ce ace wee Seen ceniey NEM EYELET a ; ° : ieinctcm i ~ F “ Rs - " " oom 2 ‘ a 2 Neer Hc outed scbndii pt Sie * ra Peet - . = naa Hotes tahiy an ahashatitiof mest We ‘ c . * PERE NOE ME REN Ae abet 2 € 7 Peatyan , Bi Sane cies Mint ntnamathe Teste Sees ental ch ‘ ete “4 ee Dra : : teh phe foe thy a Dat Michie mayne on eaocatiane Peseta EMA Zi nore, saan siesta Penner th NA Fa month PSM ne tS tie ALN SWAIN SN TE Se aoe Ue oe Metta Ro aaratiarente pata Nh qa ip Pasta nate Bede racte nat tenn te on Asm NN Rae tm S tan ag es ee PR NONE Nee Attn pak ee aN A Phe net ow es SATS en anata EAg NS No ey iney eoe ne 2 eer eta sa — oO = Ji w 2 o = ae) S) = fe) ke EE eo) Lad K- 2 5 a =) > a ea Sti 2 > Ey = re > ie a = se Fa = FE a = n m 7) O AVS o = 7) = A a RARIES SMITHSONIAN NOILNLILSNI S3IYVYUEIT LIBRARIES SMITHSONIAN x ¢ : g E : oe Z WS = = f pe A ee a QE = = VF YS = SF fie Oo ~ “& Oo NS AS zr AE S 5 PUP 3 &G4f 2S 3 3 YZ : 2 Uy * a : > = > Ue os > , Ss Miss = 2 7) Bun be y 7p) 2 77) OU ees wo LNLILSNI NVINOSHLINS S3IYVHGIT LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLINS S3INVUaT SMITHSONIAN RARIES SMITHSONIAN_INSTITUTION NOILNLILSNI NOILNLILSNI E%, be Ue ZA LIBRARIES NOILALILSNI LIBRARIES LIBRARIES Saiuyvugil LIBRARIES INSTITUTION NOILNLILSNI SAIYVYRIT LIBRARIES INSTITUTION NOILNLILSNI z r z & iz © Gy,> = B 3 EO ks E 2 3 LNLILSNI _ NVINOSHLINS S34 | avy gi etl BAG | ES SMITHSONIAN INSTITUTION NOLLALILSNI > NVINGSHEINS Ly, : : : : : : pF 3 5 6 3 ag z V fort : : : 3 : E = = ae 5 = 5 RARIES SMITHSONIAN INSTITUTION NOILALILSNI LIBRARIES SMITHSONIAN INSTITUTION LIBRARIES NOILALILSNI LIBRARIES NOILNLILSNI NOILALILSNI NOILNLILSNI INLILSNI NVINOSHLINS SSIYVYSIT LIBRARIES SMITHSONIAN & bas ‘4 RARIES SMITHSONIAN INSTITUTION, NOILALILSNI NVINOSHLIWS INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVUat saruvud! INSTITUTION saiuvugl INSTITUTION & \S INSTITUTION INSTITUTION S3SIYVYUEIT LIBRARIES NVINOSHLINS S3!YVYGI7 LIBRARIES SMITHSONIAN NVINOSHLINS S31yu¥Vvudit LIBRARI ES ” a” iz = < = 2 Dee z = = ai iz itt Lo = WS Zz Se x tif 0 QS g - ae 2 LY, ° A Z 2 = Zz ie Cy IN. fe INLILSNI NVINOSHLINS S3IuYVYaII SMITHSONIAN INSTITUTION NOILQLILSNI NVINOSHLIWS SAINVY aI: Be ee ta 2 tu Zz ws 2 | Ww x SN ” a ise Ss X8 LS As = a Z z ZX = 2X = 2 é : E 2 : 3 ra J, Yip: > = 2 Ee > = 4 > — 4 fr 2 = zu = 2 = = ” : m ” m N w” = n* m = 7) = ” fa = a) = a | HNLILSNI_NVINOSHLINS, SA1UVaGIT LIBRARIES, SMITHSONIAN INSTITUTION NOILMLILSNI_ NVINOSHLINS, Saluvuat” My, = SS ee ee = i 2 , = Y Gy, z = Z 5 z Ny 5 G&? = Ye)? i 0? i aN ee : i =e = = = (NY 2.777 & Ss z z z z Fa Ve Z = RARIES SMITHSONIAN INSTITUTION | NOILALILSNI_NVINOSHLINS Saluvu att LIBRARIES SMITHSONIAN INSTITUTION S S m) = | Vers ww RAASES TS CRP NTE UE A Pek at del mae Meet pL Se ola al Set Yate 8 8 ate § ate we ARTES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLINS S3ltyvugit @, SMITHSONIAN INSTITUTION NOLLOLILSNI_ NVINOSHLINS. S31YV¥ aI NVINOSHLIWS SMITHSONIAN ILILSNI_ NVINOSHLIWS saiuvyai LIBRARIES Zz Uns = rc z (mat oe Bt i Op ae = oO 2 ae ro) = ‘s) ie 5 2 = 5D : s Be fy = ~ = a E 5 EY » Ke = x = a ai "Uji, Ze \LILSNI SAIYVYUSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS SaluWvual ma o Ze we 22) z ee n z= 2) x S z eS Ws = x \ = wes = = Sy y z S SS rai Fa ar z > th, oO NYS : BGM ENS 3 2 a g - = 2 = > = > > > “no Zz ” BN cae o Cet ee n Zz ARIES SMITHSONIAN _INSTITUTION NOILALILSNI NVINOSHLINS S31uvYsIT_ SMITHSONIAN _ INSTITUTIOI = eae = 2 5 2 5 Gy ra = 20 Xs Ee > YW cr > = LF ra = SN ies = SS"5 2 = GY he 2 2 D WN = ~ * f . — - SS YY 7 ae 5 : 5 - 5 2 SMITHSONIAN ” wo = z = a ae ae (ep) e (ep) oO [e) Zz 7A >" > PA, Za SMITHSONIAN NVINOSHLIWS SS LIBRARI ES SMITHSONIAN SMITHSONIAN NOLLNLILSNI NOILALILSNI NOILNLILSNI LIBRARIES NOILNLILSNI LIBRARIES ARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS SAIYVYUSIT LIBRARIES SMITHSONIAN INSTITUTIOI! S2!I4YVYSIT LIBRARIES 5 = & Z iS z - 5 5 2 = 2 5 (Yt ff ® ae: E Bis = os E Gy lp 2 = E as e = + Wy = ALILSNI NVINOSHLINS S3iuvyudit SMITHSONIAN _ NOILNLILSNI NVINOSHLINS S3IYVUEI ra n z= wn = om zZ (7) Te SE ete «= = < = < = ff z Sy Ay = =| > = > =e ae en MC siGF RE LEVI ZF CG Yj iS = Gy = z = Z cs S Uy a Si meee . 2 2 oe Sone 2 Bae (72) 2 (79) ARI Eo) SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLIWS Sa 1YVYEIT_ LIBRARIES SMITHSONIAN _INSTITUTIO ul 3 w aS # j Z 5 . = cc = oc = « . = <3; z < a < eal = 2 E : 2 5 2 Ne 5 = : : : 3 : 2 SE 2 = 2 E : i = W's FARIES = SMITHSONIAN INSTITUTION NOILALILSNI_NVINOSHIIWS, Sa1uvuatt BSE AGES SMU SONIAN INSTITUTIO = = : ees iy = = ; a iy = < A eA 7S 4 PY Y G2 pa) Zz s ia > 2 z & JF 3 & 5 WX 3 § WS SS) fo) r ro) Uf Vip oe (e) Ae S \ S oO Be ~ WV NT Z = ZU aie i = = \S 2 = AS ALILSNI_NVINOSHLIWS eye) RARITIES SMITHSONIAN _INSTITUTION NOILNLILSNI_NVINOSHLIWS Sa 1YVuds a Os Sit ear RuRoecty » y Wsone eR PPS ak Meet ene, i ACN aR ae no al Ne Gi 74,05 of 3 fol, THE VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor ISSN 0042-3211 Volume 28 July 1, 1985 Number 1 CONTENTS | QL | 401 Attack mode in a predatory gastropod: labial spine length and the method of V4x prey capture in Acanthina angelica Oldroyd. | MOLL ANTE SBROMINUAT USAC fate atiy eta Goer tt are wiv tn jwise aah atedly endl eaten a ie 1 | Fluid-dynamic drag of limpet shells. j IRGPSBRGP 1D AGIOTLTD NG elie ce ea tes or TE StS Ue eae Mn eT eM ese 17061 6 The effects of aggregations on water loss in Collisella digitalis. VAVIMi, LEVRVAIB) (GOAN ULATIONS pete Neg Rea eres ca ee ca a a el UU Sd co 14 Relationship between allometric growth, with respect to shell height, and hab- itats for two patellid limpets, Nacella (Patinigera) macquariensis Finlay, 1927, and Cellana tramoserica (Holten, 1802). ROD ES INUPSO NGM OR A Pe) ook fee) Ucn MLO alll uate Ubi alvin ia MRM Re lt 18 Spatial and temporal distribution and overlap of three species of Bullia (Gastro- poda, Nassariidae) on exposed sandy beaches. aE VIG GWYNNEVAND A MGHLACHLAN 2 oo... 22 ox oa tice anton 28 Aspects of reproduction, larval development, and morphometrics in the pyram- idellid Boonea impressa (=Odostomia impressa) (Gastropoda: Opistho- branchia). Marie E. WHITE, CHRISTOPHER L. KITTING, AND ERIC N. POWELL ..... Oy) On the anatomy and fine-structure of a peculiar sense organ in Nucula (Bivalvia, Protobranchia). (CABETASZPRIWINA Retreat itl) Ux meso iain cen ites WON Geile Bi oi US BREN Se OY cide sabany te 52 CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly on the first day of July, October, January and April. Rates for Volume 28 are $22.00 for affiliate members (including domestic mailing charges) and $44.00 for libraries and nonmembers (including do- mestic mailing charges). An additional $3.00 is required for all subscriptions sent to foreign addresses, including Canada and Mexico. Further membership and subscrip- tion information appears on the inside cover. The Veliger is published by the Cali- fornia Malacozoological Society, Inc., % Department of Zoology, University of Cali- fornia, Berkeley, CA 94720. Second Class postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to C.M.S., Inc., P.O. Box 9977, Berkeley, CA 94709. THE VELIGER Scope of the journal The Veliger is open to original papers pertaining to any problem concerned with mol- lusks. This is meant to make facilities available for publication of original articles from a wide field of endeavor. Papers dealing with anatomical, cytological, distributional, eco- logical, histological, morphological, physiological, taxonomic, etc., aspects of marine, freshwater, or terrestrial mollusks from any region will be considered. Short articles containing descriptions of new species or lesser taxa will be given preferential treatment in the speed of publication provided that arrangements have been made by the author for depositing the holotype with a recognized public Museum. Museum numbers of the type specimen must be included in the manuscript. Type localities must be defined as accurately as possible, with geographical longitudes and latitudes added. Very short papers, generally not exceeding 500 words, will be published in a column entitled “NOTES, INFORMATION & NEWS”; in this column will also appear notices of meetings, as well as news items that are deemed of interest to our subscribers in general. Editor-in-Chief David W. Phillips, 2410 Oakenshield Road, Davis, CA 95616, USA Editorial Board Donald P. Abbott, Emeritus, Hopkins Marine Station of Stanford University Hans Bertsch, Universidad Autonoma de Baja California James T. Carlton, Williams College—Mystic Seaport Eugene V. Coan, Research Associate, California Academy of Sciences, San Francisco J. Wyatt Durham, University of California, Berkeley Cadet Hand, University of California, Berkeley Carole S. Hickman, University of California, Berkeley A. Myra Keen, Emerita, Stanford University David R. Lindberg, University of California, Berkeley James H. McLean, Los Angeles County Museum of Natural History Frank A. Pitelka, University of California, Berkeley Peter U. Rodda, California Academy of Sciences, San Francisco Clyde F. E. Roper, National Museum of Natural History, Washington Judith Terry Smith, Stanford University Ralph I. Smith, University of California, Berkeley Wayne P. Sousa, University of California, Berkeley T. E. Thompson, University of Bristol, England Alex Tompa, University of Michigan, Ann Arbor Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (no institutional memberships) interested in any aspect of malacology. As an afhliate member, a person may subscribe to The Veliger for US $22.00, which now includes mailing charges to domestic addresses. There is a one-time membership fee of US $2.00, after payment of which, membership is maintained in good standing by the timely renewal of the subscription; a reinstatement fee of US $1.00 will be required if membership renewals do not reach the Society on or before April 1 preceding the start of the new Volume. If a receipt is required, a self-addressed, stamped envelope (or in the case of foreign mem- bers, the envelope and two International Postal Reply coupons) should be included with the membership or subscription request. The annual subscription rate to The Veliger for libraries and nonmembers is US $44.00, which now includes mailing charges to domestic addresses. An additional US $3.00 is required for all subscriptions sent to foreign addresses, including Canada and Mexico. Memberships and subscriptions are by Volume only (July 1 to April 1) and are payable in advance to California Malacozoological Society, Inc. Single copies of an issue are US $30.00 plus postage. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: C.M.S., Inc., Post Office Box 9977, Berkeley, CA 94709. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. The Veliger 28(1):1-5 (July 1, 1985) THE VELIGER © CMS, Inc., 1985 Attack Mode in a Predatory Gastropod: Labial Spine Length and the Method of Prey Capture in Acanthina angelica Oldroyd JAMES R. MALUSA Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, Arizona 85721 Abstract. The function of the labial spine and the feeding behavior of the predatory gastropod Acanthina angelica were observed under controlled conditions. Long- and short-spined snails were presented three size classes of barnacle prey. The mode of attack was related to the length of the labial spine and the prey size. The spine was observed to function as a wedge to force apart the opercular valves of the prey (here termed wedging); drilling through the test or valves was an alternative mode of attack. As prey size increased, snails switched from wedging to drilling, with the short-spined snails switching at a smaller prey size than long-spined snails. The long-spined snails consumed medium- sized prey significantly sooner than short-spined snails. Short-spined snails are usually found in asso- ciation with small barnacles, while long-spined snails predominate among larger barnacles. However, spine length is not fixed, and available evidence indicates that prey size controls spine length. INTRODUCTION MANY PREDATORY GASTROPODS attack barnacles, bivalves, and other gastropods by drilling through the shell of the prey (CARRIKER, 1961, 1981). “Wedging” is an alterna- tive mode of attack in several of these species (PAINE, 1962; MacGINITIE & MACGINITIE, 1968); this entails the predator’s forcing its shell margin between the valves of the prey (barnacle or bivalve) and, once access is gained, inserting the proboscis to consume the prey. Some mem- bers of the neogastropod family Thaididae, which includes the genus Acanthina, have apparently taken the wedging approach to attacking prey one step further. These species are characterized by an extension of the shell margin into a labial spine or “tooth.” A variety of functions has been attributed to the spine. HeEwatt (1934) and MacGINITIE & MACGINITIE (1968) both observed several species of thaids utilizing the spine as a “pry bar” or wedge to force apart and hold open barnacle valves. PAINE (1966) concluded that the spine of Acanthina angelica Oldroyd, 1918, is not used in this fash- ion, but rather it serves as a brace to afford a firm position on the substrate while drilling. MENGE (1974), working with Acanthina punctulata (Sowerby, 1825), arrived at a similar conclusion regarding use of the labial spine. SLE- DER (1981), on the other hand, concluded that the spine of A. punctulata helps the predator to apply a fast-acting toxin to its barnacle prey. Acanthina angelica is endemic to the Gulf of California (KEEN, 1971), and is common in the rocky intertidal of the northern Gulf (TurK, 1981; Houston, 1980). Adult snails attain a total shell length of 35-40 mm, and feed almost exclusively on barnacles (PAINE, 1966). The spine length of adult snails varies considerably among individ- uals; e.g., 30-mm snails have spines ranging from 2 to 7 mm in length (YENSEN, 1979). The intertidal distributions of the long-spined and short-spined snails are skewed in a manner that reflects the intertidal size distribution of the two dominant barnacle species upon which the snails prey (PAINE, 1966; YENSEN, 1979; TuRK, 1981). Long- spined A. angelica are more common in the high intertidal zone characterized by the large barnacle Tetraclita stalac- tufera Lamarck, while short-spined snails are usually found in the lower intertidal in association with the small bar- nacle Chthamalus anisopoma Pilsbry. The correspondence between spine length and barnacle size suggests a func- tional relationship between the two. Both PAINE (1966) and YENSEN (1979) observed that snails with relatively long spines prey on Tetraclita, while short-spined snails feed primarily on Chthamalus. However, in contrast with Paine’s suggestion that the snails only drilled, Yensen’s Page 2 field observations and laboratory experiments strongly support the hypothesis that the labial spine is used directly in wedging apart the opercular valves of barnacles. The purpose of this study was to elucidate the relation- ship between labial spine length, the size of the barnacle prey, and the mode of attack of Acanthina angelica. If the spine is used to wedge open the opercular valves of bar- nacles, and drilling is an alternative to wedging when the spine is too short to effectively reach the valves, then short- spined snails should switch to a drilling mode of attack at a smaller prey size than do long-spined snails. This is assuming that wedging would be quicker than drilling, and that the snails feed in the most efficient manner pos- sible. MATERIALS anp METHODS A laboratory experiment was conducted to determine the foraging behavior of both long- and short-spined snails as they fed on three different size classes of barnacles. Spec- imens of Acanthina angelica were collected haphazardly during October of 1983 from the rocky intertidal near Puerto Penasco, Sonora, Mexico (31°18'N, 113°35’W) on the Gulf of California. The animals were brought to the University of Arizona where they were maintained with- out food in three 40-L aquaria for two weeks. Aquarium water temperature was approximately 22°C, comparable to that in the Gulf during October. Photoperiod was ap- proximately 