SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 13 Porter M. Kier Upper Miocene Echinoids from the Yorktown Formation of Virginia and Their Environmental Significance I S S U E D JW>R 1 0 1 9 7 2 SMITHSONIAN INSTITUTION PRESS CITY OF WASHINGTON 1972 A B S T R A C T Porter M. Kier. Uppe r Miocene Echinoids from the Yorktown Formation of Virginia and The i r Environmental Significance. Smithsonian Contributions to Paleobiology, number 13, 41 pages, 7 figures, 10 plates. 1972.?Five echinoid species are described from the upper Miocene par t of the Yorktown Formation of Virginia: Echinocardium orthonotum ( C o n r a d ) , Arbacia imporcera ( C o n r a d ) , Psammechinus philanthropus ( C o n r a d ) , Mellita aclinensis Kier, and Spatangus glenni Cooke. T h e assemblage probably lived in shallow, warm-temperate waters, E. orthonotum deeply buried near shore, S. glenni shallowly buried offshore, and M. aclinensis with its test just covered near shore. Arbacia improcera and P. philan? thropus presumably lived together intertidally and near shore, P. philanthropus living in holes in the indurated sediments or on the sand with its test covered with debris, whereas A. improcera probably was easily visible with nothing covering its test. Specimens formerly referred to E. orthonotum from the middle Miocene Chop- tank Formation from Maryland are referred to E. marylandiense, new species. Echinocardium gothicum (Ravene l ) , from the Bear Bluff Formation of South Carolina, is considered a junior subjective synonym of E. orthonotum. Official publication date is handstamped in a limited number of initial copies and is recorded in the Institution's annual report, Smithsonian Year. UNITED STATES GOVERNMENT PRINTING OFFICE For sale by the Superintendent of Documents, U.S. Government Printing Office Washington, D.C. 20402 - Price 55 cents (paper cover) Porter M. Kier Upper Miocene Echinoids from the Yorktown Formation of Virginia and Their Environmental Significance Introduction Although three species of echinoids have been de? scribed from the Yorktown Formation (upper Miocene lower Pliocene), these species were based on few specimens. Now that a large collection of echinoids from this formation has been studied, these species are redescribed herein and two additional species reported. This collection was made by Mr. Warren Blow, presently with the Paleontology and Stratigraphy Branch of the United States Geological Survey, who started collecting fossils from the York- town Formation eight years ago. Since that time, he has accumulated hundreds of echinoids, which he has presented to the National Museum of Natural His? tory, Smithsonian Institution (under the catalog num? bers of the United States National Museum: USNM). Many of these specimens were collected in fragments that Blow painstakingly reassembled. He collected four large slabs containing hundreds of echinoids pre? served within the sediment. Some of the specimens are extraordinarily well preserved with their apical sys? tems, spines, pedicellariae, anal plates, and lanterns still intact, and many have their original color. Speci? mens of one of the species have food particles or fecal Porter M. Kier, Department of Paleobiology, National Mu? seum of Natural History, Smithsonian Institution, Washing? ton, D.C. 20560. pellets still preserved in their tests. The study of this large collection permits a more definitive description of the species, resulting in a better understanding of their relationship to other species, and some interpre? tation of the environment in which they lived. The echinoid assemblage consists of: Echinocard? ium orthonotum (Conrad), Arbacia improcera (Con? rad), Psammechinus philanthropus (Conrad), Mel- lita aclinensis Kier, and Spatangus glenni Cooke. The first three species had been described previously from the Yorktown, but Echinocardium orthonotum was thought to include also specimens from the middle Miocene Choptank Formation from Maryland. These Maryland specimens are distinct and are referred to a new species described herein, Echinocardium mary- landiense. Four specimens of Mellita aclinensis are present, a species known heretofore only from the upper Miocene Tamiami Formation in Florida, and one specimen of Spatangus glenni, previously known only from upper Miocene or Pliocene deposits in South Carolina. Acknowledgments I thank Warren C. Blow for collecting and preparing most of the echinoids described in this paper and for his great generosity in presenting the specimens to the National Museum of Natural History. These echin- 1 S M I T H S O N I A N CONTRIBUTIONS TO PALEOBIOLOGY oids are some of the best-preserved fossil echinoids known, and their availability to science is due to his patient and preservering efforts. I am much indebted to J . E. Hazel of the United States Geological Survey, ( U S G S ) , who not only critically read the manuscript but also provided much of the information on the paleoclimatology of the Miocene of the region of the deposition of the Yorktown Formation and provided a comparison with the present climate of this area. Druid Wilson, also of the United States Geological Survey, collected many of the South Carolina speci? mens discussed in this paper and provided much in? formation on the stratigraphy of the beds from where they were collected. Blake W. Blackwelder, who is now studying the mollusks of the Bear Bluff Forma? tion, gave me his preliminary opinions on the age of this formation, and Richard E. Grant of the United States Geological Survey and David L. Pawson of the National Museum of Natura l History, Smithsonian Institution, reviewed the paper and made many pertinent suggestions. Lauck W. Ward provided valuable information on the relative age of the Vir? ginia localities. Thomas F. Phelan not only did the photography but also made the remarkable reconstruction of a dis? associated