COSTELLACEST AJ A NEW SUBGENUS OF LIMA FROM THE CRETACEOUS OF THE GULF AND ATLANTIC COAST PROVINCE ERLE G. KAUFFMAN DEPARTMEN1' OF PALEOBIOLOGY UNITED STA1'ES NA1'IONAL MUSEUM

CONTENTS Page ABSTRACT ______------· ------8 9 I. INTRO DU CTI 0 N------______------______90 I I. A CKN 0 WLEDG MEN TS ______9 0 II I. BI 0 STRATIGRAPHY______9 0 IV. PHYLOGENY AND EVOLUTION______92 v. THE Eco LOGY 0 F GIANT LIMAS ______------92 VI. SYSTEMATIC DESCRIPTIONS ______------94 Genus Lima ______.______94 Subgenus Cost ellac es ta______------__ 9 5 Lima ( Costellacesta) riddleiJ n_ sp. ______95 Lima ( C ostellacesta) sayrei Stephenson______97 Lima ( C ostellac es ta) ins olit a Stephenson______99 VII. REFERENCES CITED ______99 TABLE 1: Measurements of Lima ( Costellacesta) riddleiJ n_ sp. ______96 TABLE 2: Measurements of Lima (Costellacesta) sayrei Stephenson ______97 TABLE 3: Measurements: Holotype of Lima (Costellacesta) insolita Stephenson____ 99 TEXT FIGURE 1: Stratigraphic occurrence and phylogeny of Costellacesta______91 TEXT FIGURE 2: Depth distribution of living giant Limas ------·------93 TEXT FIGURE 3: Cardinal area of Lima ( Costellacesta) riddleiJ n. sp. ______96 PLATE ! ______------101

ABSTRACT Sandstone Member, Mississippi) , L. (C.) The new subgenus C ostellacesta is pro­ sayrei Stephenson (Corsicana Marl, Texas), posed for large Maestrichtian species of and L. ( C.) insolita Stephenson (youngest: Lima with an opisthocline shell, large lunule upper Peedee Formation, North Carolina). and byssal gape, single (posterior) auricle, These represent a single line of descent with­ a resilifer situated posterior to the beak, and out significant break. Successively younger ornamentation consisting of fine costellae species demonstrate the following evolution­ intercalated between prominent, equally de­ ary trends: change in marginal outline, par­ veloped costae. Three species of C ostellacesta ticularly an increase in concavity of the are known and described: L. (C.) riddleiJ dorsoanterior margin; increase in the rela­ n. sp., the type species (oldest; Chiwapa tive size of the lunule and byssal gape; re-

EDITORIAL COMMITTEE FOR THIS PAPER: A. LEE McALESTER, Department of Geology, Yale University, New Haven, Connec­ ticut NORMAN F. SOHL, United States Geological Survey, Washington, D. C HAROLD E. VOKES, Department of Geology, Tulane University, New Orleans, Lou­ isiana 89 90 Tulane Studies in Geology Vol. 2

duction in the strength of the costae and L. ( C.) s ayrei Stephenson (Corsicana Marl, costellae; increase in number of costellae, and Navarro Group, Texas), and L. (C.) insolita decrease in number of costae per unit length; Stephenson (Upper Peedee Formation, Owl possible decrease in maximum adult size. Creek - Providence - Prairie Bluff equivalent, Costellacesta is possibly ancestral to pli­ North Carolina) . These species occur in cate Cenozoic Limas of the subgenus Plic­ successive stratigraphic units, without over­ acesta Vokes. The habitat of C ostellacesta lapping ranges, and probably represent a was probably analogous to that of living continuous evolutionary sequence. Known giant Limas ( Acesta, Plicacesta)-cold, deep species of Lima (Costellacesta) are exclusive­ waters of the continental slope and abyss, ly Maestrichtian in age, and occur only in rarely ranging into warm, shallow shelf en­ the upper half of the Exogyra costata zone vironments. This might explain the uncom­ (see Sohl, 1960, fig. 3). mon occurrence of C ostellacesta in shallow water Cretaceous deposits of the Atlantic II. ACKNOWLEDGMENTS and Gulf Coastal Plain Province. The sub­ Specimens of Lima (Costellacesta) riddlei, genus is known only from this area, and n. sp., the type species, were collected and probably represents an endemic group. donated to the Smithsonian Institution by Mr. W . C. Riddle of Memphis, Tennessee, I. INTRODUCTION as part of a program undertaken by the Mid­ Lima is a common element of Gulf and South Earth Science Club to provide the Atlantic Coast Cretaceous pelecypod assem­ Institution with research material. To Mr. blages. Individual species are widely dis­ Riddle and his associates I offer my sincere tributed and are represented in a variety of thanks. Drs. Harold E. Vokes of Tulane lithologies, rendering them potentially im­ University, A. Lee McAlester of Yale Uni­ portant in biostratigraphic zonation and cor­ versity, John Pojeta, Jr., and Norman F. relation. The American species of Lima are Sohl of the U. S. Geological Survey reviewed divisible into several distinct lineages, some the manuscript and offered valuable criti­ of which are treated as sections or sub­ cism. Dr. Sohl was also instrumental in lo­ genera by many workers. One of the most cating specimens and provided valuable as­ unusual Cretaceous lineages, and one which sistance in stratigraphic interpretation. appears to be endemic to the Gulf and At­ Drawings are by 1vt:r. L. B. Isham and Mr. la?tic Coast Province, is the group of giant L. R. Purnell, and photography by Mr. Jack Lima comprising a new subgenus, Costell­ Scott, all of the U. S. National Museum. acesta Kauffman. This lineage is possibly ancestral to the Cenozoic and Recent sub­ III. BIOSTRATIGRAPHY genus Plicacesta Vokes ( 1963). The three known species of Lima (Co s­ C ostellacesta is characterized by its large tellacesta) occur at widely scattered localities; size, inequilateral ( opisthocline) shell, single no two species have been found in the same (posterior) auricle, large lunule and byssal area. Nevertheless, the stratigraphic rela­ gape, slightly oblique resilifer situated pos­ tionships between the species are well estab­ terior to the beaks, and by the surface orna­ lished (fig. 1 ) . The oldest form, L. ( C..) rid­ mentation, which consists of narrow, sparsely dlei, n. sp., occurs exclusively in the Chiwapa fluted costae and numerous, delicate, inter­ Sandstone Member of the Ripley Formation calated costellae. Unlike many other Cre­ (Mellen, 1958; equivalent to the Keownville taceous Lima, species of L. (Costellacesta) Limestone Member of Sohl, 1960, p. 18, fig. are rare and geographically restricted; few 3) near Ingomar and Pontotoc, M-ississippi. complete specimens are known. Cenozoic The youngest member of the lineage, L. (C. ) and Recent giant Lima of the subgenera insolita Stephenson occurs only in the upper Plicacesta and Acesta are similarly restricted. part of the Peedee Formation at the New Three species of Costellacesta are known Rocky Point quarries, Pender County, North from Upper Cretaceous sediments of the Carolina (upper part of the Exogyra costata Atlantic and Gulf Coast Province. These zone, Late Maestrichtian: Stephenson, 1927, are, from oldest to youngest: Lima (Costell­ p. 13). Here it is associated with a fauna acesta) riddlei Kauffman, n. sp. ( Chiwapa of Owl Creek - Prairie Bluff - Providence age Member, Ripley Formation, Mississippi), (Stephenson, 1927, Sohl, oral communica- No. 3 C ostellacesta, a new sub genus of Lima 91