10 L, 14 D. Two weeks after the snails were collected, barnacles were collected from the same area. To avoid the poten- tially confounding effects of using two species of barnacles for prey, as might arise from a species-specific mode of attack, I exclusively collected Tetraclita stalactifera over a range of sizes (2-40 mm basal diameter) for presentation as prey. Settlement of Tetraclita during the two months previous made it possible to collect adequate numbers of small individuals calculated to be of approximately the same size and body weight as the smaller barnacle species, Chthamalus anisopoma (MALUSA, 1983). The barnacles were brought to the University of Arizona and maintained in aquaria next to those harboring the snails. Forty-five short-spined and 45 long-spined snails were chosen for the experiment on the basis of their spine length. Long-spined snails were defined as those having a labial spine measuring more than 4 mm from base to tip; short- spined snails were defined as those having a spine mea- suring less than 3.5 mm in length. No attempt was made to control for differences in shell length between the two groups. Long-spined snails were approximately 25-35 mm in length, while short-spined snails were approximately 20-35 mm. The shells of all snails were numbered with a perma- nent felt tip marker to permit individual identification. Barnacles were sorted into three size classes based on the basal diameter along the rostral-carinal axis: less than 7 mm (small), 7-20 mm (medium), and greater than 20 mm The Veliger, Vol. 28, No. 1 (large). Inappropriately sized individuals were removed from the pieces of the substrate bearing barnacles, leaving only the desired size class on each rock. At the beginning of the experiment, each of the three aquaria received 15 long-spined snails, 15 short-spined snails, and 40 to 60 of one of the three size classes of barnacles. Within each aquarium the two groups of snails were separated by a plastic screen that allowed water passage. Approximately equal numbers of barnacles were made available on either side of the divider. The foraging activities of the snails were then observed continuously for the following 24 h, and thereafter two to three times daily for 26 days. I kept a record of (a) the mode of attack employed by each snail on its first successful feeding (wedging or drilling), and (b) the time from the start of the experiment until the first barnacle was successfully attacked and consumed (here termed the “consumption order’ —see below). The feeding behavior and the use of the spine could be observed closely in instances when the opercular valves of the barnacle were close to the opening of the shell, as is the case in small barnacles and those larger barnacles that happened to have badly eroded tests, permitting an ade- quate view. Observations of feeding behavior associated with wedging (a characteristic lunging movement de- scribed below) allowed me to infer wedging in cases where the snail’s foot and mantle obscured direct observation of the spine. In addition, wedging attacks left scratch marks on the barnacle’s valves. Drilling attacks could only be identified after the fact by the presence of a hole in the test or valves of the barnacle. I avoided handling the snails while drilling; thus, it was not possible to observe drilling directly. It is not known whether the relative hunger of Acanthina angelica modifies its foraging behavior, and con- sequently after any one snail had consumed a barnacle, both the predator and the remains of its prey were re- moved from the aquarium. RESULTS Close observation of feeding behavior established the fol- lowing sequence of events. After encountering a barnacle the snail mounted it and brought the labial spine to the barnacle’s opercular opening. The spine was then inserted into the opening, as if to “feel” for the opercular valves. The proximate stimulus for a wedging attack appears to be the contact of the spine with the opercular valves. Snails observed wedging kept the spine positioned in or above the opercular opening and, with the foot firmly attached to the barnacle, thrust the spine downward, apparently bringing the spine into forcible contact with the natural separation of the barnacle’s opercular valves at a point on the scutum near its junction with the tergum. This “lung- ing” was repeated as often as five times per minute, until either gaining access to the mantle cavity of the barnacle, or giving up. Some snails maintained an attack for up to several hours, although with diminished frequency of J. R. Malusa, 1985 Wore 1s NT X14 @ 80% } o 15 it » 60% Sa 7 o M4 3 a m 40% +10 § fo) 2 Short-spined Acanthina «——»® 5 20% Long-spined Acanthina *——* small medium large Barnacle size Figure 1 Results of the foraging experiment showing the variation in at- tack mode in Acanthina angelica relative to spine length and size of barnacle prey. Sample size is indicated. lunging. These prolonged attacks resulted in an abraded elliptical depression where the barnacle’s valves meet. This artifact of wedging could be mistaken for a drill hole, except it is not circular (as gastropod drill holes are), and it was associated only with lunging behavior. Interesting- ly, three snails that did drill failed because they entered the barnacle at a point above the opercular valves, and another snail completed a drill hole into an empty test. Most snails sequestered with the large barnacles either could not feed or chose not to feed during the entire 27 days of observation; these data were not included in the statistical analyses. Relationships between attack mode, spine length, and the size of prey are shown in Figure 1. These data were analyzed with a G-test (Table 1). The mode of attack is clearly dependent on the barnacle size (G = 20.995, df = 2, P < 0.001). One hundred percent of the small barna- cles were wedged open with the spine, as were 87% of the medium-sized barnacles, and only 43% of the large bar- nacles. Hence, successful use of the spine was dependent on barnacle size. There is also a relationship between attack mode and spine length that varies according to barnacle size. Given small or large barnacles, short-spined and long-spined snails employed similar attack modes (7.e., they both wedged small ones and drilled large ones with similar frequency; P > 0.50). However, for medium-sized bar- nacles, more short-spined snails drilled than did long- spined snails (0.025 < P< 0.05) (these tests represent partitioning of the G that is due to the spine length x Page 3 attack mode and the spine length x attack mode x prey size interaction). Figure 2 shows the time until the first barnacle was consumed for long and short-spined snails on all three size classes of barnacles. Note that this time interval in- cludes the total time from the beginning of the experiment until the first barnacle was consumed, not simply the time from initiation of feeding to completion. Data were ana- lyzed using a Mann-Whitney test for ordinal data by as- signing rank values to the observations in each time in- terval (the “consumption order’). Pairwise comparisons of consumption order show that short-spined snails con- sumed small barnacles sooner than they did medium-sized barnacles (P < 0.001); the medium-sized barnacles were consumed, in turn, sooner than large barnacles (P < 0.001). Long-spined snails showed no significant differ- ence in consumption order between small and medium- sized barnacles (P > 0.20). Large barnacles did take lon- ger to consume than either small or medium-sized bar- nacles (P < 0.001). Also using a Mann-Whitney test, I made comparisons between long- and short-spined snails on a given size class of barnacle. The results show no significant difference in consumption order between long- and short-spined snails when feeding on small barnacles (0.1 > P > 0.05), but that long-spined snails consumed medium-sized barnacles sooner than did short-spined snails (P < 0.001). DISCUSSION The results of the feeding experiment support the hy- pothesis that the labial spine of Acanthina angelica func- tions to force apart the opercular valves of its barnacle prey. Large barnacles were wedged significantly less fre- quently than either medium-sized or small barnacles, and short-spined snails switched to a drilling attack at a small- er prey size (medium-sized barnacles) than did long-spined snails. These results suggest a close relationship between spine length, barnacle size, and mode of attack. As noted above, no attempt was made to control for differences in shell length between short- and long-spined Table 1 Analysis of the relationship between spine length, bar- nacle size, and mode of attack (G-test, SOKAL & ROHLF, 1969). Comparison df G Spine length x barnacle size 2 3.294 Mode of attack x barnacle size 2 20.995* Spine length ). The Reynolds number for a particular situation of flow is defined as Re = 1U/2, where | is some characteristic length of the object, U the ‘Current address: Department of Zoology, University of Cambridge, Cambridge, England CB2 3EJ. fluid velocity, and v the kinematic viscosity of the fluid. As noted by VOGEL (1981), the only objects known that exhibit such an inverse relationship between C, and Re are those that can reshape themselves with increasing ve- locities, such as trees. In this particular range of Reynolds number (10° to 10°) C, « Re®, while for long flat plates parallel to flow, Cp « Re~°> (HOERNER, 1965). The gen- erally conical or ellipsoidal limpet shell in free-stream flow might well be expected to display a relationship more similar to that of a bluff body than that of a flat plate. The present study examines drag forces on limpet shells near a substratum. Drag is related to various morpholog- ical parameters, and the peculiar variation of Cp with Re for limpets within a relatively large boundary layer is noted. MATERIALS anp METHODS Drag forces on shells were measured in a continuously circulating flow tank (VOGEL & LABARBERA, 1978). Ve- locity was calibrated visually with ink at low velocities, and for higher velocities was derived from drag measure- ments of simple geometrical objects with known drag coef- ficients. Maximum tank velocity was 0.45 m/sec. In the working section of the tank (height 8.8 cm, width 10.4 cm), a thin plexiglas plate (20.3 by 8.7 by 0.2 cm, with a bevelled anterior edge) was fixed, parallel to flow, 4 cm below the water’s surface (Figure 1). Limpet shells were filled with hard wax, and were positioned upside-down, 0.85 mm from the lower surface of the plexiglas plate. The center of each shell was 11 cm from the leading edge of the plate. A metal rod (1.5 mm in diameter) ran up- wards from the center of the wax mass, through an open- R. Dudley, 1985 Page 7 Figure 1 Lateral view of the working section of the flow tank. ing in the plate (3 by 0.43 cm), to a thin metal strip (0.03 cm shim stock), in turn fixed firmly to the flow tank plat- form (Figure 2). A pair of strain gauges, attached to either side of the metal strip, were used as elements of a Wheat- stone bridge, the output of which was amplified and dis- played on a digital voltmeter. The voltage drop across the Wheatstone bridge was thus proportional to the drag force acting on the shell; the apparatus was calibrated by hang- ing weights from the end of the attachment rod, with the entire system held vertically. Reorientation of the shell with increased fluid velocity was observed to be minimal. For each shell, drag at a particular velocity was measured three times in each of three orientations: anterior end up- stream, posterior end upstream, and shell lateral to flow. Drag measurements, with allowance taken for drag of the attachment rod, are estimated to be accurate to within 10%, and repeatable to within 5%. The major and minor diameters, height from apex to base, and distances from the apex to the anterior and posterior shell margins were measured for 28 limpet shells with vernier calipers to within 0.05 mm. Perimeter mea- surements were taken from impressions in clay. The amount of water displaced by the shell and wax mass was taken as a measure of shell volume. Table 1 gives these data, along with geographical and ecological information for each limpet species. Also presented in Table 1 are the ratio X of the anterior to posterior distances from edge to apex (a measure of apex eccentricity), the fineness ratio R; (major diameter/minor diameter, one possible measure of pressure-drag streamlining), and the relative shell height hg, the ratio of the shell height to the geometrical mean of the major and minor diameters. The drag of an object is related to the fluid velocity by the formula: D = %CppSU?, where D is the drag, p the fluid density, S a reference area of the object, U the fluid velocity, and C, the aforementioned coefficient of drag, which for a given object and orientation is normally de- termined empirically. In practice, drag data are often re- duced to a plot of Cp versus Re, with some reference cross- sectional or projected area taken for S. However, for a given object in a given medium, C, « D/U? and Re « U. A plot of D/U? versus U will thus yield the same power dependence of C, upon Re, the function being of the form y = ax’. This latter approach is attractive in that any assumptions concerning the dimensions of often high- ly irregular biological objects are avoided. Volume to the Figure 2 Details of the force transducer used in the drag measurements. Attached to one side of the shim stock is a strain gauge, given in black. The other strain gauge is hidden from view. Water flow is perpendicular to the plane of the diagram. Page 8 The Veliger, Vol. 28, No. 1 Table 1 Species identification, locale, habitat type, and morpho- logical data for 28 limpet shells. Shell no. 1 2 14 15 16 Species and locale Scurria scurra (Lesson, 1830) Montemar, Chile Scurria scurra (Lesson, 1830) Montemar, Chile Scurria scurra (Lesson, 1830) Montemar, Chile Patelloida saccharina L. (1758) Ngeyanges, Palau Patelloida saccharina L. (1758) Ngeyanges, Palau Patelloida saccharina L. (1758) Ngeyanges, Palau Patella flexuosa (Quoy and Geimard, 1834) Salafai, Pagan Patella flexuosa (Quoy and Geimard, 1834) Salafai, Pagan Notoacmea insessa (Hinds, 1842) Madison Port, Car- mel, California Notoacmea insessa (Hinds, 1842) Madison Port, Car- mel, California Notoacmea insessa (Hinds, 1842) Madison Port, Car- mel, California Collisella digitalis (Rathke, 1833) Point Pinos, Califor- nia Collisella digitalis (Rathke, 1833) Point Pinos, Califor- nia Siphonaria javanica (Blain- ville, 1827) Rendrag, Pa- lau Siphonaria javanica (Blain- ville, 1827) Rendrag, Pa- lau Siphonaria laciniosa L. (1758) Rendrag, Palau Stphonaria laciniosa L. (1758) Rendrag, Palau Fissurella nimbosa L. (1758) Rockoy, Guadeloupe Fissurella nimbosa L. (1758) Rockoy, Guadeloupe Fissurella nimbosa L. (1758) Rockoy, Guadeloupe Fissurella nodosa (Born, 1780) Fort Point, Jamaica Fissurella nodosa (Born, 1780) Fort Point, Jamaica Nacella sp. Tierra del Fuego Nacella sp. Tierra del Fuego Nacella sp. Tierra del Fuego Patella vulgata L. (1758) Cove, Scotland Habitat type kelp stipes sheltered low intertidal, very heavily exposed kelp stipes high intertidal and splash zones sheltered sheltered highly exposed, sand scour heavy exposure intertidal moderate exposure Table 1 (Continued) Shell no. Species and locale 27 ~—~Patella vulgata L. (1758) Cove, Scotland 28 Patella vulgata L. (1758) Cove, Scotland Habitat type The shell length 1, width w, height h, volume V, perimeter p, fineness ratio Ry, eccentricity X, and the relative shell height h, for 28 limpet shells. All lengths are given in millimeters, shell volume in milliliters. Shell 1 w h Vv p R; x hp QT 2G02 3 Dime SeA 4.1 19) TENS 109 8 20872: Pete ANKE WLS) 2.0 65.0 1.12 0.83 0.56 17.5 15.4 8.4 1.0 52.5 1.14 0.85 0.51 23:2) OAR SIE 13> 70ST GSO: 8 9 OFZ Z1ESs BIS ea eZ 1.1 66.1 1.15 0.83 0.36 Wiss), 1'E)5) 5.4 0.4 48.3 1.30 0.89 0.35 AleSi 29'6R 922 33 : : : Sy) PASS) 6.5 2:9 97:9 NES OF OFA O22 9 16:2 10:2 955)" 1 0:9) 4 2°5 IES SI OlS 25a OME 10) 55 94 85 0.5 40.00 1.64 0.79 0.70 Ve > 14547 O87 8:0 0:3) 39M E49) OSA Ofer 12 20.4 16.6 8.6 1.8 57.3 1.23 0.46 0.47 Sie 212 O24 O ares 1.1 74.8 1.00 0.38 0.37 14) -°22-6 19:2, 16 1:2 65:8) SAW OSS 4E ONS 15) > 2106s S51 92, = 10:9 55.8 1.43 0.82 0.51 TO E2 5222255 6.7 1.0 75.4 1.12 0.90 0.28 17. 