lantern of Psammechinus philanthropus figured on Plate 2 : figures 7, 8. Larry Isham, the scientific illustrator for the Depar tment of Paleobi? ology in the National Museum of Natura l History made the locality m a p and the superb reconstruction of the living habits of the Miocene echinoids on Figure 2. Horace Richards of the Academy of Natura l Sci? ences of Philadelphia lent specimens and searched for one of Conrad 's lost types, and Norman F. Newell provided locality data on one of the type-specimens in the American Museum of Natura l History ( A M N H ) . Stratigraphy T h e Yorktown Formation consists of fossiliferous, silty sands and sandy silts that crop out in southeastern Virginia and northeastern North Carolina, from the Rappahannock River in Virginia to the Neuse River in North Carolina. Most of the echinoids (represent? ing all the species) came from four localities, two of which have been dated by Hazel (1971a) . H e has just completed a biostratigraphic study of the ostracodes of the Yorktown, identifying 230 species and dividing the formation into three ostracode assemblage zones. T h e oldest zone, the Ptery go cythereis inexpectata Zone, and the middle Orionina vaughani Zone are considered' by Hazel to be late Miocene, and his up ? permost zone, the Puriana mesacostalis Zone, is con? sidered by him to be early Pliocene. T h e two echinoid localities, and probably the third and fourth at and near Mogart 's Beach, belong to Hazel 's Orionina vaughani Zone. This zone approximates Mansfield's (in Gardner , 1943) Turritella alticostata Zone, which Mansfield also considered to be late Miocene. Gibson (1967) likewise considers these beds as late Miocene. Warren C. Blow and Lauck W. Ward (personal communication, 1971) have made extensive collec? tions from these four localities and believe that , on the basis of the molluscan assemblages, they are roughly equivalent in age. Paleoenvironment CLIMATE.?Haze l (1971b) has also made a paleo- climatological study of the Yorktown Formation based on the ostracodes, and he concludes tha t the climate was more equable and warmer in the winter during the deposition of the Yorktown than it now is in the same region. Fifty-nine of the ostracode species found in the Yorktown are still living, and, as a result, Hazel is able to make well-substantiated estimates of the Yorktown climate. T h e Orionina vaughani Zone (where the echinoids occur) con? tains 40 surviving ostracode species. These ostracodes indicate that bottom temperatures averaged no cooler than 12.5?C in the coldest month and that the sum? mer maximum averaged below 2 0 ? C in the older part of the zone and less than 25 ? C in the middle and younger parts. Hazel believes that , dur ing deposition of the Orionina vaughani Zone, the yearly range in bottom temperature was only about 5 ? C to 10?C. This equable thermal regime is markedly different from that of any province and concomitant climate zone now extant along the Atlantic coast of the United States. T h e temperatures varied from about 12?C to 15?C in the winter to about 17.5?C to 20?C and finally to 20?C to 25?C in summer. This part icular thermal regime, Hazel believes, is indica? tive of warm-temperate conditions analogous to tha t represented by the Lusitanian Province of western Europe. A warm-temperate zone is not present along the Atlantic coast of Nor th America today (Dana, NUMBER 13 1853; Hall, 1964; Hazel, 1970). The offshore climate of the study region today is mild-temperate with the bottom temperature in inner sublittoral waters aver? aging about 5?C to 7.5?C in the coldest month and about 20?C to 25?C in the warmest month. The Yorktown echinoids, like the ostracodes, indi? cate a past climate warmer than now. Psammechinus philanthropus (Conrad) is very similar to the living P. miliaris (Miiller), so much so that small specimens of both species are almost indistinguishable. This liv? ing species occurs in cold-temperate to subtropical waters off European coasts from Trondheimfjord to the northwestern coast of Africa. Spatangus glenni Cooke is very similar to S. purpureus Miiller, also oc? curring in the same waters and in the Mediterranean. Mellita aclinensis Kier has been found in the Tami- ami Formation of Florida, probably deposited in sub? tropical waters; it is very similar to M. quinquies- perforata (Leske), now living in mild-temperate to tropical waters along the eastern coast of America from Nantucket to the Brazilian coast as far south as Santos. Mellita quinquiesperforata, primarily a tropi? cal-subtropical species, is scarce north of Cape Hat- teras. Arbacia improcera (Conrad) resembles A. punctulata (Lamarck), which lives in mild-temperate to tropical waters from Cape Cod through the Carib? bean and in Yucatan. Echinocardium orthonotum (Conrad) is not useful in evaluating past climate be? cause it resembles the living species E. cordatum (Pennant), which is virtually cosmopolitan in cold- temperate to tropical waters. Although some of these living species range into cooler waters, they all occur in subtropical regions, suggesting that the fossil echinoids they resemble lived in waters warmer than the mild-temperate waters now occurring off the coast of Viriginia. LIVING HABITS OF THE ECHINOIDS.?Most of the fossil species are very similar to species now living, and it is therefore possible to make well-substantiated assumptions on the living habits of these fossil echin? oids (see Figure 1). Arbacia improcera resembles A. punctulata (La? marck) living off the coast of Virginia. Presumably its living habits were similar to A. punctulata, which is found in its greatest numbers in shallow water 5 to 20 feet deep on rock or sand bottoms. This species according to Sharp and Gray (1962) and Kier and Grant (1965:17-18) is highly variable in its response to light, with some specimens remaining exposed to the full light of the