SPECIES OF TEXAS MISSISSIPPI NORTH CAROLINA LIMA PHYLOGENY (COST£LLllCEST.4

; '

t • .l

L. (C.) INSOLITA z <( CHIWAPA MBR. ;: KEOWNVILLE LS. ::z: u ii: 0.. ::> NACATOCH .... 0 SAND RIPLEY FM . "'w a:: <( "' ::IE 0 a:: a:: a:: L (C.) SAYRE! w <( 0.. > 0.. <( ::> z

L (C .l RIDDLE I

Figure 1-StratigTaphic occurrence, and phylogeny of Costellacesta, showing change in shape, decrease in overall size, variation in size of the lunule and byssal gape, and change in ornamentation on progressively younger species of the subgenus. Byssal gape black where definitely known, otherwise approximated by blank area in lunule. Drawings for L. (C.) riddlei based on holotype (USNM 132632), for L. (C.) sayrei on paratype (USNM 132636), and for L. (C.) insolita on holotype (USNM 73431). Outl'ines reduced 14; ornamentation approximately X 1;2. tion, 1964). Near Ingamar and Pontotoc Bluff strata and those of the Chiwapa Sand­ the Prairie Bluff Chalk disconformably over­ stone Member of the Ripley (Sohl, 1960, lies the Chiwapa Sandstone Member of the fig. 3). This disconformity was recognized Ripley, and although L. ( C.) insolita does and widely traced by Stephenson and Mun­ not occur here, its age relationship to L. (C.) roe ( 193 7), p. 806), and is present at In­ riddlei, n. sp., is clearly established. gamar and Pontotoc (Sohl, oral communi­ The relative stratigraphic position of L. cation, 1964). Lima (Costellacesta) sayrei (C.) sayrei from the Corsicana Marl of Stephenson therefore appears to lie strati­ Texas is not as clearly defined. The Corsi­ graphically between L. ( C.) riddlei, n. sp., cana contains a fauna considered to be tran­ and L. (C.) insolita Stephenson. No species sitional between those of the Chiwapa of Costellacesta are known to have overlap­ Member and Prairie Bluff Chalk. In addi­ ping ranges, but they are too rare to be con­ tion, there is a certain amount of faunal sidered important stratigraphic indices. overlap between the Corsicana and each of The subgenus C ostellacesta is known only these units. As a whole, the Corsicana fauna from Maestrichtian deposits of the Gulf and is not duplicated on the eastern Gulf Coast, Atlantic Coast Province (fig. 1). Its range however, suggesting that no lithologic equiv­ lies in the upper part of the E:x:ogyra costata alent occurs in this area. The Corsicana zone, and is approximately equivalent to the probably occupies a time span represented range zone of Turritella bilira Stephenson on the eastern Gulf Coast by a widespread (Chiwapa through Owl Creek: Sohl, 1960, disconformity between Owl Creek - Prairie fig. 3. Range defined by solid and dashed 92 Tulane Studies in Geology Vol. 2