21.6 196 65 09 669 1.10 0.88 0.31 18 36:25) 241 1253 eee Ie) O79 O47 190 2756) 1925 8.3 1.7 72.8 1.41 0.80 0.36 20 17.4 103 #63 0.6 45:8 1668) 0:83) 10°47, 21 32.9 21:4 17:8 4.8 90:2) “1547 OISG ss OlGy 22 21.0) 1833 1218 2:6 71:4) IIIS OMS2R Ores 23. 48.2 40.0 24.7 24.0 245 32105 12525) 2067 To 90.5) 1525) 1090S ONS 25, O1eE4= 74254 123 2220 26 41.8 34.6 17.4 10.0 Dit 35:0) 42922 NS 3549. Paes PRY AOR MSGI 40 848 1.21 0.73 0.48 AAYADNAWNH power % could well be the most biologically relevant ref- erence area (VOGEL, 1981), but for craspedophilic organ- isms living in significant boundary layers it may not ad- equately reflect the relative importance of protrusion of the organism above the substrate. In the following anal- ysis, the dependence of C, upon Re will be determined from the equation D/U? = aU’, but, for purposes of com- parison with existing data, drag coefficients calculated on the basis of frontal (cross-sectional) area will also be pre- sented. RESULTS A typical plot of drag versus velocity for shells in anterior, posterior, and transverse orientations is given in Figure 3. R. Dudley, 1985 After a logarithmic transformation of the quantities D/U? and U, linear regression (In y = In a + b In x) was used to determine a and b in the equation D/U? = aU’. Cor- relation coefficients for this regression ranged from 0.91 to 0.99. From the two constants a and b, a “standard” drag for a particular orientation was calculated at a ref- erence velocity of 0.3 m/sec from the formula D = a(0.3)°*?. A mean drag, D, was calculated for each shell from the formula D = (D, + D, + 2D,)/4, where D, is the anterior drag, D, the posterior drag, and D, the trans- verse drag. The mean drag thus approximates from the experimental data the drag for a random orientation. Transverse drag was in all cases calculated only for the left side (with respect to the anterior-posterior) axis of the shell; limpets are generally, with the exception of the si- phonarids, bilaterally symmetric, and for the sample of shells studied (excluding Siphonaria laciniosa and S. javan- ica) there was no significant difference in the shortest dis- tance from the apex to the right and to the left side of the shell margin (x? test, P < 0.01). From the anterior drag, a drag coefficient was calculated from the formula: Cy = 2D/(pSU?), where D and p are as previously given, U equals 0.3 m/sec, and S is the cross-sectional area normal to flow. Finally, considering each of the three orientations equally, a mean b of the power b was calculated for each shell. Table 2 lists for each limpet shell D,, D,, D;, D, G,, and b. Anterior drag was in most cases only slightly less than posterior drag. No correlation between the apex eccen- tricity and the ratio of anterior to posterior drag could be found. A low value of eccentricity is generally regarded as desirable for efficient streamlining (BAYLEY, 1958); ve- locity gradients at surface-fluid interfaces (to be discussed later) and forces of selection independent of drag minimi- zation may well distort the predictive validity of such a result from main-stream fluid mechanics. In this context, it 1s interesting to note that limpets living on kelp stipes experience for the most part unidirectional flow, and might well be expected in the interests of streamlining to have an apex shifted far forward (low X). This is in fact ob- served in the kelp limpet Helcion pellucida (=Patella pel- lucida), with an apex eccentricity of 0.59 (WARBURTON, 1976). It should be mentioned that kelp limpets often pos- sess a convex base corresponding to an excavated concavity in the kelp stipe. For those kelp limpets in the present study (Scurria scurra and Notoacmea insessa), vertical de- viation of the base perimeter was less than 10% of the shell height, and was thus ignored. For a sample of three shells, apex eccentricity in S. scurra was not less than 0.83, while for N. insessa eccentricity was not less than 0.71. These values compare to an overall average (9 species, 28 shells in all) of 0.80. Those species with the lowest apex eccentricities were Collisella digitalis, a limpet found char- acteristically in the intertidal spray zone, and one of the particularly large Nacella shells. Neither species could reasonably be expected to experience only unidirectional flow. A further complication is the general trend towards Page 9 10-0 _ 80 ap) To) = 60 od) © Q 4:0 2-0 0-1 0:2 0:3 0-4 Velocity (m/s) Figure 3 Variation of drag with water velocity for shell no. 25, in the anterior (@), posterior (MM), and transverse (A) orientations. an anteriorly displaced apex among high-shore limpets with respect to their low-shore counterparts (VERMEIJ, 1978). It seems clear that apex eccentricity cannot be en- tirely explained as a consequence of a streamlined profile, and that other factors, such as predation (HOCKEY & BRANCH, 1983) and temperature and desiccation resis- tance, influence this aspect of shell form. Transverse drag for all shells was greater than drag in a longitudinal orientation. Not surprisingly, the fineness ratio R, is inversely correlated with the ratio of anterior and transverse drag (Figure 4). A relatively high R, is clearly desirable in the event of unidirectional water flow, as the limpet can always preferentially orient in the cur- rent. The wave-swept intertidal zone, which can only be loosely described as a region of upward wave motion, is but vaguely reminiscent of a unidirectional current; a very high value of R,; could well be a liability under these circumstances. Given equal frontal area, the drag of bod- ies in free-stream flow is minimized when R, = 2 (ALEXANDER, 1968). Values above two result in a higher total drag by virtue of an ever-increasing “friction drag” resulting from the forces of viscosity acting on the surface area of the body. For an object attached to a surface, the relevant value of R,; is not known, and may not be strictly comparable with the value of two given above. Nonethe- less, of the 28 shells examined, the highest value of R; was 1.68, with a mean of 1.27. Even for the stipe-dwelling limpets, minimization of drag along the anterior-posterior axis may not be the predominant factor determining the ratio of major to minor diameter. Page 10 The Veliger, Vol. 28, No. 1 0:9 O-7 — 05 0:3 0-1 Hina ze ro is Figure 4 The ratio of anterior drag to transverse drag as a function of the fineness ratio R;. The regression line is given by Y = (—0.47)X + 1.21, r= —0.62, P < 0.05. Mean drag was found to be strongly correlated with shell length, height, and volume. The ratio of mean drag to V*, where V equals shell volume, was chosen as a quantity indicative of the relative drag acting on a shell (V* being the biologically relevant reference area). This ratio increases only slightly with respect to relative shell height (D/V”* « h,°*, 7 = 0.41, P < 0.05), and is not cor- related with shell volume. There is admittedly consider- able scatter in the data, but as a rough approximation the relative drag does not appear to increase with greater rel- ative height or volume. Also of interest is the absence of correlation between relative drag and the ratio of the pe- rimeter to the geometrical mean of the major and minor diameters. The latter is a dimensionless measure of the convolution of the shell’s perimeter, and is thereby pro- portional to the amount of shell ribbing. BRANCH & MARSH (1978) reported that a slight roughening of the shell sur- face actually decreases the coefficient of drag (by inducing turbulent flow and thereby delaying flow separation), but that shells with pronounced radial striation experienced a higher relative drag (higher C,,). The present data, which again contain much scatter, cannot be sufficiently resolved so as to distinguish between the relative contributions of drag reduction by the induction of turbulence and the increase in drag brought on by an increase in cross-sec- tional area. The presence of ribbing and pronounced cos- tae does not necessarily indicate a sheltered existence, as many wave-exposed species of limpets and other gastro- pods display strongly sculptured shells (VERMEIJ, 1978). Finally, the relative drag for those species generally living in heavily wave-stressed environments (Patella flexuosa, Fissurella nimbosa, and F. nodosa) was significantly less than that of all other species considered (Mann-Whitney U test, P < 0.05). Although the sample size is small, it may not be incorrect to suggest that selection for drag- reducing features has been greater among wave-exposed species. By virtue of viscosity, fluid velocity near a surface-fluid interface is less than the free-stream velocity. Directly at the interface there is no fluid movement. The thickness of the boundary layer is commonly defined as the distance from the surface at which fluid velocity is equal to 99% of the free-stream value. Assuming laminar flow, the boundary layer thickness 6 for a flat plate parallel to flow is given by: 6 = 5(xu/pU)”, where p and U are as defined previously, u is the dynamic viscosity of the fluid, and x (11 cm) is the distance downstream from the leading edge of the plate (VOGEL, 1981). The assumption of laminar flow for the current situation is justified by the observation that the local Reynolds number (30,000), based upon 11 cm as the characteristic length, is less than the value gen- erally associated with the transition to turbulent flow. Boundary layers in the present experimental arrange- ment, using for x the invariant distance from the leading edge of the plate to the center of the shell, ranged from 3.7 mm at U = 0.2 m/sec to 2.5 mm at U = 0.45 m/sec. These values are not insignificant when compared to the heights of particularly the smaller shells studied (Table 1). Drag of smaller shells should thus increase dispropor- tionately with respect to larger shells as velocity is in- creased, because with the decrease in thickness of the boundary layer at higher velocities, a relatively greater shell area is exposed to the free-stream velocity. Figure 5 illustrates this inverse relationship between b (Cp & Re?) and shell height. For taller limpets, C, is roughly inde- pendent of Re (C, « Re®), while the very large values of b are reserved for very small shells. This finding is inde- pendently corroborated by the variation of b with shell orientation. The power b for shells in a transverse ori- entation was significantly less than that for shells in a longitudinal orientation (Mann-Whitney U test, P< 0.05). For higher absolute magnitudes of drag (transverse drag, for example, always being greater than anterior drag), the relative contribution of drag resulting from a smaller boundary layer is less. The value of b decreases correspondingly. The small values of b recorded for the larger shells may also indicate that the reported values (b = —0.95) of BRANCH & MARSH (1978) could be due to an error in calibration or in experimental design. DISCUSSION HOERNER (1965) presents coefficients of drag for protu- berances within turbulent boundary layers. A round rivet head shows a C, = 0.32, based on the maximum projected area (plan-form) of the head, while a highly streamlined protuberance has a C, of 0.07. The drag coefficients for R. Dudley, 1985 =O2 5:0 10-0 Page 11 15:0 20:0 25:0 Shell height (mm.) Figure 5 The variation of b with shell height. limpet shells (Table 2) are based on cross-sectional area, and reflect partial protrusion through a laminar boundary layer. Since in both cases the mean velocity experienced by the object is less than the free-stream velocity, drag coefficients calculated on the basis of the latter will be less than those calculated with some mean velocity. A drag coefficient could be calculated for smaller cross-sectional regions of the shell, using for U the velocity at the center of each region, and then the results summed over the en- tire cross-sectional area, but the significance of such a compound drag coefficient is not clear. Given the absence of a well-defined method of describing mean velocity rel- ative to shell area, and lacking any data against which to compare drag coefficients calculated in such a manner, all drag coefficients in the present study were calculated with the free-stream velocity. It is nevertheless interesting that the values for limpet shells are generally between 0.07 and 0.32, suggesting that limpets have adopted, in com- parison with a symmetrical round form, a better stream- lined profile. As mentioned above, relative drag is less for those limpets living in wave-stressed environments, indi- cating that there may be selective pressure for the reduc- tion of fluid-dynamic drag. There are two other forces generated by water movement that could potentially be of biological significance, namely shock pressures and accel- eration forces. The former is a transient force correspond- ing to the establishment of flow in a temporarily stopped mass of fluid, while the latter is the force exerted on an object by the acceleration of the fluid displaced by the object (the so-called added mass). CARSTENS (1968) states that, for continuous wave trains in shallow water, accel- eration forces are likely to be negligible for bodies that are small in comparison to the wave height, and the shapes that minimize fluid-dynamic drag also minimize shock pressure, which in any event only rarely reaches extreme values, so that the concomitant drag associated with these forces is likely to be small for limpet shells (see however DENNY [1982] for description of a more complicated sit- uation of flow). It should be noted that limpets are easily dislodged by wave surges if they have not anticipated, by means of clamping down in response to low velocity cur- rents, a strong current flow (WARBURTON, 1976). The hydrodynamic significance, if any, of shell sculp- ture remains unclear, given the absence of correlation be- tween relative shell drag and convolution of the shell pe- rimeter. Pronounced ribbing may of course have other roles. VERMEIJ (1973) noted the presence of increased shell sculpture among sun-exposed limpets, and JOHNSON (1975) demonstrated that shells of Collisella digitalis have slightly lower convective coefficients than shells of C. sca- bra, which are more ribbed and spinose. Strong shell sculpture reduces vulnerability to crushing predation in many gastropods (VERMEIJ, 1978), and in limpets has been specifically suggested as a deterrent to bird predation (GLYNN, 1965). The rough surfaces of many limpet shells may also be implicated in the reduction of the forces of shock pressure through the entrainment of air (CARSTENS, 1968), and in a redistribution of instantaneous drag forces. The expression of shell sculpture may thus be determined by a number of independent factors, and any explanation Page 12 Table 2 Drag forces in newtons (X10~*) for anterior, posterior, and transverse orientations (D,, Dp, D;), mean drag D, the coefficient of drag (C,), and the mean power b. Shell D, D> D; D Cp b 1.61 1.70 2.61 2.13 0.17 0.40 0.90 0.89 PA 1.08 0.18 0.53 0.49 0.44 0.90 0.68 0.17 0.85 0.93 0.91 1.27 1.09 0.24 0.52 0.56 0.70 0.77 0.70 0.19 0.80 0.35 0.33 0.48 0.41 0.22 1.14 0.87 1.00 1.70 1.32 0.14 0.51 0.62 0.85 1.06 0.90 0.16 0.50 9 0.30 0.32 0.70 0.51 0.14 1.09 10 0.21 0.35 0.68 0.48 0.12 0.82 11 0.17 0.25 0.40 0.31 0.10 155 12 0.92 0.97 137, 1.16 0.14 0.72 13 0.72 0.84 1.35 1.06 0.20 0.66 14 0.86 0.80 1.04 1.87 0.09 0.83 15 0.47 0.53 0.96 0.73 0.15 0.64 16 0.42 0.60 0.72 0.61 0.12 1.20 17 0.43 0.45 0.60 0.52 0.15 1.28 18 1.91 2.13 3.62 2.82 0.29 0.29 19 1.13 1.12 1.96 1.54 0.31 0.50 20 1.36 1.41 2.11 1.75 0.47 0.84 21 0.54 0.57 E22 0.88 0.06 0.21 22 0.25 0.29 0.55 0.41 0.05 0.37 23 7.35 7.37 10.09 8.72 0.33 0.14 24 3.81 3.83 4.80 4.31 0.32 —0.01 25 3.68 Se 8.88 6.30 0.32 —0.16 26 3.62 3.66 4.53 4.09 0.27 0.04 27 2.32 2.65 3.41 3.26 0:27 =0:03 28 2.09 2.12 2.82 2.46 0.29 —0.28 ANANDANAWND of the observed intraspecific, interspecific, and geograph- ical variation in limpet shell sculpture that is solely con- cerned with fluid-dynamic drag seems at this time un- warranted. In the present experimental design, the small gap be- tween the limpet shell and the fixed surface (0.85 mm) precludes direct identification of measured drag forces with those likely to be encountered in the field. This small distance, however, is well within the boundary layer as calculated above, and there is no reason to suspect that the very low fluid velocities in this region will distort the validity of comparisons of drag forces on different shells. Velocities required for the dislodgement of living animals are substantially greater than those used in the present work (WARBURTON, 1976), and it is possible that drag coefficients at these velocities will differ from those re- ported herein. Boundary layers in the field, however ill- defined in the context of breaking waves and rock surfaces, may well be smaller than several millimeters. In this case, the variation of b with height will be of relevance to yet smaller shells, and may only be of importance in larval settlement and survival, albeit at much lower Reynolds numbers. Intertidal boundary layers can also, under con- The Veliger, Vol. 28, No. 1 ditions of turbulent flow and steady wave trains, be much larger than the shell heights considered here. The result that the relationship between C, and Re can, in certain circumstances, vary according to the height of the object above the substrate has not been previously reported, and could represent an additional consideration in the adap- tation of organisms to drag forces in the intertidal zone. ACKNOWLEDGMENTS I am indebted to Dr. Steven Vogel, who gave valuable assistance and advice in both the construction of equip- ment and the analysis of results. Dr. Geerat Vermeij of the University of Maryland provided the majority of lim- pet shells used in this study. Elizabeth Dudley kindly supplied the shells from Cove, Scotland. LITERATURE CITED ALEXANDER, R. MCN. 1968. Animal mechanics. Sidgwick & Jackson: London. 