sun and others seeking cover. Like A. punctulata, A. improcera probably lacked sucking disks on its adapical tubefeet and, therefore, was un? able to cover itself with debris. It presumably did not excavate holes in the rocky shores (no living Arbacia can). It has many porepairs on the adoral surface, and the large number of tubefeet extending from them would have made it possible for the animal to cling tenaciously to intertidal rocks that were exposed to wave action. These tubefeet had sucking disks, as indicated by the presence of a ridge between the pores of each porepair, for the attachment of muscles used to retract these prehensile tubefeet (Nichols, 1959a: 421; Chesher, 1968:17). Psammechinus philanthropus (Conrad) is strikingly similar to P. miliaris, and presumably it lived sim? ilarly in the littoral zone on rock or sand. It probably lived with Arbacia improcera, but it doubtless had sucking disks on its adapical tubefeet (they occur in P. miliaris), and, therefore, it may have had its test covered with algae and other foreign debris. Some individuals might have lived in holes burrowed into rock. Mellita aclinensis Kier is very like M. quinquiesper- / or at a (Leske) and presumably lived like it in shal? low water on a firm sand bottom. It would have been slightly buried, with not more than 20 to 30 mm of sediment on top of its test. Spatangus glenni is very similar to S. purpureus Miiller now living off the British Isles. Owing to the research of Nichols (1959a, 1959b, 1962), the living habits of this British species are well known. Spa? tangus glenni shares enough morphological features with S. purpureus to suggest that the fossil species lived like the Recent ones in shallow water, that is, offshore, burrowing approximately 2 cm into coarse sediment. Spatangus glenni has a bilobed subanal fasciole that is also present in S. purpureus, where it is used to maintain, in the sediment, two horizontal tubes lying parallel to each other and originating at each lobe of the fasciole. These tubes provide a drain- away for the waste products of respiration and defe? cation, which are transported to the posterior region of the echinoid by cilia. This current is greatly en? hanced and directed away from the test down the twin tubes by the densely ciliated small spines in the subanal fasciole. According to Nichols (1962:116), two tubes are necessary in S. purpureus because the echinoid lives in coarse sediment that is subject to SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY caving. T h e extra tube provides oxygen for the echinoid while it reopens a blocked tube. Long pre? hensile tubefeet found within the fasciole are used to excavate these tubes and to plaster their sides with mucus. T h e presence of large pores within the fas? ciole of S. glenni indicates tha t it also had these tubefeet. Commonly, spatangoids that burrow deeply in the substrate have a tube extending from the dorsal sur? face of the test to the top of the substrate. This tube provides a channel for water currents that are di? rected over the respiratory tubefeet of the petals. Spatangus purpureus (and S. glenni as indicated by the presence of large tubercles) has large spines on the adapical surface that help keep an opening to the surface when the echinoid is burrowed only slightly, but both species lack the structures necessary to main? tain a long tube. T h e porepairs in the anterior am? bulacrum of S. glenni are very small and lack any protuberances between the pores of a pair for the at? tachment of the muscles that are used to retract the long tubefeet necessary to excavate and maintain a tube; however, S. purpureus, in spite of the lack of a dorsal tube, it able to burrow because it lives in coarse substratum that has large interstices through which the currents can pass. Because the subanal fasciole in S. glenni is less strongly bilobed than in S. purpureus and there are fewer adapical tubercles, resulting in fewer large spines, S. glenni was probably able to burrow slightly less deeply than S. purpureus. Although Echinocardium orthonotum is also a spatangoid, it probably lived somewhat differently than Spatangus glenni. I t is very similar and has all the functional morphological features of the living E. cordatum ( P e n n a n t ) . Considerable research has been done on this living species; Nichols (1959a) in par? ticular has studied its living habits and compared them to S. purpureus. Although S. purpureus (and presumably S. glenni) burrows only to a level 2 cm beneath the surface of the substratum, E. cordatum burrows much deeper, to a depth of 10 to 20 cm. Ac? cording to Nichols, E. cordatum, because it commonly lives near shore on sandy beaches, must burrow deeply in order to avoid being left high and dry by the receding tide, whereas S. purpureus lives far enough offshore to be unaffected by tides. Echino? cardium cordatum builds and maintains a long breathing tube to the surface and has an inner fasci? ole on its dorsal surface in order to draw water down the funnel and across the respiratory petals. Likewise, E. orthonotum has an inner fasciole and presumably also has constructed a funnel. Because Echinocardium cordatum and E. orthonotum do not have bilobed subanal fascioles as do S. purpureus and S. glenni, they could construct only one sanitary tube for the removal of waste products instead of two as in the two species of Spatangus. Finally, al though E. cordatum lives on beaches near shore, it is not found in areas subject to heavy wave or current action. In summary (Figure 1 ) , Arbacia improcera and Psammechinus philanthropus would be living in the intertidal and littoral zones on rock or sand, P. phil? anthropus in holes in the rock or covered with debris, but A. improcera would be easily visible with its test uncovered. Living also near shore, deeply buried in the sand, would be Echinocardium orthonotum