lines now definitely established; Sohl, 1964, sibly, this reflects a trend toward elimination oral communication). of the costellae in the lineage. Unfortu­ nately, the only known specimen of L. ( C.) IV. PHYLOGENY AND EVOLUTION insolita is worn and does not permit accurate determination of the number and total ex­ The Costellacesta lineage appears abruptly tent of its costellae. in the American Maestrichtian, and no an­ cestral stock is known. C ostellacesta may The costae exhibit additional changes have stemmed from some foreign lineage through time, becoming narrower, less introduced in the Gulf and Atlantic Coast prominent, less numerous, and more widely Province as a fully differentiated group, but spaced in progressively younger species (fig. a search of the principal foreign literature 1). The apparent decrease in size of the has not revealed any members of the sub­ adult shell, and variation in size of the lunule genus, or an obvious ancestral group. Present through time may be additional evolution­ evidence indicates that the lineage is prob­ ary trends, but this cannot be substantiated ably endemic to this area. without more material. The rarity of Costellacesta in shallow The reduction in size and partial loss of water Cretaceous Coastal Plain sediments is costellae in the youngest member of the possibly explained by analogy with the habi­ group, coupled with the retention of the tats of living giant Limas. Vokes (1963a) primary radiating elements of the ornamen­ has shown that living species of Acesta and tation ( costae) , gives rise to a form closely Plicacesta, which are structurally similar to resembling species of the subgenus Plic­ C ostellacesta (pl. 1, fig. 2) are found pre­ acesta, known only from the Cenozoic and dominantly in deep waters of the outer con­ Recent (Vokes, 1963a). Species of Plicacesta ti~ental shelf, continental slope, and abyss lack costellae and have plicae rather than ( f 1g. 2). Only one species is found in less costae. This suggests close relationship or than 300 feet of water and this species is possible ancestry of Costellacesta to Plic­ rare at this depth. It is reasonable to as­ acesta. Additional evidence for the ancestry sume that species of Costellacesta were simi­ of Plicacesta may be provided by future . larly distributed. If so, the ancestors and studies of certain large, coarsely plicate, In­ early representatives of the subgenus, and dian and European Cretaceous limids, such 1 perhaps most species of the group, would as Lima ( AcestaJJ) obliquistriata (Forbes) thus be preserved in deep water Cretaceous (India ) , and L. dujardini Deshayes or L. sediments not presently exposed on the marrotiana d'Orbigny (France). These spe­ emergent portion of the continent. The few cies resemble Plicacesta in gross shell mor­ Cretaceous species that ranged into shallow phology, and are contemporary with and sublittoral waters are the only ones found older than Costellacesta. They may represent today in the inner shelf deposits of the At­ an extension of Plicacesta into the Creta­ lantic and Gulf Coastal Plain. ceous, or may simply be coarsely ribbed Although the number of specimens avail­ Plagiostoma. Unfortunately, certain critical able for study is limited, the three known characters of these plicate Cretaceous species, species of C ostellacesta seem to form an such as the position and nature of the re­ evolvi?g lineage in which some general silifer, and presence or absence of an an­ evolut10nary trends can be noted (fig. 1). terior auricle, are not yet known. Subgeneric Notably among these is the increase in relationships cannot be determined without number, but decrease in relative size and this knowledge. strength of .the costellae in progressively younger ~pec1~s. The oldest known species, V. THE ECOLOGY OF GIANT LIMAS L. ( C.) riddlei, n. sp., has 1 to 3 prominent Ecologic data for living species of giant fluted costellae in interspaces between ad­ Lima (subgenera Acesta, Plicacesta ) have jacent costae; the middle member of the recently been assembled by H. E. Vokes lineage, L. (C.) sayrei has 3 to 6 smaller, ( 1963a, b ). The following comments are ~moo~h costellae per interspace. On L. (C. ) drawn from these and other works on Recent tnsolita the costellae are even narrower and pelecypods. fainter, and are visible only on parts of the The giant Limas have wide geographic shell where the ornament is coarsest. Pos- distribution, occurring from the latitudes of No. 3 Costellacesta, a new subgenus of Lima 93

II

10 0 > L "'c 9

=CL ~ 8 c cu > .,. 7

.,. 6 c L L ...:3 ~ 5

-0 !;; 3 .D E z:3 2

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 Depth in Fathoms Figure 2-Depth distribution of living giant Limas (subgenera Acesta, Plicacesta), Data from Vokes ( 1963a, b).