346 pp. BAYLEY, F. J. 1958. An introduction to fluid mechanics. Allen & Unwin: London. 215 pp. BRANCH, G. M. & A. C. MarsH. 1978. Tenacity and shell shape in six Patella species: adaptive features. J. Exp. Mar. Biol. Ecol. 34:111-130. CARSTENS, T. 1968. Wave forces on boundaries and sub- merged bodies. Sarsia 34:37-60. Denny, M. W. 1982. Forces on intertidal organisms due to breaking ocean waves: design and application of a telemetry system. Limnol. Oceanogr. 27(1):178-183. DurRRANT, P. M. 1975. An investigation into the effect of running water on shell dimension in Ancylus fluviatilis Mul- ler. J. Conchol. 28:295-300. GLYNN, P. W. 1965. Community composition, structure, and interrelationships in the marine intertidal Endocladia muri- cata-Balanus glandula association in Monterey Bay, Califor- nia. Beaufortia 12:1-198. GRAHAM, A. & V. FRETTER. 1947. The life history of Patina pellucida (L.). J. Mar. Biol. Assoc. U.K. 26:590-601. HELLER, J. 1976. The effects of exposure and predation on the shell of two British winkles. J. Zool. (Lond.) 179:201- 213. Hockey, P. A. R. & G. M. BRANCH. 1983. Do oystercatchers influence shell shape? Veliger 26(2):139-141. HOERNER, S. F. 1965. Fluid-dynamic drag. S. F. Hoerner: 2 King Lane, Greenbriar, Bricktown, NJ. Jounson, S. E. 1975. Microclimate and energy flow in the marine rocky intertidal. Jn: D. M. Gates & R. B. Schmerl (eds.), Perspectives of biophysical ecology. Springer Verlag: New York. KITCHING, J. A. & J. Lockwoop. 1974. Observations on shell form and its ecological significance in thaisid gastropods of the genus Lepsiella in New Zealand. Mar. Biol. 28:131- 144. Lewis, J. R. & R. S. Bowman. 1975. Local habitat-induced variations in the population dynamics of Patella vulgata L. J. Exp. Mar. Biol. Ecol. 17:165-203. STRUHSAKER, J. W. 1968. Selection mechanisms associated with intraspecific shell variation in Littorina picta (Prosobran- chia: Mesogastropoda). Evolution 22:459-480. VERMEIJ, G. J. 1973. Morphological patterns in high-inter- R. Dudley, 1985 tidal gastropods and their limitations. Mar. Biol. 20:319- 346. VERMEIJ, G. J. 1978. Biogeography and adaptation: patterns of marine life. Harvard University Press: Cambridge. 332 PP- VOGEL, S. 1981. Life in moving fluids. Willard Grant Press: Boston. 352 pp. Page 13 VOGEL, S. & M. LaBaRBERA. 1978. Simple flow tanks for research and teaching. BioScience 28:638-643. WARBURTON, K. 1976. Shell form, behaviour, and tolerance to water movement in the limpet Patina pellucida (L.) (Gas- tropoda: Prosobranchia). J. Exp. Mar. Biol. Ecol. 23:307- 325) The Veliger 28(1):14-17 (July 1, 1985) THE VELIGER © CMS, Inc., 1985 The Effects of Aggregations on Water Loss in Collisella digitalis WM. BRAD GALLIEN' University of California, Berkeley, Berkeley, California 94720 Abstract. Water loss rates were compared between isolated and aggregated Collisella digitalis of the same microhabitat. The solute concentration of the extra-corporeal water (ECW) was used as an indicator of water loss. Limpets within conspecific aggregations tended to have lower ECW solute concentrations than isolated limpets. Isolated limpets tended to orient with their head down on vertical surfaces, although this tendency was absent in the aggregations. INTRODUCTION HIGH-INTERTIDAL MARINE organisms are exposed to air with each tidal cycle. Prolonged calm periods and neap tides can increase the length of exposure, isolating some splash-zone animals in an essentially terrestrial environ- ment for several days (WOLCOTT, 1973). While exposed, marine animals are vulnerable to evaporative water loss due to wind and solar radiation. Many animals subject to these conditions have physiological adaptations, such as high desiccation tolerance, that allow persistence in the intertidal. In addition, these animals may have behaviors that affect the physical parameters governing water loss, thereby reducing desiccation stress. Collisella digitalis (Rathke, 1833), a high intertidal snail, demonstrates many behaviors that impede water loss when it is exposed to air. This acmaeid limpet occurs from the Aleutian Islands, Alaska, to the southern tip of Baja Cal- ifornia (MorRIs et al., 1980), predominately on vertical rock surfaces (HAVEN, 1971; COLLINS, 1976). Unlike the homing limpet C. scabra, C. digitalis does not fit exactly to the substrate. This leaves a gap between the shell and substrate through which water is lost. The mucous sheet formed by C. digitalis, and other limpets, significantly re- duces this water loss, acting as a physical barrier to ex- change (WOLCOTT, 1973). Collisella digitalis often aggre- gates in crevices and depressions during low tide (FRANK, 1965). Presumably the topographic relief provided by these sites reduces water loss rates, although this has yet to be demonstrated. Collisella digitalis also forms conspecific ag- ' Present address: Department of Zoology, University of Ha- wail, Manoa, Honolulu, Hawaii 96822. gregations on smooth rock surfaces which lack such to- pographic relief. HAVEN (1971) suggested that these ag- gregations reduce water loss from the animals but provided no supporting data. In this study, I have examined the effects of clumping on water loss in Collisella digitalis by comparing solute concentrations of extra-corporeal water (ECW) between clumped and isolated limpets. The extra-corporeal water, held between the foot and shell, is the source of evapo- rative water loss from limpets (SEGAL, 1956). As fresh- water is evaporated from the ECW, its solute concentra- tion increases. Although ECW solute concentration is an indirect assessment of water loss, it is valuable in that data can be collected in the field. MATERIALS anp METHODS Specimens of Collisella digitalis were sampled during May, 1983, at three sites along the northern California coast: (a) Blind Beach, Sonoma County; (b) the Bodega Marine Laboratory, Sonoma County; and (c) Albion, Mendocino County. Most of the samples were taken from the smooth chert at Blind Beach. Individuals were sampled zn situ at varying temperatures, wind conditions, and times of day. All samples were taken from smooth surfaces, well away from any crevices or depressions. Freehand sketches of aggregations and nearby isolates served as permanent references of the position of each individual sampled. Usually eight limpets were sampled: four from within the aggregation, and four isolates from the surrounding area. For comparison, an isolate needed to be at least 4 cm from its nearest neighbor, but not more than 30 cm from the aggregation. This assured that the eight limpets, hereafter treated as a class, were from a similar microhabitat. Wm. B. Gallien, 1985 Table 1 Solute concentration data for aggregated (Cl) and iso- lated (Is) individuals of Collisella digitalis. Os- Number Average es of osmotic PISS Relative change individ- pressure ae in concentration uals (mOsm/kg) ee with time? Class! Cl Is Cl Is ence P? Cl Is x4 1362 1516 156 >0.20 245 399 0.385 1500 1723 223 <0.01 385 606 0.368 1420 1730 310 0.18 303 613 0.506 1180 1370 190 0.14 63 253 0.751 LOA SS Ss li, <0'01 > = 1041 1118 77 0.02 ~-— — — 986 966 —-20 >0.20 — — — 1160 1176 16 >0.20 43 59 0.271 1187 1330 143 <0.01 70 213 0.671 1120 1243 123 <0.01 3. 126 0.976 1126 1280 154 0.18 9 163 0.945 OoMmMAIAADANHPWN 12 LATS S287 7.2 0.15 98 170 0.423 13 1135 1175 40 >0.20 18 58 0.690 14 1142 1205 63 0.17 25 88 0.716 15 1200 1375 175 >0.20 83 258 0.678 16 1230 1762 532 <0.01 113 645 0.825 17 1153 1460 307 0.09 36 343 0.895 18 1233 1355, 222) —<0:01 16 238 0:933 19 1206 1355 149 <0.01 89 238 0.626 1106 1147 41 <001 — — — 1OSZ MINA 55> <0101)- os RARRARR RHR ERKHWRARWWAUWWUA RHA AHAAAEAAWHKWAWWUDAWWWW ‘Clumped and isolated individuals occupying the same micro- habitat. 2 P-value for Student’s ¢-test. Comparisons were between clumped and isolates’ average osmotic pressure. > Derived by subtraction of estimated starting concentration (1117 + 39 mOsm/kg) from average osmotic pressure. * Solute concentration ratio for Cl:Is as calculated using Equa- tion 1 (see text). Limpets were removed from the substrate with a nar- row spatula. A 16-~L sample of ECW was quickly re- moved from the posterior foot-shell margin with a micro- capillary tube. The full tube was then capped with Seal Ease clay to inhibit further evaporation from the sample, and numbered to correspond with the sketch. The length of each individual was measured using calipers, after which the limpet was dipped in seawater and returned to the rock. The vapor pressure of each sample was measured in the laboratory using a Wescor Inc. 5130C Vapor Pres- sure Osmometer, which gave the ECW vapor pressure in milli-osmoles per kilogram (mOsm/kg). Vapor pressure of a sample is directly proportional to solute concentra- tion. All limpets on one rock (n = 211) were censused to determine their orientation. The direction of the anterior end of each limpet was scored according to an eight point Page 15 compass. An individual with its head straight up was scored as 1, straight down, as 5. One-way Chi-square tests were used for each group (clumped and isolates) to determine whether the orientation of individuals was random. RESULTS In all but one of the 21 classes sampled, the average ECW solute concentration of clumped limpets was lower than that of nearby isolates (Table 1). The range of differences between average solute concentrations was —20 to +532 mOsm/kg. In this pairwise comparison within classes, the size of the individuals was not considered. However, the volume of ECW determines the rate at which the concen- tration changes. To assess the role of size, an analysis of covariance was run using size as the covariate. Pooling the data in this way demonstrated that the ECW solute concentration of clumped limpets was lower than that of isolates (F,,,; = 19.85, P < 0.01). In addition, limpets in aggregations were arbitrarily ranked according to the de- gree to which they were surrounded. A completely sur- rounded individual was ranked 4, a limpet with no close neighbors was ranked 0. An analysis of covariance using the ranked positions as a covariate to solute concentration demonstrated that limpets with few close neighbors had higher solute concentrations than those that were com- pletely surrounded. Consequently, those limpets at the periphery of the aggregation (usually rank 2) tended to have higher ECW solute concentrations than individuals in the interior (F,,, = 4.38, P < 0.05) (Figure 1). Due to the paired nature of the sampling procedure, relative rates of water loss between the two groups can be estimated. Fourteen limpets, eight of which were aggre- gating, were sampled to estimate the ECW concentration at the beginning of the dichotomy. The ECW solute con- centration for these limpets was 1117 + 39 mOsm/kg. The concentration change over the exposure period can thus be calculated by subtraction, and the values used to compare average concentration changes of the clumped and isolated limpets given by Equation 1 as im: I x where C and I are the change in ECW solute concentra- tion for the clumped and isolated limpets respectively. The average value for X was 0.67 + 0.22 for all the classes used in the comparison (five were omitted because their final ECW solute concentrations were lower than the as- sumed starting value). Assuming that all animals in each class had been exposed for approximately the same period of time, the proportion X represents a difference in water loss rates. As such, the clumped limpets lost water 33% slower than isolates (Table 1). Analysis of orientation on the substrate (Figure 2) re- vealed that neither the clumped nor the isolated limpets were randomly situated (x? = 20.33, P < 0.05 for clumped; Page 16 The Veliger, Vol. 28, No. 1 Figure 1 Diagram of a typical aggregation of Collisella digitalis and surrounding isolates. The ECW solute concentration (mOsm/kg) and rank (in parentheses) of some of the limpets from Class 19 are given. The positions of the isolated limpets have been changed to reduce the figure size. x? = 143.58, P < 0.01 for isolates). The trend for the PUSS OON isolates was to orient with the head down. The trend for Clumping has been shown to reduce water loss rates in the clumped limpets was less obvious, but the most com- both marine and terrestrial invertebrates. WARBURG (1968) mon orientation was with the head to the right. and ALLEE (1926) showed that terrestrial isopods in ag- Z\ N NN CLUMPED ISOLATES Figure 2 The orientation of Collisella digitalis, clumped or isolated, on vertical surfaces. The percentage figures are the relative frequency of each position with respect to vertical. The striped bar extends in the same direction as the anterior of the limpet. (n = 94 for aggregated, clumped, limpets; n = 117 for isolates.) Wm. B. Gallien, 1985 gregations lose water at one-half the rate of isolates. In addition, clumped hermit crabs survived longer under des- iccating conditions than isolated crabs (SYNDER-CONN, 1981). The data presented in this paper clearly show a similar trend for Collisella digitalis: clumped limpets lost less water during the exposure period than isolated lim- pets. The reduction of water loss is most likely attribut- able to reduced wind velocities within the aggregation. Perimeter animals exposed to the wind would slow wind speed due to friction (drag). This notion is supported by the finding that perimeter animals (rank 2) have higher ECW solute concentrations than limpets within the ag- gregation. However, I did not resolve which solutes re- sulted in the ECW concentration changes. Thus, it is pos- sible that the observed concentration differences are due to the addition of other solutes such as urine to the extra- corporeal water. I have conducted preliminary studies on the biophysics of this system. The data gathered so far support the hypothesis that wind is the primary effector of water loss, and that wind velocities are lower within the aggregation. There