with a long funnel extending above its test to the surface of the substratum and a single tube extending posteriorly. T h e sand dollar, Mellita aclinensis, would be found near shore at a depth of 3 to 15 meters with its test slightly buried. Finally, farther offshore Spa? tangus glenni would be burrowed only a few centi? meters in the sand, with no funnel to the surface, but with its adapical spines holding open a passage to the respiratory tubefeet, and with two tubes extend? ing posteriorly. CONDITION OF DEPOSITION.?These echinoids are remarkably well preserved. Most of the regulars still have the parts of their lanterns intact inside their tests, and many pedicellariae are present with the specimens. Many specimens of Psammechinus phil? anthropus still have food or fecal pellets preserved in their tests (Plate 2: figure 2 ) . Because lanterns and pedicellariae are lost after the death of the animal, it must be assumed that the echinoids were covered with sediment at, or soon after, dea th ; otherwise, currents or the action of pred? ators would have broken up the tests and disas? sociated the lanterns and pedicellariae. Echinoids, however, do not occur in the sediment in their living positions. Large slabs have been collected that con? tain many specimens of P. philanthropus and E. or? thonotum (Plate 9 : figure 3 ) , and these specimens are jumbled with the tests in haphazard position. I t is, therefore, probable that these echinoids, while alive, were caught up by storm currents and waves and covered by sediment. Possibly this deposition took NUMBER 13 Psammechinus Mellita FIGURE 1.?The probable life positions of the five echinoid species from the Yorktown Forma? tion: These fossil species are so like species now living that it is possible to make well- substantiated predictions of their living habits. The figures of Spatangus and Echinocardium are based on the figures and descriptions of Nichols (1959a) for Spatangus purpureus Miiller and Echinocardium cordatum (Pennant). place near shore in shallow water perhaps 5 to 20 meters deep because the most common specimens are P. philanthropus and E. orthonotum, both species that are assumed to have lived near shore. POST-DEPOSITION CONDITIONS.?Many of these echinoids still have their color, and many of the tests do not have the interstices of their plates filled with secondary calcite. This unusual preservation indicates that little leaching and recrystallization has occurred in the sediments since their deposition. The presence of organic material still preserved in the food or fecal pellets indicates restricted bacterial activity. Relation to Other Fossil Echinoid Faunas The Yorktown echinoid assemblage has affinities with echinoid assemblages from the upper Miocene or Plio? cene of South Carolina and Florida. Echinocardium orthonotum of the Yorktown is considered herein a senior synonym of E. gothicum (Ravenel), described by Ravenel (1848) and Tuomey and Holmes (1855) from Grove Plantation on the Cooper River in South Carolina. This species also occurs in spoil deposits in the Intracoastal Waterway canal, which Druid Wilson (personal communication, 1970) considers, on the S M I T H S O N I A N CONTRIBUTIONS TO PALEOBIOLOGY basis of the mollusks, to be late Miocene and equiva? lent in age to the beds at Grove Plantation. Spatangus glenni, which is found in the Yorktown, is found also in these canal deposits. B. W. Blackwelder (personal communication, 1971), who is currently studying the mollusks, believes that these deposits are from the Bear Bluff Formation of Dubar (1969), which Dubar considers probably to be Pliocene. The Yorktown species Arbacia improcera is very similar and may be conspecific with A. sloani (Clark) from upper Miocene beds in the Duplin Marl at Bos- tick Landing, Pee Dee River, South Carolina. The Yorktown echinoid assemblage is also some? what similar to the echinoids of the upper Miocene Tamiami Formation of Florida. Mellita aclinensis is found in both assemblages, and the Yorktown A. im? procera is probably synonymous with the Tamiami A. crenulata Kier. Localities Most of the echinoids collected by Warren C. Blow came from the four localities in Virginia indicated in Figure 2 and described by him below. Hazel (1971a) 76?30' 76?I5 76?30' 76?I5' FIGURE 2. -The localities ( 1 - 4 ) where Warren C. Blow collected most of his echinoids from the Yorktown Formation. NUMBER 13 has studied the ostracodes from localities 3 and 4 and considers them to be from his Orionina vaughani Zone of the Yorktown Formation. Both Warren Blow and Lauck W. Ward, who also has collected extensively in the Yorktown, consider all four localities to be in this zone. LOCALITY 1 (USGS locality 25113) .?Future Homemakers of America-Future Farmers of Amer? ica [FHA-FFA] girls camp, approximately 3.6 miles NNE of Smithfield, south bank of James River, Isle of Wight County (see Virginia quadrangle 7.5' Mulberry Island, 1965). A block of cemented coquina containing numerous specimens of Echinocardium orthonotum (Conrad) was found adjacent to 23-foot- high face of outcropping Yorktown Formation at a point between 100 and 300 yards ESE of the present camp staircase. This material undoubtedly was de? rived from a distinctive 1- to 2-foot-thick ledge of cemented coquina occurring approximately 21 feet above the beach, which terminates the Yorktown Formation along Mogarts Beach. Collectors: Edward E. Bottoms and Warren C. Blow. Species: Echino? cardium orthonotum (Conrad), Psammechinus phil? anthropus (Conrad). LOCALITY 2.?Hunt club, south bank of James River, approximately 0.65 miles ESE of FHA-FFA girls camp at Mogarts Beach, Isle of Wight County (see Virginia quadrangle 7.5', Mulberry Island, 1965). This locality begins at the mouth of a deep ravine (distinguishable by its two small lakes), approxi? mately 100 feet WNW of the club house, and con? tinues ESE along