Norway, Iceland, and Greenland [L. ( Acesta) L. (A.) angolensis Adam and Knudson, excavata ( Fabricius)} south of Patagonia which occurs from 400 to 500 meters depth [L. (A.) patagonica Dall}. They are most off Angola. Only one species, L. (A.) goliath common in the tropical and warm temper­ Sowerby occurs uncommonly in water shal­ ate regions of the Indo-Pacific, especially in lower than 300 feet. The two living species the Philippines, East Indies (Sumatra and of Plicacesta have a combined bathymetric ;Borneo), and Japanese Islands. Plicacesta range of 337 to 2194 feet; the average depth is unknown outside the Indo-Pacific, while for L, (P.) smithi Sowerby is 1265 feet, and Acesta has two Atlantic representatives. The for L. (P.) sphoni Hertlein 1650 feet. Since Cretaceous subgenus C ostellacesta, however, dead shells were most commonly dredged in occurs exclusively in the Atlantic Province, all samples analyzed, and living specimens and is the only large coarsely ribbed group are not usually differentiated in dredge rec­ of Limas with a posterior resilifer known ords, these depths are at best an approxima­ from this area. It may have occupied an tion of the living range of the various spe­ ecologic niche in the Atlantic during the cies. However, they conclusively show a deep Cretaceous similar to that presently occupied water habitat preference in the giant Limas by Plicacesta in the Pacific. (fig. 2). Most species of Acesta and Plicacesta in­ Available temperature records for stations habit deep water of the outer continental from which Acesta was taken show an aver­ shelf and continental slope (fig. 2 ) . Vokes age water temperature of 48.4 ° F. (Range cited a bathymetric range of 16 to 1450 35-59.2 ° F.). Eliminating the warm water fathoms (96 to 8700 feet) and an average form, L. (A.) rathbuni, the average habitat depth occurrence of 336.4 fathoms (2018.4 temperature for all other species is 45 .7° F. feet) for all living species of Acesta except (Vokes, 1963a, p. 85). In some species, 94 Tulane Studies in Geology Vol. 2 depth distribution is controlled rigidly by "A swimming Lima is a most charming water temperature (Vokes, 1963a, p. 80). sight. Progress is made with a series of Temperature may be the primary control on rather languid movements with each of distribution in most species, though this which the long fringe of tentacles slowly remains to be tested. C ostellacesta occurs rises and then gently descends around the with what have been considered warm, shal­ white shell." low water faunas but it is rare in these as­ Although any individual swimming inci­ semblages, as are modern giant Limas, and dent in Lima is short, the animal is capable was probably more widely distributed in of moving a considerable distance during cooler, deeper waters during the Upper its lifetime as long as it remains unattached. Cretaceous. Further, it has the ability, as an adult, to be Bottom sediment was recorded from 3 3 environment selective, a distinct asset to the stations yielding species of Acesta. Vokes' survival potential of its offspring. Thus the analysis of these data (p. 85) shows a individual, and its larvae, are commonly strong preference for mud, or fine sand and more rapidly and widely distributed than in mud bottoms (color predominantly green; wholly sessile pelecypods or vagrant ben­ rarely gray, blue) ( 24 localities). Species thonic forms, an important consideration for of Acesta are more rarely found associated the paleontologist employing Lima in cor­ with sand ( 4 occurrences), Globigerina relation. ooze, ooze and sand, shell and coral bottom, rock and shell bottom, and on stones ( 1 oc­ VI. SYSTEMATIC DESCRIPTIONS currence each) . Species of C ostellacesta are found associated with chalky marl (sayrei), Family LIMIDAE fine argillaceous sand (insolita) and chalky Genus LIMA Bruguiere, 1797 sandstone to arenaceous chalk ( riddlei). Lima BRUGUIERE, 1797, Tabl. Encycl. Meth., These sediments are well within the range Vers Coq., 2, p. 206. (Valid according to of those inhabited by Acesta at present. International Code of Zool'ogical N omen­ clature 1961, Art. 16, a, vii; see discussion Species of Costellacesta probably had habi­ of Vokes, 1963a, p. 75, 76). tat requirements similar to those of living Lima Bruguiere. CUVIER, 1798, Tabl. Elem., giant Limas. This would account for the p. 421. rarity in the shallow shelf sediments that Lima Bruguiere. LAMARCK, 1799, Mem. Soc. d'Hist. Nat. de Paris, 1, p. 88. characterize the Cretaceous of the Atlantic Mantellum ROEDING, 1798, Mus. Bolt., p. 160. and Gulf Coastal Plain. As today, rare Cre­ Radula Klein. MORCH, 1853, Cat. Conch. taceous species lived on the shallow portion Yoldi, 2, p. 57. of the continental shelf in warmer waters; Radula Klein. ADAMS and ADAMS, 1858, Genera of Recent Mollusca, 2, p. 556, 557. L. (C.) riddlei, L. (C.) sayrei, and L. (C) Lima Bruguiere. H. E. VOKES, 1963, Tulane insolita represent this shallow water element Stud. Geol., 1, No. 2, p. 75, 76. of Costellacesta. The preservation of com­ Type species: Ostrea lima Linnaeus=Lima plete fragile shells belonging to the known squamosa Lamarck; by subsequent tautono­ Cretaceous species probably indicates they my (Lamarck, 1801) . lived near their burial site, and were not transported far after death. Diagnosis: Shell thin, small to moderate­ l'y large; outline ovate, obliquely subovate, Most species of Lima are sessile forms, or subquadrate; predominantly equivalve, attaching by byssal threads to rocks and other slightly to greatly inequilateral, with one hard objects. Some build elaborate nests of (posterior) or two small, unequally devel­ byssal threads in order to protect their non­ oped auricles. Anterior byssal gape small to large; posterior gape uncommonly devel­ retractile mantle tentacles from predators. oped. Lunule commonl'y present. Valves Vokes ( 1963a) records L. (A.) excavata as nearly smooth, with incised radiating lines, having been found attached by a byssus; the costellate, costate, or finely plicate. Com­ attachment habits of other living giant Limas missure smooth to crenulate. Hinge line short; cardinal area consisting of central are unknown. Many species of Lima, includ­ triangular resilifer, erect to oblique, beneath ing a number of those which normally at­ or posterior to beak, bounded laterally by tach, possess the added ability to swim, and flat cardinal plates for ligament attach­ lie free on the bottom when not swimming. ment. Pallial line entire, faint. Monomy­ arian, posterior adductor muscle scar large, Yonge (195 8, p. 171) has described the in some cases complex, irregular, common­ swimming: ly situated posterocentrally. No. 3 Costellacesta, a new subgenus of Lima 95

Subgenus COSTELLACESTA Kauffman, logic features suggestive of the giant Creta­ n. subgen. ceous, Cenozoic, and Recent Limas, in par­ Type species: Lima (Costellacesta) riddlei ticular the valve outline, single (posterior) Kauffman, n. sp. auricle, large lunule and byssal gape. Regali­ Diagnosis: Large, equivalve, moderately lima is distinct in having radial sculpture inequilateral opisthocline, slightl'y prosogyre consisting of incised lines, and a resilifer to pisthogyre. Outline obliquely subovate, which appears to lie directly beneath the with long straight dorso.anterior edge. Lun­ beak. Nevertheless the similarities are so ule large, well defined. Byssal gape lanceo­ late, large, dorsoanterior to mid-anterior. striking that it is not difficult to envision Posterior auricle small, dorsoposteriorly sit­ the Acesta - Costellacesta - Plicacesta complex uated, separated from main body of shell by as having arisen from such a group. well defined auricular sulcus. No anterior auricle. Surface covered by prominent, The morphologic criteria and the magni­ evenly developed, widely and subequally tude of structural differences which distin­ spaced costae, smooth to finely fluted or guish C ostellacesta from other subgenera of spinose at intersection with major growth Lima are equivalent to those which separate lines. Adult costae narrower than inter­ spaces between them, 1-6 fine, evenly far better known limid subgenera such as spaced, subequally developed costellae com­ Acesta and Plicacesta. These well studied monly intercalated between adjacent costae. subgenera probably arose from a common Commissure flat to very faintly undulating. Mesozoic stock, and are known to have Hinge line short, straight, situated mainly below and posterior to beaks. Resilifer evolved as distinct lineages through the slightly oblique to hinge line, triangular, Cenozoic. Costellacesta appears to have had moderately large, situated just posterior to a similar history. The restricted number of beak. Lateral' cardinal plates flat, triangu­ specimens available for study, and lack of lar, unequal. Pallial line and musculature unknown. Shell thin. knowledge concerning its ancestors and de­ scendants should not detract from the recog­ Remarks: Costellacesta is distinct from nition of Costellacesta as a distinct subgenus. the subgenus Acesta Adams and Adams in lacking an anterior auricle, having a more LIMA (COSTELLACESTA) RIDDLE!, n. sp. projecting, beak and narrower umbone, in Pl. I, figs. 3, 4, 6 its larger lunule and byssal gape, more promi­ Material: A single, nearly complete, large nent auricular sukus, and in having promi­ right valve showing the hinge characters; nent radial sculpture, consisting of narrow two moderately well preserved, medium size costae and intercalated costellae, covering val'ves (right and left) ; a large shell frag­ the entire valve. Plicacesta is even more ment. Description: Summary of measurements similar to C ostellacesta, but can be distin­ presented in Table 1. Shell attaining large guished by its radial plicae of only one size size, equivalve, height greater than length; which involve the entire thickness of the moderately inequilateral, opisthocline, opis­ shell and are reflected internally. In addi­ thogyre. Outline subtriangular to trigonal­ subovate; valves oblique (pl. 1, fig. 4). tion, the plicae are broader than the inter­ Posterior and ventral margins moderately spaces between them, and are coarser, broad­ and subevenly rounded; anteroventral' mar­ er, and more crowded than those on species gin more narrowly rounded; anterior mar­ of Costellacesta. Finally, Plicacesta appears gin straight to very slightly curved (convex outward), steeply inclined to hinge axis. to have at least a remnant anterior auricle, Valves moderately convex dorsocentrally, a more oblique resilifer, and commonly a on um bone; ventral, ventrolateral, posterior smaller lunule and byssal gape than found flanks slightly convex, gradually flatten­ on C ostellacesta. The subgenus C ostellacesta ing to·ward margin. Anterior sl'ope, espe­ cially dorsal half, gently convex centrally, is uncommon, and is known only from the abruptly truncated and sharply incurved at Maestrichtian deposits of the Gulf and At­ edge of lunule. Lunule poorly known, ap­ lantic Coast Province, from Texas to North parently long, moderately broad, excavated, and largely fill'ed with a lanceolate byssal Carolina. gape equal to one-third or one-half of the Many Jurassic species of Lima attain a anterior margin. Rim of lunule at perime­ size equivalent to that of Costellacesta, Aces­ ter of gape apparently narrow. Beak pointed, somewhat projecting, api­ ta, and Plicacesta. Most of these belong to cal angle slightly greater than 90 ° (pl. 1, the subgenus Plagiostoma and are not closely fig. 4) · umbone poorly defined, moderately comparable. One group, however, the sub­ convex' costate. Umbonal midline straight, genus Regalilima Cox, has many morpho- sl'ightly opisthogyre. Beak situated about 96 Tulane Studies in Geology Vol. 2