is indirect evidence that further supports the hy- pothesis that isolated limpets are under more extreme des- iccation stress than clumped limpets. WOLCOTT (1973) reported that Collisella digitalis produces a mucous sheet that impedes water loss. I observed, but did not sample, several isolated limpets with mucous sheets, and I saw no clumped limpets with the barriers. Those isolates with the barriers had little extra-corporeal water, suggesting that the mucous sheets form as the ECW dries, preventing further water loss from the body tissues. The orientation of isolated limpets may also indicate the severity of their condition. ABBOTT (1956) proposed that the head-down orientation observed in Lotta gigantea would ensure that the head and ctenidium would be the last to dry as the ECW volume decreased due to evapo- rative losses. The present work supports this hypothesis in that isolated limpets tended to orient with their head down on vertical surfaces (see also MILLER, 1968). In addition, some isolated limpets that had been subjected to acute dehydration, where almost all of the ECW was re- moved, were dried and discolored except in the head re- gion. Interestingly, the tendency to orient head down is absent in clumped limpets, suggesting that the rigors of desiccation are reduced within an aggregation. Neverthe- less, clumped limpets are subjected to evaporative water loss; however, position within the aggregation may be a more important factor governing their orientation. The behavior of Collisella digitalis is consistent with the finding that the aggregations represent areas of reduced water loss. The crevices and depressions where the limpets often occur stay moist throughout the tidal cycle, limiting water loss because of moist air, cool temperature, and perhaps reduced wind speeds. ALLEE (1926) reported that isopods form aggregations when ordinary shelter is un- available, thus satisfying the same tactile reaction. Simi- Page 17 larly, C. digitalis forms conspecific aggregations on smooth rock surfaces where no topographic shelter exists. This suggests that the aggregations are analogous to crevices with respect to shelter, and perhaps desiccation resistance. The combination of aggregation, exploitation of crev- ices, and formation of mucous sheets, probably accounts for the persistence of Collisella digitalis in the high inter- tidal. The inability to fit exactly to the substrate renders the limpets vulnerable to environmental fluctuations. By forming aggregations, C. digitalis can effectively reduce the rates of evaporative water loss, thus reducing the de- gree of physiological stress normally imposed by its en- vironment. ACKNOWLEDGMENTS I would like to thank Drs. Gary Adest and Val Conner for their advice on experimental design and Drs. Peter Frank, Ralph I. Smith, Victor Chow, E. Alison Kay, and Christopher Womersley for their careful readings of this manuscript. This research was conducted for Biology 100, a course offered by the University of California, Berkeley, through the Bodega Marine Laboratory. I am also in- debted to all those involved in the course, students and faculty, for their support, particularly my friend Chad Hewitt. LITERATURE CITED AppoTT, D. P. 1956. Water circulation in the mantle cavity of the owl limpet, Lottia gigantea Gray. Nautilus 69(3):79- 87. ALLEE, W. C. 1926. Studies in animal aggregation: causes and effects of bunching in land isopods. J. Exp. Zool. 45:255. Co.uins, L. S. 1976. Abundance, substrate angle and desic- cation resistance in two sympatric species of limpets, Colli- sella digitalis and Collisella scabra. Veliger 19:199-203. FRANK, P. W. 1965. The biodemography of an intertidal snail population. Ecology 46:831-844. HAVEN, S. B. 1971. Niche differences in the intertidal limpets Acmaea scabra and Acmaea digitalis (Gastropoda) in central California. Veliger 13(3):231-248. MILLER, A. C. 1968. Orientation and movement of the limpet Acmaea digitalis on vertical rock surfaces. Veliger 11(suppl.): 30-44. Morris, R. H., D. P. ABBoTT & E. C. HADERLIE. 1980. In- tertidal invertebrates of California. Stanford University Press: Stanford, Calif. 241 pp. RATHKE. 1833. Acmaea. In: Eschscholtz’s “Zoologischer At- las,” Heft 5, pp. 16-21, 23-24. SEGAL, E. 1956. Adaptive differences in water holding capacity in an intertidal gastropod. Ecology 37:174. SYNDER-COoNN, E. K. 1981. The adaptive significance of clus- tering in the hermit crab Clibanarius digueti. Mar. Behav. Physiol. 8:43. WarBurG, M. R. 1968. Behavioral adaptations of terrestrial isopods. Amer. Zool. 8:545. Wo cotTT, T. G. 1973. Physiological ecology and intertidal zonation in the limpets (Acmaea): a critical look at limiting factors. Biol. Bull. 145:389-422. The Veliger 28(1):18-27 (July 1, 1985) THE VELIGER © CMS, Inc., 1985 Relationship between Allometric Growth, with Respect to Shell Height, and Habitats for ‘Two Patellid Limpets, Nacella (Patinigera) macquariensis Finlay, 1927, and Cellana tramoserica (Holten, 1802) R. D. SIMPSON Department of Zoology, University of New England, Armidale, New South Wales 2351, Australia Abstract. The relationship between allometric growth of the shell (with respect to height: length) and environmental influences of water turbulence and desiccation were examined in two patellid limpet species, Nacella (Patinigera) macquariensis Finlay, 1927, and Cellana tramoserica (Holten, 1802), from two widely different climatic regimes. The allometric intensity of increase of shell height in relation to length and increases in the relative shell heights (comparisons of height:length ratios in different allometric groupings) were found to be correlated with increasing water turbulence, especially in N. (P.) macquariensis. This contrasted with suggestions that the major environmental influence on the height of limpet shells is from desiccation. The results supported an existing hypothesis that the height: length ratio of the shells of limpets is influenced by the frequency with which a limpet is obliged to adhere strongly to the substrate. Other possible influences on the allometric growth, with respect to shell height, are discussed. Different shell forms in predation middens of Dominican gulls (Larus dominicanus Lichtenstein) were used to interpret the selection of limpets by the gulls. INTRODUCTION RELATIONSHIPS BETWEEN THE shell heights of intertidal limpets and their environment have been examined in a number of studies (RUSSELL, 1907; ORTON, 1932; EBLING et al., 1962; BALAPARAMESWARA RAO & GANAPATI, 1971; WALKER, 1972; VERMEIJ, 1973, 1980; BRANCH, 1975; BANNISTER, 1975; WARBURTON, 1976; BRANCH & MArsH, 1978; LOWELL, 1984). In order to determine whether the shells of some limpets are relatively higher or lower than others, some form of standardization of the height is nec- essary. Usually shell height has been standardized against shell length, although other ratios have been employed, for example, shell height: geometrical mean of the major and minor diameters of the shell base (VERMEIJ, 1973). Early studies simply compared ratios of shell height: shell length over a range of lengths (RUSSELL, 1907; ORTON, 1932). Limpets usually exhibit allometric growth, resulting in changing proportions between shell height and length; the height:length ratio increases with increasing size, al- though an isometric pattern with increasing size has been recorded in some species (BRANCH, 1975). Consequently, in later studies, more complete comparisons between groups of limpets employ comparisons between regressions of shell height versus shell length over a range of limpet sizes. Different methods, objectives, and descriptive terms used in previous studies can lead to confusion in comparisons across them. Consequently, I will use the following terms: (1) allometric intensity of shell-height increase—the con- tinuum of shell height regressed against shell length for a range of lengths (7.e., the slope of a regression equation), (2) relative shell height—height : length proportions of shells in comparisons between allometric continuums, (3) height ratio of shells—height: length proportions of shells where allometry has not been considered. In previous studies, comparisons between limpets using the above criteria have attributed larger height ratios and relative shell heights to stress from desiccation (ORTON, 1932; Davies, 1969; BALAPARAMESWARA Rao & GANA- PATI, 1971; VERMEIJ, 1973, 1978; BANNISTER, 1975; BRANCH, 1975), increased water turbulence (EBLING et R. D. Simpson, 1985 LORD NELSON REEF °..: HANDSPIKE ga = Page 19 HASSELBOROUGH BAY oo: AN ARE. NORTH END OF MACQUARIE ISLAND SCALE OF KILOMETRES 0 1 60 METRE CONTOURS Figure 1 North end of Macquarie Island. Collections of Nacella (P.) macquariensis were made in the vicinity of the ANARE base. al., 1962; WALKER, 1972), and slower growth rates (VER- MEIJ, 1980). The aim of the present study was to investigate the associations of desiccation and water turbulence with the allometric growth of shell height of limpets from two widely separated localities in different climatic regimes. On sub- Antarctic Macquarie Island (54°38'S, 158°53’E) Nacella (Patinigera) macquariensis Finlay, 1927, was the study animal. Variations in the height ratios of shells of N. (P.) macquariensis were noted in previous collections examined by DELL (1964). The other species, Cellana tramoserica (Holten, 1802), was collected at Arrawarra, northern New South Wales (3093'S, 153°12’E). The heavy predation on Nacella (P.) macquariensis by Dominican gulls (Larus dominicanus Lichtenstein) and the resultant availability of shells from gull middens, nest sites, and roosts gave rise to a connected, subsidiary aim to the study. If limpet shell shape proved to be correlated with habitat, then shell characters of a predation sample could indicate that part of the limpet population particularly susceptible to predation by the gulls. SITES, MATERIALS, anp METHODS Specimens of Nacella (P.) macquariensis were collected from six habitats in the region of the isthmus at the north- ern end of Macquarie Island (Figure 1): (1) rock surfaces in the eulittoral zone on the west coast exposed to the Page 20 O ISO The Veliger, Vol. 28, No. 1 PACIFIC OCEAN ROCK PLATFORM =: SAND Figure 2 Arrawarra Headland. The numbers on the figure correspond to the collection localities for Cellana tramoserica, outlined in the text. open sea; (2) rock surfaces in the eulittoral zone on the east coast exposed to the open sea; (3) rock surfaces at the top of the sublittoral zone on the east coast exposed to the open sea; (4) deep rock pools in the eulittoral zone on the east coast; (5) high rock pools situated approximately 3.5 m above the waterline on the east coast; (6) at a depth of 6 m on the east coast (‘diving station’’). Shells of limpets preyed upon by Dominican gulls (Larus dominicanus) were collected and divided into two categories: (a) shells from Dominican gull feeding sites where the gulls pecked out the flesh and left the shell behind and (b) shells from nest sites and roosting areas where the gulls regurgitated the remains, having swallowed the limpets whole. Two feed- ing sites were sampled in the “pecked out” predation cat- egory. Limpets in habitats (1) and (2) were exposed to a high degree of turbulence from breaking waves and were sub- jected to potential desiccation only during emersion in windy and/or sunny conditions, the latter condition being very rare on Macquarie Island (the prevailing winds are from the west and, consequently, wave action is usually more severe on the west coast; however, for some months the turbulence on the east coast, estimated by a combi- nation of wave height and wave frequency, is comparable to that of the west coast [SIMPSON, 1976]). Limpets in habitat (3) were also exposed to a high degree of turbu- lence both from breaking waves and water flow; stress from desiccation was virtually nonexistent. Habitats (4), (5), and (6) presented no desiccation problems for limpets. At habitat (6) water currents could be strong, but the forces would be much less than that of wave action in the littoral zone. Habitat (4) was subjected to very little water movement, from flow into the pools at high water. The pools from habitat (5) represented a rare situation and, hence, provided limited data. These pools were located high on a steeply sloping face that received wave splash during moderate to heavy seas. The steep aspect prevented fouling by deposited kelp or from seals, as was the case for high rock pools (in which no limpets were found) on more gently sloping rock platforms. The effect of turbu- lence in these latter two habitats was negligible. The headland from where Cellana tramoserica was col- lected is shown in Figure 2. The prevailing seas are from the southeast, causing the eastern and southern side of the headland to be exposed to heavier wave action than the northern side. There were five collection sites: (1) the bottom of the range of the limpets (lower part of the eu- littoral zone) on the northern shore of the headland in an area partly sheltered by a fringing reef; (2) rock surfaces exposed to heavy wave action in the barnacle zone (Tes- seropora rosea [Krauss]) on the northern shores of the headland; (3) rock surfaces in the barnacle (7. rosea) zone on the eastern point of the headland; (4) a gently sloping rock platform in the upper eulittoral zone on the northern shore of the headland; and (5) rock outcrops in the upper eulittoral zone on the beach immediately south of the headland. The above habitats were classed as follows: (1) little influence from either desiccation or wave action, (2) R. D. Simpson, 1985 Page 21 moderate influence from desiccation and subjected to heavy wave action, (3) moderate influence from desiccation and heavier wave action than for (2), (4) frequently subjected to air exposure and desiccation with only moderate wave action, and (5) subjected to both desiccation and very heavy wave action. If limpets regularly moved from one habitat type to another, possible effects on shell form from particular en- vironmental factors of any one habitat would be masked. Consequently, specimens of both Nacella (P.) macquarien- sis and Cellana tramoserica were marked to record the amount of movement between habitats. RESULTS The movement of marked specimens in different habitats over a one year study period showed that Nacella (P.) macquariensis tends to live in fixed areas. In the eulittoral zone, individuals mostly remained in a small area (<1 m?); movements greater than 3 m were rare, usually being horizontal movements within the zone. Limpets in rock pools in the eulittoral zone showed a very high constancy of location. Limpets at the top of the sublittoral zone showed the greatest amount of movement. They were al- ways in the same general area, but individuals sometimes moved into the eulittoral zone above the region of Durvil- lea antarctica (Chamisso) Hariot holdfasts, which repre- sents a sublittoral fringe. Transect counts over one year showed seasonal variations in numbers, but there were no seasonal migrations of populations between eulittoral and sublittoral zones, such as shown by WALKER (1972) for the Antarctic limpet Nacella (Patinigera) concinna (Stre- bel) where migration down the shore was correlated with low temperatures and the formation of an ice film on the shore. Specimens of Nacella (P.) macquariensis at diving sta- tions were not marked for studies of movement. However, the heavy and normal encrustation of coralline algae on specimens indicated that they had been continuously sub- merged at the depth of collection. Limpets from the top of the sublittoral zone had a sparser covering of coralline algae, and limpets from the eulittoral zone either had no, a poor, or (when in shallow pools) a gnarled growth of coralline algae on the shells. The only coralline algal cover on the shells, similar to that for diving station limpets, was found on specimens from deep pools in the lower littoral and sublittoral zones. Encrustations of the tube- worm Spirorbis aggregatus Caullery & Mesnil were fre- quently but not always