the base of the bluffs overlooking the James River for approximately 200 yards. USGS 25114: Echinoids collected in situ in very macrofossiliferous, aqua, silty sand between the beach level and not more than 6 feet above the beach. Col? lectors: Warren C. Blow and others from 1963-1967. Species: Arbacia improcera (Conrad), Psammechinus philanthropus (Conrad), and Spatangus glenni Cooke. USGS 25115: In situ in sparsely macrofossilifer? ous, aqua, silty sand, approximately 6.5 to 8.5 feet above beach level. Collectors: R. H. Bailey, B. Dyer, R. V. Blow, and W. C Blow, 1965-1967. Species: Echinocardium orthonotum (Conrad). USGS 25116: 100 yards ESE of ravine between 0 and 0.5 feet above beach level. Collectors: Richard H. Bailey and W. C. Blow, 1967. Species: Spatangus glenni Cooke. USGS 25117: 158 yards ESE of ravine, 3 feet above beach level. Collector: W. C. Blow, 1968. Species: Spatangus glenni Cooke. LOCALITY 3.?Rice's Pit just N of the intersection of the Fox Hill Road (Virginia Highway 167), and the Harris Creek Road, Hampton City (see Virginia quadrangle 7.5', Hampton, 1965). USGS 24810: Western wall of pit approximately 350 feet NNW of the upper SW corner; approxi? mately 36 feet below local topography. Collectors: B. Dyer and W. C. Blow, 1969. Species: Echinocardium orthonotum (Conrad). USGS 25122, 25123: In situ from very macro? fossiliferous, buff, silty sand of the upper 6 to 7 feet of the Yorktown Formation (approximately 6 to 12 feet below local topography along central portion of the southern half of the western wall of pit). Col? lectors: W. C. Blow, Barbara Grocys. Species: Ar? bacia improcera (Conrad), Psammechinus philan? thropus (Conrad), Mellita aclinensis Kier, and Echinocardium orthonotum (Conrad). LOCALITY 4.?Lone Star Cement pit, 0.5 mile N of Chuckatuck, Nansemond County (see Virginia quad? rangle 7.5', Chuckatuck and Beens Church, 1965). USGS 24493: Western wall of pit approximately 300 yards from NW corner 1 to 3 feet above lake level, 24 feet below top of section. Collectors: W. C. Blow and R. V. Blow, 1967. Species: Psammechinus philanthropus, Echinocardium orthonotum (Conrad). USGS 24626: Western wall of pit, eastern face of cut 33. Collector: W. C. Blow, 1968. Species: Echinocardium orthonotum (Conrad). USGS 25119: Float of maroon chunks of semi- hardened, silty sand collected between the south- central and northernmost extent of the western wall of the pit. Collector: W. C. Blow, 1964. Species: Echinocardium orthonotum (Conrad). USGS 25120: Approximately 3 to 5 feet above the lake level (22 to 24 feet below local topography) along the central portion of the western wall of the pit. Collectors: W. C. Blow and E. E. Bottoms, 1964. Species: Echinocardium orthonotum (Con? rad). USGS 25121: Material collected in situ (gen? erally from a fine blue-gray hash) along the south- central and northern half of the western wall of pit, approximately 0 to 3 feet above lake level or 24 to 27 feet below local topography. Collectors: W. C. Blow and others, 1963?present. Species: Arbacia improcera 8 S M I T H S O N I A N CONTRIBUTIONS TO PALEOBIOLOGY ( C o n r a d ) , Psammechinus philanthropus ( C o n r a d ) , Mellita aclinensis Kier, and Echinocardium ortho? notum ( C o n r a d ) . Arbacia improcera ( C o n r a d ) P L A T E I Echinus improcerus Conrad, 1843a: 310. Psammechinus improcerus.?Stefanini, 1912:705. Coelopleurus improcerus.?Clark and Twitchell, 1915:180, pi. 84 : figs. 4a -c . Arbacia improcera.?Cooke, 1941:11 , pi. 1: figs. 7-9.-? Cooke, 1959:20, pi. 4 : figs. 1-3.?Kier, 1963: pi. 1: fig. 6. For many years only one specimen has (in addition to the missing holotype) , tional specimens are now extant. T h e five of these specimens are as follows: Diameter Height Diameter of peristome Greatest width of ambulacrum Height of interambu? lacral plate a t ambitus Width of interambu? lacral plate at ambitus Number of interambu? lacral plates in single row Number of porepairs in single poriferous zone Greatest width of apical system 6.7 3.5 1.3 1.3 18.5 6.9 9.5 3.6 1.4 3.8 32. 16 b e e n a v a i l a b l e b u t five a d d i - d i m e n s i o n s of 7 29.5 36.5 0 13.2 15.0 13.3 16.4 .1 6.2 7.4 .7 2.5 2.7 .7 6.0 6.7 8.9 12.0 14.0 13.0 15.0 - 37.0 52.0 51.0 57.0 .3 - 10.0 2.8 5.9 C O M P A R I S O N W I T H OTHER SPECIES.?Arbacia im? procera may be conspecific with A. sloani (Clark) from upper Miocene beds in the Duplin Marl at Bos- tick Landing, Pee Dee River, South Carolina. Only one specimen of A. sloani, the holotype, is known, and it is similar in most of its dimensions to A. improcera, having the diameter of its peristome 50 percent of the diameter of its test ( D ) , width of an ambulacrum at the ambitus 20 percent D, height of an interambula? cral plate at the ambitus 2.0 percent D, width 5.5 per? cent D , greatest width of the apical system 7.3 per? cent D, number of interambulacral plates in single row 14, and number of porepairs in a single pori? ferous zone 55. Cooke (1959:21) considered that the two species were very similar and might be synony? mous, but he also noted that they differed, with A. improcera having a lower test and a more rugose ornamentat ion. T h e only specimen of A. improcera that he had available is lower than the holotype of A. sloani, which has a height 41 percent of its di? ameter as opposed to 53 percent in A. sloani. O n e of the recently acquired specimens of A. improcera, how? ever, has a height equal to 49 percent D . This differ? ence from A. sloani is slight and, until more speci? mens have been found, its significance cannot be evaluated. Although Cooke considered the ornamen? tation in the naked adapical interambulacral areas different in the two species, I can see no distinguish? ing dissimilarities (compare Plate 1: figure 5 to Plate 2 : figure 1) . When more specimens are available, it should be possible to determine whether these two species are