five-eights the length from the anterior fine, crowded growth lines. Growth lines margin. Anterior auricle not developed. coarser on ventral one-quarter of valve, at Posterior auricle small (pl. 1, fig. 4), out­ widely spaced intervals crowded and prom­ line subcrescentic, with straight dorsal inent, associated with small concentric con­ margin, gently rounded posterior margin. strictions, particularly prominent near an­ Auricle fiat to slightly convex, separated terior and posterior margins. from main body of valve by narrow, mod­ Hinge line short, straight to slightly erately deep, well defined auricular sulcus. curved. Cardinal area broad, short, thick, Entire surface of valve covered with situated below and posterior to beaks, con­ evenl'y spaced, moderate size, subequally de­ sisting of a central triangular resilifer, pos­ veloped, straight, primary costae (pl. 1, terior to beak and somewhat oblique to fig. 3, 4) with steep flanks and flat to gen­ hinge axis, bounded by fiat, subtriangul'ar, tly rounded crests over early part of valve, lateral cardinal plates (fig. 3). Muscle scar, becoming fluted and more narro·wly rounded pallial line unknown. Shell thin, smooth ventrally and laterally near valve margins. internally. Fluting produced at intersections with coarsest growth lines. Costae slightly more crowded posteriorly and centrally than an­ teriorly. Costal interspaces shallow, flat­ based, slightly to moderately broader than costae on ventral two-thirds of valve, nar­ row dorsally. 1 to 3 costellae intercalcated in interspaces between adjacent costae on ventral one-half to two-thirds of valve (pl. Figure 3-Cardinal area of Lima, (Cos­ 1, fig. 3) ; 2 or 3 costell'ae per interspace tella,cesta,) riddlei Kauffman, n. sp. showing most common near margin. First formed position of resilifer (R) posterior to beak. costellae appear about one-third the height Beak and apex of resilifer reconstructed below the dorsal margin; second and third (dashed lines) . Ventral margin of cardinal costellae for each interspace appear at or area buried in matrix. Holotype, USNM ventral to midvalve, expanding slightly to­ 132632; enlarged X 2 % . ward margins. Ornament of auricle similar, but with more crowded costae and fewer Remarks: Lima (Costellacesta) riddlei, n. costellae. Commissure flat to gently undu­ sp., is distinct from L. ( C.) sayrei Stephen­ lating at points where costae intersect mar­ son, the most closely comparable Cretaceous gin of valve. Concentric ornamentation on dorsal three­ species, in having a subtriangular outline, a quarters of val've consisting of very faint, straight to slightly curved (convex outward)

TABLE 1 MEASUREMENTS OF LIMA (COSTELLACESTA) RIDDLE!, N. SP. (E = ESTIMATED ON BASIS OF RECONSTRUCTION OF VALVE) Holotype Paratype Paratype USNM 132632 USNM 132633 USNM 132634 Height (mm) 91.5 75 E 59 E Length (mm) 75.5 62.3 Width (mm) 17.4 11.2 11 Distance, anterior margin to beak, along line parallel to hinge axis (mm) 41.4 38.4 Length of posterior auricl'e (mm) 18 Height of posterior auricle (mm) 11 Angle of inclination: entire valve 95° 105° E Angle of inclination: beak, umbo 94° 98° E Angle between hinge axis and dorso anterior margin 48° 55° E Angle between hinge axis and auricular sulcus 37° Auricular angle (between posterior, dorsal margins) 36° Number of costae 10 mm below beak 41 36 38 Number of costae in 20 mm length, 20 mm below beak 21 19 21 Terminal number of costae 45 41 Number of costellae in 20 mm length, 20 mm bel'ow beak 5 23 25 Terminal number of costellae '78 54 E Length of hinge line (mm) 22.9 Angle of inclination of resilifer to hinge axis 81° No. 3 Costellacesta, a new subgenus of Lima 97