found on limpets in eulittoral rock pools, and provided some indication of type of habitat. The constancy of location exhibited by N. (P.) macquar- iensis was sufficient to ensure that the samples represented populations from the selected sites. For Cellana tramoserica, marked specimens were found to remain within the habitat areas selected over a period of nine months. In addition, the encrustations of living algae and living barnacles indicated that the limpets were Table 1 Regression analysis of shell height (y) on shell length (x) for Nacella (P.) macquariensis from all categories. Regression equation : (log y = log a Size range fe (lengths, BSS 2) Category mm) n log a b 17.9-47.2 16 -—0.771 1.063 NG2=4725 26" -— 0720" 128 18.0-45.1 131 —0.807 1.166 Deep rock pools 29.1-42.0 34 —0.933 1.194 Eulittoral, east coast 20.2-49.0 102 —0.974 1.322 Predation (pecked out #1) 18.5-44.8 58 -—1.039 1.336 Predation (pecked out #2) 25.7-49.5 95 —1.039 1.371 Top of sublittoral, east coast 22.6-42.1 41 —1.369 1.541 Eulittoral, west coast 20.6-60.7 66 —1.430 1.583 High rock pools Predation (regurgitated) Diving station long-term inhabitants of the lower littoral and the bar- nacle zones, respectively. Conversely, limpets from the up- per eulittoral zone had no encrustations, indicating that at least they had not recently come from the other two habitats. In studies of C. tramoserica near Sydney over a period of 20 months, FLETCHER (1984) also found that marked limpets had not moved between four shore sub- divisions: high, mid, and low intertidal and subtidal (there were no subtidal populations at the headland in the pres- ent study). As well, Mackay & UNDERWOOD (1977) have shown that a proportion of C’ tramoserica populations ex- hibits homing behavior. Two models of the allometric relationship between shell height (y) and shell length (x) were fitted for all sets of data using the Minitab statistical package (RYAN et al., 1981): (1) a linear regression model (y = a + bx) and (2) the linear regression form (log,y = log,a + b log, x) of the power relationship y = ax’. For both species and all sites, the power relationship proved to be a better fit to the data, where b = slope and log,,a = intercept for each line. For each species, particular models, ranging from a single regression for all data to separate regressions for each habitat/category, were fitted using the generalized linear interactive model computer program, GLIM (Ba- KER & NELDER, 1978). Mean deviances were compared to test the fit of particular models. Variance ratios were determined as (difference of deviances) /(mean deviance of the fuller model) on m:(n — m — 1) degrees of freedom, with m being the number of coefficients (slopes and in- tercepts) in the model and n being the total number of data considered. For the nine sites from which Nacella (P.) macquariensis was collected, the relationships between shell height and shell length are given by the regressions in Table 1. To determine the most appropriate combination of regres- sions, the following models were tested: Page 22 ithe Veligers VolyZ3 Now Table 2 Combination of categories of Nacella (P.) macquariensis into separate groups on the basis of slope. Intercepts Common Group (log a) slope (b) (1) High rock pools —0.885 Deep rock pools —0.848 1.139 Diving station —0.767 ‘ Predation (regurgitated) —0.737 (2) Predation (pecked out #1 + pecked out #2) —1.049 1.344 Eulittoral, east coast —1.008 : (3) Eulittoral, west coast + top of sublittoral, east coast —1.430 1.582 Model I: 1 slope, 1 intercept (a single regression); Model II: 1 slope, 9 intercepts; Model III: 9 slopes, 1 intercept; Model IV: 3 slopes, 9 intercepts; Model V: 9 slopes, 9 intercepts. Model II was found to be a more appropriate model than Model I (F = 49.33 on 8,659 d.f.; P < 0.001). Model IV was found to be more appropriate than both Model II (F = 24.05 on 2,657 d.f.; P < 0.001) and Model III (F = 21.40 on 2,657 d.f.; P < 0.001). The fullest model, Model V, was found not to be more appropriate than Model IV (F = 0.20 on 6,651 d.f.; P > 0.50). Model IV separated out three groups on the basis of significantly different slopes. Within two of these groups two data sets could be combined in terms of no significant difference between intercepts, resulting in Model VI (3 slopes and 7 intercepts). Model VI was tested against Model IV and it was found that the fuller model, Model IV, did not give a significant reduction in mean deviance and was, therefore, not more appropriate than Model VI (F = 0.20 on 2,257 d.f.; P > 0.50). Hence, the most ap- propriate combination of regressions was that represented by Model VI, and this is shown in Table 2 and Figure 3. The total variability in shell height attributable to the dependence of shell height on shell length in a particular model is given by the unbiased estimator p’, the adjusted coefficient of variation (ZAR, 1974). For Model VI, 6? was 0.88. In group 1 (Table 2, Figure 3), all the limpets came from continuously submersed habitats except for one set-— the predation (regurgitated) category. Group 2 included limpets from the east coast—from eulittoral rock surfaces and from both the predation (pecked out) categories. Group 3 was made up of limpets from the eulittoral zone on the west coast and from the top of the sublittoral zone on the east coast. In group 2, the height:length proportions of the shells at both the predation (pecked out) sites were not significantly different in terms of either slope or in- tercept and, hence, could be combined. The shells of lim- pets from the top of the sublittoral zone on the east coast were not significantly different, in terms of either slope or intercept, from those of the west coast eulittoral zone. These were combined in group 3. For the five sites from which Cellana tramoserica was collected, the relationships between shell height and shell length are given in Table 3. In determining the most ap- propriate combinations of linear regressions in this case, the following models were tested: Model I: 1 slope, 1 intercept; Model II: 1 slope, 5 intercepts; Model III: 5 slopes, 1 intercept; Model IV: 2 slopes, 5 intercepts; Model V: 5 slopes, 5 intercepts. Model II was found to be more appropriate than Model I (F = 57.46 on 4,390 d.f.; P < 0.001). Model IV was found to be more appropriate than both Model II (F = 10.97 on 1,389 d.f.; P < 0.01) and Model III (F = 10.86 on 1,389 d.f.; P < 0.01). The fullest model, Model V, was 30 t/ / Uf 20 V4 = £ -~ Se ©) Oo 8 WW sever, runs) + m8 4 3 10 20 30 40 506070 SHELL LENGTH (x) mm Figure 3 Regressions of height on length of categories within groups of Nacella (P.) macquariensis, plotted on log axes. Solid lines = group 1 where a = “high rock pools,” b = “deep rock pools,” c = “div- ing station,” and d = “predation regurgitated.” Dotted lines = group 2 where e = “predation (pecked out #1 + pecked out #2)” and f = “‘eulittoral, east coast.” Dashed line = group 3 (“eulittoral, west coast” + “top of sublittoral, east coast”). R. D. Simpson, 1985 Table 3 Regression analysis of shell height (y) on shell length (x) for Cellana tramoserica from all categories. Regression equation 5 (log y = log a Size range (lengths, ie lo Wess) Category mm) n log a b Upper eulittoral zone 15.6-42.4 81 = lei 1.429 Barnacle zone, northern side 17.2-42.0 93 —1.101 1.435 Lower eulittoral zone 15.7-44.6 98 —1.181 1.435 Barnacle zone, eastern side 16.1-38.4 71 —1.106 1.443 Upper eulittoral, southern rocks 115:0=52°7 53 S77 1.615 found not to be more appropriate than Model IV (F = O}OSong35386rd:f.5P' > 0:50). Model IV separated out two groups on the basis of significantly different slopes. Within one of these groups two data sets could be combined in terms of no significant difference between intercepts, resulting in Model VI (2 slopes, 4 intercepts). Model VI was tested against Model IV, and it was found that the fuller model, Model IV, did not give a significant reduction in mean deviance and was, therefore, not more appropriate than Model VI (F = 0.25 on 4,386 d.f.; P > 0.50). The most appropriate combi- nation of regressions is that represented by Model VI and shown in Table 4 and Figure 4. For Model VI, the ad- justed coefficient of variation, p*, was 0.93. Group 1 (Table 4, Figure 4) included limpets from both the upper and lower eulittoral zone on the northern side of the headland, and the limpets from the barnacle zone areas. Both sets of limpets from the two barnacle- zone areas could be combined, as there was no significant difference between the intercepts. The limpets comprising group 2 (upper eulittoral, southern rocks) were signifi- cantly different from the others on the basis of slope of the regression. There are two components to the allometric growth of the limpet shells of Nacella (P.) macquariensis and Cellana tramoserica, as shown by the regressions of height on length—slopes and intercepts. The slopes show the rate of increase of height with increasing length. When the slopes are significantly different, there is a difference in allometric intensity of height increase in relation to length during growth. If possible influence from some environ- mental factor during the growth of the limpet shell through adulthood is being examined, then consideration of slopes is the prime concern. Within a group having nonsignifi- cantly different slopes, significantly different intercepts in- dicate differences in relative shell heights; that is, limpets from different sites had different “‘starts” to their relative shell heights, and maintained the same allometric inten- Page 23 sity of height increase throughout growth. Comparing rel- ative shell heights between groups with significantly dif- ferent slopes will give different answers depending on the selected value of length. This emphasizes the futility of comparing shell heights among groups of limpets for some standard length of shell, if allometric growth is not con- sidered. DISCUSSION The climate of Macquarie Island (with precipitation oc- curring over approximately 330 days in each year, a mean relative humidity of 88%, and persistent cloud cover—see SIMPSON [1976]) poses very few desiccation problems for intertidal limpets. Desiccation stress would be encoun- tered by intertidal animals only on the occasional sunny day in summer with calm seas. COURTNEY (1972) iden- tified wind as a desiccating factor for intertidal mollusks, but the persistent precipitation, high humidity, and heavy seas would greatly reduce any desiccating effect of the wind on Macquarie Island shores. Thus, with minimal influence from desiccation, the habitats selected on Mac- quarie Island shores effectively represent a gradation of water turbulence. A number of authors have suggested that water tur- bulence has either a negligible or no effect on increasing the height of limpet shells (ORTON, 1932; BALAPARAMES- WARA RAo & GANAPATI, 1971; BERRY & RUDGE, 1973; BRANCH & MarsH, 1978). Further, it could be argued from the equations of drag forces on limpet shell shape (WARBURTON, 1976; BRANCH & MarsH, 1978) that a lower-spired shell should reduce drag and, hence, be more favorable for an environment with a high degree of water movement. If the above were generally true, shells of Na- cella (P.) macquariensis could be expected to show either no difference in allometric growth with respect to shell height or even a decrease in height across a gradient of increasing water turbulence. For Nacella (P.) macquariensis there were three group- ings on the basis of allometric intensity of height increase. These groups had significantly different slopes to their combined regressions, in ascending order from group 1 Table 4 Combination of categories of Cellana tramoserica into sep- arate groups on the basis of slope. Intercepts © Common Group (log a) slope (b) (1) Lower eulittoral zone —1.178 Upper eulittoral zone —1.116 1.433 Barnacle zone (northern + eastern sides) —1.095 (2) Upper eulittoral, southern rocks —= il ei/7/ 1.615 Page 24 30 20 = E a —~ 10 FE 9 ©) a 8 EO e7. 3 6 © mn 5 4 3 10 20 30 40 50 60 SHELL LENGTH (x) mm Figure 4 Regressions of height on length of categories within groups of Cellana tramoserica, plotted on log axes. Solid lines = group 1 where a = “lower eulittoral zone,” b = “‘upper eulittoral zone,” and c = “barnacle zone (northern and eastern sides).”” Dotted lines (d) = group 2 (“upper eulittoral, southern rocks’). through group 3 (Table 2, Figure 3). This order matched the increasing degree of water turbulence impinging on the habitat categories within the groups, with the appar- ent exception of the “top of the sublittoral, east coast” category which was combined with the category subjected to the heaviest wave action of all (‘‘eulittoral, west coast’’) in group 3. However, the longer period of lower intensity wave action at the top of the sublittoral zone on the east coast may have the same degree of impact as a shorter period of higher intensity wave action in the eulittoral zone on the west coast. For Cellana tramoserica, collections from four sites were combined on the basis of no significant difference between slopes of their regressions (group 1). Thus, in group 1 the differential effects from either wave action or desiccation could not be identified as influencing the allometric inten- sity of height increase. Group 1 limpets had a significantly lower slope for the combined regression than that of group 2 (Table 4, Figure 4). The limpets in group 2 were subject to stress from both heavy wave action and desiccation. Further, parts of the rocks in the group 2 habitat were occasionally covered by shifting sands that could encroach on the areas occupied by the limpets, thus presenting a further stress. The Veliger, Vol. 28, No. 1 For both species, there were significant differences in intercepts among categories comprising some of the groups. That is, although some categories showed the same allo- metric intensity of shell height increase, there were sig- nificant differences in relative shell heights for limpets from different habitats. In group 1 for Nacella (P.) mac- quariensis, limpets in high rock pools had the lowest rel- ative shell height followed by, in ascending order, limpets from deep rock pools, diving station, and predation (re- gurgitated). The increase in relative shell height of the three habitat categories followed a sequence of increasing water turbulence. For Cellana tramoserica, the significant- ly different intercepts for limpet populations in group 1 followed a sequence that could be associated with both increasing desiccation stress and water turbulence, with the relatively highest shells occurring in the area of great- est wave action. Thus, for Nacella (P.) macquariensis the allometric in- tensity of shell-height increase and the increase in relative shell heights were clearly correlated with increasing water turbulence. For Cellana tramoserica allometric intensity of shell-height increase was greater only where wave action was constant and very strong. Desiccation stress was not correlated with allometric intensity of shell-height in- crease. Both increasing water turbulence and desiccation stress were correlated with an increase in relative shell heights, with evidence of the most effect being associated with increasing water turbulence. The data for both species showed that, if there is a significantly greater slope for a group, then members of that group will eventually reach a greater relative shell height (Figures 3, 4). What is curious is that in those groups that have the significantly greatest slopes, the smaller (by length) shells should have such low relative shell heights, z.e., group 3 for Nacella (P.) macquariensis (Figure 3) and group 2 for Cellana tramoserica (Figure 4). Perhaps the smaller limpets in such areas of greater en- vironmental stress occupy some form of protected micro- habitat and move out onto more open rock surfaces as they grow. A number of factors have been previously suggested as influencing the shell height of limpets and these require examination here. Variation in shell-height ratios of Patella vulgata Lin- naeus