhaps darker anteriorly. A few slight differ-ences in size are revealed by examining the table of measurements.All the St. Matthew specimens have a maxillary toothrow of 19 mm.or over, while on James Island it is 18.5 or under, and in a singlespecimen from St. Luke Island it is only 17.5. But these extremesare all bridged over by intermediate specimens from other islands orfrom the mainland.A treeshrew probably related to Twpaia helangeri occurs onPreparis Island, between the Andamans and Pegu. A specimenfrom there is recorded by Anderson in the Catalogue of Mammals ofthe Indian Museum, Calcutta, part 1, page 155, listed as "hh." It issaid to be darker than the mainland specimens, and undoubtedlyrepresents an undescribed form.
62 PROCEEDINGS OF TEE NATIONAL MUSEUM. YOL. 45.Measurements of Tupaia belangeri.
Locality.
NO. 1976. TREESHREWS: FAMILY TUFAIID^?LTON. 63Measurements of Tupaia belangeri-
64 PROCEEDINGS OF THE NATIONAL 3IU8EUM. vol. 45.Diagnosis.? Tupaia chinensis is characterized by a unifromgrizzled olivaceous gray color without ferruginous on upper parts, orochraceous colors on rump; skull slightly smaller than in T. glisferruginea, with relatively shorter rostrum; mammas, 3-3=6.Color.?Upper parts of head, neck, and body and outer side of legsa fine uniform grizzle of blackish and a color that varies from buff toochraceous in certain individuals, the lighter colors predominatinganteriorly; both colors are about equally mixed, in some individualsthe lighter colors are in excess, and in others, especially in the middleline posteriorly the black admixture is sometimes in excess ; the tailabove is a coarse grizzle, sometimes showing indistinct annulationsof the same colors as have the upper parts of body, below the tailis lighter, especially in the middle line; the underparts of head, neck,and body, including the inner sides of the legs, vary from distinctwhitish, sometimes with the dark bases of the hairs showing through,to buffy; hands and feet similar to outer sides of legs but often lighterand grayer; shoulder stripe poorly developed and sometimes prac-tically obsolete. Three skins from Meechee, China, are quite lightand grayish, but are almost exactly matched by a British Museumskin from Manipur, Reg. No. 85.8.1.89. The Darjiling and two of theTura skins are rather dark, as are also the Siamese skins, none,however, are so generally dark as are Tupaia concolor and T. modesta.Slndl and teeth.?These are of the same general style as they are inT. glis ferruginea, but average slightly smaller and have a relativelyshorter rostrum, so that the distance from the end of the premaxillaryto the lachrymal notch is generally distinctly less than the distancefrom the notch to the external auditory meatus. Although T.chinensis was originally separated from T. hclangeri mainly on skullcharacters, I have been able to find no satisfactory constant char-acters to distinguish skulls of the two forms. The individual varia-tion in skulls of Tupaia is quite considerable, and with a relativelysmall number of specimens such as Anderson seems to have had itwould be comparatively easy to find distinguishing features. Iregret that I have not seen his type or cotypes. (Plate 8, fig. 1.)Measurements.?Anderson's measurements of the cotypes con-verted to millimeters: Head and body, , 165 mm.; tail, ,156; hind foot, , 40; inferior margin of foramen magnum to tipof premaxillaries, 39, 40 (making a condylobasal length of approxi-mately 42, 43) ; zygomatic width, 22, 23. The usual measurements ofadults corresponds very closely with those of the cotypes, the headand body measurement is often 5 mm. longer, but the tail in most ofthe specimens which have collectors' measurements is nearly alwaysfrom 5 to 15 mm. longer than head and body, but the skins andalcoholics of which I have taken approximate head and body, and tailmeasurements show the tail to be shorter than head and body. Themaxillary tooth row is about 17.5. For details of measurements, seetable, page 66.
NO. 1976. TREESHREWS: FAMILY TUPAIID.E?LYON. 65Specimens examined.?Forty-three, mostly in the collections of theBritish Museum, only one in the United States National Museum.For list of specimens, see table, page 66.Remarks.?^The specimens which I have included under Tupaiachinensis constitute a somewhat heterogeneous collection. With theT. glis fcrruginca group I have recognized many slightly differen-tiated geographic forms, mostly insular, but mth Tupaia chinensisI have been extremely conservative and have not A^entured todescribe some color variations that are as pronounced as some ofthe color variations in the ferruginea group. This is largely fromlack of adequate material and to the fact I have not seen the cotypesor even topotypes of T. chinensis. Many of the localities are repre-sented by only single specimens instead of adequate series and areunaccompanied by notes as to altitude. The specimens that onemight be mclined to recognize as races of T. chinensis are the threeMeechee ^ specunens very light in color, though sho\\dng degrees oflightness among themselves and the rather full-pelaged oUvaceousspechnen from Jerkalo on the Thibet boundary. Light as are theMeechee specimens compared Avith the majority of the others, yetthey are nob more different from them than two specimens fromTura, Assam (American Museum of Natural History, Cat. Nos.26843 and 26841), showing there may be considerable mdividualvariation. Until more material is at hand, with carefully workedout locahties it seems best for the present to refer all the northernuniformly grizzled grayish continental treeshrews to the singlespecies Tupaia chinensis. The relationship of Tupaia chinensis toT. lelangeri is not perfectly clear. What I have called T. helangeriis t}"pical in the Mergui Archipelago and adjacent mauiland, andis certainly a very different anunal from T. chinensis as foundaway from the seacoast, back m the interior. I am free to admitthat I have seen certain specimens from Tenasserim, particularlysome to the east of Moulmein that could Vsdth considerable pro-priety be placed in either species, and I strongly suspect thatfuture collections, with carefully identified localities and altitudes,\at11 show that Tupaia chinensis is a subspecies of T. helangeri. Ifthat should prove to be the case, the relation of them to Tupaiaglis ferruginea will be mtercsting. At present T. chinensis and T.helangeri appear to be sharply separated from T. g. ferruginea andT. lacernafa ivilkinsoni by the presence of six instead of four mammasas was pointed out by Thomas^ in 1891. In spite of that marked
* The only Meechee that I have been able to find on modem maps is Mitschi (see p. 75). The threeMeechee specimens v.'ere collected by Styau and are labeled Meechee, Yunnan. No Meechee appearson the numerous maps of Yuiman that T have examined. As to the exact locality of the Meechee speci-mens I can not say. It is very doubtful if treeshrews occur as far north in China as Mitschi. PerhapgMeechee is only a small village in Yunnan and the three specimens may be virtual topotypes,2 Arm. Mus. Civ. Stor. Nat. Gcnova, scr. 2, vol. 10, p. 920, 1890-91.S0i59??Proc.N.M.vol.45?13 5
66 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.difference, however, T. lacernata wiUdnsoni iii point of color appearsas an almost perfect intermediate between Tupaia glis ferrugineaand T. helangeri. With the certain rather doubtful specimens fromTenasserim it may ultimately develop that but a single speciesof Tupaia occurs on continental Asia ranging from Singapore toMeechee and from Nepal to Tonkin, but wherever there are sufficientclimatic or physiographic differences subspecies have been produced,of which five continental forms have been described
?
T. glis ferrugi-nea, lacernata wilkinsoni, helangeri, chinensis, concolor.Measurements of Tupaia chinensis.[A. M. N. H.=American Museum of Natural History, New York. B. M.=British Museum.]
Locality.
NO. 1976. TREESHREWS: FAMILY TUPAIIDJE?LYON. 67TUPAIA DISSIMILIS (Ellis).1860. Sciurus dissiniilis Ellis, in Gray, Ann. Mag. Nat. Hist., ser. 3, vol. 6, p. 71.Type-locality.?Pulo Condore, off south coast of Siam.Type-specimen.^iso type-specimen ever existed so far as known.This species was thought to be a pecuHar squirrel by W. Ellis, a sur-geon on Captain Cook's third voyage. The expedition stopped at PuloCondore, 1780, in the latter part of Januar}^ Ellis wrote a descrip-tion in Latin, published by Gray in 1860, paid made an excellent draw-ing of the entire animal and of its anterior teeth. Through the kind-ness of the authorities of the British Museum, Natural History, inwhose library Ellis's manuscript and drawings are now kept, a photo-grapliic reproduction of this picture appears as plate 1.Geographic distribution.?Pulo Condore. See No. 27 on map onpage 75.Diagnosis.?A geographic form of Tupaia cliinensis distinguishedby its smaller size ; hind foot, 38-40 ; T. cMne7isis usually over 40 mm.
;
mammae unknown.Color.?Based on Cat. No. 3745, Berlin Museum, originally re-ceived from the Paris ]\Iuseum, an old mounted specim.en with skullremoved. Upper parts of head, neck, and body, a grizzle of ochra-ceous and blackish, the two colors about equally mixed, tail similarbut grizzle coarser; ochraceous color on head slightly lighter thanon body; outer side of legs similar to adjacent parts of body, under-parts and inner side of legs with much hair gone and soiled, appar-ently dull buffy; underside of tail in middle line similar to rest ofunderparts; margins of tail underneath a coarse mixture of buffy andblackish.STcull and teeth.?Of similar form to those of Tupaia helangeri (com-parison made with Berlin Burma specimen), but smaller, rostrumnarrower, and brain case decidedly narrower; teeth similar to thoseof T. helangeri, but m^ and m^ shorter.Measurements.?Hind foot, 38-40 mm.; condylo-basal length about45; zygomatic width, 23-24; \vidth of brain caise, 18-19; maxillaiytoothrow 17. In mounted specimens the head and body is 180-200, and the tail 140-165 mm. See table, page 70.Specimens examined.?Three, two in Paris, and one in Berlin, allprobably collected b}^ Germain, in 1882.Remarks.?There can be little doubt as to the distinctness of T.dissimilis from T. chinensis, the rather limited material showing itto be distinctly smaller. Probably these are good color charactersas well, but at present there is not sufficiently good material of eitherspecies to point them out. Although tliis was the first species of atreeshrew to come under the observation of Europeans, this is thefirst time it has been given recognition as a species.
68 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.TUPAIA CONCOLOR Bonhote.1907. Tupaia concolor Bonhote, Abstr. Proc. Zool. See. London, p. 2, January22, 1907 (also see Proc. Zool. Soc. London, 1907, p. 7, June, 1907).Type-locality.?Nhatrang, on the coast of Aimam.Type-specimen.?In British Museum, Reg. No. 6.11.6.3, skin andskull of an adult male, collected by Dr. J. Vassal, at Nhatrang,Aimam, March 22, 1906; original number, 59; in good condition.Geograpliic distribution.?Southern Annam and northern CochinChina, probably along the coast. See No. 3 on map on page 75.Diagnosis.?Similar to Tupaia chinensis but larger, more bushytail, and larger, longer skull; mammae probably 2-2 = 4.Color.?Type: Upper parts of head, neck, body, outer side of legs,and feet a grizzle of ochraceous buff and blacldsh, the two colorsabout equally mixed, but mid-dorsal area slightly darker; anteriorlythe light color is more buffy, and posteriorly more ochraceous, butthe differences not at all conspicuous; tail above similar to adjacentparts of body, but the grizzle much coarser; tail below with outerhalf similar to upper surface, central portion dull pale buffy, mixedwith blackish, hairs of tail very conspicuously annulated and whenartificially spread out, five distinct light bands may be seen, alter-nating with as many blackish ones; under parts generally dull buffy.Slcull and teeth.?Large and heavy when compared with Tupaiachinensis, with a narrowed rostrum, distance from laclirjTiial pit topremaxUla equal to distance from pit to center of external auditorymeatus.Measurements.?^Type: Head and body, 230 mm.; tail, 140; hindfoot, 43; condylo-basal length, 50; zygomatic width, 29; width ofbrain case, 21; maxillary toothrow, 20. For measurements of aparatype and four specimens from Cochin China, see table, page 70.RemarJcs.?Tupaia concolor is at once distmguished from T. chinensisby its much larger size, especially seen m its skull measurements.At the time it was described it was known only from the type-locality.There is one specimen m the Paris Museum, No. 1149, marked
''Cochin Chma," which e\ndently belongs to this same species. Itsexternal measurements are large, and its skull measurements agreevery closely with those of T. concolor. While I have not had theopportunity of comparing the two specimens directly, my notes showthat the Cochin Chma skull differed from the usual Siam skulls ofT. chinensis in nearly the same manner that T. concolor does. TheCochin CMna skull, however, is rather narrower and has less spreadingzygomata.* Three other specimens marked Cochin China in the ParisMuseum represented by skms only, I have assigned to T. concolormainly on geographic grounds. One of them, collected by Germain,
NO. 197G. TREESHREWS: FAMILY TUPAIIDJE?LTON. 69may have come from Pulo Condore, and one of the others has a dis-tinctly small hind foot like that of the Cordore animal. See table,page 70.Si^ecimens examined.?Six. See table, page 70.TUPAIA MODESTA Allen.1906. Tupaia modesta Allen, Bull. Amer. Mus. Nat. Hist., vol. 22, p. 481, Dec.17, 1906.Type-locality.?Island of Hainan, off the south coast of China.Type-specimen.?In American Museum of Natural History, NewYork, Cat, No. 26654, collected at Lei-Mui-Mon, Hainan, January 5,1903, through agents of Mr. Alan Owston; skin well preserved, butskull damaged posteriorly.Geographic distrihution.?Island of Hainan. See No. 4 on map,on page 75.Diagnosis.?A geographic form of Tupaia cliinensis distinguishedby its generally darker coloration, externally not unlike T. concolor,but distinctly smaller; mammse, 2-2 = 4.Color.?In general coloration Tupaia modesta is essentially likeT. concolor, but the underparts are more whitish, and when the hahsof the tail are artificially spread only three distinct buffy annulationsare seen instead of five.Shull and teeth.?The skulls of Tupaia modesta available for exami-nation are considerably damaged. Apparently they are not essen-tially different from those of T. cliinensis.Measurements.?Type: Hind foot, 46 mm.; zygomatic width, 25.5;width of brain case, 19.5; maxillary tooth row, 18. The type has thelargest liind foot in the series, most of the other specimens measuringonly 43 mm. See table, page 70.Remarl^s.? Tupaia modesta is quite distinct in its generally darkercolor from most specimens of T. cliinensis, if not appearing distinctlydarker with reference to the upperparts, the tail and underpartsappear so. As to how different it is from tree shrews on the mainlandadjacent to Hainan it is impossible to say. The nearest specimensgeographically that I have seen are two from Tonkin in alcohol,young, and so useless for comparison. The two Mongtsze specimensare \evj dark above, but are distinctly whitish underneath. It is tobe observed that Tupaia concolor of the southeast coast region of Asiais also a dark-colored animal, but distinctly larger than T. modesta.The number of mammae, 2-2 = 4, in this species and the preced-ing is interesting as in all the other continental treeshrews north ofthe Malay Peninsula, the number is 3-3 = 6. As only one specimenin each species is available for determining the number of mammae,too much weight can not be attached to this peculiarity.Specimens examined.?Seven, from various localities in Hainan.See table, page 70.
70 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Measurements of Twpaia concolor, dissimilis, and modesta.
Name.
NO. 1976. TREE8I-TREWS: FAMILY TUPAIIDJE?LYON. YlSJcuU and teeiJi.?The skull and teeth of Tupaia liypoclirysa are notfundamental^ different from those of T. g. ferruginea. The skullaverages larger as a whole, the rostrum is relatively long and heavy,the distance from the lachrymal notch to tip of premaxillary is equalto distance from notch to posterior edge of external auditory mea-tus; the bullm are smaller than they are in T. g. ferruginea; zygo-matic arch wide and heavy and strongly marked anteriorly forinsertion of muscles; the teeth are larger and heavier and the tooth-row as a whole distinctly longer. (Plate 9, fig. 6.)Measurements.?Type and a specimen from Mount Salak, Cat. No.154599, U.S.N.M. Head and body, , 145 mm.; tail, , 145;hmd foot, 48, 49; condylo-basal length, 51.5,51; zygomatic width,2G, 28; width of brain case, 19.5, 21.5; maxillary tooth row, 21, 20.5.Remarks.?Y^hxXe Tupaia liypoclirysa is probably the Javan repre-sentative of T. glis ferruginea, yet it is a very distinct species.When first described specimens in alcohol from Sipora, MentaweiIslands, were regarded as being the same species. "While they seemto belong to the same group as T. liypoclirysa, I have identified themwith Tupaia clirysogaster from the geographically nearer Pagi Islands,and with which they seem to agree more closely with respect tomeasurements. Tupaia liypoclirysa has many resemblances toTupaia longipes and discolor of Borneo and Banka. Tupaia liypo-clirysa is one of the few species of treeshrews whose number of mammseis unknown, and in this particular instance knowledge on that pointis of much importance to show probable afFmities. T. clirysogasterof the Mentawei Islands has the mammae 1-1 =2, while in T. longipesand T. discolor they are 3-3 = 6.Specimens examined.?Three, the type from "Java" and a secondspecimen from "Java" and a third from 3,500 feet on Mount Salak,western Java. See table, page 72.TUPAIA CHRYSOGASTER Miller.1903. Tupaia chrysogaster Miller, Smiths. Misc. Coll., vol. 45, p. 58, November6, 1903.Type-locality.?North Pagi Island, off southwest coast of Sumatra.Type-specimen.?In United States National Museum, Cat. No.121752, skin and skull of adult female collected on North Pagi Island,November 21, 1902, by Dr. W. L. Abbott; original number, 2078; ingood condition.Geographic distribution.?North and South Pagi, and Sipora of theMentawei Islands, off the southwest coast of Sumatra. See No. 16 onmap on page 75.Diagnosis.?Above, including tail, finely grizzled, blackish andochraceous rufous; below, clear ochi-aceous rufous; mammas, 1-1=2.Color.?Upper parts of head, neck, body, taO, and outerside of legsa fine grizzle of blackish and ochraceous rufous, both colors about inequal proportions; on nose the ochraceous rufous lightens to rawsienna; underside of tail similar to upper but orange ocliraceous
72 PR0CEEDIN08 OF THE NATTONAL MUSEUM. VOL. 45,lighter and duller, and in greater proportion than the black element;entire underparts, including inner side of legs, oclu^aceous rufous,much clearer than in Tupaia hypochrysa; hands and feet blackishbrown, slightly and finely grizzled with an ochraceous color; shoulder-stripe barely indicated, sometimes practically obsolete, ochraceousrufous.Slcull and teeth.?The skull and teeth of Tupaia chrysogaster areessentially like those of T. hypochrysa, but the rostrum is distinctlyslenderer, and the teeth not so large and heavy and the bullae not soreduced ; in other respects the two skulls are similar. The slendernessof the rostrum suggests the genus Tana. (Plate 9, fig. 9.)Measurements.?Type: Head and body, 205 mm,; tail, 140; liindfoot, 46; condylo-basal length, 51.5; zygomatic width, 27; width ofbrain-case, 19.5; maxillary toothrow, 20. These measurements arequite characteristic of the series, which show little individual varia-tion. The tail is seen to be much shorter than the head and body.For individual measurements, see table below.Specimens examined.?Thirteen; six from North Pagi, four fromSouth Pagi, one marked simply Pagi Islands, and two from SiporaIsland.Remarlcs.? Tupaia chrysogaster is a very distinct species and needsno close comparison with any other member of the genus. It has fewaffinities with the treeshrews of Sumatra, the nearest land mass, but isclearly related by the size and form of its skuU and color of the under-parts to Tupaia hypochrysa of Java.Measurements of Tupaia chrysogaster and T. hypochrysa.
NO. 197G. TREE8HREW8: FAMILY TUPAIIDJE?LYO'M. 73TUPAIA DISCOLOR Lyon.1906. Titpaia discolor Lyon, Proc. U. S. Nat. Miis., vol. 31, p. 602, December 18,1906.Type-locality.?Island of Banka, east of Sumatra.Type-specimen.?In United States National Museum, Cat. No.124703, skin and skull of adult female, collected at Tanjong Rengsam,Banka, May 24, 1904, by Dr. W. L. Abbott; original number, 3262;in good condition.Geograpliic distribution.?Island of Banka. See No. 12 on map onpage 75.Diagnosis.?A treeshrew of the T. glis ferruginea build with thegeneral color effect of the anterior parts rather ferruginous, and theposterior parts rather tawny olive, underparts ochraceous; mammas,3-3 = 0.Color.^U\vpev parts of head, neck, anterior half of bod}^, and outerside of forelegs a fine grizzle of black, and ferruginous, the lattercolor in excess; posterior half of upper parts, with base of tail, andouter side of hind legs a fine grizzled mixture of oclu'aceous buff andblackish, both colors in about equal proportions; upper surface oftail a grizzle of blackish and cream color or bufi' ; the black being muchin excess; underparts of head, neck, and body, including inner side oflegSj varjdng from oclu*aceous buff to dull orange ochraceous ; under-side of tail similar to upper, but the lighter color in excess along itscenter; hands and feet blackish brown, with a very fine ochraceousgrizzle; shoulder stripe well developed, orange rufous.SlcuU and teeth.?These are distinctly of the ferruginea type, butthe teeth are noticeably smaller, the bullae larger, and brain case morerounded and inflated. (Plate 9, fig. 4.)Measurements.?Type: Head and body, 220 mm.; tail, 175; hindfoot, 48; condylobasal length, 48.5; zygomatic width, 26; width ofbrain case, 19; maxillary tooth row, 18.5. Except in length of toothrow and width of brain case the type has measurements slightly inexcess of the majority of specimens. For individual measurements,see table, page 78.Remarlcs.? Tupaia. discolor is a very distinct form, and along withT. longipes of Borneo constitutes a distinct section of the genus.Although clearly of theferruginea type it is very different in colorationand in number of mammae from that form. It is clearly a deriva-tive of the Bornean T. longipes and has no close affinities with T. glisferruginea of Sumatra. It is described here before T. longipes be-cause it is a more extrom.e development of the same type of animal.Although externally showing many affinities to T. liypoclirysa of Java,yet its skull shows none of the peculiarities of that species and is dis-tinctly of the wide rangmg ferruginea type.
74 PROCEEDINGS OF THE NATIONAL MU8EUM. vol.45.Specimens examined.?Fifteen, all from Banka?three from KlabatBay, on the northern end, and twelve from Tanjong Rengsam, upperpart of Banka Strait.EXPLANATION OF NUMBERS ON MAP FACING.1 T. chinensis: China, northern Burma.2 T. belangeri: Tenasserim.3 T. concolor: Anam.4 T. modesta: Island of Hainan.5 T. lacernata iviJHnsoni: Middle of Malay Peninsula.6 T. glis ferrurjinca: Southern Malay Peninsula, Sumatra.7 T. demissa: Northern Sumatra.8 T. siaca: Eastern Sumatra.9 T. longipes longipes: Northern Borneo.10 T. longipes salatana: Southern Borneo.11 T. hypochrysa: Java.12 T. discolor: Banka, east of Sumatra.13 T. palnwanensis: Palawan, Balabac, northeast of Borneo.14 T. mollendorffi: Culion, northeast of Borneo.15 T. cuyonis: Cuyo, northeast of Borneo.16 T. ch-ysogaster: Mentawi Islands, south of Sumatra.17 T. casianea: Bintang Island, south of Malay Peninsula.18 T. glis batamana: Batam Island, south cf Malay Peninsula.19 T. phccura: Sinkep Island, between Malay Peninsula and Sumatra.20 T. tephrura: Tana Bala, southwest coast of Sumatra.21 T. glis glis: Penang Island, west coast, southern Malay Peninsula.22 T. picta: Northern Borneo.23 T. lacernata lacernata: Lankawi, Terutau, west coast Malay Peninsula.24 T. lacernata raviana: Butang Islands, west coast, Jilalay Peninsula.25 T. montana montana: Mount Dulit, Northern Borneo.26 T. montana haluensis: Mount Kina Balu, northeastern Borneo.27 T. dissimilis: Condore Island, off Cochin China.28 T. glis sordida: Tioman Island, east of southern Malay Peninsula.29 T. glis pcmangilis: Pemangil Island, east of southern Malay Peninsula.30 T. glis pulonis: Aor Island, east of southern Malay Peninsula.31 T. liicida: Pulo I^aut, north of west end of Borneo.32 T. natmvx: Bunguran Island, north of west end of Borneo.33 T. chrysomalla: Siantan Island, between southern Malay Peninsula and Borneo.34 T. riabus: Riabu Island, between southern Malay Penin.sula and Borneo.35 T. anamhse: Jinmja Island, between southern Malay Peninsula and Borneo.36 T. carimatcv: Karimata Island, southwest coast of Borneo.37 T. splendidula: Southern Borneo.38 T. lacei'nata longicauda: Perhentian Island, east coast of Malay Peninsula.39 T. lacernata obscura: Redang Island, east coast of Malay Peninsula.40 T. nicobarica nicoharica: Great Nicobar Island, northwest of Sumatra.41 T. nicobarica surda: Little Nicobar, northwest of Sumatra.TUPAIA LONGIPES Thomas.(Synonymy under subspecies.)Geographic distribution.?Borneo. See Nos. 9 and 10 on map onpage 75.Diagnosis.?Similar to Tupaia discolor of Banka, but larger andwith less contrast in color between anterior and lower portions ofback; raammce, 3-3 = G.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 75
Map of the Malay Region, sho'Wing tue distribution of the forms of the genus Tupaiy,EXCEPTING the MEMBERS OF THE GRACILIS, JAVANICA, AND MINOR GROUPS.
76 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Color.?Upper parts of head, neck, and anterior half of body andouter side of forelegs a fine grizzle of black and tawny, the latter colorin excess; posterior portions of upper parts including base of tail,and outer side of hind legs a fine gi'izzly mixture of ocln^aceous buffand blackish, both colors in about equal proportions; upper surfaceof tail blackish brown faintly grizzled with a buff-like color; under-surface of tail similar, but the buffy color predominating in the middleline; anterior half of underparts light orange ochraceous, posteriorportions dull buffy, inner side of legs similar to adjacent portion ofunderparts, but colors duller; feet blackish brown with a very fewlight specks.Sl'uU and teeth.?Of the same general form as in Tupaia glisferruginea, but slightly larger tlu-oughout and very similar to theskull of T. discolor, with a similar inflation of the brain case and ratherenlarged balla?, but the skull and teeth as a whole decidedly largerthan in discolor.Measurements.?Head and body, 200 mm., or slightly more; tail,190; hind foot, 50-53; condylobasal length, 48-52; zygomatic width,2G-2S; width of brain case, 20-21; maxillary tooth row, 19-21.Forms.? Tupaia longipes is separable into two fau'ly well markedforms, T. longipes longipes from northern Borneo, and a form fromsouthern Borneo described below as new,TUPAIA LONGIPES LONGIPES Thomas.1893. Tupaia ferruginea longipes Thomas, Ann. Mag. Nat. Hist., ser. C, vol. 11,p. 343, May, 1893.1911. Tup>aia longipes, Lyon, Proc. U. S. Nat. Mus., vol. 40, p. 122, April 25,1911.Type-locality.?Northwestern Borneo.Type-specimen.?In British Museum, Reg. No. 7G.9.20.5, collectedin 1876, by II. Low.; a fairly well made skin in good condition; skullwith the occipital portion cut away after the manner of bird collectors.Geographic distribution.?Northern Borneo, specimens from Sara-wak to Mount Kalulong. See No. 9 on map on page 75.Diagnosis.?Tail and lower back more brownish, and less gray thanin the southern subspecies, and less contrast in color between anteriorand posterior portions of body; maxillary tooth row usually over20 mm. in length.Measurements.?Type: Hind foot, 51 mm.; zygomatic width, 27;width of brain case, 20.5; maxillary tooth row, 21. For measure-ments of other specimens, see table, page 77.Specimens examined.?Twelve, from northern Borneo. For exactlocalities, see table, page 77.
NO. 1976. TREESHREWS: FA3IILY TUPAIID.E?LYON. 77TUPAIA LONGIPES SALATANA, new subspecies.Type-locality.?Pangkallahan River, S. E. Borneo, 15 miles frommouth.Type-specimen.?In U. S. National Museum, Cat. No. 151882, col-lected along Pangkallahan Kiver February 11, 1908, by Dr. W. L.Abbott; original number, 5785; skin in good condition; skull dam-aged by shot in basal occipital region.Geographic distribution.?Southern Borneo, specimens from Ken-dawangen River region and Pangkallahan River. See No. 10 on mapon page 75.Diagnosis.?Differs from the northern race in having more contrastm color between anterior and posterior portions of back, more rufescenton the shoulders than is T. I. longipes and less than T. discolor; generaleffect of lower back and tail is much like clove brown, while in thenorthern race it is more like bistre; not so light on the lower back asis T. discolor; maxillary tooth row less than 20 mm. in length. (Plate9, fig. 5.)Measurements.?Type: Head and body, 213 mm.; tail, 185; hindfoot, 50; condylobasal length, 45; zygomatic width, 25.5; width ofbrain case, 20.5; maxillary tooth row, 19.5. For measurements offour other individuals, see table below.Specimens examined.?Five; four from southwestern and one fromsoutheastern Borneo. An old mounted specimen Is in the ParisMuseum, numbered 21, and marked simply "Borneo," collected byTemminck in October, 1842, hind foot measuring about 45 mm. Itmay possibly belong to the present subspecies.Measurements of Tupaia longipes and Tupaia discolor.
Name.
78 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Measurements of Tupaia longipes and Tupaia discolor?Continued.
Name.
NO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 79ochraceous buff; outer and inner side of legs similar to adjacentparts of body; feet similar to legs, but toes distinctly blackish; upperside of tail clear blacldsh; underside similar except for tawny ochra-ceous annulations on either side of the middle line and toward thebase.Not all specimens are as strongly characterized as the type; inmost of them the tail is not clear black above, in some cases beingblacldsh brown and nearly always with some fine ochraceous dots orannulations showing on the upper surface. Two specimens, Reg.No. 94.2.1.4 and 97.9.12.1, are very similar to the type in color, butthe specimens from Puerto Princessa in the Paris Museum are similarto Tufaia chinensis, but are darker above, and with distinctlydarker tails.SJcuU and teeth.?These do not show any distinguishing character-istics, being generally like those of Tupaia glis ferruginea.Measurements.?T;fpe: Head and body (from dried skin), 210 mm.;tail (from dried skin), 185? hind foot, 45; condylo-basal length, 50;zygomatic vddth, 26.5; width of brain case, 18.5; maxillary tooth row,18. Measurements of other specimens fully as old as the type areessentially the same except that the condylo-basal length is muchshorter in some, as small as 43 mm., and the hind foot is seldom aslong. For individual measurements see table, page 80.EemarJcs.?As represented by the type-specimen TujMia palawanen-sis is a very distinct form, and quite different from its geographicneighbor, T. longipes, of Borneo. T. longipes is finely grizzled on theupper parts, and its tail is essentially like the back in color, w^hile T.palawanensis is coarsely grizzled, and its black tail is very different inappearance from the back. TMiile I have not been able to comparethe Puerto Princessa specimens directly with the t3^pe of T. palawa-nensis, yet as far as can be told from memory they seem very differentin general style of coloration from the type. Although the tails aredarker than the upper parts they do not appear black, and the differ-ence in condjdo-basal length 50 in the type and 43 mm. in a PuertoPrincessa specimen with moderately worn teeth is considerable. Itis barely possible that more than one form of Tupaia m.ay occur onPalawan. It is to be noted also that the exact locality of the typedoes not appear to be known, the specimen being labeled simply
"Palawan." This species or a related one occurs on Balabac, rep-resented by British Museum, Reg. No. 94.7.2.55, a nearly adult femalepreserved in alcohol. Owing to its immaturity and manner ofpreservation it is not possible to say whether it is T. palawanensis orsome other form. Another specimen in the British Museum, notnumbered, collected by W. Doherty, is marked "Palawan or Basilan,"is not particularly different from T. palaivanensis, and for the time
80 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.being at least it seems best to regard it as having come from Palawan.In the Berlin Museum is a specimen collected by Mollendorf, agree-ing very well with T. palawanensis. It is labeled "CalamianesGruppe." If not coming from northern Palawan itself, it probablycame from some of the nearer islands at the northern end of Palawan.This specimen was called by Nehring/ in 1894, T. ferruginea. Thespecimen appears to be an old one and to have been mounted at onetime. The same name was applied to the British Museum speci-mens in 1889 by Everett.^Specimens examined.?Eighteen; 6 including the type from "Pala-wan"; 6 from Puerto Princessa, Palawan; 3 from Iwahig, Palawan;1 from the "Calamianes Gruppe"; 1 from Balabac; and 1 from"Palawan or Basilan." See table below.Measurements of Tupaia palaivanensis.[B. M.=British Museum, Lon-lon; P. M.= Philippine Museum, Manila; A. M. N. H. = AmericanMuseum of Natural History, New York.]
Locality.
NO. 1976. TREE8HREWS: FAMILY TUPAIID^?LYON. 81TUPAIA MOLLENDORFFI Matschie.1898. T[upaja] mollendorffi Matschie, Sitz.-Ber. Ges. nat. Freunde, Berlin, 1898,p. 39.Tijpe-locality.?Ciilion (also called Calimian), a small island northof Palawan, Philippine Islands.Type-specimen.?In the Royal Zoological Museum, Berlin, No.9858, skin and skull of adult male, collected on Culion, PhilippineIslands, by Dr. von MollendorfF; skin in good condition, but posteriorparts of the skull are lacking.Geographic distribution.?Island of Culion and possibly some of theimmediately adjacent islands. See No. 14 on map on page 75.Diagnosis.^\]^])Qi' parts finely grizzled ochraceous and blackish,tail coarsely black and buff, different, that is grayer, in color fromrest of upper parts. MammsB, 2-2 = 4.CoZor.?Upper parts ajid sides of head, neck, and body a fine grizzleof ochraceous (or ochraceous buff), and blackish, the lighter colorbeing slightly in excess, expecially along the sides and rump; under-parts, including throat and inner side of legs, usually dirty buff, orochraceous buff; in the type the chin and throat, cream buff, dis-tinctly lighter than rest of underparts; outer side of legs essentiallylike upper parts of body; feet similar, but the grizzle very fine, andthe dull ochraceous buff color predominating; shoulder stripe fairlywell defined, buffy; tail a coarse grizzle of buff and blackish, bothabove and below, the blackish color slightly in excess above, and thebuff below; tail from above with a distinctly grayer look than rest ofupper parts, and noticeably different in color.Slcull and teetli.?The skull of Tupaia mollendorffi, is smaller thanthat of T. glis ferruginea, relatively shorter, wider, with a relativelythicker rostrum which arises more abruptly from rest of skull.Aside from their slightly smaller size and relatively greater develop-ment of the central upper incisors the teeth of T. mollendorffi arenot essentially different from those of T. glis ferruginea.Measurements.?^Type: Head and body (dried skin), 200 mm.; tail,(dried skin), 160; hind foot, 43; zygomatic width, 24 ; width of braincase, 18 ; maxillary tooth row, 16.5. For measurements of three otherspecimens see table, page 83.RemarTcs.? Tupaia mollendorffi is quite different and apparentlyvery distinct from its geographic neighbor, T. palawanensis. It issmaller and its external appearance quite different from the typicalPalawan form. 'V'\'Tien compared with the Puerto Princesa (Palawan)specimens it is not so distinct, but its lighter colored tail serves todistinguish it easily. It was identified by Nehring ^ in 1894 asTupaia ferruginea.Specimens examined.?Four. See table, page 83.