dorsoanterior margin, a straighter midline, Route 6, 3.5 miles east of Pontotoc, NEl/ 4 slightly more convex valves, a smaller, nar­ SWl/ 4 sec. 35, T. 9 S. , R. 3 E., Pontotoc rower lunule and byssal gape, fewer but County, Mississippi, in the Chiwapa Lime­ coarser costellae per interspace, arising at a stone Member. later developmental stage, and straighter more Types: Holotype, a nearly complete right numerous, more crowded, and more promi­ valve, USNM 132632. Largest paratype, a nent costae (compare pl. I, figs. 3, 4 with medium size right valve, USNM 132633; a figs. 7, Lima ( C ostellacesta) insolita 9). smaller paratype, about two-thirds of a Stephenson is smaller, having faint, greatly crushed left valve, USNM 132634; a para­ reduced costellae, and smooth, low, rounded, type, a shell fragment sectioned to observe more widely spaced costae. thickness and inner valve surface, USNM Lima (Costellacesta) riddlei is chosen as 132635. the type species of C ostellacesta because it is the best known of the three species placed LIMA ( COSTELLACESTA ) SAYRE I in this group; the holotype exhibits most of Stephenson the essential characters of the subgenus, in­ (Pl. I, figs. 5, 7, 9) cluding the cardinal area. The species is Lima ? sayrei STEPHENSON, 1941, The larger named in honor of Mr. William C. Riddle invertebrate fossils of the Navarro Group of Memphis, Tennessee, who collected and of Texas: Univ. Texas Puhl. no. 4101, p. contributed the types, and many other speci­ 146, 147, pl. 23, figs. 12, 13. mens of Cretaceous mollusks, to the study Material: The primary types of the spe­ cies: 2 nearl'y complete, articulated pairs collections of the Smithsonian Institution. o.f adult valves (USNM 76476, 132636); 1 Stratigraphic and geographic position: The fragment of a large valve (USNM 76477). holotype and two paratypes were collected Description: Summary of measurements from the Chiwapa Sandstone Member ( presented in Table 2. Shell large, thin, equi­ = valve moderately inequilateral, opisthocline. Keown ville Limestone Member) of the Rip­ Outli~e subovate; dorsal margin apparently ley Formation near Ingomar, Union County, short, straight over posterior auricle; dor­ Mississippi. The third paratype (USNM soanterior margin long, straight to slightly 132634) comes from USGS locality 25418 concave, moderatel'y inclined to hinge axis; ventroanterior, ventral, posterior margins ( locality 38 of Sohl, 1960, p. 36, 3 7), moderately and unevenly rounded greatest from roadcuts on new Mississippi State curvature ventroanteriorly, decreasing pos-