separate or synonymous, but , until then, it seems best not to a t tempt to make this decision. Arbacia improcera differs from A. rivuli Cooke from the late Miocene (?) in South Carolina in hav? ing its adapical interambulacra more naked. In A. improcera only a single row of tubercles occurs on each column of interambulacral plates down to a point near the ambitus, whereas in A. rivuli a second row extends almost to the apical system. Arbacia im? procera differs from A. waccamaw Cooke also from the late Miocene (?) in South Carol ina in having wider and lower adapical interambulacral plates. Arbacia improcera may be conspecific with A. cren? ulata Kier from the late Miocene T a m i a m i Forma? tion of Florida. When I described A. crenulata, only one specimen of A. improcera was available for com? parison. O n this specimen the ornamentat ion was decidedly different from that of A. crenulata. In A. crenulata the ornamentat ion consists of fine crenula? tions, whereas in the single specimen of A. improcera they were granules; however, on the new specimens of A. improcera these granules commonly are joined to? gether into crenulations. These two species are prob? ably synonymous, but more specimens are needed in order to know the extent of the variation of this orna? mentation within one population. Arbacia improcera is similar in the following di? mensions to A. punctulata ( L a m a r c k ) , now living off the coast of Virginia: height, diameter of peristome, width of ambulacrum, width of apical system, num? ber of interambulacral plate, and number of pore? pairs. It differs in having only two large tubercles on each interambulacral plate at the ambitus, whereas there are three in specimens of the same size in A. NUMBER 13 punctulata. Moreover, in A. improcera the naked zones in the interambulacra extend farther adorally. LOCALITIES.?Blow's localities 2 (USGS 25114), 3 (USGS 25122), 4 (USGS 25121), J. T. Williams' marl pit, Smith County, 0.5 mile below the dam at Suffolk waterworks, collected by M. W. Twitchell. USGS 24817, exposed face of high bluff (approxi? mately 500 feet NW of cemetery), south bank of Nansemond River, 0.5 mile downstream from Vir? ginia Highways 10 and 32 bridge, Suffolk City, Vir? ginia, from fossiliferous, bluff, silty sand, 3 feet above beach level, collected by Brian Dyer and Warren C. Blow; Bank of Nansemond River rising from tidal flats at Suffolk, Virginia, collected by W. C. Mans? field. The holotype came from the James River near Smithfield. TYPE-SPECIMENS.?Location of holotype not known; figured specimens: USNM 166487, 17449. Psammechinus philanthropus (Conrad) FIGURE 3 ; P L A T E 2: FIGURES 2 - 8 ; P L A T E 3 ; PLATE 4 : FIGURES 1, 3 ; P L A T E 5 : FIGURES 1, 3 Echinus philanthropus Conrad, 1843a: 310.?Conrad, 1846: 220. Psammechinus philanthropus.?Meek, 1864:2.?Stefanini, 1912:705.?Clarke and Twitchell, 1915 :181 , pi. 84 : figs. Ga-c.~Cooke, 1941:16.?Cooke, 1959: 16, pi. 3 : figs. 1, 2. Echinus ruffini Forbes in Lyell, 1845a:426, figs, la?d [and Lyell, 1845b:560, 2 figs.].?Desor, 1858:121.?Emmons, 1858:306, figs. 239a-d.?Stefanini, 1912:705. MATERIAL.?Hundreds of extremely well-preserved specimens have been collected by Robert V. Blow and Warren C. Blow. Many of them are on two large slabs. These specimens are arranged haphazardly, in? dicating that they are not in life position, but the pres? ence of lanterns, apical systems, and spines on many of the specimens indicates that they were buried soon after death. Because in the past only a few specimens have been available, this species has never been ade? quately described. The description and statistics be? low are based on 30 specimens from the same locality (Blow locality 4, USGS 24493). SHAPE AND SIZE.?Diameter (D) varying from 14.0 to 39.3 mm, mean 21.1 (SD 5.4, CV 25.6, N-30), height varying from 46 to 61 percent D (SD 3.5, CV 6.6, N-30) , height not varying with size of specimen; marginal outline of test circular to slightly subpentag? onal with apices at ambulacra; test slightly depressed at peristome. APICAL SYSTEM.?All oculars exsert (Figure 3) , genital 2 larger than other genital plates, one or two tubercles on each genital, periproct oblique with long axis passing through genital 2 and ocular V. AMBULACRA.?Narrow, width 23-26 percent D, mean 24.8 (SD 0.7, CV 3.1, N - 3 0 ) ; plates trigemi? nate, 15 compound plates in single poriferous zone of smallest specimen 14.0 mm in diameter, 20 in speci? men 21 mm in diameter, 26 in largest 39.3 mm in diameter, with a mean 18.8 plates (SD 2.5, CV 13.3, N-30). INTERAMBULACRA.?Plates low, at ambitus 1.2 compound ambulacral plates for each interambulacral plate; 13 plates in single column of smallest specimen 14.0 mm in diameter, 16 in specimen 21.0 mm, 22 in largest 39.3 mm in diameter, mean 15.6 (SD 2.1, CV 13.2, N-30). PERISTOME.?Pentagonal with apices in interambu? lacra, large, diameter varying from 36 to 52 percent D, mean 44.8 (SD 3.1, CV 6.8, N-30) . TUBERCULATION.?Ambulacra; two vertical rows of primary, imperforate, noncrenulate tubercles in each FIGURE 3.?Psammechinus philanthropus ( C o n r a d ) : Apical system of U S N M 174452 from Blow's locality 4 (X 10) . (A photograph of this region is on Plate 3 : figure 4.) 10 S M I T H S O N I A N CONTRIBUTIONS TO PALEOBIOLOGY area (Plate 3 : figure 2) with two inner rows almost equally developed at ambitus in larger specimens (Plate 5 : figure 1) ; in small specimens only two rows at ambitus, inner tubercles slightly developed (Plate 4 : figure 1 ) ; pr imary tubercles occupy most of height of plate. INTERAMBULACRA.?In large specimens at ambitus 5-8 imperforate, noncrenulate tubercles arranged in 1?2 irregular horizontal rows (Plate 5 : figure 3 ) , in small specimens two vertical rows of tubercles at am? bitus with a single large tubercle in middle of each plate with a smaller tubercle on either side (Plate 4 : figure 3) ; as echinoid increases in size, these smaller tubercles increase in size relative to the central tuber? cle until they are of equal size and the two vertical rows are no longer distinguishable. PEDICELLARIAE.