1 Sitz. Ber. Ges. naturf. Frcundc, Berlin, 1894, p. 184.80459??Proc.N.M.vol.45?13 6
82 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.TUPAIA CUYONIS Miller.1910. Tupaia cuyonis Miller, Proc. U. S. Nat. Mus., vol. 38, p. 393, August 19,1910.Type-locality.?Cuyo Island northeast of Palawan, PhilippineIslands.Type-specimen.?In the collection of the Philippine Museum,Manila, Philippine Islands, No. 26, skin and skull of adult male,collected on Cuj^o Island January 15, 1903, by R. C. McGregor andA. Celestino; in good condition.Geographic distribution.?Known only from the Cuyo Island. SeeNo. 15 on map on page 75.Diagnosis.?Similar to Tupaia mbllendorffi; head and body of auniformly grizzled oclu'aceous and black coloration, but tail not dif-ferent in color from lower back. Mammse, 2-2 = 4.Color.?Type: Upper parts and sides of head, neck, and body a finegrizzle of oclu-aceous and black, the lighter color being slightly inexcess, especially along the sides and rump; under parts, includingthi'oat and inner side of legs, generally oclxraceous, darker anteriorlyand approaching buff posteriorly, the dark bases of the hairs showingthrough, giving an ill-defined grizzled appearance; outer side of legsessentially like the body, the feet similar, but grizzling finer; shoulderstripe ill defined, buff in color; tail above and below a coarse grizzleof ochraceous and blackish, the darker color slightly in excess aboveand the lighter color below.STcuU and teeth.?These are without special peculiarities, distinctlysmaller than those of T. glis ferruginea; brain case relatively wide,rostrum rather short and heavy, arising rather abruptly from rest ofskuU. The hypocones of the first and second molars are very poorlydeveloped. (Plate 9, fig. 1.)Measurements.?Type: Head and body, 154 mm.; tail, 166; hind foot,41; condylo-basal length, 43; zygomatic width, 24.4; width of braincase, 18; maxillary tooth row, 16.5. The external measurements ofthe type are somewhat less than those of the majority of specimens,but the cranial measurements are characteristic. For individualmeasurements see page 83.Remarks.?The two species just described, T. mollendorffi and T.cuyonis, are closely related forms but easily distinguished by thetail being grayer than rest of upper parts in the one case and by itsbeing of generally the same color as the lower back in the other case.They appear more closely related to one another than either of themdoes to T. palawanensis. The skulls of the two forms are essentiallyalike. There are many things about Tupaia cuyonis to suggest T.javanica. The two are not so very different in size, especially whenthe skulls are compared. The skulls have tlie same general shapein the two species. The development of the central upper incisorsand lower canines is distinctly greater than in most members of thegenus, but on the whole rather less than what one finds in good
NO. 1976. TREESHREWS: FAMILY TUPAIID.^?LYON. 83specimens of Tupaia javanica. T. cuyonis also resembles T. javo/nicain having dark under parts. T. cuyonis might with miicli proprietybe described as a large brown and not olivaceous T. javanica. Thedifference between the length of tooth row and the width of braincase is greater in Tupaia cuyonis than it is in T. chinensis or T.helangeri, and still greater than in T. glis ferruginea, where themaxillary tooth row is sometimes as long as the brain case is wide.In T. javanica the diiference in length of maxillary tooth row andwidth of brain case is even greater than in T. cuyonis. Tupaia cuyonisand T. moUendorffi are very distinct from each otlier and from othermembers of the genus, but at the same time they have no sharplyseparating characters. They have characters which on the one handally them with T. javanica and on the other with T. chinensis.Specimens examined.?Nine, all from Cuyo Island, and in the col-lection of the Philippine Museum. See table below.Measurements of Tupaia mollendorffi and T. cuyonis.
Name.
T. mdllen-dorffl.DoDo.
Locality.
Culion Island, Phil-ippine Islands,dodoDo
84 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.48.2.11.2, and a skin and skull without locality, Reg. No. frhi'-^iAi-From the account in the original description one would be inclinedto take the specimen in alcohol as the type, but because a coloredplate of the entire ammal is given, one is justified in regarding theskin as the type. This has been done by authors generally, Mivart,Anderson, and Thomas.^ Tliis course is rendered imperative by thefact that an examination of the specimen in alcohol shows it to be animmature example of Tanatana, the first upper molar not being throughthe gum, and just appearing through the alveolus. The skin, Reg. No.47.7.8.13, is mounted and in good condition, probably a female; theonly original information concerning it is *'Ex. coll. Verreaux." ^The skull, Reg. No. 48.1.27.14, is in fairly good condition, but con-siderably damaged on the right side about tlie palate.GeograpMc distribution.?Probably the entire island of Borneo, butImown records are only from southern portion. See No. 37 on mapon page 75.Diagnosis.?About the size of or slightly smaller than T. glis fer-ruginea, upper parts seal or walnut brown, tail similar to body, itshairs without amiulations. Mamma3, probably 2-2 = 4.Color.?General color effect above something between seal brown andwalnut brown, with indistinct grizzling with ferruginous posterior toneck, becoming almost obsolete on the rump, anterior to the neckcolor lighter and grizzling more distinct ; in the type, top of neck andshoulders and sides something of a color between bay and chestnut;under parts anterior to chest, buffy in the type, ochraceous buffy inother specimens, posterior to chest tawny ochraceous; tail abovesimilar to back, below tawny in center line, outer edges like back,hairs of tail without annulations; shoulder stripe rather poorly de-fined, tawny ochraceous.Slcull and teeth.?The skull and teeth of Tupaia splendidula are ofthe same general form as those of T. glis ferruginea, but are dis-tinctly smaller, with smaller and more oval incisive foramina, andrelatively larger orbits. The type has a slightly wider rostrum andwider brain case than Cat. No. 151883, U.S.N.M. (Plate 10, fig. 11.)Measurements.?Type: Head and body, 190mm. ; tail, 140 (bothfrommounted skin); hind foot, 40; condjdo-basal length, 45; zygomaticwidth, estimated, 27; width of brain case, IS; maxillary tooth row,17.5. Collector's measurements of head aiid body are 173-188; tail,130-157. Measurements of two adult skuUs are: Condylo-basallength, 43.5-44; zygomatic width, 24.5-24.5; width of brain case, IS-IS.5; maxillary tooth row, 16.5-18. See table, page 87.Remarlcs.? Tupaia splendidula is a well-characterized species andrepresentative of a rather definitely marked group. As seen onBorneo and as represented by geographic forms in the NatunaIslands, it is very different from mem^bers of the wide ranging /er-ruginea group, yet it seems to be not very distantly removed from it.
1 See Thomas and Hartert, Nov. Zool., vol. 1, p. 656, September, 1894.
NO. 1976. TREESHREWS: FAMILY TUPAIIDJ3?LY0N. 85Its chief distinguishing feature is the lack of annulations on the hairsof the tail, generally dark color, and the small size of skuU. OnSumatra, Singkep, and the Anamba Illands are Tupaias which,wliile evidently belonging to this splendidula group, are much closerto T. glis ferruginea than is T. splendidula itself. I have not seenKolilbrugge's Tupaia millleri, but the description would indicate itto be an example of T. splendidula. Neither have I seen Gliporarufescens "b" and "c" of Jentink's Catalogue,^ judging by thename they may be examples of this species, Mivart's Tupaia rufi-caudata was an accidental publication of Gray's manuscript name.Gray had evidently intended to call the species ruficaudata, but reallypublished the name as splendidula, forgetting to change ruficaudataon the label of the specimen. As Kolilbrugge points out, attentionto this species was first called by Miiller and Sclilegel' who con-sidered it a hybrid between "Hyl. tana and ferruginea." Not un-likely it is specimen "(Z" under Tupaia ferruginea fromBanjermassinin Jentink's catalogue of mammals in the Leyden Museum.^ In theBerlin Museum is a skull witliout skin from Kutei, a district on theeast coast of Borneo somewhat north of Klumpang Bay, whereDoctor Abbott collected two examples. I have identified it asTupaia splendidula, but not without some reservation, as I was un-able to make a direct comparison with known splendidula skulls.Specimens examined.?Five, four from southern Borneo, and one,the type, probably from Borneo. See table, page 87.TUPAIA NATUN.(E Lyon.1895. ? Tupaia splendidula typica Thomas and Hartert {nomen VAidum), Nov.Zool., vol. 2, p. 489, December, 1895.1911. Tupaia nahinse Lyon, Proc. Biol. Soc. Wash., vol. 24, p. 168, June IG, 1911.Type-locality.?Bunguran, Natuna Islands, north of Borneo.Type-specimen.?In United States National Museum, Cat. No.104714, skin and skull of an adult female, collected on Bunguran,Natuna Islands, June 27, 1900, by Dr. W. L. Abbott; original num-ber, 514; in good condition.Geograpliic distribution.?Bunguran Island. See No. 32, on mapon page 75.Diagnosis.?Similar to Tupaia splendidula of Borneo, but differingin a generally brighter and more reddish coloration of the upper parts,sides, legs, and tail, and more inflated braincase; mammee, 2-2 = 4.Color.?Ty]De: General color of upper parts of neck and body, infresh pelage, slightly brighter than burnt sienna, being produced bya wide band on most of the hair^, of a bright ferruginous burnt siennamixture, with blackish bases and a considerable number of longblackish hairs; on rump, in old pelage, general color darker and
1 Verb. Nat. Geseh. Nederl. Overz. Bezitt., p. 164, 1839-44.2 Mus. Hist. Nat. Pays-Bas, Cat Syst., vol. 12, 1888, p. 117.
86 PROCEEDINGS OF THE NATIONAL MUSEUM, vol.45.duller; sides of body and outer side of legs similar to upper parts,but slightly grizzled ; top of nose a grizzle of raw sienna and blackishgradually blending on top*of head with color of upper parts; underparts buff to olive buff; inner side of legs similar to sides of body, butlighter; tail above like back, under side of tail and bases of tail hairsgenerally tawny ochraceous, with outer and terminal margins of tailbelow, dark tav/ny; shoulder stripe buff, or ochraceous buff. Cat.No. 10-1715, U.S.N.M., has the central portion of the tail orangeochraceous in color, being fully as bright as the tail of T. lucida.STcull and, teeth.?These are of the same general form as they are inTupaia splendidula, but the skull averages longer and has a moreinflated braincase; the teeth are larger. (Plate 10, fig. 12.)Measurements.?Type: Head and body, 184 mm.; tail, 140; hindfoot,40; condylobasal length, 46; zygomatic width, 25; width of braincase,19; maxillary toothrow, 18. For measurements of individuals, seetable, page 87.Remarli's.? Tupaia natunx is closely related to T. splendidula, somuch so that Thomas and Hartert in 1894^ considered them thesame species. At that time there were no definite records of Tupaiasplendidula from the island of Borneo, the only available materialbeing the type of unknown locality. They arrived at the conclusionthat the type had not been obtained on Borneo, but had probablycome from the Natuna Islands. Since Doctor Abbott has obtainedspecimens on Borneo almost identical with the type there can be butlittle doubt that it was origina,lly collected on that island. ^ In work-ing with the treeshrews in 1904 Mr. Miller in manuscript notes hadcome to the conclusion that the Bunguran splendidula was distinctfrom true splendidula and had applied the name natunse to it.Specimens examined.?Six. All from Bunguran. See table, page 87.TUPAIA LUCIDA Thomas and Hartert.1895. Tupaia splendidula lucida Thomas and Hartert, Nov. Zool., vol. 2, p. 490,1S95.1901. Tupaia lucida, Miller, Proc. Wash. Acad. Sci., vol. 3, p. 133, March 26,1901.Type-locality.?Pulo Laut, North Natuna Islands.Type-specimen.?In the Tring Museum, skin and skull of female,collected on Pulo Laut, by Ernest Hose, September, 1894. I havenot seen this type.Geographic distribution.?Pulo Laut, North Natuna Islands. SeeNo. 31 on map on page 75.Diagnosis.?A member of the splendidula group of Tupaia, butupper parts, including tail, bright tawny; mammae, 2-2 = 4.Color.-?Based on topotypes in the United States National Museum,and a paratype in the British Museum, Keg. No. 95.11.8.7. Upper
1 Nov. Zool., vol. 1, p. G5G, September, 1S94.2 See remarks by Thomas in Lyon, Proc. U. S. Nat. Mas., vol. 40, p. 122, April 25, lOU.
NO. 1976. TREESHREW8: FAMILY TUPAIIDJE?LYO:^. 87parts of body, tail, and outer side of legs bright tawny, v/itli a veryslight admixture of blackish or black; head ochraceous finely sprinkledwith blackish; underparts dull ochraceous buff; underside of tailsimilar to upper surface, but brighter in color, and in the middleline lightening almost to ochraceous buff; shoulder stripe, moderatelydistinct, dull ochraceous.Skull and teeth.?These do not appear to differ essentially from thoseof Tupaia splendidula.Measurements.?Type (from original description): Head and body,210 mm.; tail, 154; hind foot (without claws), 39. Usual measure-ments of adults: Head and body, 170-180; tail, 145-155; hind foot,with claws, 40-44; condylobasal length, 44-40; zygomatic width,24-25; width of brain case, 18.5-19; maxillary toothrow, 18-19. Forindividual measurements, see table below.Remarks.? Tupaia lucida is a very distinct member of the splendi-dula group, at once distinguished from the Bornean form by its muchlighter and brighter color. It is a much more highly differentiatedform than T. natunse, probably owing to the smaller size of the islandit inhabits, and the greater distance of the island from Borneo.Specimens examined.?Eight. All from Pulo Laut. See tablebelow. Measurements of Tupaia splendidula, natunse, and lucida.
Name.
88 PROCEEDI^ati OF THE NATIONAL MUk^EUil. vol.45.TUPAIA CHRYSOMALLA Miller.1900. Tupaia chrysomalla Miller, Proc. Wash. Acad. Sci., vol. 2, p. 232, August20, 1900.Type-locality.?Pulo Siantan, Anamba Islands, South China Sea.Tyije-sjjecimen.?In United States National Museum, Cat. No.101710, sldn and skull of adult female collected on Pulo Siantan^August 24, 1899, by Dr. W. L. Abbott; in good condition.Geographic distribution.?-Known onl}^ from Pulo Siantan, AnambaIslands. See No. 33 on map on page 75.Diagnosis.?A member of the spZe7i?Zi(Z'uZa group, having the generalappearance above of a bright reddish T. glisferruginea, inner half ofthe caudal hairs viewed from below, ochraceous or ta^vny ochraceous;mammae, 2-2 = 4.Color.?Upper parts of body a grizzle of a color between ferruginousand chestnut, and blackish, brightest anteriorly; head a grizzle ofochraceous and blackish; outer sides of legs similar in color to adja-cent i^arts of body; upper side of tail like back; outer edge of under-side of tail similar to its upper surface, basal half or more of the longcaudal hairs ochraceous or tawny ochraceous; underparts, buff toolive buff, with dark bases of the hairs showing through.Slcull and teeth.?Intermediate in size between those of T. splendidulaand T. glis ferruginea, and rather more like the latter in form; in-cisive foramina relatively short and wide. (Plate 10, fig. 10.)Measurements.?Type: Head and body, 178 mm.; tail, 152; hindfoot, 43; condylo-basal length, 48; zygomatic width, 25; width ofbrain case, 19; maxillary tooth row, 19. See table, page 93.Specimen examined.?One, the type.TUPAIA RIABUS, new species.Type-locality.?Pulo Riabu, Anamba Islands.Type-specimen.?In United States National Museum, Cat. No.104881, skin and skull of adult female collected on Pulo Riabu,Anamba Islands, South China Sea, August 23, 1900, by Dr. W. L.Abbott, in good condition.Geographic distribution.?Ejiown only from Pulo Riabu. See No.34 on map on page 75.Diagnosis.?A member of the splendidula group, mtermediate incharacters between Tupaia lucida of Pulo Laut, Natuiia Islands, andT. chrysomalla of Pulo Siantan, Anamba Islands; mamma3, 2-2 = 4.Color.?Upper parts of body and back of head ferruginous ororange rufous, irregularly lined with blackish; on head anteriorlyand above shoulder stripe, the ferruginous colors replaced by ochra-ceous tints; outer side of legs similar ^to adjacent parts of body; tailabove similar to upper parts of body, but blackish element practi-cally wanting in distal three-quarters; long hairs of underside of
NO. 1976. TREEBHREW8: FAMILY TUPAIID^?LYON. 89tail almost clear orange rufous; underparts, including inner side oflegs, and short appressed hairs on underside of tail, ochraceous buff;feet blackish brown finely lined with ochraceous.Shull and teeth.?Not essentially different from those of Twpaialucida or T. splendidula, but distinctly smaller, especially the teeth,as compared with T. chrysomalla.Measurements.?Type: Head and body, 176 mm.; tail, 146; hindfoot, 43; condylo-basal length, 45.5; zygomatic width, 23.5; viidthofbrain case, 19; maxillary tooth row, 18. See table, page 93.Remarks.? Tupaia riahus is a very different treeshrew from itsgeographic neighbors, and externally and cranially appears to bemore closely related to T. lucida of the Natunas. It is interesting tonote that Pulos Laut and Riabu are about the same general area.Tupaia riahus was collected on a second visit to the Anamba Islandsby Doctor Abbott in 1900, and hence was not included in Mr. Mil-ler's account of the mammals of the Anamba and other islandspublished m the same year.^Specimens examined.?Two, the type, and a young individual alsofrom Pulo Riabu. TUPAIA ANAMB.^, new species.Type-locality.?Pulo Jimaja, Anamba Islands.Type-specimen.?In United States National jVIuseum, Cat. No.101743, skin and skull of adult male collected on Pulo Jimaja,Anamba Islands, September 23, 1899, by Dr. W", L. Abbott.Diagnosis.?Very closely allied to Tupaia chrysomalla of PuloSiantan, but distinguished by a generally less reddish coloration ofhead and body; mammae probably 2-2 = 4.Geographic distribution.?Known only from Pulo Jimaja, AnambaIslands. See No. 35 on map on page 75.Color.?The color of Tupaia anamhse is so like that of T. chrysomallathat no detailed description is necessary. The ferruginous or chestnut-like color in T. chrysomalla is much lighter in color and replaced by acolor something like tawny ochraceous ; the whole lower back, rump,and thighs are lighter; the underparts are lighter more buffy andless ochraceous than in T. chrysomalla, but the tawny ochraceous colorof the underside of the tail is of a darker shade in T. anamhse. Someof the difference in color may be due to difference in pelage, as thetype of chrysomalla appears to be in an old pelage, while the typeof T. anamhse- is mostly in a fresh pelage. There is one skin ofT. anamhse, Cat. No. 101741, which has just begun to change pelage,and while not appearing so distinct from T. chrysomalla as the typeskin of T. anamhse, it has distinct though slight color differences and
1 Mammals collected by Dr. W. L. Abbott on islands in the South China Sea, Proc. Wash. Acad, Sci.,vol. 2, pp. 203-246, August 20, 1900.
90 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.is quite as good a form as many of the recognized insular forms ofT. glis.STcuTL and teeth.?The skull and teeth of Tupaia anambx do notshow any differences from those of T. chrysomalla.Measurements?Type: Head and body, 178 mm.; tail, 152; hindfoot, 47.5; condylo-basal length, 24.5; zygomatic width, 18.5; maxil-lary tooth row, 18.5. See also table, page 93.RemarJcs.?-The treeshrews of the Anamba Islands fall into twogroups, the species on Pulo Riabu closely related to Tupaialiicida of the Natuna Islands and the form on Pulo Jimaja andPulo Siantan, closely related to one another and not being closelyallied to any other form. All three of the islands are separated bywater of about the same depth, and approximately the same depthof water is found between them and Borneo on one side and the MalayPeninsula on the other. In many respects Tupaia chrysomalla andanamhse. show many resemblances to T. glis ferruginea in color of thehead and body, but their smaller size and color of the tail serveto distinguish them.Specimens examined.?Three, all from Pulo Jimaja.TUPAIA CASTANEA Miller.1903. Tupaia castanea Miller, Smiths. Misc. Coll., vol. 45, p. 54, November 6,1903.Type-locality.?Pulo Bintang, Rhio Archipelago, East Indies.Type-specimen.?In United States National Museum, Cat. No.115608, skin and skull of adult female collected on Pulo Bintang,August 11, 1902, by Dr. W. L. Abbott, original number, 1872; ingood condition.Geographic distribution.?Known only from Pulo Bintang. See No.17 on map on page 75.Diagnosis.?Related to Tupaia splendidula, but not so dark incolor, and in size equalling T. glis ferruginea; mammae, 2-2 = 4.Color.?General color effect of upper parts of back of head, neck,and body and outer side of legs something between hazel and chestnut,but rather darker and brighter; on closer examination this effectseen to be produced by an indistinct and coarse grizzling of black anda color something like a rich dark ferruginous; front of head a finegrizzle of blackish and ochraceous or tawny ochraceous; tail abovesimilar to the back in places, especially near base or else a colorbetween orange rufous and cinnamon rufous; underside of tailbetween orange and cinnamon rufous; general color of under parts,including innerside of legs, between ochraceous and tawny ochraceous,with darker bases of hairs showing through in places, especially onthe inner side of legs; hands and feet a fine grizzle of blackish andochraceous; shoulder stripe moderately distinct, light tawny ochrace-ous or ochraceous rufous.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 91Shdl and teeth.?These are of the same general form as they are inTupaia splendid ala, but larger throughout. The incisive foraminaare rather large and less slit-like and the bullae somewhat largerthan in the case of T. glis ferruginea. (Plate 10, fig. 9.)Measurements.?Type: Head and body, 200 mm.; tail, 145; hindfoot, 46; condylo-basal length, 49; zygomatic width, 25.5; width ofbrain case, 19.5; maxillary tooth row, 19. The measurements of thetype are not materially departed from in a series of eight individuals,for measurements of which see table, page 93.Remarks.? Tupaia castanea is a very distinct form; from T. splen-didula it is at once distmguished by its larger size, equaling T. glisferruginea, and from T. glis ferruginea by its splendidula style ofcoloring. Schneider's* record from the Indragiri, Sumatra was per-fectly correct so far as the group is concerned, but his specimens arenow identified as Tupaia siaca.Specimens examined.?Eight, all from Pulo Bintang.TUPAIA SIACA Lyon.1908. Tupaia siaca Lyon, Proc. U. S. Nat. Mue., vol. 34, p. 661, September 14,1908.Type-locality.?Little Siak River, lowlands of eastern Sumatra.Type-specimen.?In United States National Museum, Cat. No.144205, skm and skull of adult female collected along the Little SiakRiver, Sumatra, November 4, 1906, by Dr. W. L. Abbott; origuialnumber, 4856; in good condition.Geographic distrihution.?Known only from the Little Siak andIndragiri River regions, probably occurring in the mtervening region,and for a moderate distance beyond, on either side. See No. 8 onmap on page 75.Diagnosis.?Very similar to Tupaia castanea, but underparts andinner side of legs buff to ochraceous buff, instead of ochraceous toochraceous rufous; hairs of tail, seen above more distinctly annulated,but seen below without annulations except beyond middle of hairs;color of upper parts not quite so dark and rich as m T. castanea,especiaflly in the region of the neck and shoulders, which are lighterand brighter, and more grizzled than in T. castanea; mammae, 2-2 =4.Color.? Tupaia siaca is m general very sunilar in color to T. cas-tanea, and the differences have been sufficiently pointed out underthe diagnosis, so that no detailed description is necessary.Skull and teeth.?The skull and teeth of Tupaia siaca are essentiallylike those of T. castanea. (Plate 10, fig. 8.)Measurements.?Type: Head and body, 205 mm.; tail, 175; hindfoot, 47; condylo-basal length, 50.5; zygomatic width, 25.5; width ofbrain case, 19.5; maxillary tooth row, 19. The tail of the type is from10 to 15 mm. longer than is the case with most of the adult specimens
J Zool. Jahrb., vol. 23, p. 87, 1905.
92 PROCEEDINGS OF THE NATIONAL MVSEUM. vol. 45.examined, otherwise the measurements of the tjjDe represent theaverage for the species. For individual measurements, see table,page 93.Remarks.? Tupaia siaca and T. castanea form a very interestingand at same time puzzling gi'oup of treeshrews. Both are veryclosely related to one another and geographically they are widelyseparated. Pulo Bintang is about as far removed from Sumatra asany island of the Rhio Archipelago, and no Tupaias of this group arefound on the intervening islands. In fact treeshrews are poorlyrepresented on the islands of the Rhio-Linga Archipelago. OnBatam, the next island to Bintang, occurs a treeshrew that is sepa-rable with difficulty from T. glis ferruginea, and on Siiikep, an islandclose to Sumatra, is another member of the glis group. Other tree-shrews in the Archipelago are without interest in this connection.Tupaia castanea and siaca as judged by color, are certainly relatedto T. splendidula of Borneo. In point of size they equal T. glisferruginea, and I have been unable to find any definite constantcharacter in the skulls or teeth, to separate them as a group from T.glis ferruginea. The question immediately arises what is theu" rela-tion to T. g. ferruginea. The two forms, T. castanea and siaca maybe geographic representatives of that widely spread species. So faras we know T. g. ferruginea does not occm- at the same localities withthem. Onl}^ on Borneo do we find T. splendidula occurring withwhat is evidentlj^ a representative of T. glis ferruginea, that is T.longipes. Those two forms are very different externally and cranially.On the Anamba Islands are found Tupaias which in point of size andlack of annulations on the tail are certainly members of the splendidulagroup, but some of them in general body color resemble T. glis ferru-ginea very closely. The available material indicates that typicallythe glis ferruginea and splendidula groups are very distinct, but formsof each occur strongly suggesting the other group. The relationof T. demissa to the splendidula group is puzzling. It occurs onSumatra just to the northeast of the range of T. siaca. In thegeneral color of its upper parts it is very similar to T. siaca and T.castanea. It has more grizzling along the sides than has the mfembersof the splendidula group and less on the thigh than usual in T. g.ferruginea. The underparts are more like those of T. g. ferruginea,so is the skull. The hairs of the underside of the tail are certainlylacldng in the annulations even more than the hairs of T. siaca. Thespecies demissa could with almost as much propriety be considered anabnormal color form of T. siaca as it is so considered of T. g.ferrugineaSchneider's specimens recorded ^ as Tupaia custanea from the Indraghiregion Sumatra are T. siaca.Specimen examined.?Eight from the Siak region and two fromthe Indragiri.
1 Zool. Jahrb., vol. 23, 1905, p. 87.
NO. 19T6. TREESHREW8: FAMILY TUPAIID^?LYON. 93Measuremenls of Tupaia siaca, castanea, chrysomalla, riabus, and anambae.
94 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.narrowly between the shoulders, 2 to 3 mm. wide and widening overthe lower back and hips to 25 or 30 mm. ; head a fine grizzle of blackand raw sienna, the latter predominating on tlie sides, the blackish onthe crown; underparts an ill-defined grizzle of dark tawny ochraceousand slate, the tawny ochraceous lighter and more yellowish anteriorly;tail above, similar to adjacent parts of body, but grizzle coarser; tail,below, with short appessed hairs, in median line, a fine grizzle ofblackish and ochraceous, the darker color predominating, the longerhairs, tawny ochraceous for their basal third, and a coarse grizzledmixture of this same color and dark brownish distally; feet and handsdark brownish with a very slight admixture of the light color of theadjoining parts of body; shoulder stripe practically obsolete.Skull and teeth.?The skull of Tupaia montana is of moderate sizeslightly smaller than that of T. glis fcrruginea, with relatively muchAvider brain case and more spreading zygomata; the rostrum has amore abrupt origin from rest of skull and is much compressed fromabove downward, especially the basal portion just posterior to thenasals. The first and second upper molars of T. montana are morequadrate m outlme than they are in T. glisfemiginea and the hypo-cone of m} is better developed. (Plate 9, fig. 2.)Subspecies.?There are two subspecies of Tupaia montana, thetypical form T. montana montana from Mount Dulit and a form fromMount Kina Balu, described below as new. Aside from some slightcolor and cranial differences the two forms are chiefly distinguished bythe presence in the Dulit specimens of a more or less well-defineddorsal patch and its absence in the animal from Mount Kina Balu.Remarlcs.? Tupaia montana is a very distinct species and easily dis-tinguished by external and cranial characters from other members ofthe genus. It does not appear to have any close resemblance to othermembers of the genus, except the later described Tupaia carimatse.The presence or absence of a black patch on specimens does not appearto be attributable to age, sex, or season. Most specimens from MountDulit show it plainly, while on Kina Balu specimens it is but barelyindicated. TUPAIA MONTANA MONTANA Thomas.1892. Tupaia montana Thomas, Ann. Mag. Nat. Hist., ser. 6, vol. 9, p. 252,March, 1892.1892. Tupaia montana, Thomas, Proc. Zool. Sec. London, p. 223, 1892.Type-locality.?Five thousand feet on Mount Dulit, Sarawak,Borneo.Type-specimen.?In British Museum, Keg. No. 92.2.7.5, skin andskull of adult male collected October 14, 1891, at 5,000 feet on MountDulit, Borneo; skin in good condition; skull with part of occipitalregion cut away.
NO. 1976. TREE8RREW8: FAMILY TUPAIID^?LYON. 95Geographic distribution.?Mount Dulit and probably neighboringmountains in Sarawak, Borneo. See No. 25 on map on page 75.Diagnosis.?Distinguished by the greater frequency of the blackback patch, heavier rostrum, slightly larger and darker feet, and facialportion of skull apparently larger and longer. (Plate 9, fig. 2.)Measurements.?Type: Head and body (dried skin), 215 mm.; tail(dried skin), 140; hind foot, 45; condylobasal length, 47 ? ; zygomaticwidth, 26.5; width of brain case, 19; maxillary tooth row, 18.5. Forindividual measurements see table, page 96.RemarJcs.?A specimen with a well-defined dorsal stripe can cer-tainly be identified with this subspecies, but if the stripe is practic-ally obsolete as in Reg. No. 92.2.7.6, one is uncertain whether toplace the specimen in this subspecies or the next. The Dulit subspe-cies appears to average larger than that from Kina Balu.Specimens examined.?Twelve from Mount Dulit.TUPAIA MONTANA BALUENSIS, new subspecies.Type-locality.?Mount Kina Balu, northeastern Borneo.Type-specimen.?In British Museum, Reg. No. 95.10.4.20, skin andskull of adult collected at 3,000 feet on Mount Kina Balu, north-eastern Borneo, March, 1887, by J. Whitehead; skin in fair condition;skull with about a third of the right side of brain case wanting.GeograpMc distribution.?Known only from Mount Kina Balu,Borneo. See No. 26 on map on page 75.Diagnosis.?Differs from Tupaia montana montana in the absenceof the distinct black back patch, smaller size; rostrum and wholefacial portion of skull smaller and shorter; hind feet not quite so darkas in the typical form.Measurements.?Type: Head and body (dried skin), 190 mm.; tail(dried skin), 90; hind foot, 39; condylobasal length, 43? ; maxillarytooth row, 17. For measurements of two other individuals, whichdiffer considerably from this type in having the tail 125 mm. long,see table, page 96.RemarJcs.?In certain lights there is a suggestion of the broaddorsal stripe, which is almost as evident as in those Dulit specimens,where the stripe is practically lacking. Mr. Thomas has this noteon the label of the type: "Of 15-20 specimens from Kina Balu(A. Everett) carefully compared, and some quite old, with worn teeth,not one had any trace of a dorsal line, not even as much as this."The absence of the dorsal stripe would appear to be very constant inthis subspecies, much more so than the presence of the stripe is inT. montana montana. The Kina Balu specimens were identified byMr. Thomas in 1889* as Tupaia ferruginea.Specimens examined.?Three from Mount Kina Balu.
' Proc. Zool. Soc. London, 1889, p. 229.
96 PROCEEDINGS OF THE NATIONAL MUSEUM. TOL. 45.Measurements of Tupaia montana.
Name. .
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 97Diagnosis.?A very distinct species of Twpaia at once characterizedby a narrow black dorsal line, and distichously bushy rufescent tail;manunse, 2-2 = 4.Color.?Type: Upper parts of head, neck, and body anterior torump a grizzle of blackish and ochraceous, with a very distinct narrowdorsal blackish line extending from region of neck to rump, thegrizzling very coarse in the dorsal region; toward rump and base oftail, ochraceous color gradually replaced by ferruginous ; tail a mixtureof ferruginous and blackish, the two colors about equally mixed onbasal half of tail above, the ferruginous predominating distally andbelow, except on outer edge, which is Uke upper surface; underpartsincluding inner side of legs, dull buff to ochraceous buff; outer side oflegs similar to adjacent parts of body, and feet similar in color to outerside of legs; shoulder stripe, well marked, dull ochraceous. The speci-mens from Mount Dulit, Mount Kulalong, and Balingean are slightlydarker than the others, with the dorsal stripe less distinct; under-parts are of a distinctly darker ochraceous about the neck and chest,and a grizzle of ochraceous or tawny ochraceous and blackish on restof underparts and inner side of legs; the feet are blackish brown, per-haps due to a preservative, as they have an unnatural look. Twoadults from Mount Mulu have the yellowish grizzling of the back, verycoarse and conspicuous; the feet are of the usual yellowish browncolor.Slcull and teeth.?The skull has about the same general build as thatof T. montana, but lacks the above downward compression at base ofrostrum, and approaches the skull of T. glis ferruginea in size; thebullse are relatively larger than they are in T. montana or T. g.ferruginea. The brain case isrelativelywide like that of r.montona. Thezygomatic perforation is slightly smaller than it is in T. glis ferrugineaThe upper molar teeth are not so quadrate in outline as they are in T.montana and show no essential differences from the teeth of T. glisferruginea. (Plate 9, fig. 3.)Measurements.?Type: Head and body, 195 mm. (from dried skin);tail, 155 (from dried skin); hind foot, 45; condylobasal length, 47.5;zygomatic width, 26; width of brain case, 19.5; maxillary toothrow,18.5. Measurements of the series do not depart materially fromthose of the type. See table, page 98.RemarJcs.?Twpaia picta is a very distinct species and needs no com-parison with any other. It is quite in a class by itself and appearsto have no near relatives. The superficial external resemblance to thelong-snouted treeshrews, Tana tana is very interesting ; both have thesame distichously bushy ferruginous tails, both have a narrowblack dorsal stripe, bordered on either side by lighter grizzled areas.The dorsal stripe, however, is very short in Tana and a careful80459??Proc.N.M.vol.45?13 7
98 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.examination of the markings in Tana tana and Tupaia picta showsthat the resemblance is more superficial than real. The skull ofTupaia picta is a typical Tupaia skull, and the arrangement of thenaked area on the end of the nose and the small size of the claws allshow it to be a true Tupaia. The distribution of Tupaia picta seemsto be quite restricted. Judging by the number of specimens exam-ined it appears to be a fairly common animal in the Baram District,and I doubt if it occurs in southern Borneo, as Doctor Abbott wouldprobably have encountered it sometime during his four expeditionsthere. In his Mammals of Borneo ^ Hose says of it: ''This tree-shrew is more common in the low country than on the mountains;it is usually found in the dense forest and is particularly^ active in itsmovement."Specimens examined.?Fourteen, all apparently from the BaramDistrict, Borneo. Measurements of Tupaia picta.