on English shores formed the basis of the earliest studies on this topic in intertidal limpets. RUSSELL (1907) observed that specimens from localities that were exposed to heavy wave action were lower-spired than those from sheltered localities. ORTON (1932) found that the shells of adult limpets living on the upper shore were taller than those of individuals near the low water level or in rock pools. Orton correlated higher shell types with desiccation. He suggested that limpets inhabiting higher levels would hold their shells closer to the substrate to prevent drying out. The consequently strong grip would pull in the man- tle margin, the site of secretion of new shell. Hence, a smaller peripheral increment of growth would be made R. D. Simpson, 1985 and continued growth would result in a steeper shell. ORTON (1932) further suggested that wave action would have a negligible influence on shell shape of P. vulgata, although he recognized that wave action would also cause a limpet to adhere more firmly. Moore (1934) found that specimens of P. vulgata, with large height ratios, developed a shelf of flatter shell growth when transferred from the shore into a fish-hatching pond. Moore attributed this to the removal of desiccation stress, but the experimental design could not entirely discount decrease in water tur- bulence as an effect. Curiously, nearly all the ‘“‘shelfed” limpets returned to the initial angle of shell growth while still in the pond. EBLING e¢ al. (1962), working on Patella species in Ire- land, found significantly greater relative shell heights for P. aspersa Roding where they were permanently sub- merged and subjected to strong currents. Ebling et al. sug- gested that water turbulence could also cause greater rel- ative shell heights by obliging the limpets to adhere firmly. In a study on the tropical limpet Cellana radiata radiata (Born), taken from different habitats and zonal levels, BALAPARAMESWARA RAO & GANAPATI (1971) concluded that desiccation was more important than wave action in influencing shell characters. From shell measurements of a number of limpet species, VERMEI (1973, 1978) found that the shell-height ratio increased in an upshore direction and suggested that this was an adaptation to desiccation stress. VERMEIJ (1973) argued that a taller shell would increase the water res- ervoir and decrease the region of water loss, 7.e., the area and perimeter of the base. BANNISTER (1975) recorded greater desiccation tolerance in the taller-shelled Patella lusitanica (Gmelin) of the upper eulittoral zone to that of the lower-shelled Patella caerulea (Linnaeus) of the lower eulittoral zone of Mediterranean shores. In comparisons across seven species of Patella on South African shores, BRANCH (1975) found a strong correlation between zonal position on the shore and tolerance to water loss; however, there was no close correlation between zonation and rel- ative shell heights. In a review of limpet biology, BRANCH (1981) noted that many authors have recorded greater height ratios and relative shell heights in drier habitats for a number of species. Also, BRANCH (1981) observed that an intraspe- cific increase in relative shell height usually occurs in lim- pets from higher on the shore but the same pattern is not always true when different species are compared. Results from interspecific comparisons will always embody wide genotypic options for morphological strategies. Conse- quently, it is perhaps not surprising that general hypoth- eses will be confounded when pooling results from a num- ber of species. On the coasts of North America, there have been a number of studies on the possible effects of desiccation on the morphology of acmaeid limpets. On the east coast, WALLACE (1972) found that, in Acmaea testudinalis (Miul- ler), tolerance of desiccation was related to size and that Page 25 limpets in a habitat with increased desiccation stress (in- tertidal rock face versus tide pools and subtidal area) did not have greater shell height ratios. On the west coast, WOLCcoTT (1973) reported no correlation between either size or shell shape and desiccation rates or tolerances in interspecific comparisons among five species of limpets (although no quantitative details of shell shape were pre- sented). Wolcott determined that the ability to form a mucus sheet between the shell margin and the substrate was the most important adaptation to desiccation. Aware that interspecific comparisons may confuse the issue, LOWELL (1984) undertook intraspecific investigations for four species of acmaeid limpets. Lowell determined that increasing size significantly reduced water loss but that variation in shell shape (as measured by volume/circum- ference) had no effect on water loss. Lowell suggested that variation in shell shape might be partially or primarily due to factors other than resistance to desiccation. In Britain, DAviEs (1969) recorded a greater desicca- tion tolerance in specimens of Patella vulgata from high levels of the shore compared to that of specimens from low levels. While Davies speculated that this may be partly attributable to shell shape, his results primarily showed that desiccation tolerance was inversely related to body size. As shown here for Nacella (P.) macquariensis, one factor associated with allometric intensity of shell-height in- crease and greater relative shell heights is an increase in water turbulence. As previously mentioned, some authors have discounted wave action as having an influence on the shell height of limpets. However, EBLING e¢ al. (1962) and WALKER (1972) found associations between increased water turbulence and relative shell height. DURRANT (1975) found significantly different height: width ratios in the freshwater limpet Ancylus fluviatilis Muller from river and lake populations, where there was no exposure to desiccation. The river populations (with greater water flow) had taller shells. BRANCH (1981) has noted the con- trasting arguments for influences on shell height in areas of strong water movement: (a) flatter shells are adaptive where wave action is strong because they cause less drag, versus (b) strong water currents cause a limpet to clamp down tightly and thereby deposit shell in a tall conical form—in the manner as postulated by ORTON (1932). Further, BRANCH (1975) and BRANCH & MARSH (1978) have reasoned that higher-domed shells will allow greater muscle development and insertion that, in turn, would strengthen tenacity. GRENON & WALKER (1981) found no significant dif- ferences between the tenacity of low and high shore level Patella vulgata on both exposed and sheltered shores. ‘Thus, the taller-shelled populations (from the upper shore) did not display a greater tenacity. BRANCH & MArsH (1978) reported that relative shell height was not correlated with tenacity in six Patella species. However, in the two species with strong allometric intensity of height increase against length (P. argenville: Krauss and P. granatina Linnaeus), Page 26 relative shell height was significantly correlated with te- nacity per unit area of the foot. VERMEIJ (1978) noted that a pattern in allometric growth intensity of mollusks for different parts of the shore may be a function of an adaptive trend or a by-product of variation in growth rate. VERMEIJ (1980) went on to state ‘“Various lines of evidence have led me to believe that many instances of gradual changes in shell allometry (es- pecially doming) are geometrically tied to growth rate.” In experimental manipulations of Collisella limpets in California, HAVEN (1973) noted that rapid growth in C. scabra resulted in new shell being deposited at a flatter angle. Growth rates of intertidal limpets have often been found to decline in an upshore direction (BRANCH, 1974; Lewis & BOWMAN, 1975; PHILLIPS, 1981) and this cor- responds with the previously mentioned trend for relative shell heights of limpets to increase in an upshore direction. Presumably, the slower growth rates are related to lesser abundance of food or time available for grazing. LEwIs & BowMaN (1975) showed that Patella vulgata had different growth rates in different habitats. Growth rates were higher at low tidal levels compared to high tidal levels, and superimposed on this was the biological influence of growth rates being lower in sites inhabited by barnacles and/or mussels, whose presence reduced the surface area that could be easily grazed. In five of these habitats, Lewis & Bowman recorded a matching sequence between de- crease in growth rate and increasing allometric intensity of shell-height increase. This presents strong evidence for shell allometry being a function of growth rate. THOMPSON (1980) also found growth rates of P. vulgata to be highest on bare rock and lowest on areas with a dense population of barnacles. Of particular relevance to the present study, FLETCHER (1984) found that a mid-tidal population of Cellana tra- moserica had a lower growth rate, a higher density, and a significantly greater allometric intensity of shell-height in- crease than that for a subtidal population. However, the correlation with growth rate did not hold true for all of the study sites investigated by Fletcher (high, mid-, and low intertidal and subtidal). The order of allometric in- tensity of shell height increase across the four populations was “high” = “mid-” > “low” > “sub,” while that for the growth rates was ‘“‘mid-” < “high” = “low” < “‘sub.” A reverse trend of higher growth rates at upper levels has been found where densities of limpets at higher levels are lower, which would result in more food being available (SUTHERLAND, 1970; CREESE, 1980). Unfortunately, there are no corresponding data on shell heights in these cases. Water turbulence and desiccation could be acting with different emphases upon intertidal limpets in different cli- matic regions and different parts of the shore. The data from the present study and the mixed findings from other works show that no one environmental factor has a uni- versal relationship with allometric growth of shell height. If, as ORTON (1932) suggests, an obligation to adhere The Veliger, Vol. 28, No. 1 firmly increases the steepness of shell formation, then any factor (e.g., water turbulence or desiccation) that causes a limpet to clamp down frequently enough will have such an ultimate effect. Although alternative explanations have been put forward in seeking relationships between relative shell height and tenacity (BRANCH, 1975; BRANCH & MarsH, 1978), Orton’s argument would apply, whatever the tenacity capability of a species. The assigning of adap- tive trends to allometric growth in limpets relies almost entirely on correlative evidence. Indeed, correlations be- tween such features as allometric intensity of shell-height increase and degree of stress from water turbulence and desiccation may simply be because such stresses reduce growth rates rather than because of any adaptive advan- tage. Experimental proof for time-related features such as shell growth is difficult to obtain. Field manipulation of animals, for example, between habitats of differing inten- sities of desiccation and water turbulence would also have to account for possible differences in food availability, grazing capabilities, or densities, which, in turn, would affect growth rates. A useful addition to the present data would be to apply an immediate aging technique (ze., from shell growth lines) to limpets from several habitats, and this will form the subject of further studies. The predation categories of Nacella (P.) macquariensis present an interesting result, subsidiary to the central aim of the study. The similarity in the shells of limpets from the two predation (pecked out) sites on the east coast showed that gulls feeding in this way were selectively taking limpets with respect to shell shape in terms of height: length proportion. The similarity in slope to that for lim- pets in the eulittoral zone indicated that the gulls were taking the limpets from that region, but they were select- ing limpets that had lower relative shell heights, as indi- cated by the significantly different lower intercept for the predation category. Limpets taken by gulls and later re- gurgitated were combined with limpets from pools and the diving station into the one group, on the basis of slope. This implies that the gulls took limpets that they could swallow from pools. As they floated among intertidal rocks during calm seas, Dominican gulls were observed diving their heads under water to pick off limpets. A similar situation was reported for Dominican gulls feeding on Nacella (P.) delesserti (Philippi) on Marion Island by BLANKLEY (1981). However, in regard to rock pools at Macquarie Island, Dominican gulls were observed taking limpets only from the edges of pools. Also, it is highly unlikely that pool populations could supply the number of limpets swallowed whole by gulls—not only in the area covered in this study but also for the whole island. It is more likely that the inclusion of the predation (regurgi- tated) category in group 1 is an artifact from the gulls’ selection of limpets. That is, their selection of smaller limpets for swallowing (the majority of the limpets in this category were in the lower end of the length range) biased the result, placing these limpets in with group 1. R. D. Simpson, 1985 Page 27 ACKNOWLEDGMENTS This study was supported by the Antarctic Division, De- partment of Science and Technology, Australian Govern- ment, and by internal research grants from the University of New England. I am particularly grateful to Dr. S. Cairns by way of assistance in computing and statistical analysis and criticism of the manuscript. Dr. V. Bofinger provided valuable statistical advice, and Dr. D. Woodland gave helpful comments on the manuscript. Mr. R. Hobbs and Ms. S. Harrington assisted in various stages of the work. LITERATURE CITED BAKER, R. J. & J. A. NELDER. 1978. The GLIM system manual, release 3. Numerical Algorithms Group: Oxford. BALAPARAMESWARA RAo, M. & P. N. GANAPATI. 1971. Eco- logical studies on a tropical limpet, Cellana radiata. Struc- tural variation in the shell in relation to distribution. Mar. Biol. 10:236-243. BANNISTER, J. V. 1975. Shell parameters in relation to zona- tion in Mediterranean limpets. Mar. Biol. 31:63-67. Berry, R. J. & P. J. RuDGE. 1973. Natural selection in Ant- arctic limpets. Br. Antarc. Surv. Bull., No. 35:73-81. BLANKLEY, W. O. 1981. Marine food of Kelp Gulls, Lesser Sheathbills and Imperial Cormorants at Marion Island (Subantarctic). Cormorant 9:77-84. BRANCH, G. M. 1974. The ecology of Patella Linnaeus from the Cape Peninsula, South Africa. 3. Growth-rates. Trans. Roy. Soc. S. Afr. 41:161-193. BRANCH, G.M. 1975. Ecology of Patella species from the Cape Peninsula, South Africa. IV. Desiccation. Mar. Biol. 32: 179-188. BRANCH, G. M. 1981. The biology of limpets: physical factors, energy flow, and ecological interactions. Oceanogr. Mar. Biol. Ann. Rev. 19:235-280. BRANCH, G. M. & A. C. MarsH. 1978. Tenacity and shell shape in six Patella species: adaptive features. J. Exp. Mar. Biol. Ecol. 34:111-130. CourTNEy, W. A. M. 1972. The effect of wind on shore gas- tropods. J. Zool. (Lond.) 166:133-139. CREESE, R. G. 1980. An analysis of distribution and abun- dance of populations of the high-shore limpet Notoacmea petterd: (Tenison-Woods). Oecologia 45:212-260. Davies, P. S. 1969. Physiological ecology of Patella. III. Des- iccation effects. J. Mar. Biol. Assoc. U.K. 49:291-304. DELL, R. K. 1964. Marine Mollusca from Macquarie and Heard Islands. Rec. Dom. Mus., Wellington, N.Z. 4:267- 301. DurRRANT, P. M. 1975. An investigation into the effect of running water on shell dimensions in Ancylus fluviatilis Miller. J. Conchol. 28:295-300. EBLING, F. J., J. E. SLOANE, J. A. KITCHING & H. M. Davies. 1962. The ecology of Lough Ine XII. The distribution and characteristics of Patella species. J. Anim. Ecol. 31:457-470. FLETCHER, W. J. 1984. Intraspecific variation in the popula- tion dynamics and growth of the limpet, Cellana tramoserica. Oecologia 63:110-121. GRENON, J.