TABLE 2 MEASUREMENTS OF LIMA (COSTELLACESTA) SAYRE! STEPHENSON (L = MEASURED ON LEFT VALVE; R = MEASURED ON RIGHT VALVE; C = MEASURED ON BOTH ARTICULATED VALVES AND VALVES SAME; E = ESTIMATED) Holotype Paratype USNM 76476 USNM 76477 Height (mm) 95 c 84 CE Length (mm) 84 c 79.2 c Width (mm) 16.6 R 14.9 R 32.6 c 30 L Distance beak to anterior margin, along Tine parallel' to hinge axis (mm) 32.5 c 49 .2 LE Length of lunule, along margin (mm) 55.5 c 54.6 c Width of lunule; maximum (mm) 14.4 c 15.7 c Length of byssal gape, alon~ margin (mm) 40.3 CE Width of Byssal gape; maximum (mm) 6.1 c Angle of inclination, entire valve 110 c 108 c Angle of inclination, beak and umbone . . 88 CE 86 CE Angle between hinge axis and dorsoanteri?r marg1"!1 40 c 42 L Angle between hinge axis and dorsopostenor margm 57 c 52 L Number of costae 10 mm below beak 29 L Number of costae in 20 mm l'ength, 20 mm below beak 14 L 19 L Total number of costae at margin 35 L 39 L Number of costellae in 20 mm length, 20 mm below beak 30 L 44 L Total number of costellae at margin 130 L 165 L 98 Tulane Studies in Geology Vol. 2 teriorly; margin concave and notched dor­ costae like those on main body of shell, sep­ soposteriorly beneath auricle. Central dor­ arated by equally wide or narrower, round­ sal parts of valve slightly to moderately ed interspaces rarely bearing a single, inter­ convex, ventral, ventrolateral, posterior calated, fl'uted costella. flanks slightly flattened, dorsoanterior Remarks: Lima (Costellacesta) sayrei flanks steeply inclined into lunule. Lunule poorly preserved on all specimens (pl. 1, Stephenson is the largest and most ornate fig. 5); lanceolate, l'ong, occupying dorsal species of the lineage. It is known only one-half of anterior margin, excavated, mod­ from Stephenson's 3 type specimens. Al­ erately deep, apparently sharply defined though he partially exca\'ated them, Stephen­ and partially over hung by abruptly trun­ cated dorsoanterior margin. Anterior byssal son did not note the lunule with its large gape long, moderately narrow, lanceolate byssal gape. Consequently, he reversed an­ (pl. 1, fig. 5) ; commissure slightl'y upturned terior and posterior throughout his descrip­ around gape, flat to very faintly undulat­ ing over remainder of valve. tion, considering the left valve the right, Beaks not known. Umbone slightly to and the auricle as being anterior. Compari­ moderately convex, apparently erect to son of the species of Costellacesta with the slightly prosogyre. Anterior auricle absent; similar, living, Lima (Plicacesta) smithi (pl. anterior edge of cardinal area rounded, slightly projecting and incurved at dorsal I, figs. 1, 2) clearly shows that the major extremity of byssal gape, similar to ill'us­ byssal gape is dorsoanterior and situated in trated specimen of L. (Plicacesta) smithi a prominent lunule, the anterior auricle is (pl. 1, fig. 1). Posterior auricle small, flat, absent, and the valves are opisthocline. An­ incompletely known on both specimens (see Stephenson, 1941, pl. 23, fig. 12) apparently terior and posterior are determined on L. (P.) straight dorsally, rounded posteriorly, with smithi by the resilifer (oblique and pos­ distinct ornamentation. Posterior auricular terior to the beak) and the well developed sulcus narrow, moderately deep, well' de­ posterior adductor muscle scar. A parallel fined, an enlarged and much deepened in­ terspace between costae. Hinge line not pre­ orientation in C ostellacesta is determined by served, assumed similar to L. (C.) riddlei, the position of the lunule and byssal gape n. sp. Pallial line, musculature, unknown. on all species, and by the position of the Ornamentation consisting of 34 to 38 resilifer in L, ( C.) riddlei, n. sp. simple primary costae separated by 3 to 6 fine intercalated costellae per interspace The distinction between L. ( C.) sayrei and (pl. 1, fig. 7). Costae, prominent, subevenl'y L. (C.) riddlei have been discussed under the spaced, equally developed, narrow, steep­ "Remarks" section for the latter species. sided, with rounded crests, originating- on early part of umbone, their trace straight L. (C.) sayrei is distinguished from L. (C.) to slightly curved over most of shell, gently insolita Stephenson by its larger size, more sinuous ventrally and ventrolaterally at in­ rounded outline, more prominent and slight­ tersections with major concentric elements. ly more numerous fluted costae, and espe­ Costae slightly more crowded posteriorly than anteriorly; 2 to 3 costae cl'osely crowd­ cially by its more numerous, more extensive ed on inturned edges of lunule and on steep beaded costellae. L. (C.) sayrei also appears part of dorsoposterior flank descending into to have a larger byssal gape, and better de­ auricular sulcus. Costae fluted to subspi­ nose at somewhat regular intervals where fined lunule. Stephenson illustrated the holo­ intersected by major growth lines (pl. 1, type ( 1941, pl. 23, fig. 12). fig. 7) ; 4-10 flutes per centimeter, best de­ Stratigraphic and geographic distribution: veloped ventrally. Costellae fine, thread­ like, evenly spaced and equall'y developed in Lima (Costellacesta) sayrei is known only broad flat inters paces between costae; from the Corsicana Marl, Navarro Group, straight to slightly sinuous, beaded due to at U.S.G.S. locality 15621, in a ravine east irregularities produced at intersections with of Medio Creek, 0.8 mile south of Castro­ majority of growth lines (pl. 1, fig. 7). 1 to 2 costellae present between each pair of ville road, near Castroville, Texas; at U.S.G.S. costae on earliest part of umbone preserved; locality 15622, from a bluff on Medio Creek number increasing by intercalation through­ about 0.8 mile "below" (south of) Castro­ out growth of shell, with 5 to 6 commonly ville road; and from 6 miles east of Castro­ found in interspaces near margin of adult shell (pl. 1, fig. 7, 9). Concentric ornament ville, Texas (Texas Bureau 53; locality of consisting of faint to well defined but small, holotype). crowded growth lines over entire shell, and Types: Holotype, the largest set of rela­ 2 to 3 scattered, narrow zones of moderate tively complete articulated valves USNM to coarse growth lines, and more rarely lamellae, near margin ( <;mter 1 inch or less). 76476. Paratypes, a well preserved set of Ornamentation of posterior auricle dis­ articulated valves (U.S.G.S. loc. 15621), tinct, consisting of crowded, coarsely fluted USNM 132636 (pl. I, figs. 5,9) (formerly No. 3 Costellacesta, a new subgenus of Lima 99 a part of USNM 76477, unfigured paratype TABLE 3 lot); a shell fragment showing the orna­ MEASUREMENTS: HOLOTYPE OF ment (USGS loc. 15621, USNM 76477). LIMA (COSTELLACESTA) INSOLITA STEPHENSON LIMA ( COSTELLACESTA,.) INSOLITA Height, including reconstruction Stephenson of beak (mm) · 63 Length (mm) 53.3 Pl. I, figs, 8, 10 Width (mm) 9.8 Lima inso.Zita STEPHENSON' 1927' Additions Distance, anterior margin to to the Upper Cretaceous Invertebrate fau­ beak, along line parallel to nas o.f the Carolinas: Proc. U. S. Natl. hinge axis (mm) 26.4 Mus., v. 72, art. 10, p. 13, pl. 5, fig. 10. Angle of inclination, entire valve 105° Material: A single, nearly complete, worn Angl'e of inclination; beak, umbo 85° adult left valve, the holotype (USNM Angle between hinge axis 73431). and dorsoposterior slope 42° Description: Summary of' measurements Angle between hinge axis and presented in Table 3. Shell moderately dorsoanterior slope 46° large, thin, presumably equivalve; inequilat­ Number of costae 10 mm eral', moderately opisthocline; height greater below beak 28-30 than length. Outline subovate, with long, Number of costae in 20 mm slightly concave, dorsoanterior margin. Ven­ length, 20 mm below beak 17 troanterior margin strongly curved; ven­ Total terminal number of costae 28-30 tral, posterior margins moderately, and sub­ evenly rounded (pl. 1, fig. 10) . Dorsal mar­ of costae, fainter discontinuous costellae, and gin not preserved. Umbone slightly to mod­ erately convex; ventral, ventrolateral, pos­ the smaller, less rounded valve are sufficient terior flanks slightly convex medially, be­ to distinguish the species from L. (C.) sayrei coming flatter toward margin. Dorsoan­ and L. (C.) riddlei, n. sp. The species ap­ terior flank abruptly truncated, moderately pears to be most closely related to, and de­ to narrowly rounded at edge of lunule. Lunule poorly defined, depressed, occupying rived from L. (C.) sayrei. It is the youngest two-thirds of dorsoanterior margin. Byssal known member of the lineage. Apparent gape unknown, probably long and narrow. reduction of the costellae, and retention of Beaks unl•nown. Umbo suberect, broad, well defined costae, may reflect ancestry to not well defined. No anterior auricle devel­ oped. Posterior auricle, if present, not pre­ the Cenozoic - Recent subgenus Plicacesta served. Hinge l'ine, internal structures un­ Vokes. known. Stratigraphic and geographic position: The Radial ornamentation consisting of 28 to only known specimen is from the upper part 30 straight simple primary costae extend­ ing from early part of umbone to margin. of the Peedee Formation, in sediments bear­ Costae narrow, low, rounded, smooth equally ing an Owl Creek-Prairie Bluff fauna (upper developed, subevenly spaced (more crowded part of the zone of Exogyra costata Say: late anteriorly), separated by much larger, flat Maestrichtian). It was found at U. S. Geo­ interspaces, a few of' which bear very faint, fine, scattered, discontinuous costellae (pl. logical Survey locality 13585, in the New 1, fig. 8), 1 to 2 per interspace preserved, Rocky Point quarries, one mile northeast of but probably much more extensive on un­ Rocky Point Station, Pender County, North worn shells of species. Costellae smooth, Carolina. straight, very low, rounded. Concentric or­ namentation consisting of numerous, crowd­ Type: Holotype, USNM 73431; a nearly ed, very faint growth l'ines over entire valve, complete left valve. and a few, scattered, moderately prominent growth lines near margin. Commissure VII. REFERENCES CITED smooth to very broadly undulating. ADAMS, HENRY, and ARTHUR ADAMS, 1858, The genera of Recent Mollusca: v. 2, p. Remarks: As in L. (C.) sayrei, Stephenson 556-558, London, J'ohn Van V oorst. did not note the position of the lunule and BRUGUIEREI, J. G., 1797, Tableau encyclope­ main byssal gape in this specimen, and con­ die methodique, OU par ordre de matiere: sidered it a right valve; anterior and pos­ Histoire naturelle des Vers: v. 2, p. 206, Paris & Leige. terior are reversed throughout his descrip­ CUVIER GEORGES, 1798, Tableau elemen­ tion. taire' de l'histoire naturelle des animaux: The holotype is worn, and thus the precise p. 421, Paris. LAMARCK, J. B. P. A. DE, 1709, Prodrome nature of the delicate surface ornamentation d'une nouvelle classification des coquilles: cannot be determined. In spite of this, the Mem. Soc. d'Hist. Nat. de Paris, v. 1, p. wide spacing of the costae, lower number 88. 100 Tulane Studies in Geology Vol. 2