Locality. Number. Sex. Molar teethworn.
Borneo, BaramBorneo, Mount BatuSang.Borneo, Mount Dulit,2.000 feet.Balingean, Sarawak..Mount KalulongMount Mulu, 1,000 feetMount MuluDoBatu Sang, 3,000 feet..Baram RiverBaram, ApohBaram, SarawakSarawakMoimt Dulit, 2,000 feet(?)
84506669/4, Geneva.92.9.4.15.3.1.593.6.1.28.1.27.394.6.2.294.6.2.392.11.8.192.2.7.2192.2. 8.1<0.8.4.8 ,
Male...FemaleMale...FemaleMale...
..do...
...do...
...do...
..do...
...do...
ModeratelyNone 1
99. 12. 9. 6Berlin11882, Phila*. Male....do.
(^)Moderately(.')(2)ModeratelyNone 3doModerately(2).".-?.v.::;Slightly....(2)
190?220?210?235 ?220?240?195 ?210?215 ?195 ?205 ?No skin210?| 140+170? 180?
165?155 ?135?155 ?150?150?140?155 ?155 ?155 ?1C0?
mm.4643 mm.4848? mm.2724 mm.2019 mm.1815
47.5
'48.'5
25.5 18 2-2
47.546.'
5
2018.51919.519.5
191716.51818.519.5 184640?
' Milk premolars still in place.s Skin with skull inside.
' m^ just appearing.* Type.s Skinned from alcohol and very abnormal in appearance. Academy of Natural Sciences, Philadelphia.TUPAIA CARIMATiE Miller.1906. Tupaia carimatx Miller, Proc. TJ. S. Nat. Mus., vol. 31, p. 61, July 23, 1906.Type-locality.?Telok Edar, Karimata Island, off west coast ofBorneo.Type-specimen.?In United States National Museum, Cat. No.125123, skm and skull of adult male, collected on Karimata Island,September 2, 1904, by Dr. W. L. Abbott; original number, 3716; ingood condition.
J Page 32, 1893.
NO. 1976. TREESHREW8: FAMILY TUPAIIDJE?LYOU. 99Geographic distribution.?Known only from Karimata Island, butprobably occurring on other islands of the same group. See No. 36on map on page 75.Diagnosis.?A very distinct species intermediate in general char-acteristics between TwpUia splendidula and T. montana, slightlysmaller than either, without a black dorsal area, basal portion of hairsof underside of tail, almost without annulations," underparts notdistinctly grizzled.Color.?Type: Upper parts of head, neck, body, outer side of legsand tail a fine distinct grizzle of ochraceous and blackish, the twocolors about equally mixed except on dorsal area posterior to shoul-ders where the black is in excess and suggests the condition found inTupaia montana haluensis; underparts, including inner side of legs,varying from buff to ochraceous buff with dark bases of hairs showingthrough in places, but not finely grizzled as in T. montana; undersideof tail ochraceous, distinctly grizzled with blackish along margins,and along the area of short appressed hairs; shoulder stripe, ochra-ceous, well developed.STcull and teetli.?Skull slightly smaller than in either Tupaia mon-tana or T. splevdidula, relatively wide, zygomata spreading, brain-case inflated, rostrum shortened. The teeth are essentially Uke thoseof T. montana, but first and second upper molars not quite so quad-rate. As a whole the skuU seems to have more affinities with that ofTupaia montoTia than with that of T. splendidula. (Plate 10, fig. 6.)Measurements.?Type: Head and body, 176 mm.; tail, 147; hind foot40; condylobasal length, 42.5; zygomatic width, 24; width of brain-case, 18; maxillary toothrow, 16. For individual measurements.see table, page 100, none of which differ essentially from those of thetype.RemarJcs.? Tupaia carimatse is a very distinct species and wouldscarcely be confused with any other form. It appears to possessmore characteristics of T. splendidula than it does of any otherspecies, namely, moderately well developed shoulder stripe, gen-erally ochraceous underparts, without fine distinct grizzhng, andbasal portion of hairs of tail on underside practically clear ochraceous.Its montana characteristics are its generally finely grizzled upperparts, slight tendency to a dark dorsal area, and skull as a wholeapproaching that of T. montana more than that of T. splendidula.On geographic grounds it seems most likely that Tupaia carimatse,is an insular form of the Bornean T- splendidula. The mammalianfauna of Karimata has apparently been derived from forms similarto those occurring in southern Borneo,^ a region where we know T.splendidula certainly occurs.Specimens examined.?Seven, all from Karimata Island.
1 See Lyon, Proc. U. S. Nat. Mus., vol. 40, p. 81, April 25, 1911, remarks oa squirrels onhe prevostii group.
100 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Measurements of Tupaia carimatse.
NO. 1976. TREESHREWS: FAMILY TUPATID^?LTON. 101indistinct lateral stripe on either side to meet the dark blackish-brown area of the back, in many specimens, an indistinct medianstripe extending from the dark area on top of head to the blackish-brown area of the back; tail, in unworn conditions, very similar toseal brown, both above and below except along the center line belowwhere the color is much lighter. In some specimens the little mid-dorsal stripe from head to the dark area of the back is obsolete, andin others it is more conspicuous than the lateral stripes from ears todark dorsal area, very rarely are the mdistmct stripes from ears toback lacking. In worn condition the tail becomes much lighter incolor, approaching cinnamon or russet. Also m worn pelages thedark area of the back becomes lighter and duUer. (Plate 3, fig. 1.)Slcull and teeth.?Unlike most members of the genus the skull ofTupaia nicoharica is quite characteristic and would hardly be con-fused with the skulls of other species. The slmU has about the samegeneral size as that of Twpaia glisferruginea, but appears more solidlybuilt, and more angular with spreading zygomata and with a moreabruptly arising rostrum; the palate is better ossified, the fenestra inzygoma less elongate, and more oval, and often much reduced in size;space between the external and internal pterygoid plates greater andbuUag slightly smaller; the impression for muscular attachment at thean tern-inferior angle of the zygoma distinctly smaller, and its smallsize more noticeable in view of the otherwise greater angularity of theskull; the coronoid process of the mandible is heavier and wider. Theteeth of Tupaia nicoharica are for the most part similar to those ofTupaia glis ferruginea, but the central pair of upper incisors greatlyexceed the second pair of upper incisors and are very much larger andmore recurved than the corresponding teeth of T. glisferruginea ; theyhave about the same relative degree of development that is seen inTupaia javanica. The upper canmes are more slender and pomtedthan they are in T. glis ferruginea. In comparison with other mem-bers of the genus, except T. javanica, the lower canines are greatlydb7eloped, approaching the enormous development of the lowercanines in the genus Urogale. The third incisor in front and the firstpremolar behind the lower canine are correspondingly diminished insize, and the length of the lower canine exceeds or at least equals twicethe length of incisor in front and often more, while in the speciespreviously considered the lower canine is only one and one-half timesthe length of the incisor m front and often less.Measurements.?Usual measurements of adults: Head and body,180-195 mm; tail, 200-225 ; hmdfoot, 45-49; condylo-basal length,47-50; zygomatic width, 27-29; width of bram case, 19-20; maxillarytooth row, 18-19. For individual measurements see tables, pages 104and 105.
102 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Subspecies.?Two; Twpaia nicoharica nicobarica, page 103, andTwpaia nicobarica surda, page 104.Remarks.?As shown above, Twpaia nicobarica is a remarkablydistinct species and sharply separated from all other members ofthe genus by its combination of coloration, large size, long tail,large sharp claws, angular skull, large central upper incisors, andlarge lower canmes.Tupaia nicobarica was probably first observed m 1846 by CaptainLewis,^ who reported large squirrels as occurring on the NicobarIslands. As no squirrels were collected or reported by Abbott andKloss in 1901,- the inference is that Lewis's squirrels, like thoseseen by Captain Cook's party on Pulo Condore,^ must have beenTupaias. Zelebor mentions this treeshrew as occurring on all thelarger islands of the group, but particularly abundant on GreatNicobar; but as no specimens were collected or reported on otherislands than Little and Great Nicobar, by Abbott and Kloss, itwould seem that Zelebor's statement is too general. ApparentlyTupaia nicobarica is a rather conspicuous and noisy treeshrew,otherwise it would not have called forth the few comments thathave been made regarding it. Zelebor says it cries frequently andprolonged ' ' Danh-Danh ' ' when disturbed or pursued. Doctor Abbottremarks of them, ''Common m the heavy jungle which covers theisland (Little Nicobar); very active and generally in the tree tops."EHoss ^ writes : " They were very common (Little Nicobar); but unliketheir representatives in the Malay Peninsula, etc., which are groundanimals, we saw them only in trees. Tupais were plentiful (GreatNicobar). These appear to be entirely arboreal in habits and arequite as active as squirrels in ruxming along branches, or climbingabout amongst smaller twigs in search of msects. Their cry is a sortof trilling squeak, which is easily confounded with the call of abu'd." The rather large, compressed, sharp claws and the long tailalso lead one to believe it more arboreal than most of its relatives.As to the origin of this treeshrew on the Nicobars it is impossibleto say. It is so unlike any of the other treeshrews at present knownthat it seems highly improbable that it can be descended from anyof them, should they have been accidentally carried to the islands,or have been brought there by man. On the other hand, if Tupaianicobarica is the survivor of an old wide-ranging species, one wouldexpect to find other mammals surviving with it. As Mr. Miller shows,the mammals of the Nicobars, "with the single exception of Tupaianicobarica^ are all types well known to be closely associated with
1 Joum. Asiat. Soc. Bengal, vol. IS, p. 368.2 Proc. U. S. Nat. Mus., vol. 24, pp. 751-795, May 29, 1902.3 See account of Tupaia dissimilis in the present paper, p. 67.
* In the Andamans and Nicobars, 1903, pp. 122, 136.
NO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 103man throughout the Malayaxi region." ^ I can not agree with himthat any of the existing treeshrews introduced into the Nicobars byman would have had sufficient time to develop into such a strikingspecies as Tupaia nicoharica. Its origin and relationship must forthe present remain unsolved.TUPAIA NICOBARICA NICOBARICA (Zelebor).1861. Cladobates nicoharicus Fitzinger, Sitz. Akad. Wiss. Math. Nat. Wien, vol.42, 1860, p. 392 {nomen nudum).1869. Cladobates nicoharicus Zelebor, Reise Novara, Zool. Theil, vol. 1, p. 17,pi. 1, fig. 1, entire animal natural size in colors; figs. 2 and 3, soles of foreand hind feet; pi. 2, skull, skeleton, and teeth.1879. Tupaia nicoharica, Anderson, Zool. Res. West. Yunnan, p. 136, pi. 7, fig.3, skull.1902. Tupaia nicoharica nicoharica, Miller, Proc. U. S. Nat. Mus., vol. 24, p. 773,May 29, 1902.Type-locality.?Great Nicobar, of the Nicobar Islands.Type-specimens.?According to Fitzinger/ these are in the Impe-rial Zoological Museum at Vienna. I have not seen them. In theoriginal account are mentioned an alcoholic specimen, a skeleton, andfour stuffed individuals.Geographic distribution.?Great Nicobar, of the Nicobar Islands.See No. 40 on map on page 75.Diagnosis.?Distinguished by having the light areas brighter andmore yellowish, and more strongly contrasted with the dark areasthan in the case of the form from Little Nicobar Island; mammae1-1 = 2Color.?With the differences noted in the diagnosis, the color ofTupaia nicoharica nicoharica is sufficiently described in the generalaccount of the species.SliuU and teeth.?There are no characters by which these may bedistinguished from those of the other subspecies.Measurements.-?Usual measurements of adults: Head and body,180-195 mm; tail, 200-225; hind foot, 45-50; condylo-basal length,47-50; zygomatic width, 26-29; width of brain case, 19-20; maxillarytooth row, 18-19. See table, page 104.Specimens examined.?Twenty-four.Remarks.?There are a few individuals in the series of specimens ofthis, the typical subspecies, that cannot be distinguished with cer-tainty from the form T. n. surda that follows.
1 Proc. U. S. Nat. Mus.^ vol. 24, p. 791, May 29, 1902.2 Sitz. Akad. Wiss. Math. Nat. Wien, vol. 60, 1869, pt. 1, p. 279,
104 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Measurements of Tupaia nicobarica nicobarica.
Locality.
NO. 1976. TREESHREWS: FAMILY TUPAIIDjE?LYON. 105and base of tail, blackish brown, not contrasting with the light areaof the pelage so noticeably as in the typical subspecies; top of headlight blackish brown, this color spreading down to behind ears, andusually posteriorly as an indistinct lateral stripe, to meet dark areaof back; in many specimens an indistinct median streak extendingfrom top of head to dark area of back.Slcull and teeth.?These do not differ from those of Tupaia nico-harica nicolarica. (Plate 10, fig. 7.)Measurements.?Type: Head and body, 190 mm.; tail, 220; hindfoot, 48; condylo-basal length, 49.5; zygomatic width, 28.5; width ofbrain case, 20; maxillary tooth row, 19. The measurements of thetype are shghtly in excess of the usual measurements. See tablebelow.Remarlcs.? Tupaia nicobarica surda is not a highly differentiatedform, but it is quite as distinct as some of the insular forms of Tupaiaglis to which binomial names have been given. The present form isa member of such a distinct species group of which there are onlytwo members so closely related geographically, it would seem a matterof convenience to employ trinomial names even though the two formswere more highly differentiated than they are. As noticed in theoriginal description there are a few specimens in each series of the twosubspecies which can not be distinguished with certainty one fromthe other.Specimens examined.?Sixteen.Measurements of Tupaia nicobarica surda.
Locality. Number. Molar teethworn.
106 PROCEEDINGS OP THE NATIONAL MUSEUM. vol. 45.TUPAIA JAVANICA Horsfield.1821. Tupaia javanica Horsfield, Zool. Res. Java, No. 3. Pages not numbered.Plate of entire animal and figures of fore foot and teeth on another unnum-bered plate. Date on title page of entire work, 1824. (In the next referencebelow the third part of Horsfield 's Researches is referred to. This reference isdated December, 1821.)1821. Sorex-Glis Javanica or Tupaya Javanica, Geoffrot and Cuvier, Hist. Nat.Mamm., vol 3, livr. 35, p. 1 and plate, December, 1821.1842. Cladobates javanica, Cuvier, Hist. Nat. Mamm., vol. 7, Tab. gen. Meth.,p. 2, 1842.1843. Hylogalea javanica, Schlegel and Muller, Verb. Nat. Gesch. Nederl.Overz. Bezitt., p. 165, pi. 26, fig. 4; pi. 27, figs. 11 to 16.1879. Tupaia javanica, Anderson, Zool. Res. West. Yunnan, p. 134, pi. vii,figs. 14 and 15.Type-locality.?Java, Province of Blambangan, probably near thepresent town of Banyu-wangi at extreme eastern end of Java. SeeHorsfield's Plantae Javanicae Rariores, 1838-1842, map in front, andpage V of postcript.Type-specimens.?In British Museum. This species was basedon two specimens collected by Thomas Horsfield in 1806, in extensiveforests in the Province of Blambangan, Java, One of these was sentto the Museum of the Honorable East India Co. in 1812. It is nowin the British Museum, Reg. No. 79.11.21.574; it is not mounted,is in poor condition; skull immature with the milk incisors stUl inplace, zygomata broken away, as well as occipital bones and anteriorthird of right half of mandible. The other specimen was brought toEngland by Horsfield in 1819, and is now in British Museum, No.52a; it is an old specimen, was formerly mounted, and is in poorcondition; parts of the cranium posterior to the tooth rows arebroken away; mandible perfect. A third specimen marked " cotype "is in the British Museum, No. 52b, Java, "pres. by E. F. Comp";neither its skin nor skull are in good condition, I do not considerthis specimen a cotype and see no reason why it should be so marked.Geographic distribution.?Java, western part of Sumatra back fromcoast, and islandof Nias. See No. 1 on map on page 111.Diagnosis.?^Tail longer than head and body; skull shaped likethat of Tupaia nicolarica, but much smaller, condylobasal length notexceeding 41 mm.; central upper incisors and lower canines enlargedin the manner of those of T. nicoharica; general coloration of upperparts and tail olivaceous finely grizzled, under parts moderatelydark and grizzled; hind foot about 38 mm.; mammae, 2-2 = 4.Color.?Upper parts and sides of head, neck, body, tail, outerside oflegs and feet, with the general color effect of something between oliveand bister, produced by a fine distmct grizzling of blackish andlight tawny olive or raw sienna; the two colors about equallymixed, the grizzling on the tail cop.rser and on the feet finer;underparts of body always rather dark, the bases of the hairs being
NO. 1976. TREESHREW8: FA3IILY TUPAIIDJE?LYON. 107extensively slate-color, which shows through usually to a considera-ble extent; general effect of underparts varying from an almostclear gray, not unlike olive gray slightly tinged with buffy, to analmost uniform ochraceous, with more or less of dark bases of hairsshowing through, the ochraceous colors of the underparts morepronounced anteriorly and the gray better developed posteriorly;shoulder stripe conspicuous, buff or cream buff in color.SJcull and teeth.?The skull and teeth of Tupaia javanica are almostan exact miniature of those of Tupaia nicoharica, but the zygomataare relatively less spreading, the brain case relatively more inflated,the bullae relatively larger, and the palate less completely ossified.The enlargement of the central upper incisors, and lower canmes, andreduction in size of the third lower incisors are in the same relativedegree in Tupaia javanica as they are^in T. nicoharica. (Plate 1 0, fig. 1 .)Measurements.?The available measurements of cotype Reg. No.79.11.21.574, a specimen not fully adult, the milk incisors still beingin place, are: Hind foot, approximately, 36 mm.; width of brain case,17.5; maxillary tooth row, 14; the other cotype, 52a has the maxillarytooth row 14.5; the specimen is in such poor condition that the otherusual measurement can not be taken. Usual measurements of adults
:
Head and body, 145-155 mm; tail, 160-175; hind foot, 37-39; condylo-basal length, 38-41; zygomatic width, 22-23; width of brain case,17-18; maxillary tooth row, 14-15. For individual measurementssee table, pages 108, 109.Rennarlcs.? Tupaia javanica is a very distinct species with no nearrelatives; its large size, dark underparts, and nicoharica shape of skullat once separates it from the other small olivaceously colored mem-bers of the genus, wliile the smaller size and differences in color serve,of course, to distinguish it from Tupaia nicoharia, and its smaller sizeand longer tail to distinguish it from the cJiinensis group. Whetheror not it is true Tupaia javanica that occurs on Sumatra and IKTiasitis impossible to say with the available material. The mounted skinin Genoa from Pulo Nias appeared unusually dark, especially the tail,which is quite blackish. I have little doubt that more material willshow it to be a distinct geographic form of Tupaia javanica.^ Thespecimens from Sumatra,, Si Rambi, Mount Singalang, and Pajoresemble the Javan specimens, but the material is not sufl&cient toestablish their true status. Apparently Tupaia javanica on Sumatrais confined to the higher regions. Although Doctor Abbott hasvisited many places on the Sumatran coast, he found no examples of it.Specimens examined.?Java, 58; Sumatra, 12; Nias, 1. Althoughthis material seems abundant, yet an examination of the list of speci-mens, pages 108, 109, will show most of it is made up of immatureindividuals, and specimens preserved in alcohol.
1 Recorded by Modigliani, Ann. Mus. Civ. Stor. Nat. Genoa, ser. 2, vol. 7 (27), p. 239, 1889.
108 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Measurements of Tupaia javanica
.
Locality.
NO. 1976. TBEE8HBEW8: FAMILY TUPAIIDJE?LYON. 109
Measurements of Tupaia javanica?Continued.
Locality.
110 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Slcull and teeth.?The skull and teeth of Tupaia minor are thesmallest of any member of the genus. The rostrum is relativelyvery short and stubby, and the brain case appears large and roundedin contrast. The distance between the two lachrymal notches isabout equal to the distance from that notch to the front of thecentral upper incisors. In Tupaia javanica the distance betweenthe two lachrymal notches is equal to the distance from the notchto the space between the first and second upper incisors. The wholeshape of the skull of Tupaia minor is quite unhke that of Tupaiajavanica or other species of Tupaia except Tupaia gracilis, whichhas a skull of the same general style as Tupaia minor but slightlylarger. The teeth of Tupaia minor are of the same form as those ofT. javanica and T. nicoharica. The central upper incisors are largeand heavy in comparison with most species of Tupaia, and the lowercanines strongly developed and the third lower incisor correspond-ingly reduced in size. Quite a little variation in the shape and sizeof the skull of Tupaia minor appears to exist, more so than is usuallyseen in most species of Tupaia; most of it, however, maybe accountedfor by differences in age.Subspecies.?Three geographic races of Tupaia minor are hererecognized?typical minor, the most distinct of them confined toBorneo; malaccana, occuning in Sumatra, the Malay Peninsula,and the island of Linga; and what is in reality a subspecies of thisform, occurring on the island of Singkep.RemarJcs.? Tupaia minor is a well-marked species. It differs fromT. javanica in being smaller, in having light, clear-colored underparts,and in its peculiar form of skull. It differs from Tupaia gracilis inits smaller size, less bushy tail, and in having teeth like those of T.javanica and not like those of T. glis ferruginea. The forms of Tupaiaminor recognized here are perhaps as distinct as some forms oftenrecognized as species, but the three are so closely allied to one anotherand they form such a compact group that it is a matter of greatconvenience to regard them as all belonging to one species. Theavailable material of this species is not entirely satisfactory, althougha large number of specimens exists in various museums, much of itis unsuited for systematic work, and many of the specimens areimmature. TUPAIA MINOR MINOR Giinther.1876. Tupaia minor Gunther, Proc. Zool. Soc. London, 1876, p. 426.1888. Gliporaleucogaster Jentink, Cat. Syst. Mamm. Mus. Hist. Nat., vol.12, p.116, 1888. A manuscript name of Diard, published by Jentink in a list ofspecimens of Tupaia javanica. As T. javanica does not occur in Borneo, andthe two specimens have white bellies, they are probably examples of thisspecies. Type-locality Pontianak, western Borneo.Type-locality.?Borneo, opposite the island of Labuan.Type-specimens.?In the British Museum, five cotypes, Reg. Nos.76.5.2.21, 76.5.2.22, 76.5.2.23, 76.5.2.24, 76.5.2.25, skins and skulls,
NO. 1976. TREE8HREW8: FAMILY TUPAIID^?LYON. Ill
112 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.
all collected on the Boraean mainland, opposite the island of Labuan,by Mr. Hugh Low; skins in fair condition; skulls all considerablydamaged in occipital region.Geographic distrihution.?Throughout the island of Borneo, and theisland of Banguey at the northeastern extremity, and the island ofLaut at the southeastern extremity of Borneo. See No. 2 on mapon page 111.Diagnosis.?This subspecies is distinguished from the othersm hav-ing a distmct russet wash over the lower back and slightly extendedon to base of tail and in having the tails very dark and blackish.Shull and teeth.?There are no characters by which the skulls andteeth of Tupaia minor minor may be distinguished from those of theother subspecies. (Plate 10, fig. 3.)Measurements.??or measurements of the five cotyj^es see table,page 113. Usual measurements of adults: Head and body, 125-130mm. ; tail, 145-160; hind foot, 32-33 ; condylobasal length, 32-34; zygo-matic width, 19-20; width of brain case, 16-17; maxillary tooth row,12-12.5. For individual measurements, see table, page 113.RemarTcs.? Tupaia minor minor is widely distributed throughoutBorneo and appears fairly constant in its characters. There is somevariation in the distinctness of the shoulder stripe. The shoulderstripe in the cotypes is rather dull and inconspicuous, but in a speci-men from the Trusan River, close to the locality of the cotypes, theshoulder stripes are quite white and conspicuous, nearly as much soas are those in the specimens from Mount Dulit, which have very con-spicuous shoulder stripes. The four mounted Sarawak skins in Genoaalso have very distmct whitish shoulder stripes. The smgle specimenfrom Palo Laut has the tail darker than usual, while in the specimenfrom Banguey the tail is hghter.Miiller and Sclilegel * did not distinguish between the three groups?javanica, minor, gracilis?of the small long-tailed olivaceous treeshrews, but grouped them all as Tupaia javanica. The skull of ayoung individual illustrated on plate 27,^ figures 13 and 14, is appar-ently an example of Tupaia minor and not of T. javanica. The otherfigures on that plate intended for T. javanica undoubtedly representthat species.The skms recorded as Tupaia javanica from Pontianak, Borneo, byJentink ^ are probably examples of T. m. minor.Concermngits habits, Hose ? remarks: ''This little tree shrew is fairlycommon, both on the mountains to a height of 4,000 feet and m thelow country. It breeds in a nest in an old stump covered withcreepers, but I am not sure whether it makes the nest itself or occupiesthe nest of a bird. I have found two of these nests, but the materiarused was different." 'Specimens examined.?Twenty-jwo. See table, page 113.
1 Verb. Nat. Gesch. Ned. Overz. Bez., 1839-44, p. 165, pis. 26 and 37.
' Cat. Syst. Mamm. Mus. Hist. Nat., vol. 12, 1888, p. 116.3 Mammals of Borneo, 1893, p. 30.
KO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 113Measurements of Tupaia minor minor.
Locality.
114 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.TUPAIA MINOR MALACCANA Anderson.1879. Tupaia malaccana Anderson, Zool. Res. West. Yunnan, p. 134, pi. 7,fig. 16.Type-locality.?Malacca, Malay Peninsula.Type-specimen.?The two specimens on which Anderson's descrip-tion was based are said by him to have been "procured at Malaccaand referred by Blyth to T. javanica." They are without doubt No.242, "A. B. Specimens from Malacca, and C. Skull, presented by Mr.Frith (1846)," (p. 82), in the Catalogue of the MammaHa in theMuseum Asiatic Society, by Edward Blyth, 1863. I have not seenthe specimens.Geographic distribution.?Southern end of Malay Peninsula, Selangor,Malacca, Johore, Sumatra, and the island of Linga. See No. 3 onmap on page 111.Diagnosis.? Tupaia minor malaccana differs from the other sub-species of Tupaia m having, the tail more like the general color of thehead and body, less blacldsh, and in having less of a chestnut or russetwash on posterior half of body.SJcull and teeth.?These show no characters but slightly larger sizeto distinguish them from the other subspecies. (Plate 10, fig. 4.)Measurements.?Usual measurements of adults: Head and body,125-140 mm.; tail, 150-165; hindfoot, 32-34; condylobasal length, 33-35; zygomatic width, 20-21; width of braincase, 16-17; maxillarytootkrow, 12-13. The measurements are nearly all very slightlylarger than the corresponding ones in Tupaia minor minor.RemarTcs.? Tupaia minor malaccana is not a very different formfrom T. m. minor. Although T. m. minor was described in 1876,Anderson in describing T. m. malaccana in 1879 did not seem to beaware of Giinther's work. If he had, he would probably have iden-tified the Malaccan specimens with it. The specimens that I haveidentified as T. m. malaccana show a few variations. The Lingaspecimens seem to be slightly larger than the others, and are ratherintermediate in color between the subspecies from Singkep andtypical T. m. malaccana, but generally more like the Malaccan speci-mens. The Sumatra skins as a whole, especially shown in one fromthe Indragiri River, have a greater tendency to a russet wash on thelower back than the Peninsular skins. Specimens of T. m. malaccanaare recorded by Schneider ^ from Unter Lankat, Sumatra. Thespecimen "ge," recorded by Jentink^ as Tupaia javanica from Deli,Sumatra, is probably an example of T. minor malaccana. Theother Sumatra specimen, c, from Upper Padang, and c-i, may be thepresent species or may be examples of T. javanica.Specimens examined.?Ten from ^lajay Peninsula, 6 from Sumatra,and 4 from Linga Island.
1 Zool. Jahrb., vol. 23, Heft 1, 1905, p. 88.
? Cat. Syst. Mamm. Mus. Hist. Nat., vol. 12, 1888, p. 117.
NO. 1976. TREE8HREW8: FAMILY TUPAIID^?LYON. 115Measurements of Tupaia minor malaccana.
Locality.
116 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.34.5; zygomatic width, 20.5, 19.5; width of brain case, 16.5, 16.5;maxillary tooth row, 13.5, 13. See table below.Remarks.? Tupaia minor sincipis is a fairly distinct form; it issuJOGiciently distinguished by its slightly larger size and more russetcolor of the lower back. It is rather closely approached by thesmall tree shrews of the neighboring island of Lin^a, which specimensare rather intermediate between T. m. sincipis and T. m. malaccana,but in general more like the latter form. The Sinkep specimenshave previously been identified as T. malaccana.^Specimens examined.?Six from Sinkep Island.Measurements of Tupaia minor sincipis.
NO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 117in Tupaia javanica and usually of a less clear whitish or buff colorthan in T. minor; tail, long, soft, and more distichously haired thanin either T. javanica or T. minor, the upper surface a fine grizzle ofblackish and buffy or sometimes russet, the darker color in excess;underside of tail similar, but the dark element less conspicuous, ex-cept along the edges and at tip, on either side of middle line a clearor nearly clear area of the buff or russet color; shoulder stripe lightbuffy in color, less conspicuous than it is in Tupaia minor.Shull and teeth.?The skull of Tupaia gracilis is of the same generalshape as that of T. minor, with a rather large inflated brain case and ashort stubby rostrum, but as a whole the skull is larger and the braincase relatively more enlarged. The teeth are strildngly differentfrom those of Tupaia minor or T. javanica in the moderate develop-ment of the central upper incisors, the lack of special development ofthe lower canines, and the relatively greater size of the third lowerincisor. The teeth are essentially of the same form as those ofTupaia glis ferruginea, while the skull as a whole is essentially ofthe form of that of T. minor.Subspecies.?Three subspecies of Tupaia gracilis are here recog-nized, the typical form on Borneo, a race on Karimata and another onBanka and Billiton.Remarlcs.? Tupaia gracilis is a very distinct species and its com-bination of external, cranial, and dental characters serve to distin-guish it very clearly from T. javanica and T. minor, which resembleit superficially. The available material of T gracilis is even moreunsatisfactory than that of T minor. It is apparently a rare animaland many of the specimens are not fully adult, or are not in the bestcondition for systematic work. The distribution of this species israther limited. Outside of Borneo the only islands on which itoccurs are islands whose fauna is very closely related to that of Borneo.TUPAIA GRACILIS GRACILIS Thomas.1893. Tupaia gracilis Thomas, Ann. Mag. Nat. Hist., ser. 6, vol. 12, p. 53, July,1893.Type-locality.?Apoh River, base of Mount Batu Song, BaramDistrict, northern Borneo.Type-specimen.?In British Museum, Reg. No. 92.9.6.2. Collectedon Apoh River, base of Mount Batu Song, northern Borneo, in Sep-tember, 1891, by A. H. Everett; skin in good condition, but skuUwith occipital region and basal parts of brain case broken away.Geographic distribution.?Probably generally distributed through-out Borneo; specimens in collection are known only from BaramDistrict and southwestern Borneo. See No. 5 on map on page 111.Diagnosis.? Tupaia gracilis gracilis is the largest of the tlu'ee sub-species, hind foot over 40 mm., without special peculiarities of color,less russet color present posteriorly, and slightly larger skull.
118 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.Measurements.?^TyiDe: Head and body (from dried skill), 155 mm.;tail (from dried skiii), 160; hind foot, 40; zygomatic width, 20.5; widthof brain case, 17.5 ? ; maxillary tooth row, 14. Usual measurementsof adults: Head and body, 140-150; tail, 155-170; hmd foot, 40-42;condylo-basal length, 36-36.5; zygomatic width, 20.5-21; width ofbram case, 17.5; maxillary t^ooth row, 13-14. See also table, page 119.Specimens examined.?Ten, including the type from Baram Dis-trict, northern Borneo, and one from southwestern Borneo. Seetable, page 119. TUPAIA GRACILIS INFLATA Lyon.1906. Tupaia inflata Lyon, Proc. U. S. Nat. Mus., vol. 31, p. 600, December18, 1906.Type-locality.?Tanjong Rengsam, Island of Banka.Type-specimen.?In United States National Museum, Cat. No.124709. Skui and skuU of adult male, collected at Tanjong Reng-sam, Banka, May 20, 1904, by Dr. W. L. Abbott; original number,3241; skin and skuU m good condition.Geographic distribution.?Islands of Banka and Billiton. See No. 6on map on page 111.Diagnosis.?Of the same general color as Tupaia gracilis gracilis,but with more of a russet tmge posteriorly, hmd foot slightly smaller,not exceeding 40 mm.; skuU with relatively larger buUag. (Plate 10,fig. 2.)Measurements.?Type: Head and body, 150 mm.; tail, 172; hmdfoot, 37; condylo-basal length, 37; zygomatic -svidth, 21; width ofbrain case, 17.5; maxillary tooth row, 13.5. Usual measurementsof adults: Head and body, 140-150; tail, 160-170; hmd foot, 37-39;condylo-basal length, 36-37; zygomatic width, 20-21; width of braincase, 16.5-17.5; maxillary tooth row, 13-13.5. See table, page 119.RemarTcs.? Tupaia gracilis injiata is only shghtly differentiatedfrom T. g. gracilis. At the time it was originally described it wascompared with T. javanica and T. minor, and of course appeared tobe a very distmct form. It is weU established now that it has nothingin common -with those two species. Two of the specimens. Cat. No.124985, Billiton, and Cat. No. 124909, Banka, have the tails verydifferent m color from other spechnens, the usual buffy annulationsbeing replaced by russet or tawny ochraceous; it appears to be adifference in pelage.Specimens examined.?Three, including the type, from Banka, andthree from Billiton.TUPAIA GRACILIS EDARATA, new subspecies.Type-locality.?Telok Edar, Karimata Island, off west coast ofBorneo.Type-specimen.?In United States National Museum, Cat. No.153859, skin and skuU of young adult female, collected on Karimata
NO. 1976. TREESHREW8: FAMILY TUPAllDJE?LYON. 119Island by Dr. W. L. Abbott, October 5, 1908; original number, 6226;skin in good condition; skull imperfect.GeograpTiic distribution.?Known only from Karimata Island. SeeNo. 7 on map on page 111.Diagnosis.?Characterized by having the general olivaceous color ofthe upper parts strongly approaching bister, the effect produced by afine grizzHng of blackish and ochraceous; tail more brownish and lessdifferent in color from head and body than in the case of the othersubspecies; light colors of underside of tail distinctly ochraceous.Slcull aTid teeth.?There are no characters by which the skull andteeth of Tupaia gracilis edarata may be distinguished from those ofT. g. injiata.Measurements.?Type: Head and body, 135 mm.; tail, 156; hindfoot,38; condylo-basal length, 36 ?; zygomatic width, 19; width of^braincase, 17; maxillary tooth row, 13.RemarTcs.? Tupaia gracilis edarata is more like the subspecies injiatathan the typical form and I identified it as Tupaia injiata in 1911.'It is browner than any of the Banka-Billiton specimens, and althoughbased on the examination of but a single sldn it is apparently afairly well defined geographic form.Specimen examined.?One, the type.Measurements of Tupaia gracilis.