-?Tridentate and ophicephalous pedicallariae found (Plate 2 : figures 3 - 6 ) . LANTERN.-?Preserved on most specimens (Plate 2 : figures 7, 8 ) . COLOR.?Preserved on all specimens, dark brown band running down center of ambulacra ; lighter or a dark brown band running down center of interam? bulacra. C O M P A R I S O N W I T H OTHER S P E C I E S . ? T h i s species is very similar and obviously closely related to Psam? mechinus miliaris (Mii l ler) , now living along the European coasts from the Trondheimfjord and Ice? land to the northwestern coast of Africa. I measured the diameter, height, diameter of peristome, width of ambulacrum, and number of interambulacral and am? bulacral plates of a collection of both species and tested them statistically to see whether or not they differed significantly. No significant difference is pres? ent in the height of the test, width of ambulacra, and number of ambulacral plates, but the two species dif? fer in the diameter of their peristomes and the num? ber of interambulacral plates. Psammechinus philan? thropus has a larger peristome and more interambu? lacral plates than P miliaris, with a peristome whose diameter is 44.8 percent the diameter of the test as opposed to 39.7 percent in P. miliaris, and an average of 15.6 interambulacral plates in each half-column as opposed to 16.1 in the living species. A student t test was run on the difference of these means, and they are significant to .001. T h e tuberculation of P. philanthropus (Plate 5 : figures 1, 3) differs in large specimens in having at least five tubercles on each interambulacral plate at the ambitus of approximately equal size, whereas in P. miliaris one tubercle is larger than all the others (Plate 5 : figures 2, 4 ) . This difference, however, is not present in small specimens. Specimens of the two species 14 'mm in diameter are indistinguishable in their tuberculation. T h e small P. philanthropus (Plate 4 : figures 1, 3) has one larger tubercle on each plate just as in P. miliaris (Plate 4 : figures 2, 4) ; however, the difference in size of the peristome is persistent in all the specimens regardless of their size. T h e test of P. philanthropus has brown bands run? ning down the ambulacra and interambulacra, whereas in P. miliaris the test is generally green with dark green bands. This difference is probably signifi? cant, but perhaps the present color of the fossil speci? mens is not the same as it was originally. Most previous authors have considered Forbes' Echinus ruffini a synonym of P. philanthropus. T h e type (probably in the Lyell Collection) came from a locality near Williamsburg, Virginia, where beds of Yorktown age are known to occur, and although Forbes' illustrations are not very clear, his specimen appears to be P. philanthropus. Cooke (1959:16) considered P exoletus McCrady a synonym of P. philanthropus; however, the type- specimen, which is lost, was only a small fragment, and it is not possible from the illustration of it to see enough specific characters to be able to determine whether or not it belongs to P. philanthropus. T h e fragment came from "Smith 's ," Goose Creek, South Carolina, from beds that Cooke (1941:16) considered to be the Pliocene Waccamaw Format ion bu t that later (1959:16) he thought were late Miocene (Du? plin M a r l ) . T Y P E - S P E C I M E N S . ? L o c a t i o n of holotype not known; it is not with Conrad 's other types at the Academy of Natura l Sciences of Philadelphia (H . G. Richards, 1971, personal communica t ion ) ; figured specimens: U S N M 562264, 559495 (formerly Johns Hopkins University T 1 0 0 1 ) , U S N M 174450, 1774451, 1774452, 174453, 174454. STRATIGRAPHIC POSITION AND GEOGRAPHIC LOCALI? T I E S . ? U p p e r Miocene Yorktown Formation. Vir? ginia: Holotype from James River near Smithfield; south side of James River at J. A. Mogarts ' residence 5 miles N of Smithfield, U S N M 166501, collected by W. M. Twitchel l ; Rock Wharf road on Days farm 1.5 miles N E of Smithfield, collected by W. M. Twitchel l ; Bluff W of Days Point from uppermost NUMBER 13 11 bed, James River, USGS 14065, collected by W. C. Mansfield and C. W. Mumm; Yorktown, USNM 373038, collected by Julia Gardner; 2.5 miles NW of Suffolk, about 0.5 mile below Calhoun Bridge (on right bank of creek), USGS 10198, collected by W. C. Mansfield. Warren C. Blow collected specimens from his localities 1-4 at USGS 25113, 25114, 25121, 25122. The holotype of Echinus ruffini Forbes came from Burwell's Mill near Williamsburg. All the above localities are probably in Hazel's Orionina vaughani Zone, according to Warren C. Blow (1971, personal communication). Mellita aclinensis Kier PLATE 6; PLATE 7: FIGURE 1 Mellita aclinensis Kier, 1963:40-45, figs. 36-41; pi. 15: figs. 1-3; tbls. 3,4. Leodia caroliniana Cooke [not Ravenel], 1959:47. Three specimens are referred to this species, which previously was known from the late Miocene Tamiami Formation in Florida. The Virginia specimens differ only in their larger size. The Florida specimens are less than 73 mm long, whereas the Virginia speci? mens are 128-147 mm long. Because echinoids com? monly occur together in groups of individuals of the same age, the Florida population probably was only more juvenile. The figured Florida specimens have wider lunules, but this is because they are smaller, and in smaller specimens the lunules are wider rela? tive to their length. A larger fragment from Florida is figured herein (Plate 6: figure 3) showing the nar? rower lunules typical of the larger specimens from Virginia. The dimensions of the Virginia specimens are as follows (in mm) : Length Width Height Length of petals: I I