LAMARCK, J. B. P. A. DE, 1801, Systeme des the Carolinas: Proc. United States Natl. animaux sans vertebres: v. 1-8, p. 1-432, Mus., v. 72, art. 10, p. 13, pl. 5, fig. 10. Paris. STEPHENSON' L. w., 1941, The larger inver­ MELLEN, F. F., 1958, Chiwapa Sandstone tebrate fossils of the Navarro Group of Member of Ripl'ey Formation in Creta­ Texas: Univ. Texas Publ. No. 4101, p. ceous shelf sediments of Mississippi: Mis­ 146, 147, pl. 23, figs. 12, 13. sissippi State Geol. Surv., Bull. 85, p. STEPHENSON, L. w., and w. H. MUNROE, 49-54, figs. 28-33. 1937, Prairie Bluff and Owl Creek For­ MORCH, 0. A. L., 1853, Catalogus conchyli­ mations of eastern Gulf region: Ameri­ orum quae reliquit D. Alphonso d'Aguirva can Asscc. Petrol. Geol. Bull., v. 21, no. & Gadea Comes de Yoldi: Pt. 2, Acephala, 6, p. 806. Echinodermata Hafniae, p. 57. VOKES, H. E., 1963a, Studies on Tertiary ROEDING, P. F., 1798, Museum Boltenianum and Recent Giant Limidae: Tulane Stud. sive catalogus cimeliorum e tribus regnis Geol., v. 1, no. 2, p. 73-92, pls. 1, 2. naturqe: pt. 2, p. 160, Hamburg. SOHL, N. F., 1960, Archeogastropoda, Meso­ VOKES, H. E., 1963b, Additions to a cata­ gastropoda and Stratigraphy of the Rip­ logue of the described Recent and Terti­ ley, Owl Creek, and Prairie Bluff Forma­ ary species of Acesta and Plicacesta : Tu­ tions: United States Geol. Surv. Prof. lane Stud. Geol., v. 2, no. 1, p. 18-20. Ppr. 331-A, p. 18-20, 22, fig. 3. YONGE, C. M., 1958 (2nd printing), The sea STEPHENSON, L. w., 1927, Additions to the shore: p. 171, London, Collir.s, St. James Upper Cretaceous invertebrate faunas of Place. June 30, I964

EXPLANATION OF PLATE I Figures Page I, 2, - Lima ( P licacesta) smithi Sowerby______98 I, Interior and 2, exterior views (XI) of a left valve from Echigo, Japan; USNM 344887; Hirase Collection, Division of Mollusks, USNM. Illustrated for comparison with members of L. (Costellacesta).

3, 4, 6 - Lima ( Costellacesta ) riddlei Kauffman, n. sp. ____ ·------·------____ 95 3, Portion of right valve (fig. 4) showing adult ornamentation ( X2 ) . 4, Lateral view (XI), holotype (USNM I32632), a right valve. 6, Anterior view (XI) of holotype showing margin of lunule.

5, 7, 9 - Lima ( Costellac esta) sa yrei Stephenson______97 5, Anterior view (XI ) of Stephenson's unfigured paratype (USNM I32636) showing limits of lunule, position of byssal gape (space inside byssal gape retouched to improve contrast), infolded anterior margin of cardinal area. 7, Portion of the same adult shell near margin (X2) showing details of orna­ mentation: left valve. 9, Lateral view of left valve (paratype, USNM 132636, of figs. 5, 7) with auricle broken off.

8, I 0 - Lima ( Costellacesta ) insolita Stephenson ------99 8, Portion of left valve shown in fig. I 0 ( X2) , showing costae and faint, dis­ continuous, intercalated costellae. IO, Lateral view (XI ) of left valve, the holotype (USNM 7343I) and only known specimen. No. 3 Costellacesta, a new subgenus of Lima 101

PLATE I