Name.
120 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.ANATHANA, new genus.Type.? Tupaia ellioti Waterhouse, Proc. Zool. Soc. London, 1849,p. 107, Plate, mammalia, 13.Diagnostic cliaracters.?^A member of the mammalian familyTupaiidse differing from typical Tupaia as restricted on page 30 inhavino- larger and better haired ears, coarser reticulations on nakedarea of nose; unusually well-developed hypocones on the uppermolars, lower canines not projecting above the level of the adjacentteeth; relatively short andheavy rostrum ; and a smalland inconspicuous fenestrain zygoma.External cliaracters.?^Ex-ternally thegenus^naif?ionais very simuar to Tupaia,more like it than any othermember of the family. Themost tangible differencesare the distinctly greatersize of the lower portion ofthe external ear in compari-son with the upper portion,due to a widening out of thelower lobe, which projectsbackward to a considerableextent, and slightly for-ward. The interior of theear is much better hairedthan is the ear of Tupaia.The rhinarium is squarelycut across posteriorly as inTupaia, but is apparentlymore coarsely reticulatedthan in Tupaia and thereticulations are more regu-larly arranged. The footpads in Anathanu are essentially as they arein Tupaia, but on the hind foot the internal proximal pad is per-fectly distinct from the first interdigital pad in the few specimensthat have been suitable for making the observations; but in Tupaiaitself this distinctness of the two pads is often indicated, but usuallythe two are fused to form one long pad. The pelage of Ajuithanaappears distinctly coarser and harsher than that of Tupaia, and thegrizzling of the colors is coarser, differences something like those seenbetween Sciurus and Xerus. Nearly all the skins of Anathanadepending, however, a good deal upon the manner of arranging and
Fig. 6.?Anathana wkoughtoni, Type x li- Keg. No.11. 7, 1, Bkitish Museum, Mandvi, India.
NO. 1976. TREESHREW8: FAMILY TUPAllD^?LYON. 121
?
stuffing the head, show ill-defined but still evident light Imes over andbelow the eye, thus suggesting the more pronounced face markingsseen in most species of Dendrogale. A shoulder stripe is present.Mammae are 3-3 = 6, 1 axillary, 1 inguinal, 1 ventral near the ingui-nal one. (Plate 4.)Cranial characters.?The skull of Anathana shows little devia-tion from the typical Tupaia skull. The rostrum is short and heavy,does not arise from the skull abruptly as in the Twpaia minor group,where the rostrum is also relatively short. The distance from thelachrymal notch to end of premaxilla is about equal to the distancefrom that notch to end of external pterygoid plate. In Tupaia, withthe exception of the minor group, the dis-tance from the lachrymal notch to end ofpremaxilla equals the distance from thenotch to the external auditory meatus.Correlated with the shortened rostrumthe premaxilla is correspondingly short,antero-posteriorly and relatively high,supero-inferiorly. Usually in Tupaia thenaso-premaxillary suture is considerablylengthened or apparently obliquely pushedbackward, making the premaxUla a ratherobliquely elongated bone. In Anathanathe premaxilla is somewhat quadrate inshape and only slightly distorted. Ex-treme obliquity of the premaxilla is seenin the genus Tana. The fenestra in thezygoma in AnatJiana is reduced to a smalloval foramen quite different from thelarge opening found in Tupaia. Whileit appears a trivial character, yet m theTupaiidse the size of this openmg is animportant feature correlated with other pecuKaritiespage 120.Dental characters.?The most evident of the dental characters inAnathana is the relatively small size of the lower canine, so that whenthe lower jaw is viewed from the side this tooth does not stand con-spicuously higher than the adjacent incisor and premolar. The devel-opment of hypocones on the first and second upper molars is very con-spicuous in Anathana; they are larger than in any other member ofthe family though nearly equalled by the hypocones on the first twoupper molars of Urogale and Ptilocercus. All the upper molariformteeth of Anathana are relatively wider and shorter than they are inthe other genera. This is especially well shown m the third andfourth premolars, where the protocones are better developed andextend inward toward the median line more than is the case in othergenera of Tupaiidse. Looked at on its grindmg surface the fourth
Fig. 7.?Upper and lower tooth-rows OF Anathana wkoughtoniType x 3, Reg. No. 96. 11. 7. 1,British Museum, Mandvi, India.See fig. 6 on
122 PROCEEDINGS OP THE NATIONAL MUSEUM. vol. 45.premolar is distinctly four-sided in outline, while in Tupaia it israther triangular. The third premolar in Anathana has a very evi-dent protocone, while in Tupaia and Tana it is quite rudimentary.Corresponding with the conspicuous development of the protoconesof the third and fourth premolars the metaconid of the fourth lowerpremolar is much increased in size as compared with its developmentin Tupaia. See iig. 7 on page 121.A cecum about 25 mm. long is perhaps one of the generic characters.See page 14.GeograpMc distribution.?Anathana is confined to the Indianpeninsula, south of the River Ganges. The northeastern part of itsrange almost meets the southwestern limit of Tupaia cMnensis. Sofar as our present knowledge goes the two genera do not overlap inthen- distribution. See map, page 125, Nos. 1, 2, 3.Number of forms.?The genus Anathana contains but a singlespecies group, easUy separable into three forms. Perhaps they arenothing more than subspecies, but material is at present lacking toshow intergradation.Anathana ellioti, eastern India; upper parts and tail reddishbrown, feet and hind legs buff or ochraceous, page 122.Anathana pallida, northeastern India; upper parts a less conspicu-ous reddish brown, different in color from tail, feet and hind legsgrizzled buffy, page 124.Anathana wroughtoni, western India; upper parts dull grizzledbrownish, tail slightly dissimilar, feet and hind legs grizzled grayish,page 123.Eemarks.?Anathana is clearly closely related to Tupaia inmost of its characters. In the small size and oval shape of thezygomatic fenestra it resembles Dendrogale, as well as in the barelymdicated face markings, which are even lacking in one species ofDemlrogale. In the development of the hypocones of the uppermolars and the protocones of the premolars it is approached byVrogaJe and Ptilocercus. It is quite distinct from any of the familyin the relatively smaU size of the lower canines, and is the antithesisof Urogale, which has the lower canines exceedmgly weU developed.ANATHANA ELLIOTI (Waterhouse).1849. Tupaia ellioti Waterhouse, Proc. Zool. Soc. London, 1849, p. 107, plate,mammalia, 13.1879. Tupaia ellioti, Anderson, Zool. Res. West Yunnan, 1879, p. 124, pi. 7,figs. 12 and 13.1888. Tupaia ellioti, Blanford, Fauna Brit. Ind. Mamm., p. 209, 1888.Type-locality.?Hills between Cuddapah and NeUore, eastern Ghats,India, not far from Madras.Type-specimen.?In British Museum, Reg. No. 50.1.21.5, adultmale, collected by W. Elliot, who procured specimens "from the hUlsbetween Cuddapah and NeUore in what may be termed the eastern
NO. 1976. TREESHREW8: FAMILY TUPAIID,^?LYON. 123Ghats." The label of the specimen reads "Madras," the nearest largecity to the locality mentioned by Waterhouse. The specimen israther old, mounted, and without skull, but in a good state of preser-vation and apparently not injured as to color by exposure to light.There are two other specunens, paratjq^es, collected by Elliot at thesame locality: 50.1.21.7, a young individual recently made into amodern study skin, in good condition and color, and with fragmentsof skull present, and 50.1.21.6, with skull, 50.8.21.16, an adultfemale, long mounted and much exposed to the light and so bleachedthat the grizzling is lost, and the brownish color of the back merelyindicated.GeograjjJiic distribution.?Eastern Ghats, and Sheveroy HiUs to thesouth, India. See No. 1 on map on page 125.Diagnosis.?Upper parts of body and tail distinctly reddish brown;feet buff to ocliraceous.Colo7\?Upper parts of body and tail a coarse grizzle of tawnyoclu"aceous and blackish, the former much in excess and the grizzlecoarser on the tail; head, neck, and sides of body a grizzle of oclira-ceous and blackish, both colors about equally mixed; outer side oflegs similar to adjacent parts of body; underparts and inner side oflegs buffy; upper surfaces of hands and feet, dark buff or ocliraceousbuff; shoulder stripe whitish cream color; underside of tail similarto upper side, except in the median hne where the short appressedhairs have the general color of the underparts,STcull and teeth.?There are apparently no characters by which theskull and teeth of Anatliana ellioti may be distmguished from those ofother species in the genus.Measurements.?Type: All from skm, head, and body, 180 ? mm.;tail, 180 ? ; liindfoot, 40. Skull of an adult paratype, Reg. No.50.1.21.6, condylobasal length, 42; zygomatic width, 22.5; width ofbraincase, 17.5; maxillary toothrow, 16. External measurements ofspecmiens preserved in alcohol: Head and body, 160-180; tail, 185-190; hindfoot, 44-45. See table, page 126.Remar-ks.?The Sheveroy Hills specmien, represented by a flatsldn with skull, is not typical, the head bemg darker, the back,and especially the tail being less reddish than in the Cuddapahspecimens, but the underparts, feet, and underside of tail beingdistinctly buffy, as in the others.Specimens examined.?Eight: 5 skins and 3 preserved in alcohol.ANATHANA WROUGHTONI, new species.Type-locality.?Mandvi, near Bombay, India.Type-specimen.?In British Museum, Reg. No. 90.11.7.1, skin andskull of adult female, collected at Mandvi, near Bombay, India,March 21, 1896, by R. C. Wroughton; in good condition.Geograpliic distribution.?Region of Satpura Hills, and Dangs, nearBombay, western India. See No. 2 on map on page 125.
124 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Diagnosis.?Upper parts of body dull grizzled brownish, tail slightlydissimilar in color from body, feet and hind legs, grizzled grayish.Color.?Same general pattern as in Anathana ellioti, but upper partsof body not a conspicuous reddish brown, but a grizzle of blackish andochraceous ; tail a coarse grizzle of buff and blackish, the latter colormexcess; sides of head, neck, and body, and outer side of legs and feet agrizzle of pale or cream buff and blackish; underparts and inner sidesof legs whitish; shoulder stripe whitish, but not conspicuous.Shull and teeth.?There are no tangible characters by which theskull and teeth of Anathana wroughtoni may be distinguished fromthose of A. ellioti.Measurements.?Type: Head and body, 177 mm.; tail, 187; hindfoot, 44; condylobasal length, 42; zygomatic width, 22.5; width ofbraincase, 18; maxillary toothrow, 15.Specimens examined.?Three, one each from Mandvi, Dangs, andMatheran.Remarlcs.?^The "Indian Tupaia," recorded in the Satpura HiUsby Ball m 1874,^ is undoubtedly this species.ANATHANA PALLIDA, new species.Type-locality.?Munbhum, northeastern India.Type-specimen.?In British Museum, Reg, No. 66.12.28.2, skin andskull of young adult female, collected at Munbhum, India, in 1865by R. C. Beavan; in good condition.Geographic distribution.?Northeastern India, rangmg from Raipurnortheastward as far as the Ganges River. See No. 3 on map onpage 125.Diagnosis.?Generally paler in color than is Anathana ellioti,upper parts only slightly reddish brown; tail not concolor with body;hind feet and legs grizzled buffy.Color.?Of the same general pattern as that of Anathana ellioti,but the reddish brown of the back is not nearly so dark or conspicuous,and is not continued on to the tail; tail similar in color to sides ofbody; shoulder stripe quite conspicuous and nearly clear white; under-
J
parts and inner side of legs whitish. 1SlcuU and teeth.?These show no evident characters by which theycan be distinguished from those of Anathana ellioti.Measurements.?Type : Head and body (from dried skin), 185 ? mm.
;
tail, from dried skin, 165 ?; hind foot, 41; zygomatic width, 21;width of braincase, 17; maxillary toothrow, 15. For measurementof a second specimen see table, page 126.Remarlcs.?Several specimens are recorded by Anderson in theIndian Museum from Monghyr or near there. They are probablyexamples of this species. The Monghyr specimens bring the range ofAnathana very close to that of Tupaia chinensis, to the northeast ofthe Ganges Valley.Specimens examined.?Two, the type and one from Raipur.
? Proc. Asiat. Soc. Bengal, 1874, p. 95. J
NO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 125
126 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Measurements of the genus Anathana.
Name.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 127portion. Some members of the genus have conspicuous face markings,caused by a dark streak extending from half way between nostriland eye through the eye to the ear, and a lighter stripe above and onebelow this line. In the dark species Dendrogale melanura these mark-ings ftre only barely traceable. Similar face markings are also faintlyindicated in Anathana. The shoulder stripe present (sometimesalmost obsolete in Tupaia) in all the other members of the familyexcept Ptilocercus is absent in Dendrogale. There is probably onlyonly one pair of mammge and they are inguinal.Cranial characters.?The skull of Dendrogale is in general propor-tioned as in Tupaia. Theonly very striking differ-ence is the reduction ofthe zygomatic fenestra toa small foramen. In thecomparatively few speci-mens that I have ex-amined the skull, m ad-dition to being generallysmall, has a more roundedbrain case, is less angular,has less conspicuous tem-poral ridges as comparedwith Tupaia. Unfortu-nately none of the speci-mens have been oldadults . The skuU figuredby Sclilegel and Miiller ^is quite as angular as themajority of Tupaia skills.See below under type-specimens of Dendrogalemurine, page 130.Denial characters.?Theteeth of Dendrogale arevery similar to those of Tupaia. The hypocones of the upper molarsare considerably reduced, so that practically none are found on thesecond and third molars. The anterior teeth, i^, i^, c\ Cj, pm2, pm^,are more trenchant in character than they are in Tupaia, especiallyseen in the second upper incisor. The dental characters, however,are relatively unimportant and would not be of themselves sufficientto warrant the generic distinctness of Dendrogale, but in conjunctionwith other characters serve to emphasize the validity of the genus.
Fig. 8.?Dendrogale melanuea melanura, Type x li-Reg. No. 92. 2. 7. 11, British Museum, 5,000 feet onMount Dulit, northern Borneo.
1 Verb. Nat. Gesch. Nederl. Overz. Bezitt., 1843, pi. 27, figs. 17-18.
128 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Geographic distribution.?So far as known Dendrogale occurs onlyin Borneo and southeastern French Indo-China. As members of thegenus Dendrogale are very rare in collections, it is not at all probablethat this represents the extent of its distribution. When explora-tions of the Malayan region are more Com-plete it will probably be found elsewhere,particularly Sumatra, where the rare genusPtilocercus was unknown for some time.(A M'' See also remarks under Dendrogale murina^ # below, page 131. See A, B, M, N, on mapI on page 133.\i ^ Number offorms.?Four named species orsubspecies are recognized. They fall intotwo very distinct almost subgeneric groups
:
1. Murina growp; color light, face-markingspresent, claws small, plate 5.Dendrogale murina, Pontianak, Borneo,page 129.Dendrogale frenata, Cambodia, page128.2. Melanura group, color dark, face-mark-ings absent, claws long.Dendrogale melanura melanura, MountDulit, Borneo, page 132.Dendrogale melanura baluensis, MountKina Balu, Borneo, page 132.
Fig. 9.?Upper and l wer tooth-rows OF Dendrogale melanu-ra MELANURA, X SJ. REG. NO.92. 2. 7. 10, British Museum,5,000 feet on Mount Dxn.iT,northern Borneo. KEY TO THE SPECIES OF DENDROGALE.Color light, face marking conspicuous, claws small, 1.5-2 mm.Hind foot more than 25 mm. (27-29) (southeastern French Indo-China)f. D. frenata, p. 128.Hind foot less than 25 mm. (22) (Borneo) D. murina, p. 129.Color very dark brown, face markings absent, claws large, 4 mm.Upper parts dai"ker, base of tail and rump distinctly rusty, tail and feet darker(Borneo; Mount Dulit) D. melanura melanura, p. 132.Upper parts lighter, base of tail and rump less rusty, tad and feet lighter(Borneo; Mount Kina Balu) D. melanura baluensis, p. 132.DENDROGALE FRENATA (Gray).1860. Tupaia frenata Gray, Ann. Mag. Nat. Hist., ser. 3, vol. 6, p. 217, August,1860.1879. Dendrogale frenata, Anderson, Zool. Res. West. Yunnan, 1879, p. 110.Type-locality.?Cambodia.Type-specimen.?In British Museum, Reg. No. 60.8.28.11, skin andskull, collected in Cambodia by M. Mouhot; skin is mounted and infair condition; skull considerably damaged between the palate andforamen magnum, not fully adult, dpm^ still in place and pm^ justappearing beneath it.
NO. 1076. TREESHREWS: FAMILY TUPAIID^?LYON. 129Geographic distribution.?Cambodia and Anam. See A on map onpage 133.Diagnosis.?Color light, face markings conspicuous, claws smalland blunt, 1.5-2 mm.; hind foot 25 mm. or over.Color.?Type, with rather soft and close fur, upper parts of neck,body, and base of tail a fine grizzle of brownish black and ochraceous,(the ochraceous inclining to buff anteriorly and to tawny posteriorly)the two colors about equally mixed; outer side of legs similar to adja-cent parts of body; upper parts of head a fine grizzle of blackish andbuflf, the former in excess; underparts and inner side of legs palebuffy ; side of head with a blackish line beginning at base of whiskers,gradually becoming wider, and running through eye to ear, borderedabove by a buffy line and below by a similar line; tail above similarto back, but darker for distal three-fourths, below, a dark line in themiddle, bordered by an ochraceous line, and on the extreme outeredge by the color of the tail above. Plate 5.A paratype is essentially like the type, but is more tawny on therump and base of tail. Reg. No. 7.1.1.1, a topotype collected in1861 (?), is ochraceous buff on the underparts instead of pale buffy.Two modern skins, Reg. Nos. 6.11.6.5 and 6.11.6.6, from Nhatrang,Anam, are in general similar to the original series, but are slightlymore yellowish anteriorly, and one of them, 6.11.6.5, has rather longsoft fur like that of Dendrogale melanura.ShuU and teeth.?These are practically indistinguishable from thoseof other species of Dendrogale that I have examined ; the brain caseis not quite so inflated or arched as that of D. melanura.Measurements.?Type: Head and body (from mounted skin) , 135mm.; tail (frommountedskin),95; hind foot, 27; condylobasal length,31; zygomatic width, 17; width of brain case, 15.5; maxillary toothrow, 12. For individual measurements, see table, page 134.Specimens examined.?Four from Cambodia and two from nearNhatrang, Anam.DENDROGALE MURINA (Schlegel and Mflller).1843. Hylogalea murina Schlegel and Muller, Verb. Nat. Gesch. Nederl.Overz. Bezitt., p. 167, pi. 26, fig. 5; pi. 27, figs. 17 and 18; entire animal incolors, and lateral and dorsal views of skull both apparently very good.1879. Dendrogale murina, Anderson, Zool. Res. West. Yunnan, p. 110, pi. vii,figs. 18 and 19, skull.1888. Glipora murina, Jentink, Mus. Hist. Nat. Pays-Bas, Cat. Syst. Mamm.,vol. 12, 1888, p. 118. Publication of Diard's manuscript name under thebeading Typaja murina.Type-locality.?Pontianak, western Borneo.Type-specimen.^?In the Leyden Museum, mounted skin and skull;male, skin somewhat faded and dust stained; skull broken away
J I have not seen this specimen, which appears to bo the only represcntaiive of the species known. Myinformation regarding the specimen and species is obtained from notes made by Mr. G. S. Miller, jr., in theLeyden Museum in 1905.80459"?Proc.xN.M.vol.45?13 9
130 PROCEEDINGS OF THE NATIONAL MUSEUM. tol. 45.behind orbits but perfect in front and with complete lower jaw.(The skull figured by Schlegel and Miiller is perfect. It is possiblethat the posterior portion of their figure was made up by analogyfrom a Tupaia skuU. Anderson figures a perfect skull but I suspecthe copied Schlegel and Miiller's figures, transposing their lateral viewfrom right to left, and apparently the dorsal view also in the processof lithographing. The skull was very incomplete in 1888 accordingto Jentink.^ (See Plate 5.)Geographic distribution.?Known only from the type-locality, butundoubtedly occuring elsewhere in Borneo. See B on map on page133.Diagnosis.?Differs from Dendrogale frenata mainly in the smallersize of hind feet, and if the illustrations of the skull are accurate, inhaving much narrower nasals; and tail above different in color frombody.Color.
^
?"Underparts and under surface of tail dull light ochra-ceous-buff. Upperparts the same but much darkened dorsally byadmixture of a dark broccoli brown. Tail an indefinite dark tawnybrown above, line of demarkation on tail sharp, but colors not form-ing any strong contrast. Median line of tail below not differentfrom rest of its lower surface. At middle the hairs alongside of tailare 5 mm. in length, at tip they are 10 mm. Outer surface of legsslightly less yellowish than sides of body. Feet dark. A faint darkshade passes from muzzle through eye to ear; above, it is bordered bya light area 3 mm. wide behind eye, less distinct in front. Below,the light border may be detected behind eye, but not in front."Slcull and teeth.?If the illustrations are to be relied on the skull ofDendrogale murina would appear to bo much more angular and withbetter developed temporal ridges than that of D. frenata, and tohave much slenderer nasals. The teeth in the two species are prob-ably essentially the same; their measurements agree almost exactly.Measurements.'? ' ' Head and body, 115 mm. ; tail, 110; hind foot (dis-torted), 22 (all from mounted specimen); least distance from orbitto tip of premaxillary 13.6 (13.8);^ least interorbital breadth 9.8 (10);zygomatic breadth (approximately) 17; mandible, 22 (22); max-illar}^ tooth row (entire including incisors) 17.4." The maxillarytooth row, including incisors, in the illustration is the same or nearlyso, and the measurement of the maxillary tooth row without the in-cisors in the illustration is 12, the same as the maxillary tooth rowof D. frenata.RemarTcs.-?It is unfortunate that direct comparisons of Dendrogalefrenata and D. murina have not been made. The two animals
1 Mus. Hist. Nat. Pays-Bas Cat. Osteol. Mamm., vol. 12, 18S8.s Quoted from manuscript notes made by Mr. G. S. Miller, jr.3 Measurements in parentheses are those made by Mr. Miller or the type of T. frenata.
NO. 1976. TREESHBEW8: FAMILY TUPAIID^?LYON. 131appear to be very similar. Their most apparent differences have beenpointed out above, and from what is known of the fauna in generalof Borneo and Anam, the two animals would certainly be expectedto be different; in fact much more different than they appear to be.Both species have probably some peculiarity of habit making themdifficult to secure. So far as I Imow only one specimen of Dendro-gale murina has been collected. It is curious that no subsequentspecimen has been obtained since the original. As the geographicdistribution of the genus is so peculiar and so totally unlike that ofany of the related genera or in fact of most genera of mammals thereis just a possibility that Dendrogale murina is an example of D.fre-nata wrongly labeled as coming from Pontianak, Borneo. Dr. W. L.Abbott, with much careful collecting in the region of the type-locality,failed to secure it.Specimens examined.?None.DENDROGALE MELANURA (Thomas).(Synonymy, type-specimens, etc., under the subspecies.)Geographic distribution.?High mountains of northern Borneo. Nospecimens recorded below 3,000 feet. See M and N on map on page133.Diagnosis.?Distinguished from the other members of the genus byits dark-brown color, lack of face markings, and by its large sharpclaws, about 4 mm. in length, fur long and soft; mammae 1-1 = 2.Color.?Upper parts of head, neck, and body a fine grizzle ofblackish and ochaceous buff, anteriorly, and cinnamon rufous, pos-teriorly and at base of tail, the darker color in excess; bases of hairsslate black; outerside of legs similar to adjacent parts of body; feetdark brownish; underparts ochraceous with slate bases of hairs show-ing through; inner side of legs similar to adjacent underparts; im-mediately above and below eye, a short ochraceous line, both togetlierappearing like an eye ring with indistinct corners; tail a mixture ofblack or blackish, and cinnamon rufous with the black very prominentwhen viewed from above; tail, seen below, with the short appressedhairs in middle line black or brownish black, bordered on either sideby other short hairs with more or less evident cinnamon rufous bases.SJcull and teeth.?These are practically indistinguishable from thoseof Dendrogale frenata, except that the braincase is slightly moreinflated and arched.Subspecies.?Two races of Dendrogale melanura are known, one fromMount Dulit and the other from Mount Kina Balu, neither of themvery highly differentiated.
132 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.IXENDROGALE MELANURA MELANURA (Thomas).1892. Tupaia melanura Thomas, Ann. Mag. Nat. Hist.,'ser. 6, vol. 9, p. 251,March, 1892.1892. Tupaia melanura, Thomas, Proc. Zool. Soc. London, 1892, p. 224, pi. 19,figs. 4, 5, skull.Type-locality.?Mount Dulit, northern Borneo, 5,000 feet altitude.Type-specimen.?In British Museum, Reg. No. 92.2.7.10, skin andskull of adult female, collected at 5,000 feet on Mount Dulit, Borneo,October, 1891, by Mr. Charles Hose; skin well preserved, but some-what damaged about the head; skull in fair condition, slightlydamaged in supraoccipital region. (See figures 8 and 9 on pages 127and 128.)Geographic distribution.?Known only from Mount Dulit, Borneo.See M on map on page 133.Diagnosis.?Upper parts of head and body dark, due to a deepershade of the ochraceous buff element of the color; base of tail andrump with the cinnamon rufous element of the color more conspic-uous, so that that part of the animal appears distinctly rusty; feetand especially the tail darker, the latter with more black and the och-raceous on the underside less in evidence.STcull and teeth.?These are without distinguishing characteristics.Measurements.?See table, page 134.Specimens examined.?Three from Mount Dulit.Remarl^s.?Hose, in Mammals of Borneo (p. 33, 1893), remarksof this animal: ''The type of this pretty little treeshrew was obtainedby me on the top of Mount Dulit at 5,000 ft., living amongst the moss-covered stunted jungle, and it is apparently a true mountain species,as I have since obtained other specimens, none of which were foundbelow 3,000 ft."DENDROGALE MELANURA BALUENSIS, new subspecies.Type-locality.?Mount Kina Balu, northeastern Borneo.Type-specimen.?In British Museum, Reg. No. 92.9.6.3, skin andskull of adult female, collected on Mount Kina Balu, Borneo, byMr. A. Everett; skin in good condition; skull imperfect.Geographic distribution.?Js^nown only from the type locality. SeeN on map on page 133.Diagnosis.?Distinguished from Dendrogale melanura melanura byhaving the ochraceous buff of the anterior upper parts lighter andmore in evidence, and the cinnamon rufous posteriorly lighter, feetlight brownish; underparts ochraceous buff, rather than ochraceous:tail with black hairs less in evidence and with the appearance of anarrow line of ochraceous on either side of the middle line.STcvll and teeth.?These show no distinguishing characteristics.Measurennents
.
?See table, page 134.Specimens examined.?Three, all from Mount Kina Balu.
xo. 1976. TREESHREWS: FAMILY TUPAIID.l^?LYON. 133
134 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Mdisurements of specimens of the genus Dendroqale.
Name.
NO. 1976. TREESHREW8: FAMILY TVPAIID^?LYON. 135Craniul characters.?Rostrum considerably elongated and attenu-ated, so that the distance from the lachrymal notch to the end ofpremaxilla is equal to the distance from the notch to the occipitalcondyles. The lengthof the rostrum is mainlybrought about by alengthening of the pre-maxiUary and nasalbones and not by anyappreciable increase inlength of the maxilla.Fenestra in zygomalarge and conspicuous,elongated oval in shape.Dental cTiaracters.?The teeth of Tana do notdiffer in form or numberfrom those of Tupaia.Owing to the elonga-tion of the rostrumthe anterior teeth aresomewhat differently
Fig. 10.?Tana tana besaea. Type x 1- Cat. No. 142247,U.S.N.M., Kapuas River, Borneo.spaced. Thus the lower canine stands closerto the last lower incisor than it does to theadj acent premolar, while in Tupaia it standsmidway between the two teeth. The com-bined length of the three upper molars isabout one-third the length of the entire up-per tooth row in Tupaia, but in Tana it isdistmctly less than one-third the length ofentu'e tooth row.The absence of a cecum may be one of thegeneric characters. See page 14.Geographic distribution.?The genus Tanaso far as known, is confined to the land masses of Borneo and Sumatraand certain of the small adjacent islands. Unlike Tupaia it is not
Fig. 11.?Upper and lower tooth-rows OF Tana tana besara.Type X 2. Cat. No. 142247, U.S.N.M. Kapuas River, Borneo.