I II I Width of petals: I I I II I Number of porepairs I I I USNM 174457 128 135 16.0 - - - - 14.2 - USNM 174458 130 129 12.3 33.0 30.5 38.2 13.1 12.7 12.5 in single poriferous zone of - 88 USNM 174468 147 148 - - - - - - - petals: - USNM USNM USNM 174457 174458 174468 II 90 I - 107 - Width (maximum) of interporiferous zones: III - II 4.8 4.7 I 4.1 Distance of apical system from anterior margin: 62.2 60.2 Length of posterior lunule: 26.4 32.2 Distance of peristome from anterior margin: 64.6 62.0 71.0 Considerable evidence suggests that M. aclinensis is the ancestor of M. quinquiesperforata (Leske). The species differ mainly in that M. aclinensis has six lunules and M. quinquiesperforata commonly has five; however, as pointed out by Cerame-Vivas and Gray (1964), some specimens of M. quinquiesperfor? ata have been found with six lunules. Probably the species evolved from the form with six lunules in the late Miocene and Pliocene to a form commonly with five lunules in the Recent. (Two specimens were collected by Richard H. Bailey and Warren C. Blow from Hazel's (1971a) Puriana mesacostalis Zone of the Yorktown Formation at Colerain Landing, North Carolina, which Hazel considers to be early Pliocene.) Although there is a sand dollar, Leodia sexiesper- forata Leske, with six lunules living today off the southeastern United States, I do not believe that it is a descendent of M. aclinensis. The presence of six lu? nules in Leodia is not really the important morpho? logical character distinguishing Leodia from Mellita. Cerame-Vivas and Gray (1964) assumed this when they considered that L. sexiesperforata, the type- species of Leodia, should be referred to Mellita. Leodia sexiesperforata differs from M. quinquiesper? forata, the type-species of Mellita, in having its pos? terior lunule not extending far anteriorly between the posterior petals, in having its paired interambulacra separated from the basicoronal row by two pairs of ambulacral plates, in having its periproct slightly in? denting the basicoronal plate, and in having a short pair of post-basicoronal plates in the paired inter? ambulacra. Cooke (1959:47) tentatively referred a specimen from the Yorktown Formation at Days Point to L. caroliniana (Ravenel), but this specimen is M. aclin? ensis. The test of M. aclinensis is much flatter than 12 S M I T H S O N I A N CONTRIBUTIONS TO PALEOBIOLOGY this South Carolina species, which has a sharper mar? gin, and the lunules are commonly, but not always, narrower. Although the holotype of L. caroliniana has been lost, there are many specimens in the national col? lections from localities near the type-locality (Grove Plantation, Cooper River), from beds of the same age and slightly younger, that appear to be conspecific with Ravenel's holotype. These specimens are very variable in the shape of their petals and lunules, and they include specimens that have longer and curved posterior petals with narrow lunules, as in the speci? men figured by Ravenel (1842), and the straight pos? terior petals and short lunules of the specimen figured by Tuomey and Holmes (1857: pi. 1: figs. Aa-c). Leodia caroliniana should be referred to Mellita not Leodia because the periproct occurs partially within the basicoronal plate (Figure 4) , the posterior lunule extends far anteriorly between the posterior petals, the paired interambulacra are separated from the basicoronal row by one pair of ambulacral plates, the FIGURE 4.?Mellita caroliniana (Ravenel) : plate arrange? ment around the peristome of USNM 174465 from beds con? sidered to be probable late Miocene by Cooke (1959:80) (X 1.3). Dredged from the Intercoastal Waterway canal 1 to 1.5 miles SW of the bridge on US Highway 17 near Nixon's Crossroads about 15 miles NE of Myrtle Beach, Horry County, South Carolina (collected by L. C. Glenn). first pair of post-basicoronal plates in the paired in? terambulacra are elongate, and the lunules are formed by the closing of marginal notches and not by re? sorption of the test. It resembles Encope in its thick test and in the fact that a few specimens have five instead of four genital pores, but the presence of the periproct in the basicoronal plate separates it from Encope, in which the periproct is more posterior and occurs between post-basicoronal plates. Mellita caro? liniana in general appearance strongly resembles the living E. emarginata (Leske), suggesting that Mellita and Encope may have had a common ancestor. This resemblance caused Cooke (1942:20, pi. 3: figs. 14, 15; 1959:49) to refer these South Carolina speci? mens, together with a few from North Carolina, to E. emarginata. T Y P E - S P E C I M E N S . ? H o l o t y p e : USNM 648136; figured paratypes: USNM 648189-648193; figured specimens: USNM 174457, 174458. STRATIGRAPHIC POSITION AND LOCALITIES.?Florida: Tamiami Formation (barnacle-echinoid-oyster facies = Murdock Station Member of Hunter, 1968:442- 443), spoil banks from group of pits (sec. 29, T. 41 S, R. 23 E) about 1 mile SW of Acline, Charlotte County. Virginia: Yorktown Formation, Orionina vaugh? ani Zone, at Blow locality 3, USGS 25122; one from Blow's locality 4 (USGS 25121) ; and one specimen in a collection made by G. A. Cooper, J. Cooper, Druid Wilson, and H. B. Roberts; specimen referred to by Cooke (1959:47) probably from same zone at Days Point, James River, W of mouth of Pagan Creek, about 4 miles N of Smithfield, USGS 16920. North Carolina: Yorktown Formation, Puriana mesacostalis Zone; Colerain Landing, west bank of Chowan River, Bertie County, in a buff, silty sand approximately 100-150 yards S of the Colerain Beach Club, from layer containing Pectin eboreus Conrad, occurring approximately 7 feet above beach level, USGS 25118. Echinocardium orthonotum (Conrad) FIGURE 5; PLATE 8: FIGURES 3-7; PLATE 9 Spatangus orthonotus Conrad, 1843b: 327. Amphidetus orthonotus.?Tuomey and Holmes, 1855: pi. 2: figs. l-l