136 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.found in Java, the Malay Peninsula, or elsewhere on the Asiatic main-land. See map on page 143.1. Dorsalis group; delicate build, short claws, pi. 6, fig. 2.Tana dorsalis, Borneo, pi. 11, fig. 1, page 152.2. Tana group; heavy build, stout claws, pi. 6, fig. 1.Tana tana tana, Sumatra, southern Borneo, pi. 11, fig. 4, page 139.Tana tana besara, Kapuas River, Borneo, pi. 11, fig. 8, page 141.Tana tana utara, northern Borneo, page 141.Tana tana sirhassenensis, Sirhassen Island, pi. 11, fig. 3, page 142.Tana tana bunose, Tambelan Islands, pi. 11, fig. 5, page 144.Tana tana tuancus, Banjak Islands, page 145.Tana lingse, Lioga Island, pi. 11, fig. 2, page 145.Tana cervicalis cervicalis, Tana Bala, Batu Islands, page 147.Tana cervicalis masse, Tana Masa, Batu Islands, page 148.Tana chrysura, Borneo, opposite Labuan, page 149.Tana paitana, northeastern Borneo, Banguey Island, page 150.KEY TO THE SPECIES OP TANA.Size small, a narrow dorsal stripe extending the entire length of the back; claws smalland short; hairs of tail of medium length not strongly distichous. Dorsalis group.Borneo, T. dorsalis, p. 152.Size lai^e, a narrow dorsal stripe on anterior half of back, and bordered by a lightarea different in color from rest of back; claws large and well developed; hairsof tail long and strongly distichous Tana group.Tail, buff, very different in color from rest of body .North Borneo, T. chrysura, p. 149.Tail, reddish brown, not noticeably different in color from rest of body.Shoulder stripe bordered above and below by the light or grayish or buffy colorof the anterior back Northeast Borneo, T. paitana, p. 150.Shoulder stripe bordered above and below by the reddish brown color of the sidesof body.Size small; hind foot, 40 mm.; tail, 135; rostmm very narrow and slender.Linga Island, T. lingse, p. 145.Size medium or large, hind foot over 45 mm. and tail over 150; rosti'um not sonarrow and slender.Reddish colors bright, and back with a very large well marked brilliant blackarea T. <mrvicalis, p. 147.Light areas beside the dorsal stripe, cream buff slightly mixed with black.Tana Bala Island, T. cervicalis cervicalis, p. 147.Light areas beside the dorsal stripe tawny ochraceous slightly mixed withblack Tana Masa Island, T. cervicalis masse, p. 148.Reddish color usually not so bright, generally rather dull, without a largewide brilliant black area on back, the lower back,, however, is alwaysdarker than rest of upper parts and sometimes black, but the black color isnever widely spread T. tana, p. 137,General coloration decidedly dull.Size very large; hind foot 53 mm.; condylo-basal length 59.Western Borneo, T. tana besara, p, 141,Size medium; hind foot about 48 mm.; condylo-basal length about 55;tail about 175 Sumatra, southern Borneo, T. tana tana, p. 139.Size small; hind foot about 44; condylo-basal length about 53; tail about155 Tambelan Islands, T. tana bunoas, Tp. 144.General coloration relatively bright.Light areas beside the dorsal stripe very light and grayish; condylo-basallength 53 mm. or more .... Northern Borneo, T. tana utara, p. 141.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 137Light areas beside the dorsal stripe duller and darker, more ochraceous;condylo-basal length 53 mm. or less.Size slightly smaller; reddish color duller; underside of tail lighter;shoulder stripe buff, well marked.Sirhassen Island, T. tana sirhassenensis, p. 142.Size slightly larger; reddish color brighter; underside of tail darker;shoulder stripe ochraceous; rather poorly developed.Tuangku Island, T. tana tuancus, p. 145.TANA TANA (Raffles).(Synonymy, type-specimens, etc., under the subspecies.)Geographic distribution.?Borneo, Sumatra, and adjacent smallislands. See Nos. 3, 4, 5, 6, 1, and 8 on map on page 143.Diagnosis.?General color of sides and shoulder region hazel; lowerback with an area distinctly darker than rest of upper parts, butsometimes presenting a distinct but rather small black patch; shoul-der stripe bordered above and below by the hazel of the sides; sizelarge.Color.?Sides of body and shoulder region and outer side of legs,hazel, somewhat brighter than Ridgway's, and more or less lined withblackish annulations of the hairs, especially posteriorly, where thedarker color gradually spreads upward to concentrate itself in a darkarea on the lower back, which sometimes appears as a small distinctblack patch, dorsal stripe beginning faintly at occiput but becomingvery distinct over shoulders, blackish reddish brown, and graduallylosing itself m with the dark area of lower back; dorsal stripe bor-dered on either side by a light area sometimes poorly and sometimeswell defined, made up of a rather coarse grizzling of cream buff (whenlight area is well marked) or ochraceous buff (when light area is lessconspicuous) and blackish brown; top and sides of head and neckgenerally like the light areas on each side of dorsal stripe, but ratherlighter and the grizzling finer; underparts, including inner side oflegs, orange rufous to ferruginous; feet generally a dull and oftendarker extension of the color of the outer side of legs; tail above, brighthazel or ferruginous, more or less, and irregularly suffused wdthblackish; underside of tail similar to upper side for its outer half, butthe inner portion orange rufous; shoulder stripe well marked and ofthe same general colors as light area on each side of dorsal stripe, butlighter in tint.SlcuU and teeth.?There are no apparent characters by which theskull and teeth of Tana tana may be distinguished from other speciesof equal size.Measurements.?Usual measurements of adults: Head and body,200-230 mm.; tail, 160-190; hind foot, 45-50; cond^dobasal lengtli,53-60; zygomatic width, 25-28; width of brain case, 20-22; maxillarytooth row, 20-22.Subspecies.?The following geographic forms may be recognized:Tana tana tajia, Sumatra, southern Borneo; T. t. utara, northern
138 PROCEEDINGS OF fUE NATIONAL MtlSEUM. vol.45.Borneo; T. t. hesara, lower Kapuas River, Borneo; T. t. sirhasse-nensis, Sirliassen, Natuna Islands; T. t. hunose, Tambelan Islands,and T. t. tuancus, Banjak Islands.Remarks.? Tana tana is distinguished from T. cervicalis by itsduller and less brilliant colors and less conspicuous black area on lowerback, from T. paitana by having the shoulder stripe bordered aboveand below by general color of sides, and from the Linga species byits greater size. The material for making a comparison of theSumatran and Bornean tanas is not entirely satisfactory. GoodSumatran specimens are rather scarce, as an examination of the listof specimens on page 140 will show. There is not one good fullylidultmodern stud}^ skin of Tana tana tana from the island that is its type-locality. Probably the best material is in the United States NationalMuseum (represented by two specimens from Deli almost as far fromthe exact type-locality, Bencoolen, as one can find) and in the Phila-delphia Academy of Natural Sciences. One of the National Museumspecimens is in alcohol, the other a skin made from one in alcohol.The colors of both of them are not beyond question, although appar-ently unchanged. Their underparts are distinctly darker than theanderparts of Bornean examples. Most of the Philadelphia skinsare immature, and none shows any tangible character by which todistmguish them from Bornean skins. The skulls and teeth show nodistmguishing features between the Bornean and Sumatran speci-mens. For the present at least the same subspecies of Tana tanamust be regarded as occurrmg in southern Borneo and Sumatra. InParis, Berlm, and London are a few specimens of Tana tana simplylabeled "Borneo." I have not mcluded them in the lists and measure-ments of specimens, and have made no attempt to assign them toany particular subspecies. Many of them are old and mounted.A fair number of specimens of Tana tana exist m museums fromdefinite locaUties m Borneo, and while they show considerable indi-vidual variation, yet as a whole they fall into two groups, those fromSarawak behig brighter colored and those from southern Borneoduller, with a large member of the dull form on the lower KapuasRiver. The brighter-colored specimens seem to be associated withthe mountams while all those from southern Borneo have been takenm the lowlands. Future explorations may show that on Borneothe bright colored form is an inhabitant of the higher regions ofthat island and the duller form are inhabitants of the lowlands of thecoastal region. The present division into northern and southernraces may sunply be owing to the fact that the lowlands of southernBorneo have been explored, mainly by Abbott and by Shortridge;while in the north Hose worked in the higher altitudes. Our knowl-edge of this species on both Sumatra and Borneo is very elementar}^at present. On SMiassen of the Natuna Islands is a well-markedgeographic race closely related to the lowland Bornean form.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 139TANA TANA TANA (RaflBes).1821. Tupaia tana Raffles, Trans. Linn. Soc. London, vol. 13, p. 257, May, 1821.1825. [Cladobates] tana, Cuvier, Dents Mamm., 1825, p. 61.1840. Erinaceus {Glisorer) tana, Blainville, Osteogr. Mamm., vol. 1, p. 112,pi. 6, fig. 1.1841. Cladobates speciosus Wagner, Schreber's Saugthiere, Supplementband.pt. 2, p. 43, 1841. (Type-locality, Borneo.)1843. Hylogalca tana, Schlegel and Mxjller, Verh. Nat. Gesch. Nederl. Overz.Bezitt., p. 161, pi. 26, fig. 2; pi. 27, figs. 1-6, 1843.1879. Tupaia tana, Anderson, Zool. Res. West. Yunnan, 1879, p. 136, pi. 7, figs.1 and 2.Type-locality.?Sumatra, probably Bencoolen.Type-specimen.?In British Museum, Reg. No. 95.3.21 .4, collected bySii' Stamford Raffles in Sumatra, probably at Bencoolen, a badly dam-aged skull (no skin can be found), nearly everything posterior topalate lacking, as well as right premaxilla, several of the maxillaryteeth lost; mandible perfect; not fuUy adult, the permanent upperincisors just appearmg.Geographic distribution.?Sumatra, southern Borneo, in the lowcountry. See No. 5 on map on page 143.Diagnosis.?A relatively dark and dull member of the species,especially in region of sides and shoulders, light area on either side ofdorsal stripe a mixture of ochraceous buff and blacldsh brown, neithercolor m excess.Slcull and teeth.?Without definite subspecific characters. ("Plate11, fig. 4.)Measurements.?^Usual measurements of adults: Head and body,200-220 mm.; tail, 160-180; hindfoot, 47-50; condylo-basal length,54-56; zygomatic width, 25-27; width of braincase, 20-21; maxillarytoothrow, 19.5-21.5. For details of measurements see table, page 140.Remarks.?Two Sumatran specimens of Tana tana tana in theUnited States National Museum differ slightly from the Bornean onesin having the underparts ferruginous instead of orange rufous; butboth spechnens have been subjected to alcohol and one is still pre-served in that fluid. The bullae of one of them, Cat. No. 174611, arevery small in comparison with the bullae of Bornean skulls, but thelatter show marked variation in the size of the bullae. The skull ofthe second Sumatran specimen has bullae equaling in size the smallerbuUae of the Bornean specimens. Specimens from the LampongDistrict, Sumatra, in the Philadelphia Academy of Natural Sciences,do not show tangible differences from the Bornean specimens.Wagner's name speciosus w^as applied by him to a specimen in theErlangen Museum. His description applies more closely to thesouthern Bornean form than to the brighter colored northern form.Until the distmctness of the Sumatran longnosed treeshrew from thelowland Bornean animal is established, speciosus must stand as asynonym of tana, but in the event of their distmctness it will be avail-able for the Bornean form. It is rather unfortunate that the exactlocality for Wagner's specimen is not known, but as most of the old
140 PROCEEDINOS OF THE NATIONAL MUSEUM. VOL. 45.collecting in Borneo was done in the southern part, it is probable thatit came from southern Borneo, especially as the description seems moreapplicable to the southern form. Wagner did not have a skin of theSumatran tana when he applied the name speciosus, hence his errorof considermg the two forms different.The specmien in the British Museum from Pajo is not typical, andif the characters it presents are constant for the longnosed treeshrewsof that region, it represents a distinct form. It is very dark and dullin color and the red element much suppressed. Some of this differencein color may be the result of preservatives. (See remarks underTana tana utara, p. 142.)The Bornean specimens m the United States National Museumhave been previously called Twpaia speciosa.^Specimens examined.?From Sumatra 16, and 11 from Borneo. Seelist of specunens below.Measurements of Tuna tana tana.
Locality.
NO. 1976. TREESHREW8: FAMILY TUPAIID.^?LYON. 141TANA TANA BESARA, new subspecies.Type-locality.?Opposite Pulo Jambii, below Tyan, on north orright bank of Kapuas River, western Borneo.^Type-specimen.?In United States National Museum, Cat. No.142247, skin and skull of adult male, collected opposite Pulo Jambu,Kapuas River, on the north or right bank of the river, September 17,1905, by Dr. W. L. Abbott; original number, 4458; skin and skullin good condition. (See figures 10 and 11 on page 135.)Geographic distribution.?Known only from the type-locality, butprobably found throughout the low swampy area of western Borneo,north of the Kapuas River.^ See No. 4 on map on page 143.Diagnosis.?Similar in general coloration to Tana tana tana ofsouthern Borneo, but even darker and duller in color and with con-siderably more black on the lower back, forming a fairly well-definedblack patch, underparts darker and duller; size very large; hindfoot, 53 mm.; condylo-basal length, 59.5.Slcull and teeth.?Aside from their distinctly larger size these do notdiffer from those of the related subspecies. (Plate 11, fig. 8.)Measurements.?Type, the only adult specimen know^n: Head andbody, 229 mm.; tail, 196; hind foot, 53; condylo-basal length, 59.5;zygomatic width, 29 ; width of brain case, 21 ; maxillary tooth row, 23.RemarJcs.?This subspecies is at once distinguished from the othersby its large size and dull color. Its distribution would appear to becoincident with that of Sciurus horneoensis palustris,"^ and like thatsquirrel its color appears darker than it does in related subspecies.Doctor Abbott in his field catalogue refers to the large size of thistreeshrew. It has hitherto been caUed by me Tupaia speciosa.^Specimens examined.?Two, the type, and a young mdividual fromnear Tyan. TANA TANA UTARA, new subspecies.Type-locality.?Thi'ee thousand feet on Mount Dulit, Baram dis-trict, northern Borneo.Type-specimen.?In British Museum, Reg. No. 99.12.9.5, skin andskuU of adult male, collected October, 1898, at 3,000 feet altitude onMount Dulit, Borneo, by Charles Hose; in good condition.Geographic distribution.?Northern Borneo, probably confined tothe more elevated regions. See No. 3 on map on page 143.Diagnosis.?A bright-colored subspecies of Tana tana, with sides,shoulder, and underparts lighter, and the light areas on either side ofthe dorsal stripe very light and gray in appearance, black area onback better defined.
1 Proc. U. S. Nat. Mus., vol. 33, map opposite p. 547, Dec. 24, 1907.! Idem, p. 553, Dec. 24, 1907.'Idem, p. 562, Dec. 24, 1907, and vol. 40, p. 121, Apr. 2o, 1911
142 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.Slcull and teeth.?These appear to be a trifle larger in this subspeciesthan they do in T. tana tana, but the difference is so slight as to bepractically negligible.Measurements.?Type: Hind foot, 50 mm.; condylo-basal length,55.5; zygomatic width, 28; width of brain case, 20; maxillary toothrow, 21. Usual measurements of adults: Head and body, 225-240
;
tail, 160-175; hind foot, 47-51; condylo-basal length, 54-57; zygo-matic width, 27-28; width of brain case, 20-21; maxillary tooth row,21-22. For individual measurements see table, page 146.Remarks.?The brighter colors of this subspecies, and especiallythe grayish area on either side of the dorsal stripe, are well-markedcharacters for T. t. utara. The most characteristically marked speci-mens are from Mount Dulit. A single specimen from Mount Mulu,1,000 feet, British Museum, Reg. No. 94.6.2.1, is larger than theMount Dulit specimens, is generally ''redder," with narrower andgrayer areas on either side of dorsal stripe, has a more anterior exten-sion of hazel color of the sides over the shoulder stripe, the dorsalstripe rather reddish and less black on the lower back. If thesecharacters should prove to be constant in Mount Mulu specimens, theywould of course represent another race of Tana tana. The specimensfrom British North Borneo are provisionally referred to the presentrace. The material representing them is poor. Three skins fromthere in the United States National Museum have been so alteredin color by pickling fluid that they are of no systematic valuewhatever. The reddish elements of the pelage have everywhere beenturned to brown, and if one were unaware of their altered conditionthey would appear to be the most distinct of any of the forms in thegenus except Tana dorsalis.Specimens examined.?Fifteen from the Baram district and ninefrom elsewhere in northern Borneo. See table, page 146.TANA TANA SIRHASSENENSIS (Miller).1901. Tupaia sirhassenensis Miller, Proc. Wash. Acad. Sci., vol. 3, p. 133, March26, 1901.Type-locality.?Sirhassen, Natuna Islands.Type-specimen.?In United States National Museum, Cat. No.104712, skin and skull of adult male, collected on Sirhassen, NatunaIslands, June 5, 1900, by Dr. W. L. Abbott; original number, 442.Geographic distribution.?Sirhassen Island. See No. 7 on map onpage 143.Diagnosis.?Slightly smaller than either Tana tana tana or T. t.utara and rather intermediate in color between the two but morelike the form from northern Borneo, having the same bright colorof the sides, shoulders, and tail, but the upper parts in the region ofthe dorsal stripe duller and approaching the condition found inT. t. tana; in some specimens, including the type, the head, and thelight areas on either side of dorsal stripe being quite as dark and
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 143
144 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.dull as in the south Bornean race, and in some of the others nearlyas bright and light as in the north Bornean race. (Plate 11, fig. 3.)Measurements.? Type: Head and body, 203 mm. ; tail, 152; hind foot,47; condylo-basal length, 51.5; zygomatic ^^'idth, 25; width of braincase, 19.5; maxillary tooth row, 20. Maximum and minimum meas-urements of four adult specimens : Head and body, 203-203 mm.; ^ tail,152-162;^ hind foot, 45-47; condylo-basal length, 51.5-53; zygomaticwidth, 25-26; width of brain case, 19-19.5; maxillary tooth row,19-20.5. See table, page 149.Remarl's.? Tana tana sirhassenensis is a fairly well-marked formof long-nosed treeshrew, but hardly distinct enough to warrantspecific separation. In color it closely resembles the north Borneanrace and its maximum measurements easily extend into the rangeof the mainland forms. In 1894 Thomas and Hartert - consideredit identical Avith that form under the name Twpaia tana.Specimens examined.?Eight, all from Sirhassen Island.TANA TANA BUNO.E (MUler.)1900. Tupaia hunose. Miller, Proc. Wash. Acad. Sci., vol. 2, p. 229, August20, 1900.1904. [Tupaia] bunoai (sic) Trouessart, Cat. Mam. Suppl., 1904, p. 123.Type-locality.?Pulo Bunoa, Tambelan Islands, South China Sea.Type-specimen.?In United States National Museum, Cat. No.101640, skin and skull of adult female collected on Pulo Bunoa,Tambelan Islands, August 5, 1899, by Dr. W. L. Abbott, in goodcondition.Geographic distribution.?The Tambelan Islands, Pulos Bunoa, andBig Tambelan. See No. 6 on map on page 143.Diagnosis.?^Yery much like Tana tana tana in general coloration,but sides slightly brighter in color, and under parts lighter and paler,tending more toward ochraceous; tail very distinctly shorter, hindfoot shorter and skull slightly smaller. (Plate 11. fig. 5.)Measurements.? Type: Head and body, 210mm. ; tail, 152; hindfoot,45; condylo-basal length, 53.5; zygomatic width, 26; wddth of braincase, 20; maxillary tooth row, 19.5. The rest of the specimens donot differ materially from the type. See table, page 149.RemarJcs.?The shortness of tail and hind foot of this race wouldbe sufficient to warrant its specific distinction from Tana tana tanawere it notfor T. t. sirhassenensis, which bridges the gap between themso far as size is concerned. It differs from T. t. sirhassenensis in itsduller color of the upper parts, and its very much duller and lightercolor of the underparts tending to dull ochraceousf or ochraceousbuff instead of ochraceous rufous.Specimens examined.?Four, two from Big Tambelan and two fromBunoa, Tambelan Islands.
? Vhese mea-surements were made by Dr. W. L. Abbott in the flesh in inches, the head and body ofall the adults being 8 inches; fractions of an inch were probably disregarded.
' Nov. Zool., vol. 1, p. 657, September, 1S94.
NO. 1976. TREESHBEW8: FAMILY TLPAIID^?LYON. 145TANA TANA TUANCUS, new subspecies.Type-locality.?Piilo Tuangkii, Banjak Islands, west of Sumatra.Type-specimen.?In United States National Museum, Cat. No.114412, skin and skull of adult male, collected on Pulo Tuangku,Banjak Islands, January 29, 1905, by Dr. W. L. Abbott; originalnumber, 1489; in good condition.Geographic distribution.?Known only from Pulo Tuangku, butprobably occurring on other islands of the Banjak group. See No. 8on map on page 143.Diagnosis.?Most like TaTiatonas'irAassr/ieTiS'is, butbody colors ratherbrighter and redder, underparts darker and redder, especially under-side of tail ; light elements on either side of dorsal stripe, more ochra-ceous-buff ; in T. t. sirhassenensis, buff; in T. t. tana, cream buff.Measurements.?Type: Head and body, 215 mm.; tail, 160; hindfoot, 46; condylo-basal length, 53; zygomatic width, 26.5; width ofbraincase, 20.5; maxillary tooth row, 20.5. See table, page 149.Remarks.?This is a rather well-marked subspecies, easily distin-guished by its rich dark colorings, and the ochraceous elements on eitherside of the dorsal stripe. The specimens on which this form is basedwere regarded as true T. tana tana in 1903.^Specimens examined.?Two, both from Pulo Tuangku.TANA LING^, new species.Type-locality.?Linga Island, between Sumatra and Malay Pen-insula.Type-specimen.?In United States National Museum, Cat. No101597, skin and skull of adult male, collected at 2,000 feet altitudeon the peak of Linga Island, July 16, 1899, by Dr. W. L. Abbott; ingood condition.Geographic distribution.?Known only from Linga Island. See No.11 on map on page 143.Diagnosis.?Almost identical in color with Tana tana sirhassenensis,the only noticeable difference being the greater narrowness and lessconspicuousness of the dorsal stripe, but as the only specimen has asmall albinistic spot at about the middle of the dorsal stripe, thatdifference may be more apparent than real. The species is at once dis-tinguished from other members of the genus by its small-size, hindfoot 40 mm., condylo-basal length of skull 50.5. In addition to its smallsize the skull is further distinguished by its slender attenuate rostrum,havhig a rather abrupt origin from the rest of skull, and by its smallermore rounded bullae. (Plate 11, fig. 2.)Measurements.?Type and only known specimen: Head and body,191mm.; tail, 133; hindfoot,40; condylo-basal length, 50.5; zygomaticwidth, 25; width of brain case, 19.5; maxillary tooth row, 19.5.Remarks.?Owing to the absence of specimens of the genus Tanafrom Sumatra until within the last year, this specimen from Linga
' See Miller, Proc. U. S. Nat. Mus., vol. 2G, p. 472, February 3, 1903.80459"?Proc.N.M.vol.45?1.3 10
146 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.has previously been identified ^ with Tana tana. While it is clearlyrelated to that species, yet its small size and narrow rostrum serve todistinguish it very clearly. Additional material, however, may showthat it grades in with the smaller subspecies of Tana tana.Specimen examined.?One, the type.Measurements of Tana tana utara and T. t. besara.
Name. Locality. Number. Sex. Molar teeth
Ttana utaraDo.Do.
Do.Do-Do.Do.Do.Do.Do.Do.Do.Do.Do.Do.Do.Do.Do.DoDoDoDoDoDoT. tana be-sara.Do
British NorthBorneo.
....doBritish NorthBorneo, Kiua-batanganRiver.British NorthBorneo.
.ao.British NorthBorneo, Suan-alamba River.Sandakan,Northwest Bor-neo.Borneo, Sarawak
.
MountDulit,1,000 feet.NorthBorneo.MountDulit,
. 3,000 feet.NorthBorneo.Mount Dulit,4,000 feet,NorthBorneo.Mount Mu lu,1,000 {eet,NorthBorneo.Mount Penrisen,Borneo.Base of MountSons, Borneo.Bukit Sebau,Sarawak, Bor-neo.Borneo, MountDulit.Borneo, Sarawak.
-do.
....do
....do
....do
....dododoBorneo, KapiiasRiver, T y a nDistrict.Borneo, KapuasRiver, oppositePulo Jambu.
19225
.
34944.
78.4.26.2..
8393899.12.9.4...
99.12.9.5 i.
92.2.7.4...
94.6.2.1...
90.6.25.2. .
.
92.9.6.30...5.3.1.4
^, Gen.eva.Berlin,4453.8Berlin,72S3.Genoa 8...
....dos....tlos....Genoa 5a..do 5c..Genoa 6a.,142246
142247 4.
Female.
..do.Male. Moder-ately.M o der.ately.
Slightly..Unworn..M o d e rately.None
mm4242?45
mm26? mm22 mm20.521.521
26.525 20.51
Female.Male....
..do...
Slightly.Much . .
.
.do.
Moder-ately.do....
.do.... Slightly.
Male....
Female.Male
do.None 5
.
None 6.
55.5
54
00
23.528
Female.Male....Female.Male
...do...,Male.
.do....
None'..
Adult..None 6.Adult..None 6.do..
225?225230?230155240205?230?170?190
Slightly..
170?175 50?53
47.5
56.6
:0
21
21
2119.520
22
160? 49175 49140 47160 51155? 52?165?50?155? 45?155 50
196
43.5
4447 23.5
.59.5
2220.521
2120.5
21
20.5
22
2219
I Miller, Proc. Wash. Acad. Sci., vol. 2, p. 229, August 20, 1900.Miller. Proc. U. S. Nat. Mus., vol. 31, p. 271, September 11, 1900.Lyon, Proc. U. S. Nat. Mus., vol. 36, p. 490, June 1, 1909.* Skin only, no skull.3 Skull only, no skin.Type.
? tos just appearing.6 771' last tooth in place.
' 7)77*1 half way through alveolus.8 Preserved in alcohol.8 Skull in skin.
NO. 1976. TREE8HREWS: FA2IILY TVPAIIDJE?LYO'S\ 147TANA CERVICALIS (Miller).(Synonymy, type-specimens, etc., under the subspecies.)Geographic distribution.?The Batu Islands off the west coast ofSumatra. See Nos. 9 and 10 on map on page 143.Diagnosis.?The two members of this species differing considerablyfrom each other are distinguished from Tana tana by their brightruddy style of coloration accompanied with a large and well definedbrilliant black area on the lower back; mammae, 2-2 = 4.CoZor.?Sides of body, tail, outer side of legs dark, rich ferruginous,washed or lined with black about base of tail, middle of sides andthighs, the black becoming conspicuous on the lower back where itmay form a well defined black patch of moderate area or spread overalmost the entire posterior portion of body in a brilliant black pattern,underparts of body including inner side of legs bright cinnamon rufous;underside of tail rich ferruginous, darkening to burnt sienna alongthe edges ; light area on either side of the dorsal stripe varying froma grizzled mixture of blackish and cream-buff to a mixture of blackishand tawny ochraceous; shoulder stripe cream buff to ochraceous.Slcull and teetli.?There are no characters by which the skull andteeth of Tana cervicalis may be distinguished from those of relatedspecies, nor the different fotms from each other.Measurements.?Head and body, 205-215 mm.; tail, 155-165; hindfoot, 43-47; condylo-basal length, 53-54; zj^gomatic width, 25-26;width of brain case, 19.5-20; maxillary tooth row, 20-21.5.Subspecies.?Two subspecies of Tana cervicalis may be recognized:T. cervicalis cervicalis, Tana Bala; T. cervicalis masse., Tana Masa,Batu Islands.Remarlcs.?Tana cervicalis is not distantly removed from T. tana,but is immediately distinguishable by the large, conspicuous briUiantblack area of the lower back. This character, however, is only oneof degree, for in T. tavM, many examples are seen where a darkenedor even black area exists on the lower back, but it is never so intensein color, or so large, extending so far on the sides, or so far anteriorly.TANA CERVICALIS CERVICALIS (MiUer).1903. Tupaia cervicalis Miller, Smiths. Misc. Coll., vol. 45, p. 59, November 6,1903.Type-locality.?Tana Bala, Batu Islands, off west coast of Sumatra.Type-specimen.?In United States National Museum, Cat. No.121754, skin and skull of adult male collected on Tana Bala, BatuIslands, February 14, 1903, by Dr. W. L. Abbott; original number,2294 ; in good condition.
148 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Geographic distribution.?Tana Bala Island. See No. 9 on map onpage 143,Diagnosis.?Distinguished by the great extent and intensity ofthe black area of the lower back, extending forward to meet in strik-ing contrast the light area beside the dorsaL stripe, and also welldown on the sides of body and on base of tail; light areas beside thedorsal stripe very light and conspicuous, a mixture of buff and black,the former in excess; shoulder stripe cream buff. (Plate 11, fig. 7).Measurements.?Type: Head and body, 210 mm.; tail, 165; hindfoot, 47; condylo-basal length, 54.5; zygomatic width, 25; width ofbraincase, 20; maxillary tooth row, 21.5. For measurements of asecond specimen see table, page 149.RemarTcs.? Tana cervicalis cervicalis is a very well marked form,and conspicuous animal. The rich ferruginous tones of the sides andtail, the brilliant black of the lower back, and the Hght areas of theneck and bordering the dorsal stripe, make T. cervicalis cervicalis thehandsomest animal in the family Tupaiidse.Specimens examined.?Two, both from Tana Bala.TANA CERVICALIS MASiE, new subspecies,Type-locality.?Tana Masa, Batu Islands, west of Sumatra.Type-specimen.?In United States National Museum, Cat. No.121835, skin and skull of adult female, collected on Tana Masa, BatuIslands, Februaiy 18, 1903, by Dr. W. L. Abbott; original number,2302; in good condition.Geographic distribution.?Tana Masa Island. See No. 10 on mapon page 143.Diagnosis.?A member of the species Tana cervicalis, character-ized by having the hght areas on either side of the dorsal stripetawny ochraceous slightly mixed with blackish, the black area of theback not quite so well developed, nor extending so far forward, andthe reddish colors of maximum intensity, head relatively dark.Remarlcs.?This race appears to be a well-defined one and quitedifferent in appearance from Tana cervicalis cervicalis on the nearbyisland of Tana Bala so far as the anterior portions of the body areconcerned, but the posterior halves of the two animals are essentiallyalike except that T. c. masse has the darker coloration of the two..Anteriorly T. c. masse more closely resembles the form from theBanjak Islands. This form was identified by me in 1908 as identicaljwith the Tana Bala form.^Specimens examined.?Two, both from Tana Masa.
> Ann. Mag. Nat. Hist., ser. 8, vol. 1, p. 138, Febraary, 1908.
NO. 1070. TREE8HREWS: FAMILY TVPAIIDM?LYOy. 149
Measurevieyits of insularforms of the genus Tana.
150 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.SJcull and teeth.?They do not differ materially from those ofTana tana; the rostrum is, perhaps, a little wider in T. chrysurathan in species commonly seen on Borneo.Measurements.?Type : Head and body (from dried skin) , 240? mm
;
tail (from dried skin), 160? ; hind foot, 52; zygomatic width, 29;width of brain case, 21.5; maxillary tooth row, 22. For individualmeasurements see table, page 151.Remarlcs.? Tana chrysura is a well-marked form. Its buff-coloredtail serves at once to distinguish it from other members of the genus.In addition to the difference in color of the tail, it also differs in itsgeneral color from T. tana. This to my mind shows it to be nomere color phase of T. tana, but a perfectly distinct species. Thisview is further advanced by the fact that Tana chrysura has not yetbeen secured in Borneo other than in the vicinity of the type-locality,and that no specimens of T. tana have been taken at the type-locality of T. chrysura. Its analogy with Tupaia demissa, page 58,is interesting.Specimen examined.?Seven, all from the vicinity of the type-locality. TANA PAITANA, new species.Type-locality.?Paitan Kiver, northeastern Borneo.Type-specimen.?In British Museum, Reg. No. 93.4.1.1, skin andskull of adult, collected along the Paitan River, northeastern Borneo,July, 1892, by A. Everett; in good condition, but skull somewhatdamaged posteriorly.Geographic distribution.?Known only with certainty from the type-locality, but represented on Banguey Island by the same or a closely ,allied form. See No. 2 on map on page 143.Diagnosis.?Differs conspicuously from Tana tana in the greaterdevelopment of yellowish gray liglit areas on either side of the dorsalline so as to embrace the area of the shoulder stripe, instead of havingit embraced by the reddish brown color of the sides. Mammas un-known.Color.?Much brighter and "redder" than is Tana tana on the sides,legs, and lower back; the lower back without any evident black patchand nearly as clear ferruginous as are the sides; the Hght areas onanterior half of back more extensive and hghter than in any otherform, with the general effect of a yellowish gray instead of the ratherclear gray seen in Tana tana utara, the light shoulder stripe beingentirely surrounded by this yellow gray area, and not bordered aboveand below by a forward extension of the ferruginous color of thesides, underparts not essentially different from those of Tana tana.SJcull and teeth.?These show no special distinguishing features andare apparently indistinguishable from those of Tana tana.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 151Measurements.?Type: Head and body (dried skin), 260 mm.; tail(dried sldn), 160; hindfoot, 50; zygomatic width, 28; width of braincase, 21; maxillary tooth row 22.Remarks.? Tana paitana appears to be a very distinct species,nearly as much so as T. chrysura, the surrounding of the shoulderstripe by the grayish of the anterior back being a feature unknown inother species; the distinct ferruginous instead of blackish color ofthe lower back is also distinctive. It appears to be more nearlyrelated to Tana chrysura than to T. tana as far as its general colora-tion is concerned, excepting of course the tail. T. chrysura has ratherlarge grayish areas on the anterior back, but they do not embrace theshoulder stripe. The specimen from Banguey Island is provisionallyreferred to T. paitana. The specimen is immature, and it does notappear advisable to found a new name on its smaller size and m?reblackish lower back.Specimens examined.?Two, the type, and one from BangueyIsland, off northeastern Borneo.Measurements of Tana chrysura and paitana.
Name.
152 PROCEEDINOS OF THE NATIONAL MUSEUM. vol.45.TANA DORSALIS (Schlegel).1857. Tnpaja dorsaUs Schlegel, Handl. Beoef. Dierk., p. 59, pi. 3, fig. 31, two-fifths nat. size, in black and white, 1857.1890. Typaia dorsalis, Jentink, Notes Leyden Museum, vol. 12, p. 228, 1890.Type-locality.?^Lower Kapuas Kiver, western Borneo.^Type-specimens.?Cotypes in Leyden Museum,^ mounted skins "&"and "^f," under Tupaja tana of Jentink's Catalogues, of which theyare there considered young examples, and skull "/" under T. tana.Skin "gr" was collected b}^ M. Schwaner in Borneo, along the KapuasRiver. I have not seen these cotypes.Geograpliic distribution.?Known from western and northernBorneo, but probably occurring elsewhere in the island. See No. 12on map on page 143.Diagnosis.?^A very well-marked species of the genus Tana char-acterized by its small size, short claws, narrow dorsal stripe extendingfrom nape almost to base of tail. Mammae 2-2 = 4.Color.?Upper parts and sides of anterior parts of body, "svith thegeneral effect of olive, produced by a fine grizzling of blackish andbuffy, upper parts and sides of posterior parts of body, with thegeneral effect of burnt umber, produced by a fine grizzling of blackishand tawny. The line of demarcation between the two colors is notsharp, but they gradually blend one with the other; the olive color inmost specimens occupies more than the anterior half of body, but insome tile olive and burnt umber are about equally divided. Extend-ing through the middle of both colors from the nape almost to theroot of tail is the narrow (2-3 mm.) black dorsal line, shghtly widerat the middle than at the ends. Top and sides of head, intermediatein color between the olive of the anterior parts of body and the burntumber wash of the posterior parts. Outer side of hind legs similarto adjacent parts of body; outer side of fore legs intermediate in colorbetween the head and the anterior parts of body. Tail above the5ame color as the lower back, at base, becoming dark brownishtoward the end; underside of tail similar to the color above for theouter haK of hairs, russet for the inner half of hau's ; underparts, includ-ing innerside of legs, dull buffy to ochraceous buff, clearest in theregion of the throat, elsewhere obscured by the slaty bases of thehairs showing through; shoulder stripe fairly conspicuous, creambuff or buff. (Plate 6, fig. 2.)Sl'ull and teeth.?The skull of Tana dorsalis is mainly distinguishedby its smaller size, but the rostrum is relatively less slender andattenuate. The difference in size between Tana dorsalis and thesmallest member of the tana group, T. lingx, is not greater than
? See Jentink, Notes Leyden Museum, vol. 12, 1890, p. 228, and also Cat. Syst. Mamm. Mus. Hist. Nat.Pays-Bas, vol. 12, 1888, p. 116.2 See Jentink, Notes Leyden Museum, vol. 12, 1890, p. 228, and also Cat. Ost. Mamm. Mus. Hist. Nat.Pays-Bas, vol. 9, 1887, p. 240.
NO. 1976. TREE8IIREWS: FAMILY TUPAIID2E?LYON. 153between the largest, T. tana hesara, and the smallest members of thetana group. The fii-st and second upper molars of T. dorsalis aremore quadrate in outline than they are in the tana group and haverelatively better developed hypocones; the last upper molar is rela-tively larger and with a better developed metacone, and the hypoconeof 'pm^ is better developed, resembling that of pm'^ on a small scale.(Plate 11, fig. 1.)Measurements.?Usual measurements of adults: Head and body,175-200 mm.; tail, 140-150; hindfoot, 42-45; condylo-basal length,45-46.5; zygomatic wddth, 22-24; width of brain case, 18-19; maxU-lary toothrow, 17-18. See table below.Remarks.? Tana dorsalis is one of the best marked forms in thesubfamily Tupaiinfe, and needs no comparison with any other species.It is so different from the members of the tana group in the genusTana that it might almost be made the type of a distinct subgenus,mainly on the basis of the small size of the claws, as well as upon thetooth differences, but the latter are relatively slight. Nothingappears to be published regarding its habits, but judged by the struc-ture of its feet, it probably leads a rather different life than does thetana group.Its association with the long-snouted treeshrews was pointed outby Jcntinkin 1890.^Specimens examined.?Fifteen, from various localities in westernand northern Borneo. Measurements of Tana dorsalis.
Locality.
154 PROCEEDIXGS OF THE XATIOXAL MU8EU2I. VOL. 45.Genus UROGALE Mearns.1905. Urogale Meakns, Proc. U. S. Nat. Mus., vol. 28, p. 435, May 13, 1905.Type.? Urogale cylindrura Mearns= U. everetti (Thomas).Diagnostic characters.?A large member of the family Tupaiidse,easily distinguished externally by its elongated snout and close-hau-ed rounded tail, and craaially by its long rostrum, small zygo-matic fenestra, and dentally by the large size and canine-Kke appear-ance of i~ and the small rudimentary condition of i^.External characters.? Urogale, externally, is Hke the genus Tana^with the exception that the tail is not bushy or distichous, but ratherclose haired. The tailhas the same relativelength to the head andbody that it has inTana, but because of theshorter hairs appearsmuch smaller. Thenaked area on the nose,the ear, the shoulderstripe, and the arrange-ment of footpads showno differences in thegenera Tana and Uro-gale. The claws, espe-cially those of the forefeet, are particularlylong and sharp likethose of Tana tana. Thecolor pattern does notshow a dorsal stripe.The manmise are 2-2 = 4.(Plate 6, fig. 3.)Cranial characters.?The skull of Urogale isbuilt on thesame generalplan as that of Tana, butdiffers in many impor-tant features. The skullof Urogale on the whole ishea^aer and more angu-lar, although it is scarcely larger than that of Tana; the rostrum is rela-tively heavier, has a more abrupt origin from the skull, and is enlargedjust back of the extremity to accommodate the canine-like incisors;the temporal ridges are more prominent, but shorter, so that they
Fig. 12.?Ueogale everetti; Type X 1; Reg. No. 79.5.3.11,British Museum, Zamboanga, Mindanao, rnaiPPiNE Is-lands.
NO. 1976. TREE8HBEWS: FAMILY TUPAIID.E?LYON. 155meet at a more anterior point, forming a sagittal crest or ridge quiteas long as each temporal ridge. In Tana the sagittal ridge is muchlower and only about a third or a fourth the length of the less con-spicuous temporal ridges. The surface for the attachment of thetemporal muscles is thus much greater in Urogale than in Tana orTupaia. The coronoid process of the mandible is correspondinglyincreased in size. In Urogale the lambdoid region of the skull pro-jects further posteriorly than in Tana or Tu])aia, and when viewedfrom behind, the two ridges make a more acute angle than they doin the the other genera. The fenestra in the zygoma of Urogale isreduced to an almost invisible slit. A rather conspicuous groovedsurface is found on the underside of the maxiUary zygomatic root.In Tana this surface is much smaller andless conspicuous. The zygomata are more ^ ^spreading in Urogale than in Tana or ^ ImTupaia. The bony palate is more ossified ^in Urogale and without the vacuities moreoften seen than not in Tanu and Tupaia;the interpterygoid space is slightly nar-rower; the external plate larger and plate-Uke instead of forming a short wide hook,as in Tana. The gi-eatest width of thebrain case is about the same in Urogaleand the large species group of Tana, butit rapidly narrows anteriorly, so that thepostorbital constriction is distinctly lessthan the preorbital; in Tana the reversecondition holds. The orbit is relativelyand absolutely slightly smaller in Urogalethan in Tana. When the skulls of the two Fig. 13.?Upper and lower tooth-Eows of Urogale everetti; Type,X 2; Reg. No. 79.5.3.11, BritishMUSEITM, ZaMBOANGA, MINDANAO,Philippine Islands.genera are looked at from above, the pos-terior bar of the orbit divides the spacebetween the zygoma and the rest of theskull into two approximately equal parts,one anterior and one posterior, in Urogale; in Tana the anteriorportion is distinctly the larger of the two. The mandible of Urogaleis distinctly heavier and more massive than that of Tana or Tupaiaand the ascending ramus larger and more upright, especially seen inthe coronoid and angular processes; the cond^T'le is also somewhatlarger. See figure 12, page 154. (Plate 11, fig. 6.)Dental characters.? Urogale differe more in respect to its teeth fromthe other members of the subfamily Tupaiinte than in respect to itsskulL The second pan* of upper incisors are enormously developedand are rather canine-like in form, but straighter than the ordinarycanine tooth, so that instead of having.a backward curving fang the axis
156 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.
of this large tooth is so placed that the straight fang is directeddownward and backward. The canuie tooth itself is slightly betterdeveloped than it is in Tana or Tupaia, but owing to the great sizeof the incisor just in front it is scarcely noticed. In the mandiblethe first two incisors are essentially as they are in Tana or Tupaia,but the third lower incisor is very small, functionless, and sometimesdeciduous. The lower cariine is well developed, to match with thecanine-like second upper incisor. While the lower canine and thesecond upper incisor undoubtedly perform the functions of upperand lower canines, yet it is interesting to notice that the positions ofthe two teeth are reversed, the lower canine being placed posteriorto the tooth that functions as the upper canine, whereas in the caseof true canines the lower tooth cuts in front of the upper tooth.The upper and lower series of premolars in TJrogale are essentiallyas they are in Tupaia or Tana, but both of the last premolars arerelatively better developed in TJrogale and apparently of better serviceto the animal. This is particularly well shown in the protocone ofpm *, which is quite large and has nearly the same relative degree ofdevelopment as is found in that tooth in Anathana. The lower molarseries of Urogale and Tupaia are indistinguishable. The upper seriesare nearly alike in the two genera, but the hypocones of m* and m^are much better developed in Urogale than they are in Tupaia orTana, being nearly as large as in Anathana. (See figure 13, page 155.)Geographic distrihution.?So far as known TJrogale occurs on onlyone island, Mindanao, of the Philippines. See E on map on page 143.Number of forms.? Urogale contains but a single species, U.everetti.RemarliS.? Urogale, while clearly belonging to the subfamily wdthTupaia, Tana, Anathana, and Dendrogale, is more different from themthan any of them differs among themselves. With which one it hasthe closest affinity it is hardly possible to say. In most respects it hasmany points of real resemblance to the genus Tana and in some waysmay be looked upon as the Tana type carried to an extreme. Myown view is that both have been derived from some common ancestordifferent from Tupaia, and that owing to its isolation and smallnessof the land area on which it is found, Urogale went farther than didTana. The habits and food of Urogale probably differ considerablyfrom those of the rest of the subfamilj^ From the development ofits teeth, elongated rostrum, generally heavy build, one would sup-pose it to be a more predatory and carnivorous animal than any othermember of the subfamily.The two specimens collected by Dr. E. A. Meams on Mount Apowere snared in trees by natives, and the one from Mount Malindangwas shot on a tree stump. It had been observed several days before,and in its actions resembled a chipmunk.
xo. 1976. TREESHREWS: FAMILY TVPAIID^?LYON. 157UROGALE EVERETTI (Thomas).1892. Tupaia everetti Thomas, Ann. Mag. Nat. Hist., ser. 6, vol. 9, p. 250, March,1892.1905. Urogale cylindrura Mearns, Proc. U. S. Nat. Mus., vol. 28, p. 435, May 13,1905. (Type-specimen, in United States National Museum, Cat. No. 125287,collected by Doctor E. A. Mearns at 4,000 feet on Mount Apo, Mindanao,July 12, 1904; original number, 5727; skin and skull in good condition.)Type-locality.?Zamboanga, Mindanao, Philippine Islands.Type-specimen.?In British Museum, Reg. No. 79.5.3.11, adultmale, collected at Zamboanga, Mindanao, Philippine Islands. Theoriginal entry in the Register says "in spirit," but the specimen hasfor some time been made into a modern study skin in good condition.Aside from some cracks about braincase, the skuU is in good condition.With the principal exception of the pelvis, most of the skeletonexists.Geographic distribution.?^Mindanao, Philippine Islands. Speci-mens from Mount Apo, Zamboanga, and Mount Malindang. See E,on map on page 143.Diagnosis.?^As for the genus above.Color.?General effect of upper parts of head, neck, body, and tailand outer side of legs a color between mummy and seal brown, pro-duced by a fine grizzlmg of blackish and tawny, the blackish brownin excess, about nose and sides of head, the tawny color replaced byraw sienna and more of it; feet similar to legs but darker, and withalmost none of the lighter color; underparts, including inner sideof legs, varying from ochraceous to ochraceous rufous brightest inregion of the chest, the slaty bases of the hairs showing tliroughin places to a considerable extent; underside of tail similar to upper-side, but the light and dark color elements about equally mixed;shoulder stripe rather poorly defined, ochraceous. The type-speci-men is dull and rusty hi comparison with the modern skins, thedifference probably being due to its original preservation in alcohol;its underparts are duU brown and tail quite rusty.Slcull and teeth.?Skull large and angular, with relatively heavyrostrum rising abruptly from rest of skull, enlarged just back ofextremity to accommodate roots of the large second pair of incisors;temporal ridges short but prominent, sagittal crest rather long andhigh; fenestra m zygoma reduced to a minute slit; bony palate wellossified, usually without vacuities; postorbital constriction less thanpreorbital constriction. Second pair of upper incisors enormouslydeveloped, functioning as canines, third lower incisors, small, func-tionless and sometimes deciduous; last upper and lower premolarswell developed, especially protocones on upper tooth; lower canineconspicuously developed. (See fig. 12, p. 154; fig. 13, p. 155; alsoplate 11, fig. 6.)Measurements.?Type of U. everetti and of Z7. cylindrura: Head andbody, 235 mm. (dried skin), 182; taO, 170 (dried skin), 163; hind-
158 PROCEEDINGS OF THE NATIONAL MUSEVM. VOL. 45.foot, 51, 50; condylobasal length, 58, 58.5; zygomatic width, 29,29; width of braincase, 21, 20.5; maxillary toothrow, 22, 21. Forindividual measurements see table below.Remarks.?There can be but little doubt that Urogale cylindrurais perfectly synonymous with U. everetti. The relatively slightdifference in color between the type of U. everetti and the MountApo specimens is due only to the fact that the former is an old speci-men skinned out of alcohol,^ while the latter are modern freshlyprepared skins. There is almost as much difference between mem-bers of the Apo series as there is between the type of U. everetti andthe Apo series as a whole. The skull and teeth of Zamboanga speci-men are essentially like those from Mount Apo. The Mount Mahn-dang specimen is represented by a skin brought back by DoctorMearns in 1906. The information on the label reads: ''MalindangMts., summit of Mt. Bliss, 5,750 ft., June 3, 1906. Shot on trunkof an oak tree by Private D. W. West, Co. B, 19th Infantry, U. S. A."It has every appearance of having been preserved in alcohol orformalin at some time. Apparently the skull was not saved. Ahind foot and tibia removed from the skin show it to be a youngindividual. It has the general color characters of Urogale everettiQombined with the long sharp claws and Tana-like style of nose.The entire pelage, however, including that of the tail, is much longerand softer than it is in Mount Apo specimens. It is barely possiblethat it represents an entirely distinct species of Urogale or evenanother genus.Specimens examined.?Nine. Seven from Mount Apo, one fromZamboanga, and one from Mount Malindang, the last doubtfullyreferable to the genus and species.Measurements of Urogale everetti.
Locality.
NO. 1976. TREESHREW8: FAMILY TUPAIID^?LYON. 159Genus PTILOCERCUS Gray.1848. Piilocercus Gray, Proc. Zool. Soc. London, 1848, p. 23.1864. Ptilocerus, Brehm, 111. Thierl., vol. 1, pp. 663-665. A misspelling ofPtilocercus.1876. Ptilocerus, Wallace, Geographical Distribution of Animals, vol. 1, p. 337.A misspelling of Ptilocercus.Type.?Ptilocercus lowii Gray,^ the only species included in thegenus.Diagnosis.?An aberrant member of the family Tupaiidse, form-ing the subfamily Ptilocercmse, distinguished externally by its largethin ears, naked tail basally, distichously tufted distally, craniallyby absence of supraorbital foramen, and approximately paralleltemporal ridges; dentally by the relatively larger size of the firstupper incisors over the second pair, double-rooted upper canine,without conspicuous diastema between second upper incisor andupper canine, and absence of mesostyles on upper molars.External characters.?Ptilocercus differs very strildngly from anyother members of the family Tupaiidce m the form of its tail, which isnaked and scaly for about its proximal half or a little more, and dis-tichously tufted for a trifle less than the distal half. The extremebasal eighth of the tail is covered with soft furry hair like the adja-cent parts of the body. Each scale of the naked portion of the tailis subtended by tliree short hairs about equal to a scale and a quarteror a scale and a half in length. The hands and feet of Ptilocercus arerelatively larger than in the other genera of the family, and the foot-pads relatively larger, with an appearance of greater softness. Thefifth finger is as large as the second, and relatively much larger thanit is in Tupaia. The four interdigital pads are larger and morerounded; the thenar and hypothenar are relatively larger and sit-uated at about the same anteroposterior level; posteriorly the twopads are connected by a slight bridge. Of the toes, the third, fourth,and fifth are equal and the second nearly as large. The first toe iswell developed but much shorter than the others. Of the pads ofhind foot, the interdigitals are large and rounded; the small padaccessory to the fourth interdigital is wanting. Corresponding tothe relatively shorter foot the proximal external pad is relativelyshorter than in Tupaia, and the proximal internal is much shorterand wider and perfectly distinct from the fii'st mterdigital pad.Mammae, 2-2=4.The ear of Ptilocercus is large, thin, and membranaceous in con-trast to the small thick ear of Tupaia, with a fairly well developedtragus and a ridgehk? antitragus. Viewed by the unaided eye,both inside and outside of the ear is essentially naked, but under aglass a few scattered hairs are found interiorly, and slightly more
1 Proc. Zool. Soc. London, 1S4S, p. 23.
160 IROCEEDIXGS OF THE yATIOy^lL MUSEUM. vol. 4.3.on the outer surface, gradually increasing in quantity toward thebase, whore the ear is as well furred as the body.No shoulder stripe is present on Ptilocercus, but distinct markingsare found on sides of head, consisting of a distinct black patch,almost a stripe, extending from near the tip of the nose, posteriorlythrout^h the eye a little more than half way to base of ear. Thisblack patch is bordered above and below not by distinct light stripes,but by the rather light color of top of head and the distinctly lightcolor of the underparts.At the beginning of theblack face stripe arisewell developed vibrissselacking in the Tupaiinae.(Plate 7.)Cranial cJcaracters.?Some of the most funda-mental differences be-tween Ptilocercus andthe other Tupaiidse arefound in the region ofthe orbit, certain struc-tures in one having nodirect counterpart in theother. Twpaia has adisdnct supraorbitalforamen, which is con-tinued backward as awell-marked groove onthe underside of the up-per rim of the orbit.There is no counterpartof this foramen orgrooveiaPtilocercus. The infra-orbital is a long canal inTwpaia, the proximalend in the orbit aiid thedistal end of the canalover the penultinate upper premolar: m Ptilocercus the canal is muchshorter and is situated over the middle of the last upper premolar.In the most anterior corner of the orbit is the posterior opening ofanother foramen leading mto the nasal fossa. In Tujpaia this open-ing is situated in a distinct notch, 2 mm. or so in size, and the upperend of this notch forms a distinct blunt spine. This notch and spineare entirely lacking in Ptilocercus. The position of the orbits inPt^Xocerais" "13, quite different from that in the Tupaiin^. They arerelatively larger, and placed relatively farther forward; the posterior
PiG.-i4.?FnLOCEECUs LOWD cosTDrExns, X U; Cat. No. 112611,U.S.K.1I., Skjtbeoxg KrsxE, Johobe, Malay Pexiksula.
NO. 1976. TREESHREWS: FAMILY TUPAIID.E?LYOX 161edge of the orbit is placed only a trifle behind the level of the unionof olfactory bulb ^vith the cerebrum, whereas in Tupaiinag the pos-terior edge of the orbit is placed very much behind the level of thepoint of union of olfactory bulb and cerebrum. The general planeof the orbits in PtUocercus looks more to the front of the animal, theplane of each orbit making with the long axis of the skuU an angle ofabout 40?. In the case of Tupaiinse the orbits look more to the side,each making with the long axis of the skull an angle of about 20?.Other conspicuous features about the skull of PtUocercus are the widespreading zygomata, temporal fossa larger than orbit, approximatelyparallel temporal ridges, a better ossified palate and simpler auditalbuUee; that is, not divided by internalsepta, as is frequently the case in theTupaiinae. The comparatively smallantero-extemal segment of the buUain Tupaia is very much reduced inPtUocercus. At the apex of the orbitalca-sHty in both Tupaia and PtUocercusare two foramina, (1) the optic and(2) sphenoid fissure. In Tupaia, justbelow and external to these is theforamen rotundum, situated at thebase of the external pterygoid plate,communicating with the cranialcavity as weU as with a canal, thealisphenoid, running lengthwise inthe base of the external pterygoidplate. In PtUocercus the foramenrotundum is confluent with thesphenoidal fissure. Posterior to theexternal pterygoid plate in PtUocercus,near the bulla, is a distinct oval fora-men leading into the cranial cavity;this foramen is almost entirely overlapped by the bulla in Tupaia.The postorbital constriction in PtUocercus is slightly less than theinterorbital, being sunilar in this respect to VrogcUe. In Tupaia thepostorbital constriction is greater than the interorbital. The twolimbs of the lambdoid crest in PtUocercus do not meet in a back-wardly projecting obtuse angle as in Tupaiinje, but each limb runs upto meet its corresponding temporal ridge. Between the posteriorextremities of the temporal ridges the lambdoid exists as a slightlyconvex (backwardly) ridge.The coronoid process of the mandible is relatively larger and widerin. PtUocercus than in Tupaiei. and the notch between the condyle andcoronoid shallower. See figure 14, page 160.S0459'?Proc.X.M.vol.45?13 11
Fig. 15.?Upper and lower toothbovs ofpxilocekcrslo-kticoxtixextis, x 4; cat.No. 112'jll, U.S.X.M., Sembkoxg Ritee,JOHOKE, ItALAT PE^^^-SX^,A.
162 PROCEEDTNOS OF THE NATIONAL BIUSEUM. vol.45.Dental characters.?There are not so many fundamental differencesbetween the teeth of Ptilocercus and Tupaia as there are betweenthe skulls, perhaps no more than generic differences. The uppercanine is distinctly two-rooted and premolariform in shape. Attimes the upper canine in Tupaia is also two-rooted, but is notpremolariform in shape. The first upper incisors are very muchlarger than the second pair, usually disproportionately more so thanis usual in Tupaia. Both pairs of upper incisors are somewhat differ-ently shaped in Ptilocercus and Tupaia. In the latter they are simplecone-like recurved teeth ; in Ptilocercus these teeth are more trenchantwith a moderately well-developed posterior cuttmg edge, and thesecond incisor almost has a distinct posterior cusp. Owing to theshortness of the rostrum, the incisor-canine-premolar series of teethstand closer in tke tooth row than they do in Tupaiinse. The secondupper premolar is three-rooted, as is usually the case in Tupaiinee,and also is more triangular and more like the last upper premolar inshape. The last upper molar is more compressed antero-posteriorlyin Ptilocercus than in Tupaia. The upper molars of Ptilocercus lackthe mesostyle, which is quite conspicuous in Tupaiinse; the cusps onthe teeth appear blunter and more rounded. The molars are sur-rounded by a distinct cmgulum, absent in Tupaia. In the lower jawthe first and third incisors are relatively smaller with respect to thesecond incisor in Ptilocercus than in Tupaia. The lower canines areabout the same relative sizes in the two genera, but in Ptilocercusthey are less canine-like in shape and look like enlarged first pre-molars. In Ptilocercus the middle lower premolar is much smallerthan either of the two other premolars and is below the general level ofthe lower tooth row, while in Tupaia it is the first which is thesmallest of the premolar series and which stands below the level ofthe tooth row. The lower molar teeth have a well-defined cingu-lum on their outer aspect, lacking in Tupaiinae, but otherwise noessential differences are found in the two genera. See figure 15,page 161.Geograpliic distrihution
.
?Ptilocercus is found in Borneo, Banka,Sumatra, and Pulo Pinie of the Batu Islands, and in the southernpart of the Malay Peninsula. It is said to exist on Sirhassen of theNatuna Islands.' See Nos. 2 and 3 on map, page 133.Nuonber offorms.?Ptilocercus contains but a single species group,loivii, of which one geographic race has been described.RemarJcs.?Ptilocercus differs in external and particularly in cran-ical characters from any of the other genera of the Tupaiid.-ic morethan any of them differs from each other, and seems well worthy ofsubfamily rank. The most striking differences are found in the ears,tail, feet, supraorbital foramen, and certain foramina at base of skull,
I See Thomas and Hartert, Nov. Zool., vol. 1, p. 656, September, 1894.
NO. 1976. TREE8HREWS: FAMILY TUPAIIDyE?LYON. 163some of them having no counterpart in the other genera. Dentallyit differs conspicuously from the other genera in lacking mesostyleson the upper molars and in having distinct cingula on both upperand lower molars. PTILOCERCUS LOWII Gray.(Synonymy, type-specimens, etc., under the subspecies.)Geographic distribution.?The same as that of the genus, page 162.Diagnosis.?The same as for the genus, page 159.Color.?General color of upper part of head, neck, and body andouterside of legs, isabclla color variously shaded with drab or drab-gray, top of head usually lighter, sides of body with a more or lessdistinct buff wash; underparts including innerside of legs varyingfrom gray to a rather dark buff, the latter more conspicuous anteri-orly; a black or blackish mark extending posteriorly through the eye,but not reaching base of ear, the black mark contrasting stronglywith the lighter colors of the head; distal portion (75-80 mm.) of plumeof tail, white or cream color, proximal portion (about 10 mm.) blackishbrown; the well-haired basal portion of the tail colored like adjacentparts of body. (See Plate 7.) .Skull and teeth.?The same as for the genus, pages 160 to 162.RemarJcs.?Ptilocercus lowii is one of the most strongly character-ized of mammals and can be instantly identified by its external aswell as cranial characters. It is quite rare in collections. Dr. W. L.Abbott has collected over 6,000 mammals in regions where it isknown or supposed to occur, and has only encountered it twice. Ihave seen 10 examples of the pentailed treeshrew, and know of twoothers in the Leyden Museum which I have not seen. Ptilocercusand Dendrogale seem to be of about equal rarity, at least in collec-tions. This probably is due not so much as an actual rarity in natureas to some peculiarity of habit, rendering them seldom observed anddifficult to trap. Observations on the habits of this animal arescarce. Both of Doctor Abbott's specimens were caught in traps inheavily forested hills. The Selangor ^ specimen was '^ captured inits nest in a hollow bough. * * * The nest was merely a tunnelabout 3 inches in diameter and 18 inches in length, roughly linedwith fibrous material and green leaves." The original representa-tive was caught in a house. Schneider's ^ remarks are interestingand worth quoting in full:The natives did not know this little animal and simply called it, in response to myquestions, Tikuskaju?that is, Treemouse. The male of this pentailed treeshrewhappened to be brought alive to me by a Battak man employed in the vicinity inwoodcutting. Thereupon I hastened at once with the man to the exact spot wherehe had obtained it, in the hope of perhaps yet capturing the female, and my joy wac
1 Robinson, Journ. Straits Brit. Royal Aslat. See, No. 44, 1905, p. 225.?Zooi. Jahrb., vol. 23. pt. 1, 1905, p. 84.
164 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.indescribable when after a quarter hour I had the luck to discover the female in theleafy summit of a large tree lying on the ground, amongst the parasitic vegetation whichthickly covered the tree, and to capture it with the help of the woodcutters. I thenkept the pair alive for several hours in order to observe them. The long pentail theycarried hanging or lightly outstretched, and at the same time they constantly movedit to and fro like the pendulum of a clock. It appeared to me as if they used the tailas an organ of touch. If I merely touched lightly the hairs of the plume with myfinger, they moved away, but they permitted themselves to be gently stroked andhandled without maldng an attempt to bite. They sniffed the bananas which I laidbefore them, but they did not eat any. As I feared through some accident these rareand interesting little animals might escape from me, I chloroformed them at evening,measured them at once, and preserved them in alcohol.Of this pair the male is preserved in the Natural History Museumat Basel and the female at Strassburg.Subspecies.?Two forms of Ptilocercus lowii have thus far beenrecognized, true lowii on Borneo, and continentis from the MalayPeninsula. The specimen from Banka is probably the same form asthe Bornean one. The Sumatran and Batu specimens are hereregarded as the same as the peninsular form. The two subspeciesare not highly differentiated from one another, and it would be wellif their distinctness could be established by the examination of morematerial or the examination of all the known examples together.The only material that I have carefully examined is that in theUnited States National and British Museums.KEY TO THE SUBSPECIES OF PTILOCERCUS LOWII.Toes not different in color from rest of hind foot; muzzle and palate slightly broader.Borneo, Banka; P. lowii lowii, p. 164.Toes distinctly lighter in color than rest of hind foot; muzzle and palate slightly nar-rower. Malay Peninsula, Sumatra, Batu Islands P. lowii continentis, p. 165.PTILOCERCUS LOWII LOWII Gray.1848. Ptilocercus lowii Gray, Proc. Zool. Soc. London, 1848, p. 23, pi., mammalia,2, entire animal in colors, very good.1848. Ptilocercus lowii, Gray, Ann. Mag. Nat. Hist., ser. 2, vol. 2, p. 212, Septem-ber, 1848. The same account as appeared in Proc. Zool. Soc. London, 1848.1850. Ptilocercus loioii. Gray, Zool. Voy. Samarang, p. 18, pi. 5, entire animal incolors, and views of skull and teeth.Type-locality.?Sarawak (the town), northern Borneo.Type-sjyecimen.?In British Museum, mounted skin, Reg. No.47.12.30.1, and skull, Reg. No. 48.5.12.3, collected in the Rajah'shouse, Sarawak, Borneo, by ^Ir. Hugh Low. The mounted skin isvery old and exhibition worn, and was probably made from analcoholic specimen; considerable patches of hair have slipped fromthe left side and from belly. The skull is perfect and in good condi-tion, but is not fully adult, as the deciduous incisors are still retained.
NO. 1976. TREE8HREWS: FAMILY TUPAIID^?LYON. 165GeograpMc distribution.?Northern Borneo, Sarawak, Baram River,Lawas Mountains,^ islands of Banka, and vSirhassen.^ See No. 2 onmap on page 133.Diagnosis.?Toes not distinctly lighter than rest of hind foot, andskull with broader muzzle and palate, teeth slightly larger, generalcoloration a clearer Isabella color, with less tendency to drab ordrab gray.Measurements.?Type: Head and body (from mounted skin) 140 mm.,tail (from mounted skin) 160; hind foot, 27; condylo-basal length,35; zygomatic width, 20,5; width of braincase, 14.5; maxillarytoothrow, 14. Usual measurements of adults: Head and body,135-150; tail, 160-180; hind foot, 27-28; condylo-basal length,37-38; zygomatic width, 22-23; width of braincase, 14.5-15; max-illary toothrow, 13-14.Remarks.?Jentink ^ records in the Leyden Museum a specimen ofPtilocercus from Banka, a mounted skm and incomplete skull of anearly adult animal. Owing to the general similarity of the mammalsof Banka and Borneo, this specimen is here regarded as belonging tothe typical subspecies.Specimens examined.?Five. See table, page 166.PTILOCERCUS LOWII CONTINENTIS Thomas.1910. Ptilocercus lowi continentis Thomas, Ann. Mag. Nat. Hist., ser. 8, vol. 5,May, 1910.Type-locality.?Vicinity of Kwala Lumpur, Selangor, MalayPeninsula.Type-specimen.?In British Museum, Reg. No. 10.4.17.1, skin andskull of an adult male, collected 10 miles from Kwala Lumpur,Selangor, December 27, 1903, by a Dyak collector employed bySelangor Museum; in good condition.Geographic distribution.?At present known from *Selangor andJohore of the Malay Peninsula, DeU-Lankat region of Sumatra, andPulo Pinie of the Batu Islands off west coast of Sumatra. See No. 3on map on'page 133.Diagnosis.?Differs from Ptilocercus lowii lowii in having more drabin the colors of the upper parts, the metatarsal region of hind feetbrown in contrast with the lighter toes, and in the possession of asomewhat narrower rostrum and palate, and slightly smaller teeth.(See Plate 10, fig. 5.)Measureme7its.?Type: Head and body, 133 mm. ; tail, 167 ; liind foot,28; condylo-basal length, 27; zygomatic width, 22.5; width of braincase, 15; maxillary tooth row, 12.5. For individual measurementssee table, page 166.
' See Hose, Mammals of Borneo, 1S93, p. 34.2 See Thomas and Hartert, Nov. Zool., vol. 1, 1894, p. 656.
? Cat. Mus. Hist. Nat. Pays-Bas, Osteol. Mamm. vol. 9. 1887. p. 242; Cat. Mus. Hist. Nat. Pays-Bas,Mamm,. vol. 12. 1888, p. 118.
166 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 45.Remarks.?The chief distinguishing mark of this subspecies is thebrown color of the metatarsal region of the foot in contrast with thebuff-colored toes. The differences pointed out by Mr. Thomas inregard to the "naked portion" of the tail is simply due to the mannerin which the skin of the tail of the type is stretched. This animalwas first recorded on the Malay Peninsula by Mr. Robinson in 1905.^Although the Selangor specimen was the first to be recorded in theliterature, Dr. W. L. Abbott was the first person to take Ptilocercuson the Asiatic Continent, his Johore specimen being collected July 5,1901, about two and a half years before the Selangor animal wassecured.Specimens examined.?Two from the Malay Peninsula, one fromPulo Pinie. The two Sumatran specimens. I merely saw in theirexhibition cases without critically examining them. See table below.In addition to these a skeleton of a specimen collected at DeliSumatra is recorded by Jentink,^Measurements of Ptilocercus lotvii.
Name.
NO. 1976. TREESHREWS: FAMILY TVPAIID^?LYON. 167BIBLIOGRAPHY.This bibliography contains a list of all the works which I haveconsulted in preparing this paper, and in addition many titles of pa-pers in which reference is made to treeshrews, often in a rather sub-ordinate way. I believe that it contains all the important worksbearing on the subject, but necessarily does not contain every refer-ence to treeshrews in literature. Certain publiQations which I haveseen referred to as probably mentioning the family, I have beenunable to find, and of course there must be many numerous minorarticles or references which I have not Encountered.Allen, J. A. The Geographical Distribution of the Mammalia considered in relationto the principal ontological regions of the earth, and the laws that govern the distri-bution of animal life. Bull. U. S. Geol. Geogr. Surv. Terr., vol. 4, 1878, pp. 313-378.Tupaia, one of the characteristic mammals of the Indian iffegion.
. Mammals from the Island of Hainan, China. Bull. Amer. Mus. Nat. Hist.,vol. 22, pp. 463-490, December 17, 1906.Original description of Tupaia modesta.
. Mammals from Palawan Island, Philippine Islands. Bull. Amer. Mus. NatHist., vol. 28, pp. 13-17, January 29, 1910.Records the specimen of Tupaia from Palawan in the American Museum of Natural History, onpage 17, as T.fcrruginca palawanensis.A.NDERSON, John. On the osteology and dentition of Ilylomys. Trans. Zool. Soc.London, vol. 8, 1874, pp. 453-467, plate 1.Shows that affinities of Hylomys are with Erinaceus and Oymnura and not with Tvpaia.
. Anatomical and Zoological Researches: Comprising an account of theZoological Results of the two Expeditions to Western Yunnan in 1868 and 1875,vol. 1, text, and vol. 2, plates, 1879.Pages 107-137, a very excellent anatomical and systematic account of the Tupaiidse with numerousbibliographic references. Genus Dendrogale, D. murina, D.frenata; Gemis Tupaia, T.ellioti, T.belan-geri, T. chinensis, new; T. ferruginea; T. splendidula, T. javanica, T. malaccana, new; T. tana, T.nicobarica. Plate 7, dorsal and lateral views of skulls of all.
. Catalogue of Mammalia in the Indian Museum, Calcutta, pt. 1, 1881, pp.153-157.Lists, often with exact localities, Tupaia eUioti, belangeri, chinensis, ferruginea, malaccana. tana,nicobarica.Ball, V. On the occurrence of Tupaia EUioti, Waterhouse, in the Satpura Hills,Central Provinces. Proc. Asiat. Soc, Bengal, 1874, pp. 95-96.Records Tupaia ellioti from Satpura Hills i=Anathana wroughtoni of this paper).Bartholomew, J. G., Clarke, W. Eagle, and Grimshaw, Percy H. Atlas ofZoogeography= vol. 5, Bartholomew's Physical Atlas, 1911.Page 15, plate 4, map 4. Remarks on and map of the geographic distribution of the Tupaiidse, 34species recognized, but none named.Beddard, Frank Evers. Mammalia. Cambridge Natural History, vol. 10, 1902.Brief account of the family Tupaiidee, with the genera Tupaia and Ptilocercus on p. 511.Blainville, H. M. Duerotayde. Osteographie des Mammiferes Insectivores, pp.31-35. Osteographie ou Description Iconographique compar^e du Squelette et duSysteme dentaire des cinq classes d'Animaux Vertebres, 1840.Accomit of the osteology, uses the generic terms Glisorcx and Cladobates, pi. 3, lower figures, and intable of plates uses the combination Erinaceus {Glisorex) tana.
168 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Blakford, W. T. Fauna of British India, including Ceylon and Burma, Mammalia,1888, Tupaiidse, pp. 207-212.General account of lamily, genus, and species ellioti, fcrruginea, nicobarka, quoted accounts ofhabits, etc.Blyth, E. Cuvier's Animal Kingtlom, etc., London, 1840, pp. 78 and 79.A brief account of the genus Tupaia, with included species T. tana, sumatrana, and ferruginea. Su-matrana occurs as a new name, and is a nomen nudum.
. Catalogue of M^ammalia in the Museum Asiatic Society, Calcutta, 1863.pp. 81-82.Lists a few specimens and mentions T. ellioti as questionablj- occurring in Ceylon.
. Catalogue of Mammals and Birds of Biu-ma. Journ. Asiat. See. Bengal,part 2, Extra Number, August, 1875, pp. 31, 32.Describes T. bdangeri and quotes authors on habits.Boas, J. E. V. Ohrknorpel und Aussere.s Ohr der Saugetiere, 1912, pp. 60-61, pi. 3,fig. 27.Description and figure of the ear cartilage of Tupaia tana {=Tana iana).Bonaparte, C. L. Vertebratorum Systematis. Nuov. Ann. Sci. Nat. Bologna,vol. 2, 1838, p. 113.Uses the name Cladobatidina as a group of the Soricidae.Prodromus Systematis Mastozoologiae. Nuov. Ann. Sci. Nat. Bologna, vol.3, 1840.Catalogo Metodico dei Mammiferi Europei, pp. 1-35, 1845.On p. 3 uses Cladobatina as a subfamily of Soricidas.BoNHOTE, J. Lewis. On a collection of Mammals from Siam made by Mr. T. H.Lyle. Proc. Zool. Soc. London, 1900, pp. 191-195.Records Tupaia belangeri from Nan and Chengmai. In this paper these specimens are identified asT. chinensis.On the Mammals collected during the "Skeat Expedition" to the MalayPeninsula, 1899-1900. Proc. Zool. Soc. London, 1900, pp. 869-883.Tupaia ferruginea from Belimbing Kelantan, Gunong Inas, Kota Bhara. Good lists of localities.Fasciculi Malayenses. Zool., vol. 1, 1903, pp. 1-^5.Records Tupaia belangeri from Biserat, and T. malaccana from Perak-Pahang boundary.On Mammals from South Johore and Singapore collected by Mr. C. B. Kloss.Proc. Zool. Soc. London, 1900, pp. 4-11.Tupaia ferruginea, T. malaccana.[Original Description of Tupaia concolor.] Abstr. Proc. Zool. Soc. London,1907, p. 2, January 22, 1907.On a Collection of Mammals made by Mr. Vassel in Annam. Proc. Zool. Soc.London, 1907, pp. 3-11, pi. 2.Tupaia concolor, Dendregale frenala.Report on the Mammals, pt. 1 of Report on the Gunong Tahan Expedition,May-September, 1905. Journ. Fed. Malay States Mus., vol. 3, 1909.Tupaia ferruginea recorded from Gunong Tahan, 3, 300 feet, and Pahang River.Bourns, Frank S., and Worcester, Dean G. Preliminary Notes of the Birds andMammals collected by the Menage Scientific Expedition to the Philippine Lslands.Occ. Papers Minn. Acad. Nat. Sci., vol. 1, No. 1, pp. 1-64, 1894.On p. 61, Tupaia javanica, recorded on Palawan, Calamianes. Probably misidentification of T.palawanensis and T. mollendorffi.Brehm, a. E. Illustrirtes Thierleben, 1864, vol. 1, pp. 663-665.Cladobates tana, C.ferrugineus, Ptilocerus (sic) lotoii, not very good wood cuts of the tlu^e forms.Die Saugethiere, 1877, vol. 2, pp. 223, 224.Cladobates tana. {Sorex glis, Tupaya, and Hylogalea ferruginea apparently considered synonymouswith it.) A fair wood cut of Tana which is copied by Gill and by Dallas. Ptilocercus not mentioned.
NO. 1976. TREE8HREWS: FAMILY TUPAIID^?LYON. 169Cantor, T. Catalogue of Mammalia inhabiting the Malayan Peninsula and Islands.Journ. Asiat. Soc. Bengal, vol. 15, 1846, pp. 188-190.Remarks on habits, young, mamma;, and cecum (| in. long) of Twpaia.Carus, J. V. Handbuch der Zoologie, 1868-75, p. 89.Brief account of Twpajx; Cladobates, Dendrogale, Ptilocercus, Hylomys.Chapman, Henry C. Observations on Tupaia, with reflections on the origin otPrimates. Proc. Acad. Nat. Sci. Phila., 1904, pp. 148-156.Observations on the anatomy of cecum, stomach, and liver of Tupaia fcrruginea and T. pictum.Cecum said to be absent.CuviER, Frederic. Des Dents des Mammiferes consid^rees comme caractereszoologiques, pp. i-lv, and 1-259, pis. 1-103, 1825.First use of the term Cladobates, pp. 60-61, and descriptions, and illustrations of teetli of [ Cladobates]tana, ferruginea, javanica.
. Table G^nerale et M^thodique at end of Hist. Nat. Mamm., vol. 7, dated1842.106 Cerp ou Banxring {Cladobates javanica), 107 Press {Cladobates ferruginea).CuviER, Georges. Regne Animal, vol. 1, 1829, pp. 125-126.The genus Cladobates or Tupaia briefly mentioned with the included species jat)a?fca, tana,ferruginea,
. Legons d'Anatomie Compar^e, ed. 2, vol. 2, 1837.In article on osteology of the skull much scattered information regarding Cladobates.Dallas, W. S. Insectivora. Cassel's Natural History, vol. 1 (1880?).A compiled account of the family; genera: Tupaia, Dendrogale, Ptilocercus, Hylomys (sic), pp 347-350. Woodcut of Tana chrysura after Giinther.Desmarest, a. G. Mammalogie ou Description des Espfeces de Mammiferes, vol. 2,supplement, 1822, pp. 535-536.Proposes the generic name Glisorex, but adopts Tupaia: T. tana , javanica, ferruginea.Dlard. Report of a meeting of the Asiatic Society of Bengal for March 10, 1820.Asiat. Journ. Month. Reg., vol. 10, pp. 477^78, November, 1820.The first published account of a treeshrew and original description oiSorex glis (= Tupaia glis glis)from Penang.DiARD and Duvaucel. On the Sorex Glis, communicated by Major General Hard-wicke. Notice.?Sur une nouvelle espece de Sorex.?Sorex Glis. Asiat. Res.,vol. 14, 1822, p. 472, pi. 9.a republication of the preceding article.DoRAN, Alban H. G. Moi-phology of the Mammalian Ossicula auditiis. Trans.Linn. Soc. London, ser. 2, vol. 1, 1879, August, 1878, pp. 371-197, pis. 58-64.On p. 441 and following, description of auditory ossicles of Tupaia, and illustrated on pi. 62, fig. 17.Elera, Castro de. Catalogo Sistemdtica de toda la Fauna de Filipinas, etc., vol.1, 1895, p. 17.Tupaia ferruginea recorded from Culion (probably= T. mollendorffi), Paragua (probably= T. pala-wanensis), Java, Sumatra, Borneo, Singapore, Malacca; and T. javanica from Calamianes, Culion (prob-ably= T. mollendorffi, Paragua (probably= T. palawanensis), Borneo, Java, Sumatra.Elliot, Daniel Girard. A Catalogue of the Collection of ilammals m the FieldColumbian Museum. Field Columbian Museum, Pub. No. 115, Zool. Ser., vol. 8,1907.On p. 464, Tupaia tana, one specimen from "Sumatra."Ellis, William. Manuscript Journal and drawings of animals observed during Capt.Cook's third voyage, 1780. In Library of British Museum (Natural History).Description and colored illustration of Tupaia dissimilis. The description published by Gray, Ann.Mag. Nat. Hist., ser. 3, vol. 5, 1860, p. 71. A copy of Ellis's colored drawing forms Plate 1 ot this paper.
. Voyage of Capt. Cook and Capt. Clerke in Ships Resolution and Discovery,1776, 1777, 1778, 1779, and 1780, published 1782.On page 340 of vol. 2 the treeshrews of Pulo Condore are referred to as squirrels
170 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Everett, A. H. A list of the Birds of the Bomean Group of Islands. Journ. StraitsBranch, Royal Asiat. Soc, No. 20, 1889, pp. 91-212.Map of Borneo showing roughly the distribution of highlands and lowlands, witli collector's localitiesmarked in red; map of Palawan and adjacent islands.
. Remarks on the Zoo-geographical Relationships of the Island of Palawan andsome adjacent Islands. Proc. Zool. Soc. London, 1889, pp. 220-228, pi. 23, mapshowing sea depths.On. p. 223 mentions Tupaia javanica SJid ferruginea as occurring on both Borneo and the Palawangroup.
. A Nominal List of the Mammals inhabiting the Bomean Group of Islands.Proc. Zool. Soc. London, 1893, pp. 492-496.On p. 495, a list of the Bomean Tupaildse.
. In Thomas and Hartert, Nov. Zool., vol. 1, p. 656, September, 1894.Records Ptilocercv^ on Sirhassen, Natuna Islands.Fischer, J. B. Synopsis Mammalium, 1829.Pages 2o9-2G0, describes Tupaia (mentioning four other terms that had been proposed) and threespecies tana, javanica, ferruginea.FiTziNGER, Leop. Jos. Die Ausbeute der osterreichischen Naturforscher an Sauge-thieren und Reptilien wahrend der Weltumsegelung Sr. Majestat Fregatte Novara.Sitz. Akad. Wiss. Math. Nat. Wien, vol. 42, year 1860, p. 392.Cladobates nicobaricus nomen nudum.
. Die natiirliche Familie der Spitzhornchen (Cladobatas). Sitz. Akad. Wiss.Math. Nat. Wien, vol. 60, 1869, pt. 1, pp. 263-289.A systematic accoimt of the fatoily, in which Fitzinger included the genus Hylomys: Cladobates tana,speciosuSiferrugineus, belangeri, ellioti, nicobaricu^s,javanicus, Dendrogale murina, Ptilocercus lowii.Flower, Stanley Smyth. On the Mammalia of Siam and the Malay Peninsula.Proc. Zool. Soc. London, 1900, pp. 306-379.Tupaia ferruginea, ranging from Nepal to Java (embraces many forms), T. javanica, from Peninsulato Java (embraces many forms). Good lists of geographic localities.Flower, W. H. Mammalia, pp. 347-446. Encyk. Brit., 9th ed., vol. 15, 1883.Tupaiidse, p. 401.
. An Introduction to the Osteology of the Mammalia, 1885.Tupaia treated nowhere as a whole, but most of the essential osteological characters pointed out atvarious places in the work.Flower and Lydekker. An Introduction to the Study of Mammals Living andExtinct, 1891.Pages 617 and 618 devoted to Tupaiidae, two genera recognized, Ptilocercus and Tupaia. Figiue ofPtilocercus after Gray.Garrod, a. H. Notes on the Visceral Anatomy of the Tupaia of Burmah {Tupaiabelangeri). Proc. Zool. Soc. London, 1879, pp. 301-305, figs. 1-3 of the brain.A brief account of the visceral anatomy of a specimen that lived in the garden of the Society fromFebruary 8, 1875, to December 18, 187G, with observations on the anatomy of Tupaia tana and T. splen-didula.Geofroy St. Hilaire, Etienne. Dictionnaire des Sciences Naturelle Strasbourg,Paris, 1828, vol. 56, p. 77. Article Tupai.Genus called Tupaia Raffies, Cladobates F. Cuvier, latter apparently preferred.Species.?Java: Le Banxrings; C. javanica, F. Cuv., Hist. Nat. Mamm., liv. 35. Sumatra: LeTana; C. m7wRaf!l.,Tran. linn., 1. 13. Le Press; C.ferruginea'RatR., Press, T. Cuv. ,'Rist. Hat. Uamm.liv. 36.Geoffroy St. Hilaire, Isadore. In Belanger, Voyage aux Indes-Orientales,Zoologie, 1835. Text, pp. 103-107, and Atlas, pi. 4.Original description at Tupaia belangeri, and rather poor colored illustrations of same.
NO. 1976. TREESHREWS: FA2IILY TUPAIIDJE?LYON. 171Geoffrot-Saint Hilaire and Cuvier, Frederic. Hist. Nat. Mamm., vol. 3, livr.35, Cerp ou Banxring, dated December, 1821. Description of Sorex-Glis Javanica(Horsfield) and fair colored plate; first nse of Sorex-Glis as a generic term; livr. 36.Le Press, dated January, 1822. Description of and fair colored plate of Tupayaferruginea Raffles.Gervais, Paul. Histoire Naturelle dea Mammiferes, vol. 1, 1854, pp. 226-229.Tupaia ferruginea, tana, javanica, murina, ellioti, peguana, Piilocercus lowii. Figures of skull andteeth of T. ferruginea, external appearance of same, of T. murina, and of Ptilocercus, all in black andwhite.GiEBEL, C. G. Die Saugethiere, 1855, pp. 913-915.A brief review of the Tupaiidse as known at the time of publication. Cladobates murinus, tanaferrugineus, ellioti, javanicus, Piilocercus lowi.Odontographie, p. 18, pi. 5, figs. 6 and 15-18, 1855.Descriptions of teeth, illustrations, copied evidently from Horsfield and Owen; and original publica-tion of Glisosorex.GtLL, Theodore. Arrangement of the Families of Mammals. Smiths. Misc. Coll.No. 230, November, 1872.On page 19, Tupayidse given as the 91st family. No included genera.Synopsis of Insectivorous Mammals. Bull. Geol. Geogr. Surv. Terr., No. 2,eer. 2, May 14, 1875.On page 21, group Tupaioidea, fam. Tupaiidae, genera Tupaia (including Dendrogale) and Ptilocercus.Article Insectivora. The Standard Natural History, 1886, pp. 134-158.A general account of the family on p. 141 (compiled), and wood cut of Tana opposite p. 148.Gray, John Edward. An outline of an attempt at the disposition of Mammalia intoTribes and Families, with a list of the genera apparently belonging to each tribe.Ann. of Philos., new ser., vol. 10=Thomson's Ann. Philos., vol. 26, 1S25, p. 339.Tupaina as a subfamily of Talpidae.List of the Specimens of Mammalia in the collection of the British Museum.1843, pp. 76-77.Eight specimens listed, 3 species: Tupaia tana, T. javanica, T. ferruginea.Description of a new genus of Insectivorous Mammalia, or Talpidae, fromBorneo. Proc. Zool. Soc. London, 1848, pp. 23-24, pi. 2, in colors; very good.Original description of the genus and species Ptilocercus lowii, and remarks on treeshrews in general;original description of the genus Dendrogale.Description of a new geuus of Insectivorous Mammalia, or Talpidse, fromBorneo. Ann. Mag. Nat. Hist., ser. 2, vol. 2, p. 212, September, 1848.A republication of preceding account.Vertebrata. Zool. Voyage H. M. S. Samarang, pp. 18-20, 1850.Detailed account of Ptilocercus, and summary of the other genera and species, Tupaia javanica,ferruginea, tana, Dendrogale murina, and Ptilocercus lowii. Entire animal in colors, and skull and teethon plate 5.Early notice of the Tapaia (sic) found in Pulo Condore. Ann. Mag. Nat. Hist.,ser. 3, vol. 5, 1860, p. 71.Original publication of W. ElHs's account oC'Sciurus dissimilis" (= Tupaia dissimilis).Ann. Mag. Nat. Hist., ser. 3, vol. 6, p. 217, August, 1860.Original description of Tupaia frenata.Notice of a species of Tupaia from Borneo in the collection of the BritishMuseum. Proc. Zool. Soc. London, 1865, p. 322, pi. 12.Original description of Tupaia splendidula, and a fairly good colored plate.Gregory, W. K. The Orders of Mammals. Bull. Amer. Mus. Nat. Hist., vol. 27.1910.Pages 269-280, a detailed consideration of the family Tupaiidae. Two genera recognized: Tupaia,Ptilocercus, illustrations of the feet and skeleton of Tupaia, and of skull of Ptilocercus. Special attentionis paid to affinities of the family, structure of teeth, and anatomy of skull.
172 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Griffith, E. Animal Kingdom, vol. 2, 1827, pp. 211-216.A rather vague account of the genus Tupaia and the species tana,javanica, and ferruginea; a betteraccount, vol. 5, p. 106, and there occurs this combination, .Tupaia Tanaia Tana, ascribed to Raffles.Evidently a misprint, as the combination is not found in Raffles.GiJNTHER, A. Remarks on some Indian and more especially Bornean Mammals.Proc. Zool. Soc. London, 1876, pp. 424-428, pi. 36, Tupaia tana var. chrysura in colors.A review of the genus based on the specimens in the British Museum; 12 forms recognized, and Tanachrysura characterized for first time.Haeckel, Ernst. Systematische Phylogenie der Wirbelthiere, vol. 3, 1895.On p. 582 uses Cladobatida as a group name for the treeshrews.Hardwicke, Thomas. [Introductory Remarks. | On the Sorex Glis. Trans. Asiat.Soc. Bengal, vol. 14, 1822, pp. 471-472. See Diard and Duvaucel, on page 169.Observations on habits.Heilpuin, Angelo. The Geographical and Geological Distribution of Animals, 1887.On p. 345, Tupaiidse (genera Tupaia, Ptilocercus) briefly mentioned as inhabitants of Oriental Region.HoLLisTER, N. A List of the Mammals of the Philippine Islands, exclusive of theCetacea. Philippine Journ. Sci., vol. 7, No. 1, pp. 1-64, February, 1912.A list of the Tupaiidce in the Philippine Islands, pp. 6-7.HoRSPiELD, Thomas. Zoological Researches in Java and the neighboring islands,1824.An account of the members of the genus Tupaia known at that time: T. ferruginea, T. tana, and T.javanica new. Rather poorly colored illustrations of the latter two, and a plate in black and whiteshowing head and teeth of the three forms, and feet T. tana and javanica.
. Planteg Javanicse Rariores, 1838-1842.Contains an excellent old-time map of Java, showing Horsfield's journeys in Java, and map with p.V of postscript enables one to determine location of type-localityof Tupaia javanica.
. Catalogue of Mammalia in the Museum of the .Hon. East India Company,1851.On pp. 130-134 lists all the then known species of Tupaia: javanica, ferruginea, tana, murina, peguana,helangeri, and gives a very clear summary of the history, discoverj', and habits of the animals. Exam-ples only of the first two in the museum.Hose, Charles. Mammals of Borneo, 1893.A good list of all'of the Bornean forms of the family Tupaiidse, the descriptive matter mostly com-piled, but some good original notes. Forms mentioned: T. ;avanica, longipes, tana, minor, dorsalis,splendidula, picta, montana, melanura, gracilis, D. murina, P. lowii.Hubrecht, a. a. W. Ueber die Entwicklelung der Placenta von Tarsius undTupaja nebst Bermerkungen ueber deren Bedeutung als haematopoietische Organe.Proc. Fourth Internat. Cong. Zool., 1899, pp. 343-382, pis. 4-15.Huschke. Ueber die Zahne von Cladobates. Isis, vol. 20, 1827, pp. 758-759. pi. 10.A rather good description of the teeth of Tupaia and in comparison with Sorex, Talpa, Erinaceus, andClinjsocliloris.Huxley, Thomas H. A Manual of the Anatomy of Vertebrated Animals, 1872.On pp. 3S3, 384, a few facts about anatomy and osteology of the treeshrews, which are designatedTupayae.Ingersoll, E. Life of Animals (Mammals), 1906, p. 75.Barely mentions Tupaias and their distribution, but refening to their resemblance to squinels says:
"This has often been adduced as a case of ' mimicry,' which is very rare among mammals; but it seemsto me rather an instance of 'convergence '?that is, the result of two animals coming to be like one another,because they have followed the same manner of life under identical circumstances."Jentink, F. a. On some rare and interesting mammals. Notes Leyden Museum,vol. 7, p. 37, 1885.Mentions occurrence ot Ptilocercus on Sumatra and Banka and makes a few remarks on the skeleton.
. Catalogue Ost^ologique de Mammifferes. Mus. Hist. Nat. Pays-Bas., vol. 9,1887.List of the osteological material of the family Tupaiidse in the Leyden Museum, pp. 240-242.
NO. 1976. TREE8HREWS: FAMILY TVPAIID^?LYON. 173Jentink, F. a. Catalogue Systematique des Mammifferes. Mus. Hist. Nat. Pays-Bas., vol. 12, 1888.List of the skins, all mounted, and specimens preserved ia alcohol of the family TupaiidJE in theLeyden Museum, pp. 116-118. Contains the new generic term Glipora and the new specific namesleucogaster and rufesccns.
. On a collection of Mammals from East Sumatra. Notes Leyden Museum,vol. 11, pp. 17-30, 1889.Records Tupaiajavanica (pToha,hly= T. 7niTiorvialaccana),Tand}ongMord\vii, 7\ tana Deli, T.ferru-ginea Deli, T.fcrruginca var. chrysura (= T. demissa), and Ptilocercu.s lowii, Tandjong Morawa.
. On a collection of Mammals from Billiton. Notes Leyden Museum, vol. 12,pp. 149-154, 1890.Records Tupajajavanica from Billiton; it is probably an example of T. gracilis injlata.
. On two very rare, nearly forgotten and often misunderstood mammals fromthe Malayan Archipelago. Notes Leyden Museum, vol. 12, 1890, pp. 222-230.Considers in detail the cotypes of Tupaia dorsalis Schlegel, and points out the similarity ia form ofthe skulls of that species with those of Tupaia tana, both now in the genus Tana.Jerdon, T. C. Mammals of India, 1867, pp. 64-66.Gives characters of the family and of T. clliotti and T. chinensis, remarks on food.Kloss, C. Boden. In the Andamans and Nicobars, 1903.An account of Abbott and Kloss's visit to those islands in 1901. Describes habits of Tupaia nico-barica, pp. 122, 136.
. A Provisional List of the Mammals of the Peninsula Region. Journ. Fed.Malay States Mus., vol. 2, No. 3, pp. 147-150, September, 1908.Tupaia fcrruginca, T. f. belangeri, T. sordida, T. pulonis, T. malaccana, Ptilocercus lowi.
. Diagnoses of new Mammals from the Trengganu Archipelago, east coast ofMalay Peninsula. Ann. Mag. Nat. Hist., ser. 8, vol. 7, pp. 115-119, January, 1911.Original descriptions of Tupaia obscura and T. ferruginea longicauda.
. On Mammals and Birds from Trengganu. Journ. Fed. Malay States. Mus.,vol. 4, No. 2, pp. 135-143, April, 1911.Tupaiaferruginea recorded.
. On Mammals and Birds from the lowlands of Pahang. Journ. Fed. MalayStates Mus., vol. 4, No. 2, pp. 144-166, April, 1911.Tupaiaferruginea and T. malaccana mentioned.
. On a Collection of Mammals and other Vertebrates from the TrengganuArchipelago. Journ. Fed. Malay States Mus., vol. 4, pp. 175-212, November, 1911.Detailed descriptions and measurements of Tupaia obscura, and T. longicauda; remarks on habits.
. On Mammals and Birds from the Hills of Negri Sembilau. Journ. Fed.Malay States Mus., vol. 4, pp. 219-229, November, 1911.Tupaia ferruginea recorded from Bukit Tangga.KoHLBRUGGE, J. H. F. Bijdragen tot de natuurlijke Geschiedenis van Menschenen Dieren, III. Zoogdieren van Zuid-Oost Borneo. Nat. Tijdschr. Ned.-Ind.,vol. 55, ser. 9, vol. 4, pp. 176-200, 1896.Original description of Tupaja Miilleri (p. 196)= T.splendidula.Leche, W. Zur Anatomie der Beckenregiou bei Insectivora, mit besonderer Beriick-sichtigung ihrer morphologischeu Beziehungen zu derjenigen anderer Siiugethiere.Kongl. Svenska Vet.-Akad. Handl., vol. 20, No. 4, 1883, pp. 1-113, pis. 1-10.Descriptions and illustrations of the bony pelvis, the pelvic nerves, and muscles of Insectivora, amongthem a male Tupaia ferruginea and a female T. javanka.
. Zur Morphologic des Zahnsystems der Insectivoren. Anat. Anzeiger, vol.13, 1897, pp. 528-529.Brief account of milk and permanent teeth of Tupaiidae ( Tupaia tana,belangcvi (sic),and melanura).
174 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Lesson, Rene-Primeverre. Manuel de Mammalogie, 1827, pp. 122-123.CladobatesfcTTugineus , tanajavanica.
. Nouveau Tableau de Regne Animal, Manamiferes, 1842, p. 93.Tupaiaferruginea, T.javanica, T.tana T. pet/Manus, original descripti&n of latter.Lewis, Capt. Observations of in E. Blyth. Journ. Asiat. Soc. Bengal, vol. 15, p. 368.Mentions squirrels on the Nicobar Islands, probably = Tupaia nicobarica.LicHTENSTEiN, H. Uber die Verwandtscliaft der Kleinen (Insectenfressenden)Raubthiere mit den Nagern. Abhandl. kon Akad. Wissensch, Berlin (1831), 1832,pp. 345-360.On p. 356 mere mention of resemblance between Tupaya and Sciurus.Lydekker, Richard. The Tree-Shrews, or Tupaias. Royal Natural History, vol. 1,1893-94, pp. 312-315.A good popular account of the family with woodcuts of Tana and Ptiloccrcus.
?-? . A Geographical History of ]\Iammals, 1896, pp. i-xii, 1-400.On p. 270 gives distribution of family and a rather poor woodcut on p. 271, " Tupaia tana,"apparently taken from Giinther's plate of Tana clmjsuTa.
. Insectivora (pp. 638-644), Encyl. Brit., ed. 11, vol. 14, 1910.Brief account of the Tupaiidac, genera Tupaia, Ptilocercus, and Urogale. U. everetti erroneouslyreferred to Borneo.Lyon, Marcus Ward Jr. Mammals of Banka Mendanau, and Billiton, islands be-tween Borneo and Sumatra. Proc. U. S. Nat. Mus., vol. 31, pp. 575-612, Decem-ber 18, 1906.Original descriptions of Tupaia inflaia and T. discolor.
. Mammals of Batam Island, Rhio Archipelago. Proc.U. S. Nat. Mus., vol.31, pp. 653-657, January 16, 1907.Original description of Tiipaia fcrruginea batamana.
. Mammals collected in western Borneo by Dr. W. L. Abbott. Proc. U. S.Nat. Mus., vol. 33, pp. 547-572, December 24, 1907.Records Tupaia dorsalis and T.spcciosa from Lower Kapuas River (both nowi n genus Tana and thelatter= T. t. bcsara).
. On a collection of mammals from the Batu Islands, west of Sumatra. Ann.Mag. Nat. Hist., ser. 8, vol. 1, pp. 137-140, February, 1908.Records Tupaia ccrvicalis from Tana Masa Island (now Tana cervicalis masx).
. Mammals collected in eastern Sumatra by Dr. W. L. Abbott during 1903,1906, and 1907, with descriptions of new species and subspecies. Proc. U. S. Nat.Mus., vol. 34, pp. 619-679, September 17, 1908.Original description of Tupaia siaca; T.fcrruginea recorded from Aru Bay.
. Additional Notes on Mammals of the Rhio-Linga Archipelago, with descrip-tions of new species and a revised list. Proc. U. S. Nat. Mus., vol. 36, pp. 479-491,June 1, 1909.List of troeshrews occurring in the Archipelago.
. Mammals collected by Dr. W. L. Abbott on Borneo and some of the small 'adjacent Islands. Proc. U. S. Nat. Mus., vol. 40, pp. 53-146, April 25, 1911.Records Tupaia speciosa (now Tana tana besara) and T.t. tana from Kapuas River and southernBorneo; T'. dorsoZis, Kapuas River; T. splendidula, southern Boraeo; T.longipes (now T. hsalatana),southern Borneo; T.canmatx, Karimata Island; T.inflata (now T. gracilis edarata), Karimata Island;T. grocilis Kendawangan River; T. minor, southern Borneo and Pulo Laut.
. Descriptions of four new treeshrews. Proc. Biol. Soc. Wash., vol. 24, pp.167-170, June 16, 1911.Original descriptions of T.raviana, T.pemangilis, T.natunx, T.sincipis,
NO. 1976. TREESHREWS: FAMILY TUPAIID^?LYON. 175Lyon, Marcus Ward, Jr., and Osgood, Wilfred Hudson. Catalogue of the Type-specimens of Mammals in the United States National Museum, including theBiological Survey Collection. Bull. U. S. Nat. Mus. 62, January 28, 1909.The type-specimens of species in the family Tupaiidfe in the U. S. National Museum, listed, pp.251-254.Mason, Rev. Francis. The natural productions of Burmah, 1850.On p. 224 mentions Tupaiajavanica and pcguana (=chinensis and belangeri).Matschie, Paul. Uber Saugethiere von der Philippinen. Sitz-Ber. Ges. nat.Freunde, Berlin, 1898, pp. 38, 43.Original description of Tupaia moUendorffl.Mearns, Edgar A. Descriptions of new Genera and Species of Mammals from thePhilippine Islands. Proc. U. S. Nat. Mus., vol. 28, pp. 425-460, May 13, 1905.Original description of genus Urogale and species cylindrura {=cvcreUi).Meijere, J. C. H. De. iJber die Haare der Saugethiere, besonders iiber ihre Anord-nung, pp. 312-424. Morph. Jahrb., vol. 21, 1894.On p. 398, the arrangement of the hairs of Tupaia javanica and of Ptilocercus lowii, and schematicfigure of the scales and hairs of tail of Ptilocercus (fig. 3, p. 319). Hairs arise, singlj', and not in definitegroups, 3 hairs to a scale in Ptilocercus. Hairs of Tupaia liave a thickness of 0.016 to 0.020 mm., occa-sionally 0.065 mm. Hairs of Ptilocercus 0.012 mm. in diameter.Miller, Gerrit S., Jr. Mammals collected by Dr. W. L. Abbott on Islands in theSouth China Sea. Proc. Wash. Acad. Sci., vol. 2, pp. 203-246, August 20, 1900.Original descriptions of Tupaia hunox, T . sordida, T. chrysomalla. T. tana (now TarM lingx)recorded from Linga Island, T.malaccana, from Linga, with illustrations of skull of latter.
. Mammals collected by Dr. W. L. Abbott on Pulo Lankawi and the ButangIslands. Proc. Biol. Soc. Wash., vol. 13, pp. 187-193, December 21, 1900.Records Tupaia fcrruginea on Pulo Lankawi and Butang Islands, and also at Trong, Lower Siam.At present these represent 3 forms of Tupaia laccrnata. At the time of Miller's paper examples of trueTupaia glis ferruqinea were not in the United States National Museum collection.
. Mammals collected by Dr. W. L. Abbott on the Natuna Islands. Proc. Wash.Acad. Sci., vol. 3, pp. 111-138, March 26, 1901.Original description of Tupaia (now Tana) sirhassenensis. T. splendidula (now natunx) recordedon Bunguran, and T. lucida on Laut.
. The Mammals of the Andaman and Nicobar Islands. Proc. U. S. Nat. Mus.,vol. 24, pp. 751-795, May 28, 1902.Original description of Tupaia nicobarica surda, Little Nicobar, and record of T. n. nicobarica onGreat Nicobar.
. Mammals collected by Dr. W. L. Abbott in the region of the Indragiri River,Sumatra. Proc. Acad. Nat. Sci. Phila., 1902, pp. 143-159, June 11, 1902.Original description of Tupaia phxura, Sinkep Island . T. malaccana recorded from Linga Island andIndragiri River, and from Sinkep (now T. minor sincipis), and T. tana from Linga (now Tana lingx).
?? . Mammals collected by Dr. W. L. Abbott on the coast and islands of north-west Sumatra. Proc. U. S. Nat. Mus., vol. 26, pp. 437-484, February 3, 1903.Records Tupaia ferruginea, Loh Sidoh and Tapanuli Bays; T. tana (now Tana tana tuancus) fromTuangku Island, T. malaccana from Tapanuli Bay.
. Seventy New Malanyan Mammals. Smiths. Misc. Coll., vol. 45, No. 1420,pp. 1-73, pis. 1-19, text fig. 1, November 6, 1903.Original descriptions of Tupaia castanea, Bintang Island; T. pulonis, Pn\o Aor; T. tephrura, TanaBala Island; T. chrysogastcr, Pagi Islands (skull, fig. 1, pi. 10); T. cervicalis, Tana Bala Island. Views ofskull of Tupaia ferruginea from Tringanu, fig. 2, pi. 10. Key to species of splendicula group= T. lucida,T. splendidula, T. chrysomalla, and T. castanea.Mammals collected by Dr. W. L. Abbott in the Karimata Islands, DutchEast Indies. Proc. U. S. Nat. Mus., vol. 31, pp. 55-66, July 23, 1906.Original description of Tupaia Qanmatx.
176 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Miller, Gerrit S., Jr. The Mammala collected by Dr. W. L. Abbott in the Rhio-Linga Archipelago. Proc. U. S. Nat. Mus., vol. 31, pp. 247-286, September 11, 1906.Records rwpaiacosfaneafromPuloBintang; 7\ tana {now Tana lingx)iTomLinga.; T. phxura tromSinkep; T. fenuginea (now T. glis batamana) from Batam; T. malaccana from Linga and Sinkep (nowpart T. minor sincipis).
??? . Descriptions of two new genera and sixteen new species of mammals fromthe Philippine Islands. Proc. U. S. Nat. Mus., vol. 38, pp. 391-404, August 19,1910.Original description of Tupaia cuyonis, p. 393.MivART, St. George. Notes on the Osteology of the Insectivora, Jouxn. Anat.Physiol., vol. 1, 1867, pp. 292-295, and vol. 2, 1868, pp. 145-146.Describes skull, osteology, and teeth of Tupaia, and illustrates trunk vertebrae, and grinding surfaceof upper molars of Tupaia and Ptilocercus and lower of Tupaia.
. Notes suf rost6ologie des insectivores. Ann. Sci. Nat., Paris, ser. 5, vol. 8,1867, pp. 221-284, and vol. 9, 1868, pp. 311-372.A publication in French of the above.On Hemicentetes, a new genus of Insectivora, with some additional remarkson the osteology of that order. Proc. Zool. Soc. London, 1871, pp. 67-79.A general account of the families and genera of Insectivora. Osteological characters of Tupaiideewell set forth. Throe genera recognized in family, Tupaia, Ptilocercus, Hylomijs.MoDiGLiANi, E. Appui\ti intorno ai mammiferi dell' isola Nias. Ann. Mue. Civ.Stor. Nat. Genoa, ser. 2, vol. 7 (27), 1889, pp. 238-245.Records Tupaia javanica on Nias.Murray, A. Geographical Distribution of Mammals, 1866, p. 233, map 65.Briefly gives distribution, not mentioning Ceylon, but the colored area of distribution on the mapincludes Ceylon.Nehring, Alfred. Uber Saugethiere von den Philippinen, namentlich von derPalawan-Gruppe. Sitz.-Ber. Ges. naturf. Freunde, Berlin, 1894, pp. 179-193.On p. 184 record^ Tupaia fcrruginea from the Calamines (= T.palawanesis) and probably came fromPalawan.OsBORN, Henry Fairfield. The Age of Mammals in Europe, Asia, and NorthAmerica, 1910.On p. 522, imder classification of Mammals: Suborder Menotyphla, Family Tupaiidas, genera Tupaia,Ptilocercus, and family Macroscelididae.Owen, Richard. Odontography, vol. 1, p. 419, vol. 2, pi. Ill, fig. 3, 1840-1845.lUiostrations and descriptions of teeth of Tana.Palmer, T. S. Index Genera ]\Iammalium=North American Fauna, No. 23, 1904.On page 875 are gathered together the family, subfamily names, genera, and subgenera, the latterfully discussed in the body of the work. The only omission I have noticed is Glipora Jentink.Parker, W. K. Development of the skull in the Mammalia; a few remarks on anadult skull of Tupaia javanica. Philos. Trans. Royal Soc. London, vol. 176, year1885, pp. 267-8, pi. 1, 1886.A rather brief account, showing Tupaia rather primitive, with certain marsupial aflinities, but inbrain capacity approaching lemurs.Paulli, Simon. Uber die Pneumaticitat des Schadels bei den Saugethieren, Morph.Jahi-b., vol. 28, pp. 483-564.On p. 486, endo and ecto turbinals of Cladobatcs discussed, peculiar among the Insectivora in ha\Tngonly 2 ecto-turbinals instead of 3.Pechuel-Lo;esche. Brehms Tierleben, 1890, vol. 2, pp. 382-383.Brief account of Tupaiida;, but Ptilocercus omitted.Woodcut of Tana.Pelzeln, August von. Uber die malayische Saugethier-Faima, K. k. zool. bot,Ges. Wien, 1876, pp. 53-74, and map.Tupaia montionod as occurring in his "tibotanische hinterindien and sundaishe Unterabteilung."
NO. 1976. TREE8HREWS: FAMILY TUPAIID^?LYON. 177Peters, Wilhelm. tJber die Saugethier-Gattung Solenodon. Abh. kon Akad. Wiss.,Berlin, 1863, p. 20.Outline classification of Insectivora, under Tupayse: 1. Cladobates; 2. Ptilocercus; 3. Hylogale.PoMEL, A. Etudes sur les camassiers insectivores (extrait), Seconde-Partie.?Classi-fication des insectivores. Arch. Sci. Phys. Nat. Genfeve, vol. 9, 1848, pp. 244-251.On p. 250 the treeshrews, called Hylogaliens (one genus Sorexglis),th.e first type of the tribe Glisori-ciens (Glisoricina), the second type being Dipogaliens with the genera Macroscelis and PetrodTomus.PousARGUES, E. DE. Mammiferes de I'lndo-Chine. Mission Pavie Indo-Chine,1879-1895, Etudes diverses, III, Recherches sur I'Histoire Naturelle de I'lndo-Chiae orientale, pp. 510-549.Mentions as occurring in Indo-China Tupaiachinensis, T.belangeri, T.fcrrngineajDendrogalefrenafa(p. 520).Raffles, Sir Thomas Stamford. Descriptive Catalogue of a zoological collectionmade on the account of the Honourable East India Company, in the Island of Sumatraand its vicinity, etc. Trans. Lion. Soc. London, vol. 13 (1822), pp. 239-274 (mam-mals). May, 1821,Original description of the genus Tupaia and species ferrtiginea and tana; remarks on habits.Ridley, H. N. On the Dispersal of Seeds by Mammals. Journ. Straits BranchRoyal Asiat. Soc, No. 25, pp. 11-32, 1894. .On p. 21 mentions eating of fruits by Tupaia ferruginea.
. List of Mammals recorded from Pahang. Journ. Straits Branch Royal Asiat.Soc, No. 25, pp. 57-60, 1894.Tupaiaferruginea and T.javanica (probably = T. minor malaccana.)
. The Mammals of the Malay Peninsula. Natural Science, vol. 6, January,1895, pp. 23-29.Excellent remarks on the habits of Tupaia glis ferruginea.Robinson, H. C. A List of a Small Collection of Mammals and Birds from the Moun-tains of Ulu Langat, Selangor. Journ. Fed. Malay States Mus., vol. 4, November,1911, pp, 235-241.Records Tupaia ferruginea ferruginea.H. C. R[obinson] and C. B[oden] K[loss]. In Thomas and Wroughton, Journ. Fed.Malay States Mus., vol. 4, No. 1, December, 1909, pp. 111-112.Notes on the habits of Tupaia ferruginea.
. On Six New Mammals from the Malay Peninsula and Adjacent Islands.Journ. Fed. Malay States Mus., vol. 4, No. 2, pp. 169-174, April, 1911.Original description of Tupaia ferruginea wilkinsoni.
. On New Mammals from the Malay Peninsula and Adjacent Islands. Journ.Fed. Malay States Mus., vol. 4, pp. 241-246, November, 1911.Original description of Tupaia ferruginea penangensis? Tupaia glis glis, Penang Island.ScHiNz, H. R. Naturgeschichte und Abbildungen der Menschen und der Sauge-thiere, p. 54Briei description ol "Gladobates (sic) ferrugineus." "It lives in Java." Black and white figure onplate 11. Evidently compEed from Cuvier.
. Naturgeschichte und Abbildungen der Saugethiere, 1824, pp. 87-88, pi. 62.Brief description and black and white illustration of " Chladobates (sic) javanicus." Evidentlycompiled from Cuvier. Among the known species in the genus are "Der gestreifte Tupaja. Cladob.vittatus, in Sumatra. Der Tana, Cladob. tana, in Sumatra. Der rostfarbe Tupaja. Cladob. ferrugineus,in Java."ScHLEGEL, Hermann. Handleiding to de Beoefening der Dierkunde, 1857, vol. 1,pp. 58-59, pi. 3, fig. 31 (f).Describes the genus Tupaia and mentions in it javantca, tana, dorsalis (new), black and whitefigure of latter.80459??Proc.N.M.vol.45?13 12
178 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.ScHLEGEL, Hermann. Die Dierentuin van het Kon. Zool. Gen. Natura Artis Magis-tra, Mammalium, 1872.Rather poor woodcut of Tarm dorsalis on page 62; short description of genus; species mentioned,javanka, taim, dorsalis.ScHLEGEL, Hermann, and Muller, Sal. Over de op de oostindische eilandenlevende soorten van het geslacht Hylogalea. Verh. Nat. Gesch. Nederl overz.Bezitt., 1843, pp. 159-168, pis. 26 and 27.Descriptions of the 4 then known Malayan species, tana, ferruginea, javanka, murina (new); coloredillustrations of the heads of the others, and entire animal of murina; drawing of skulls and bones of feetof all; original publication of Hylogalea as the generic term for the treeshrews.Schneider, Gustav. Ergebnisse zoologischer Forschnngsreisen in Sumatra 0. K.Zool. Anzeiger, vol. 27, pp. 722-724, July 12, 1904.Contains original description of Tupaia ferrxiginea demissa. See Oldfield Thomas.Ergebnisse zoologischer Forschnngsreisen in Sumatra. Saugetiere (Mammalia).Zool. Jahrb., vol. 23, 1905, pp. 1-172.Records specimens collected in the Deli and Indragiri regions, enumerating Ptilocercus lowii, Tupaiaferruginea, T. /. demissa, T. splendidula (probably T. siaca), T. castanea (= T. siaca), T. javanica, T.malaccana, T. tana, and T. taria var. speciosa (the latter two probably represent the same form, tana).Interesting remarks on living specimens of Ptilocercus. Illustrations in colors, plates 1 and 2 ofPtilocercus and Tupaia demissa.ScLATER, W. L. and P. L. Geography of Mammals, 1899, p. 145.Mentions Tupaia and Ptilocercus as being very characteristic of the Malayan Subregion of theOriental Region.ScUDDER, Samuel H. Nomenclator Zoologicus. Bull. U. S. Nat. Mus., No. 19,1882.In part 2, p. 130, original publication of Glirisorex.Sterndale, R. a. Natural History of the Mammalia of India and Ceylon, 1884, pp.99-104.Describes the mainland and Nicobar species and includes Ptilocercus, thinking it may be foimdin Tenasserim. Remarks on habits, etc., mostly quoted.Stone, Witmer, and Rehn, J. A. G. A collection of Mammals from Sumatra, with aReview of the genera Nycticebus and Tragulus. Proc. Acad. Nat. Sci. Phila.,1902, pp. 127-142.lupaia tarm recorded from Gunong Sugi, Lampong District, Sumatra.Temminck, C. J. Monographies de Mammalogie, vol. 1, 1827.On p. 19 first use of Hylogale for the " barbarous " Tupaia.Thomas, Oldfield. On the Mammals presented by Allan O. Hume, Esq., C. B., tothe Natural History Museum. Proc. Zool. Soc. London, 1886, pp. 54-79.Records Tupaia belangeri (now called chincnsis) from Aimole and Machi Manipur; T. helangeri and T.ferruginea frotn Tenasserim; T. ferruginea from Malacca and Selangor; T. javanica{j}Tohah\y T. malaccana)from Selangor and Johore.
. On the Mammals of Mount Kina Balu, North Borneo. Proc. Zool. Soc. Lon-don, 1889, pp. 228-236.On p. 229 records 3 specimens of Tupaia ferruginea from Mount Kina Balu, collected by John White-head. They were probably examples of T. montana baluensis.
. On a Collection of Mammals obtained by Dr. Emin Pasha in Central andEastern Africa. Proc. Zool. Soc. London, 1890, pp. 443-450.Remarks on the teeth of the Macroscelididfe and figure of the milk dentition of Petrodromus.
. On some new Mammalia from the East Indian Archipelago. Ann. Mag.Nat. Hist., March, 1892.Original description of Tupaia everetti (now Urogale everetfi) ajid T.pkta, T. montana, T. malanura(now Dendrogale melanura).
. On some Mammals from Moimt Dulit, North Borneo. Proc. Zool. Soc. Lon-don, 1892, pp. 221-227.Among them Tupaia tana, montana, minor, melanura, and on p. 227, T. dorsalis.
NO. 1976. TREE8HREW8: FAMILY TUPAIID^?LYON. 179Thomas, Oldfield. On the Mammalia collected by Signer Leonardo Fea in Burmaand Tenasserim. Ann. Mus. Civ. Stor. Nat. Genoa, ser. 2, vol. 10 (1890-91), pp.913-949, 1892.On p. 920 records Tupaia ferruginea belangeri (embracing both helangeri and chinensis of the presentpaper). Discusses number of mammse in T. ferruginea, tana, ellioti, and javanica.
. On some new Bomean Mammalia. Ann. Mag. Nat. Hist., ser. 6, vol. 11,pp. 341-347, May, 1893.Original description of Tupaia ferruginea longipes.
. Description of a new Bomean Tupaia. Ann. Mag. Nat. Hist., ser. 6, vol. 12,pp. 53, 54, July, 1893.Original description of Tupaia gracilis.
. On the Palawan Representative of Tupaia ferruginea. Ann. Mag. Nat.Hist., ser. 6, vol. 13, p. 367, April, 1894.Original description of Tupaiaferruginea palawanensis.
. On some Mammals collected by Dr. E. Modigliani in Sipora, MentaweiIslands. Ann. Mus. Civ. Stor. Nat. Genoa, ser. 2, vol. 14, pp. 661-672, January,1895.Original description of Tupaiaferruginea hypochrysa, type-locality, Java.
. In Schneider, Gustav. Ergebnisse zoologischer Forschxmgsreisen in Sumatra,0. K. Zool. Anzeiger, vol. 27, pp. 722-724, July 12, 1904.Original description of Tupaia ferruginea demissa.
. The Duke of Bedford's Zoological Exploration in Eastern Asia.
?
III. OnMammals obtained by Mr. M. P. Anderson in the Philippine Islands. Proc. Zool.Soc. London, 1907, pp. 140-142, June 12, 1907.Remarks on status of Urogale everetti and U. cylindrura, p. 140.
. On Mammals collected by Mr. H. 0. Robinson on Tioman and Aor Islands,S. China Sea. Joum. Fed. Malay States Mus., vol. 2, No. 3, pp. 101-106, 1908.Records Tupaia sordida on Tioman.
. Two new Mammals from the Malay Peninsula. Ann. Mag. Nat. Hist., ser. 8,vol. 5, pp. 424-426, May, 1910.Original description of Ptilocercus lou'i continentis.Thomas, Oldfield, and Hartert, Ernst. List of the first collection of Mammalsfrom the Natuna Islands. Nov. Zool., vol. 1, pp. 652-660, September, 1894.Record Ptilocercus lowi on Sirhassen and Tupaia splendidula (now natunx) on Bunguran and T. tana(now sirhassenensis) on Sirhassen.
. On a second collection of Mammals from the Natuna Islands. Nov. Zool.,vol. 2, December, 1895, pp. 489-492.Tupaia splendidula typica recorded from Bunguran (now T. natunsf) and original description of Tupaiasplendidula lucida.Thomas, Oldfield, and Wroughton, R. C. Diagnoses of new Mammals collectedby Mr. H. C. Robinson in the islands of the Straits of Malacca. Ann. Mag. Nat.Hist., ser. 8, vol. 4, pp. 534-536, December, 1906.Original description of Tupaia lacernata.
. On Mammals from the Rhio Archipelago and Malay Peninsula collected byMessrs. H. C. Robinson, C. Boden Kloss, and E. Seimund, and presented to theNational Museum by the Government of the Federated Malay States. Joum. Fed.Malay States Mus., vol. 4, No. 1, pp. 99-129, December, 1909.Tupaia castanea on Bintang T. ferruginea, S. E. Johore and Singapore; T.f. batamana on Batam.Notes by Robinson and Kloss on TJerruginea.
180 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 45.Trouessart, E.-L. Catalogue Mammalium tarn viventium quam fossilium, vol. 1,1897, pp. 167-169.Ptilocercus, 1 form; Tupaia, subgenus Dendrogale, 2 forms; subgenus Tupaia, 20 forms.Vol. 2, 1899, appendix, pp. 1286, 1287. Two additional species of Tupaia.Quinquennale Supplementum Anno 1904, pp. 120-123. The species and subspe-cies of subgenus Tupaia now 28.[Vigors.] Catalogue of Zoological Specimens. Memoir of the Life and PublicServices of Sir Thomas Stamford Raffles, etc., by Sophia Raffles, London, 1830.Pages 637-638, genus rttpoia. Tup. Tana, Sumatra. Tup. Javanica, Java. Tup./eTruginea,Sum.atra.VoGT, Carl, and Specht, Friederich. Die Saugetiere in Wort und Bild, 1883.On p. 87 a short account of treeshrews and wood cut. Cladobates, Ptilocercus, (Hylomys alsoincluded), family Tupajee.Wagler, J. G. Naturliches System der Amphibian mit vorangehender Classifica-tion der Situgthiere und Vogel, 1830.On p. 15, genus 9, Hylogale, with the sp^cv&s fenuginea, taria, javaniea.Wagner, J. A. Schreber's Saugthiere, Supplementband, 2. Abtheilung, 1841, pp.37^4, p. 553."Systematicaccount of the known Tupaiidae, Cladobates tana, p. 40; C. ferruginev^, p. il; C.belangeri(new), p. 42; C.speciosus (new), p. 43; C.javanicus,p. 44; C.murinus,i>. 553.
. Schreber's Saugthiere, Supplementband, 5. Abth. 1856, pp. 524-529, andpis. 34 and 35, in colors.Descriptions of C7o(fo6a<es tona, p. 525; Cl.ferTugineus,p. 52&,Tpl. 34; CI. ellioti, p. 526; Cl.javanicus,p. 527; CI. belangeri, p. 527; CI. murinus (subgenus Dendrogale), p. 528; Ptilocercus lowii, pi. 35.Wallace, A. R. Geographical Distribution of Animals, 1876, vol. 1, p. 337, pi. 8,vol. 2, p. 187.Mentions in a very general way the distribution of the family. Hylomys is regarded as a member ofthe Tupaiidfe.
. Island Life, 1881.On p. 345, list of the then known Bornean Tupaiidae, Tupaia, Dendrogale, Ptilocerus (sic).Waterhouse, G. R. Catalogue of the Mammalia preserved in the Museum of theZoological Society of London, 1838.On page 19 four specimens listed, Tupaia tana (probably the original specimen), T.javanica, T.ferruginea.
. Description of a new species of Tupaia discovered m Continent of India byWalter Elliot, Esq. Proc. Zool. Soc. London, 1849, pp. 106-108, pi. 13.Original description of Tupaia eZiJoii in comparison with T. tana, ferruginea, and.javanica; very goodplate in colors.Weber, Max. Die Saugethiere, 1904.Pages 376 and 377, family Tupaudse defined; two genera recognized, Tupaia, Ptilocercus.WiLLiNK, T. Mammalia voorkomende in Nederlandsch-Indie. Nat. Tijdschr. Ned.-Ind., vol. 65, pp. 296-300, 1905.A list of Tupaiidae known up to 1905 and occurring in the Dutch East Indies, compiled and non-critical.Winge, Herluf. Jordfundne og nulevende Flagermus (Chiroptera) fra Lagoa Santa,Minas Geraes, Brasilien.Pages 41-51, various osteological and anatomical observations on treeshrews, usually designatedas Cladobates, sometimes as Tupaia.Zelebor, Johann. Saugethiere. Reise der osterreichischen Fregatte Novara umdie Erde in den Jahren 1857, 1858, 1859. Zool. Theil, vol. 1, 1869, pp. 1-42, pis. 1-3.Original description of Tupaia nicobarica, with a plate in colors of entire animal and a plat? illus-trating skeleton, skull, and teeth.
EXPLANATION OF PLATES.Plate 1.Tupaia dissimilis, Pulo Condore. Reproduction of the original figure of WilliamEllis' Sciurus dissimilis, in his natural history journal, written during Capt. Cook'sthird voyage, 1776-1780; now in the British Museum (Natural History). Repro-duced by the permission of the authorities of the British Museum. A scale of 100mm. was laid on the page when photograph was made.Plate 2.Skeleton of Tupaia lacernata ivilkinsoni; Cat. No. 49468, U.S.N.M., Tarang, LowerSiam; collected by Dr. W. L. Abbott. About one-half natural size.Plate 3.External appearance of Tupaia nicobarica, T. javanica, T. minor, and T. glis. Aboutone-third natural size.Fig. 1. Tupaia nicobarica surda; Type; Cat. No. 111757, U.S.N.M., Little NicobarIsland; Dr. W. L. Abbott, collector.2. Tupaia javanica; Cat. No. 154598, TJ.S.N.M., Mount Salak, Java; Bryant andPalmer, collectors.3. Tupaia minor minor; Cat. No. 145575, U.S.N.M., Sempang River, westernBorneo; Dr. W. L. Abbott, collector.4. Tupaia glis ferruginea; Cat. No. 114548, U.S.N.M., Tapanuli Bay, west coastoi Sumatra; Dr. W. L. Abbott, collector.Plate 4.External appearance of Anathana ellioti (Waterhouse). After the original plate incolors by Waterhouse in the Proceedings of the Zoological Society of London for 1849.Plate 5.External appearance of Dendrogale miinna (Schlegel and Miiller). After the originalfigure in Verhandelingen over de Natuurlijke Geschiedenis der NederlandecheOverzeesche Bezittingen, 1843. Plate 6.External appearance of Tana cervicalis, T. dorsalis, TJrogale everetti. About one-thirdnatural size.Fig. 1. Tana cervicalis cervicalis; Type; Cat. No. 121754, U.S.N.M., Tana Bala, BatuIslands, off west coast of Sumatra; Dr. W. L. Abbott, collector.2. Tana dorsalis; Cat. No. 142245, U.S.N.M., Kapuas River, western Borneo;Dr. W. L. Abbott, collector.3. Urogale everetti; Cat. No. 125287, U.S.N.M. (Type of U. cylindrura), MountApo, Mindanao, Philippine Islands; Dr. E. A. Mearns, collector.Plate 7.External appearance of Plilocercus lowii Gray. After the original plate by Gray inthe Proceedings of the Zoological Society of London for 1848. 181
182 PROCEEDINGS OF THE NATIONAL MUSEUM. yol. 45.Plate 8.Skulls of Treeshrews of the genus Tupaia. All figures natural size.Fig. 1. Tupaia chinensis; Cat. No. 37384, U.S.N.M., Carin Hills, Burma; LeonardoFea, collector.2. Tupaia belangeri; Cat. No. 124284, U.S.N.M., Telok Besar, Tenasserim; Dr.W. L. Abbott, collector.3. Tupaia lacernata wilkinsoni; Cat. No. 83254, U.S.N.M., Trong, or Tarang,Malay Peninsula; Dr. W. L. Abbott, collector.4. Tupaia lacernata raviana; Type; Cat. No. 104355, U.S.N.M., Pulo Rawi,Butang Islands, west coast Malay Peninsula; Dr. W. L. Abbott, collector.5. Tupaia lacernata lacernata; Cat. No. 123901, U.S.N.M., Pulo Lankawi, westcoast Malay Peninsula; Dr. W. L. Abbott, collector.6. Tupaia glis ferruginea; Cat. No. 141074, U.S.N.M., Tarussan Bay, west coastof Sumatra; Dr. W. L. Abbott, collector.7. Tupaia tephrura; Type; Cat. No. 121752, U.S.N.M., Tana Bala, Batu Islands,southwest coast of Sumatra; Dr. W. L. Abbott, collector.8. Tupaia glis glis; Cat. No. 1444/11, Selangor Museiun, Penang Island, westcoast Malay Peninsula.9. Tupaia glis batamana; Type; Cat. No. 142151, U.S.N.M., Batam Island,Rhio Archipelago; Mr. C. Boden Kloss, collector.Plate 9.Skulls of treeshrews of the genus Tupaia. All figures natural size.Fig. 1. Tupaia cuyonis; Type; Cat. No. 26, Philippine Museiun, Cuyo, PhilippineIslands; McGregor and Celestino, collectors.2. Tupaia montanxi montana; Cat. No. 84507, U.S.N.M., Mount Dulit, northernBorneo; Charles Hose, collector.3. Tupaia picta; Cat. No. 84506, U.S.N.M., Baram District, northern Borneo;Charles Hose, collector.4. Tupaia discolor; Type; Cat. No. 124703, U.S.N.M., Island of Banka; Dr.W. L. Abbott, collector.5. Tupaia longipes salatana; Type; Cat. No. 151882, U.S.N.M., PangkallahanRiver, Klumpang Bay, southeastern Borneo; Dr. W. L.Abbott, collector.6. Tupaia hypochrysa; Cat. No. 154599, U.S.N.M., Mount Salak, western Java;Bryant and Pahner, collectors.7. Tupaia phseura; Type; Cat. No. 113148, U.S.N.M., Sinkep Island; Dr. W. L.Abbott, collector.8. Tupaia glis sordida; Type; Cat. No. 101747, U.S.N.M., Pulo Tioman, offsoutheast coast Malay Peninsula; Dr. W. L. Abbott, collector.9. Tupaia chrysogaster; Type; Cat. No. 121572, U.S.N.M., North Pagi Island,southeast coast of Sumatra; Dr. W. L. Abbott, collector.Plate 10.Skulls of treeshrews of the genera Tupaia and Ptilocercus. All fignres natural size.Fig. 1. Tupaia javanica; Cat. No. 121488, U.S.N.M., near Batavia, Java.2. Tupaia gracilis injlata; Type; Cat. No. 124709, U.S.N.M., Island of Banka;Dr. W. L. Abbott, collector.3. Tupaia minor minor; Cat. No. 153857, U.S.N.M., Kendawangan River, south-western Borneo; Dr. W. L. Abbott, collector.4. Tupaia minor malaccana; Cat. No. 112618, Sembrong River, Johore; DrW. L. Abbott, collector.5. Ptilocercus lowii continentis; Cat. No. 112611, U.S.N.M., Sembrong Riv?r,Johore; Dr. W. L. Abbott, collector.
NO. 1976. TREESHREWS: FAMILY TUPAnOM?LYOti. 183Fig. 6. Tupaia carimatse; Type; Cat. No. 125123, U.S.N.M., Karimata Island. Dr.W. L. Abbott, collector.7. Tupaia nicobarica surda; Type; Cat. No. 111757, Little Nicobar, NicobarIslands; Dr. W. L. Abbott, collector.8. Tupaia siaca; Type; Cat. No. 144205, U.S.N.M., Little Siak River, easternSumatra; Dr. W. L. Abbott, collector.9. Tupaia castanea; Type; Cat. No. 115608, U.S.N.M., Pulo Bintang, RhioArchipelago; Dr. "W. L. Abbott, collector.10. Tupaia chrysomalla; Type; Cat. No. 101710, U.S.N.M., Pulo Siantan, AnambaIsland, South China Sea; Dr. W. L. Abbott, collector.11. Tupaia splendidula; Cat. No. 153856, U.S.N.M., Kendawangan River, south-western Borneo; Dr. W. L. Abbott, collector.12. Tupaia natunse; Type; Cat. No. 104714, U.S.N.M., Bunguran, Natuna Is-lands; Dr. W. L. Abbott, collector.Plate 11.Skulls of treeshrews of the genera Tana and Urogale. All figures natural size.Fig. 1. Tana dorsalis; Cat. No. 142245, U.S.N.M., Kapuas River, western Borneo;Dr. W. L. Abbott, collector.2. Tarm lingse; Type; Cat. No. 101597, U.S.N.M., Linga Island; Dr. W. L.Abbott, collector.3. Tarm tarm sirhassenemis; Type; Cat. No. 101712, U.S.N.M., Sirhassen, NatunaIslands; Dr. W. L. Abbott, collector.4. Tana tana tana; Cat. No. 174612, Landak region, Sumatra; Gustav Schneider,collector.5. Tana tana bunose; Type; Cat. No. 101640, Bunoa, Tambelan Islands, SouthCliina Sea; Dr. W. L. Abbott, collector.6. Urogale everetti; Cat. No. 125287, U.S.N.M. (type of U. cylindrura), MountApo, Mindanao, Philippine Islands; Dr. E. A. Mearns, collector.7. Tana cervicalis cervicalis; Type; Cat. No. 121754, U.S.N.M., Tana Bala, BatuIslands, southwest coast of Sumatra; Dr. W. L. Abbott, collector.8. Tana tana besara; Type; Cat. No. 142247, U.S.N.M., Kapuas River, westernBorneo; Dr. W. L. Abbott, collector.
i
3.
\
V. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 45 PL. 2
Malay Peninsula Treeshrew, Tupaia lacernata wilkinsoni.For explanation of plate see page 181.

U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 45 PL. 3
NICOBARICA. Skins of Treeshrews of the Genus Tupaia.For explanation of plate see page 181.

U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 45 PL. 4
Indian Treeshrew, Anathana ellioti from Waterhouse.For explanation of plate see page 181.

U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 45 PL. 5
Pigmy Treeshrew, Dendrogale murina from Schlegel and Muller.For explanation of plate see page 181.

U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 45 PL. 6
T. DORSALIS. U. EVERETTI.T. CERVICAUIS.Skins of Treeshrews of the Genera Tana and Urogale.For explanation of plate see page 181.

U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 45 PL. 7
Pentailed Treeshrew, Ptilocercus lowii after Gray.For explanation of plate see page 181.

U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 45 PL. 8
1., CHINENSIS
4. RAVIANA
7 TEPHRUF
?
5. LACERNATA
8. GLIS
6. FERRUGINEA
9. BATAMANA
Skulls of Treeshrews of the Genus Tupaia.For EXPLANATfON OF PLATE SEE PAGE 182.

U. S. NATIONAL MUSEUM PROCEEDINGS. VOL. 45 PL. 9
4. DISCOLOR 5. SALATANA
\
7. PHAEURA 8. SORDIDA
\
6. HYPOCHRYSA
%
9. CHRYSOGASTER
Skulls of Treeshrews of the Genus Tupaia.For explanation of plate see page 182.



U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 45 PL. 11
6. U. EVERETTI 7. T. CERVICALIS 8. T. BESARA
Skulls of Treeshrews of the Genera Tana and UrogaleFor Explanation of plate see page 183.

INDEX OF THE TECHNICAL NAMES.
[References of major importance and names of new forms have the page numbers printed in black-facedtjrpe. For the sake of simplicity trinomial names have been disregarded and indexed as though theywere binomial.] Page.anambas, Tupaia 89Anathana 120ellioti 122pallida 124wroughtoni 123baluensis, Dendrogale 182Tupaia 95batamana, Tupaia 46belangeri, Cladobates 59Tupaia 59besara, Tana 141bunoae, Tana 144Tupaia 144bunoai, Tupaia 144carimatse, Tupaia 98castanea, Tupaia 90cervicalis, Tana 147Tupaia 147chinensis, Tupaia 63Chladobates 25chrysogaster, Tupaia 71chrysomalla, Tupaia 88chrysura, Tana 149Tupaia 149Cladobatse 25Cladobates 25, 30belangeri 59ferruginea 41javanica 106nicobaricus 103speeiosus 139tana 139vittatus 177Cladobatida 25Cladobatina 25Cladobatinida 25concolor, Tupaia 68continentis, Ptllocercus 165cuyonis, Tupaia 82cylindrura, Urogale 157demissa, Tupaia 58Dendrogale 26, 126baluensis 132frenata 128melanura 131,132mm^ina 129discolor, Tupaia 73dissimilis, Sciurus 67Tupaia 67dorsalis, Tana 152Tupaia 152
Page.edarata, Tupaia 118Elephantulus 9ellioti, Anathana 122Tupaia 122Erinaceus 26, 30tana 139everetti, Tupaia 157Urogale 157ferruginea, Cladobates 41Hylogalea 41Tupaia 41, 45frenata, Dendrogale 128Gladobates 26Glipora 26, 30leucogaster 110murina 129Glirisorex 26, 30glis, Sorex 45Tupaia 40, 45Glisorex 26, 30tana 139Glisoricina 25Glisoricinse 25Glisosorex 27, 30gracilis, Tupaia 116,117Herpestes 27Hylogale 27,30Hylogalea 27, 30ferruginea 41javanica 106murina 129tana 139Hylomys 5hypochrysa, Tupaia 70inflata, Tupaia 118javanica, Cladobates 106Hylogalea 106Sorex-Glis 106Tupaia 106lacemata, Tupaia 51,53leucogaster, Glipora 110lingse. Tana 145Lipotyphla 5longicauda, Tupaia 56longipes, Tupaia 74,76lowii, Ptilocercus ? 163,164Ptilocerus 164lucida, Tupaia 86Macroscelididse 5malaccana, Tupaia 114masse, Tana 148185
186 INDEX.Page.melanura, Dendrogale 131,132Menotyphla 5minor, Tupaia 109,110modesta, Tupaia 69mollendorffi, Tupaia 81montana, Tupaia 93,94miilleri, Tupaja 83muriaa, Dendrogale 129Glipora 129Hylogalea 129Tupaja 129natunse, Tupaia 85nicobarica, Tupaia 100,103nicobaricus, Cladobates 103obscura, Tupaia 55paitana, Tana 150palawanensis, Tupaia 78pallida, Anathana 124Pegou, Tupaia du 59peguanus, Tupaia 59pemangilis, Tupaia 48penangensis, Tupaia 45Petrodromus 5, 178phseura, Tupaia 49picta, Tupaia 96press, Sorex-Glis 41Tupaya 41Ptilocercinae 3Ptilocereus 27, 159continentis 165lowil 1 63, 164Ptiloceras 27,159lowii 168pulonis, Tupaia 47raviana, Tupaia 54riabus, Tupaia 88rufescens, Glipora 85, 173rufieaudata, Tupaia 83salatana, Tupaia 77Sciurus 28, 30dissimEis 67siaca, Tupaia 91sincipis, Tupaia 115sirhassenensis, Tana 142Tupaia 142Sorex 28,30glis 45Sorexglis 28, 30Sorex-Glis javanica 106press 41sordida, Tupaia 48speciosus, Cladobates 139splendidula, Tupaia 83sumatrana, Tupaia 168surda, Tupaia 104Tana 134besara 141bunose 144eervicalis 147ehrysura 149dorsalis 162lingse 145masse 148paitana 150sirhassenensis 142tana 187,139
Page.Tana, tuancus 145utara 141tana, Cladobates 139Erinaceus 139Glisorex 139Hylogalea 139Tana 187,189Tupaia 139Tanaia 172Tapaia 28, 30teplirura, Tupaia 50tuancus, Tana 145Tupaia 28,30anamba3 89baluensis 95batamana 46belangeri 59bunose 144bunoai 144carimatEe 98castanea 90eervicalis 147chinensis 63chrysogaster 71chrysomalla 88ehrysura 149concolor 68cuyonis 82demissa 58discolor 73dissimUis 67dorsalis 152edarata 118eUioti 122everetti 157ferruginea 41,45glis 40,45gracUis 116, 117hypochrysa 70inflata 118javanica 106laeemata 51, 53longicauda 56longipes 74, 76lucida 86roalaccana 114minor 109,110modesta 69mollendorffl 81montana 93,94natunae 85nicobarica 100,108obscura 55palawanensis 78Pegou du 59peguanus 59pemangilis 48penangensis 45phseura 49picta 96pulonis 47raviana 54riabus 88rufieaudata 83salatana 77siaca 91sincipis 115
INDEX. 187Page.Tupaia, sirhassenensis 142sordida 48splendidula 83sumatrana 168surda 104tana 139tephrura 50typica 85wjlkinsoni 52Tupaiadse 25Tupaiidse 5, 25Tupaiince 3Tupaina 25Tupaioidea 5Tupaja 29
Page.Tupaja , miilleri 83murina 129Tupajidae 25Tupaya 29press 41Tupayae 25Tupayidse 25typica, Tupaia 85Urogale 29, 154cylindrura 157everetti 157utara, Tana 141vittatus, Cludobates 177wiLkinsoni, Tupaia 52wroughtoni, Anathana 128ADDENDA.The following description appeared too late to be placed in thebody of this paper. It should be read after the account of Tupaiajavanica on page 107. (See also page 15.)TUPAIA JAVANICA BALINA Thomas.1913. Tupaia javanica balina Thomas, Ann. Mag. Nat. Hist., eer. 8, vol. 11,p. 505, May, 1913.Type-locality.?Island of Bali, east of Java.Type-specimen.?In British Museum, skin without skull, Reg. No.13.3.6.12, collected at Gunong Bratan, Bali, 4,000 feet altitude,January 27, 1911, by Mr. E. Stresemann. Origmal number, 10.Geographic distribution.?Island of Bali. (See map on page 111.)Diagnostic characters.?Essentially like Tupaia javanica javanica,but smaller; hind foot shorter; color, browner.Color.?Upper parts grizzled brown, nearest mummy brown; bellyclay color, the hairs slaty basally, buffy on throat and axillae; shoulderstreak well marked ; hands and feet bro^vn, not so gray as in javanicajavanica; inner side of hind limbs washed with buffy whitish ; tail likebody, middle of under surface suffused with cinnamon.Measurements.?Type. Head and body, 117 mm.; tail, 153; hindfoot (without claws), 34. Skull lost.Specimens examined.?None. The above account is based on theoriginal description. The type-specimen is unique.The United States National Museum has received from the Kuteior Mahakkam River, eastern Borneo (see map on page 75), 30 speci-mens of treeshrews, too late to be mcorporated with the precedingaccount. They are identified as follows:TUPAIA MitLLERI KoWbrugge. (See page 83.)1896. Tupaja miilleri Kohlbrugge, Natuur. Tijdschr. Nederl. Indie, vol. 55(ser. 9, vol. 4), 1896, p. 196.Type-locality.?Environs of Banjarmassin, southeastern Borneo.Type-specimen.?I have not seen the type-specunen and do notknow where it is. It may be in the Leyden Museum.Geographic distribution.?Southeastern Borneo.
188 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.45.Diagnosis.?A member of the splendidula group, characterized bylighter neck and shoulders, and by a dark stripe extending from napeonto lower back much as in the case of Tana tana; tail with a distinctgrizzle, resembling that of Twpaia carimatse; mammae, 2-3 = 5 ?Color.?Based on two individuals from the Kutei River region.Upper parts and sides of neck and body, upper side of tail, and outerside of legs, a fine grizzled mixture of chestnut or dark cimiamonrufous and blackish, the reddish color slightly in excess except alongmiddle line of neck and body where the blackish color forms a stripeabout 2 millimeters wide anteriorly and nearly 20 millimeters wideover lower back. Anteriorly the reddish color inclines towardochraceous. Under parts and inner side of legs varying from buffyto ochraceous with darker bases of the hairs showing through. Under-side of tail similar to upper, but lighter color more tawny or tawnyochraceous and more conspicuous on either side of the middle linethan is the brownish black color. Shoulder stripe well marked.Slivll and teetli.?Essentially as in Twpaia splendidula, but braincase slightly wider, approaching that of T. picta.Measurements.?Female with moderately worn teeth and femalewith last molars appearing: Head and body 166, 169 mm.; tail, 143,157 ; hmd foot, 40, 41 ; condylobasal length, 46, 43.5 ; zygomatic width,26, 23; width of braincase, 19, 19; maxillary toothrow, 18, 18.Specimens examined.?Two, both from Loa Bamban, near Samar-inda, Kutei River, eastern Borneo.Remarlcs.?With the exception of the dark dorsal stripe, this speciesclosely resembles Tupaia carimatse (page 98) in color, but the skullis larger and wider. The presence of the dorsal stripe is somewhatsuggestive of T. picta, but the stripe is not so pronounced or narrow,and the whole animal is much duller in color and apparently closelyrelated to T. spendidula. I have not examined the type of T. milUeri,and the above identification is made with some reservation.TUPAIA MINOR MINOR Giinther. (See page 110.)Twelve specimens; 2 from Batu Panggah, 2 from Segah River, and8 from Birang River; Kutei River region.TUPAIA GRACILIS GRACILIS Thomas. (See page 117.)Four specimens; 1 each from Loa Bamban, Karang Tigau, BirangRiver, and Segah River; Kutei River region.TANA TANA TANA (Raffles). (See page 139.)Twelve specimens, 7 from Birang River, 3 from Segah River, 1 fromDomering, and 1 from Talisaian Mountain; Kutei River region.