-F. & G. WALKER. 1981. The tenacity of the lim- pet, Patella vulgata L.: an experimental approach. J. Exp. Mar. Biol. Ecol. 54:277-308. HaveEN, S. B. 1973. Competition for food between the inter- tidal gastropods Acmaea scabra and Acmaea digitalis. Ecology 54:143-151. Lewis, J. R. & R. S. BowMaANn. 1975. Local habitat-induced variations in the population dynamics of Patella vulgata L. J. Exp. Mar. Biol. Ecol. 17:165-203. LowELL, R. B. 1984. Desiccation of intertidal limpets: effects of shell size, fit to substratum and shape. J. Exp. Mar. Biol. Ecol. 77:197-207. Mackay, D. A. & A. J. UNDERWOOD. 1977. Experimental studies on homing in the intertidal patellid limpet Cellana tramoserica (Sowerby). Oecologia 30:215-257. Moore, H. B. 1934. The relation of shell growth to environ- ment in Patella vulgata. Proc. Malacol. Soc. Lond. 21:217- 222. OrTON, J. H. 1932. Studies on the relation between organism and environment. Proc. Liverpool Biol. Soc. 46:1-16. PHILLIPS, D. W. 1981. Life-history features of the marine intertidal limpet Notoacmea scutum (Gastropoda) in central California. Mar. Biol. 64:95-103. RUSSELL, E. S. 1907. Environmental studies on the limpet. Proc. Zool. Soc. Lond. 11:856-870. RYAN, T. A., Jr., B. L. JOINER & B. F. Ryan. 1981. Minitab Reference Manual. Duxbury Press: Boston. 154 pp. Simpson, R. D. 1976. The shore environment of Macquarie Island. ANARE Rep. Ser. Bl, No. 125:1-41. SUTHERLAND, J. P. 1970. Dynamics of high and low popu- lations of the limpet Acmaea scabra (Gould). Ecol. Monogr. 40:169-188. TuHompson, G. B. 1980. Distribution and population dynam- ics of the limpet Patella vulgata L. in Bantry Bay. J. Exp. Mar. Biol. Ecol. 45:173-217. VERMEIJ, G. J. 1973. Morphological patterns in high-inter- tidal gastropods: adaptive strategies and their limitations. Mar. Biol. 20:319-346. VERMEIJ, G. J. 1978. Biogeography and adaptation. Patterns of marine life. Harvard University Press: Cambridge, Mass. 352 pp. VERMEJJ, G. J. 1980. Gastropod shell growth rate, allometry, and adult size: environmental implications. Pp. 379-394. In: D. C. Rhoads & R. A. Lutz (eds.), Skeletal growth of aquatic organisms. Plenum: New York. WALKER, A. J. M. 1972. Introduction to the ecology of the Antarctic limpet Patinigera polaris (Hombron and Jacqui- not) at Signy Island, South Orkney Islands. Br. Antarc. Surv. Bull., No. 28:49-69. WALLACE, L. R. 1972. Some factors affecting vertical distri- bution and resistance to desiccation in the limpet, Acmnaea testudinalis (Miller). Biol. Bull. 142:186-193. WARBURTON, K. 1976. Shell form, behaviour and tolerance to water movement in the limpet Patina pellucida (L.) (Gas- tropoda: Prosobranchia). J. Exp. Mar. Biol. Ecol. 23:307- 325% Wo coTT, T. G. 1973. Physiological ecology and intertidal zonation in limpets (Acmaea): a critical look at “limiting factors.” Biol. Bull. 145:389-422. ZAR, J. J. 1974. Biostatistical analysis. Prentice-Hall Inc.: New Jersey. 620 pp. The Veliger 28(1):28-36 (July 1, 1985) THE VELIGER © CMS, Inc., 1985 Spatial and ‘Temporal Distribution and Overlap of Three Species of Bullia (Gastropoda, Nassariidae) on Exposed Sandy Beaches by L. E. Mc GWYNNE anpb A. Mc LACHLAN Zoology Department, University of Port Elizabeth, P.O. Box 1600, Port Elizabeth, 6000, South Africa Abstract. Three species of plough shell, Bullia rhodostoma, B. digitalis and B. pura, coexist on dy- namic sandy beaches in the Eastern Cape, South Africa. The robust B. rhodostoma is most abundant in the zone of swash and backwash, and exhibits great efficiency in foraging behavior, feeding on carrion stranded on the shore. The entire life cycle of this species is spent intertidally, while only adults of B. digitalis and B. pura appear on the beach, where they prefer a more subtidal habitat. The latter species exploit their surfing ability in the quest to capture food on its path shorewards and to migrate offshore during winter probably to spawn their eggs in deeper water. All three Bullia species move horizontally with the tides, a behavior facilitating continuous access to food. Although zonation is evident among the three whelk species, some spatial overlap exists. Bullia rhodostoma is the most widely dis- tributed while B. digitalis shows the greatest overlap with the other two species. INTRODUCTION THE EXPOSED SANDY BEACH, despite little spatial hetero- geneity and much physical instability, harbors a marine fauna of some ecological diversity. Eighteen species have been recorded in the Eastern Cape, South Africa, with four mainly intertidal forms constituting 99.5% of the bio- mass. The plough shell Bullia rhodostoma Reeve, 1847, makes up 1.7%, while the other two species, B. digitalis (Dillwyn, 1787) and B. pura Melvill, 1885, account for only 0.4% of the total biomass. The large white sand mus- sel Donax serra Botten, 1848, dominates the macrofauna comprising 97.3%, while the smaller mussel D. sordidus Hanley, 1845, contributes only 0.1% of the biomass (Mc GWYNNE, 1980; Mc LACHLAN ef al., 1981). Although their contribution to the macrofaunal biomass is small, the ecological importance of Bullia is indicated by their abundance (particularly B. rhodostoma) and their position in the food chain (BROWN, 1964, 1971). Snails of B. rhodostoma with shell lengths ranging from 3 to 52 mm have been recorded (Mc LACHLAN et al., 1979a) while a much narrower size range has been encountered in both B. pura (11-29 mm) and B. digitalis (22-42 mm) (Mc GWYNNE, 1980). The snails function as predator/scav- engers, feeding on a wide range of organisms stranded in the swash. The most important predator of Bulla is the swimming crab Ovalipes (DU PREEZ, 1981). The holo- cephalan Callorhynchus, elasmobranch Rhinobatus, and te- leosts Coracinus, Lithognathus and Rhabdosargus prey on the plough shells during high tide. The sanderling C7ro- cethia feeds on Bullia while they are exposed at low tide (Mc LACHLAN et al., 1981; BRowN, 1982). Brown described the mode of life of Bulla on beaches in the Western Cape and conducted extensive research into their ecophysiology. Recently, BROWN (1982) re- viewed the existing knowledge on the biology of Bullia in South Africa. In the Eastern Cape, the general ecology of B. rhodostoma has been described (Mc LACHLAN e¢ al., 1979a, 1979b) and aspects of the physiology of the genus recorded (ANSELL & Mc LacHLAn, 1980; DYE & Mc Gwynne, 1980; Mc Gwynne, 1980; Mc LACHLAN & YOUNG, 1982). On beaches in the Eastern Cape, Bullia rhodostoma, B. digitalis, and B. pura appear to coexist in a relatively un- structured habitat where little niche differentiation is ev- ident. The aim of this paper is to describe aspects of the ecology of only the latter two species, such as distribution and abundance, and to determine how the three whelk species apparently coexist successfully in the “uniform” environment of the sandy beach. L. E. Mc Gwynne & A. Me Lachlan, 1985 SOUTHERN AFRICA Page 29 PORT ELIZABETH Figure 1 Map of Southern Africa, showing the locations of the three study beaches: Maitlands (M), Kings (KB) and Bluewaterbay (BWB). THE STUDY AREA Three beaches were selected as study sites (Figure 1). Maitlands beach is a southfacing beach lying 30 km west of Port Elizabeth; Kings beach and Bluewaterbay beach face northeast and east, respectively, into Algoa Bay. The main features characterizing the three beaches have al- ready been described by Mc LacHLaNn (1977, 1979) and are summarized in Table 1. These represent average con- ditions, which may differ markedly during extreme calm or storms. Maitlands beach supports a high macrofaunal biomass. The total ash-free dry biomass approaches 7 kg/m of shoreline (Mc LACHLAN et al., 1981), the major contrib- utor being the large white sand mussel Donax serra. Less common inhabitants of the beach zone include the swim- ming crab Ovalipes, sand-burrowing mysid Gastrosaccus, and isopods, chiefly Eurydice. Donax serra is absent on Kings beach and present in low numbers on Bluewaterbay beach. Populations of beach macrofauna on the latter two beaches are much smaller than on the high energy beach at Maitlands. METHODS Distribution and Abundance Sampling of Bullia digitalis and B. pura was conducted approximately every six weeks from January 1978, to December 1979, on Maitlands beach, and from January to December 1979, on Kings and Bluewaterbay beaches. A dredge (0.5 m wide with 1.5-mm mesh) was used to complete a series of five hauls for each sample. Each haul was 10 m long and cut 5 cm into the sand; thus, a 25-m? area was sampled. The hauls were continuous, with no overlap between each haul. The sampling procedure cov- ered a 50-m line from a point 5 m below LWS (low water springs) to just above the mean tide level. This was essen- tial as B. digitahs and B. pura inhabit deeper water than the intertidal resident, B. rhodostoma. The appearance of Bullia rhodostoma in dredged sam- ples was ignored, as its distribution and abundance has already been described (Mc LACHLAN et al., 1979a). The occurrence of this species was, however, noted in five sam- ples taken from Maitlands beach from April to October 1979, to measure the degree of overlap in the distribution Page 30 The Veliger, Vol. 28, No. 1 Table 1 Characteristic features of the three East Cape beaches, Maitlands (M), Kings (KB), and Bluewaterbay (BWB). Feature Maitlands Average slope of beach gentle 4, concave Average width of intertidal 100 m Width of surf zone 150-400 m Swash periods 20-30 sec Wave periods 14 sec Volume of seawater filtered 9 through interstices (m?/ m/day) Grade of sand very well-sorted medium quartz particles Particle size range (median 268-308 um diameters) Gentle sloping beach backed by extensive dune system. Strong wave action. General appearance of the three beach populations. All animals collected were measured to 1 mm greatest shell length using vernier cal- lipers. Due to the absence of juvenile snails of both Bullia digitalis and B. pura from the beach zone, sampling was extended into the offshore region beyond the breakers. A dredge 0.33 m wide with a mesh size of 4 mm was dragged by a motorboat and guided by a diver using SCUBA. The series of hauls was approximately 10 m in length and in water depths ranging from 3 to 8 m. Tidal Migrations The tidal movements of Bullia digitalis and B. pura were monitored during spring tides in March and again in No- vember 1979. Mc LACHLAN et al. (1979b) recorded the tidal migrations of macrofauna resident in the intertidal zone just below ELWS to EHWS (extreme low water springs, extreme high water springs). Preliminary obser- vations showed B. digitalis and B. pura to prefer a more subtidal habitat than the intertidally dominant B. rhodo- stoma. During March of this study, the sampling zone was extended from ELWS seawards to the line of break- ers initiating the dynamic surf zone. In November, the first line of breakers marked the seaward extent of the sampling procedure. Steel rods, at 20 m intervals, marked each sampling site. A second series of rods, 10 m from and parallel to the first, marked the length of each strip of beach to be dredged. Samples were taken every 3 h for 12 h over two tidal cycles using the smaller 4-mm mesh size dredge. No area was sampled more than once in 6 h. The dredge proved to be efficient on dry sand and at water depths to approximately 4 m. Kings Bluewaterbay moderate 4, concave steep 4, concave 50m 40m 30-120 m 50-200 m 10-23 sec 12-20 sec 10 sec 10 sec 7 12 clean fine well-sorted well-sorted fine to medium quartz quartz sand with large shell fragments and pebbles around LWS 200-220 um variable size, median 250 wm Berm 1.25 m above MTL, dunes poor. Wave action moderate to strong. Berm present 2 m above LWS, dunes poorly de- veloped. Moderate wave action. Niche Breadth and Overlap In terms of the spatial model of the niche by HUTCHINSON (1958), niche breadth is the “distance through” the niche along some particular line in niche space. The distribution of whelks on the beach was taken to represent this line, and its vertical range and evenness of spread were calculated using an index of niche breadth, B. We used data collected at Maitlands beach over five months (from April to October 1979) when all three species were sampled through a series of five tidal levels. Num- bers collected in 5-m? dredged samples at each level were used to calculate the following quantity: where B is the niche breadth of a species, P, is its pro- portion (in numbers) in the 7th habitat unit of the envi- ronment, and 7n is the number of units. The sandy beach was the habitat and the series of five dredges taken each month made up the units. The index was calculated for each species for every month over the five-month period. The theoretical maximum niche-breadth value was esti- mated by assuming an even distribution of the three species over the five units. Niche overlap is simply the joint use of a resource by two or more species. As in niche breadth, the resource is space. Niche overlap is calculated as: L. E. Mc Gwynne & A. Me Lachlan, 1985 50 40 N E wo N oc c 30 a ep) jag Ww my 20 = =) Zz 1978 Page 31 MAITLANDS BEACH Sime KING'S BEACH — — — BLUEWATERBAY BEACH TIME (MONTHS) Figure 2 Temporal abundance of Bullia digitalis on three beaches in the Eastern Cape, South Africa. where a is the niche overlap or probability of species k overlapping species j, and P,, and P,, are the proportions of species 7 and & respectively in the Ath unit of the hab- itat. If the distribution for the two species is identical (z.e., complete overlap), a=1 (LEvINs, 1968). Interspecific niche overlaps were calculated monthly and estimated for the overall sampling period. RESULTS Temporal Abundance and Distribution Two peaks of abundance were evident for both Bullia digitalis and B. pura on all three beaches (Figures 2, 3). Maximum densities of B. digitalis occurred during mid- summer (November to January) with a smaller peak in and around March. Bullia pura was most numerous in December and January, and then again during March and April. Numbers of both species were low, reflecting the absence of juvenile snails. No distinct year classes were discernible. Snails appeared to be patchily distributed in groups. The sampling procedure, and probable inefficien- cy of the dredge, did not always reflect the true distribu- tion and abundance patterns on the beach. The samples containing Bullia rhodostoma show this population to occur intertidally and higher on the shore than B. digitalis and B. pura, which were confined to lower tidal levels and deeper waters. The offshore hauls located juvenile snails of both Bullia pura and B. digitalis coexisting with juveniles of four other Bullia species, namely B. callosa Wood, 1828, B. annulata (Lamarck, 1816), B. laevissima (Gmelin, 1791), and B. tenuis Gray, 1839. Notable differences in shell morphol- ogies eliminated confusion over their taxonomic identity. A t-test indicated a significant difference in the mean shell lengths between the beach and offshore populations of B. digitalis (P < 0.005) and B. pura (P < 0.005). Two other gastropods, Ancilla albozonata Smith, 1904, and Melapium lineatum (Lamarck, 1822), were also found in the deep- water samples. Tidal Migrations Figures 4 and 5 show the profile of Kings beach along with the pattern of tidal migration undertaken by the whelks during both monitoring sessions. Only the spatial ranges and not changes in abundance of the snails are shown. One snail in a dredged sample was taken as rep- resentative of the presence of a species at a particular location and time. Despite low numbers, migration pat- terns on both occasions revealed similar trends. All three species showed distinct movements with the tides. Bullia rhodostoma kept abreast and sometimes slightly ahead of the incoming swash, while B. pura and B. digitalis failed to penetrate the swash, always remaining in deeper water. Page 32 NUMBERS PER 25m 2 he Veliger Vole 28Nowl MAITLANDS BEACH Soma KING’S BEACH —-—-— BLUEWATERBAY BEACH TIME (MONTHS) Figure 3 Temporal abundance of Bullia pura on the three South African East Cape beaches. Bullia pura, however, extended into shallower water than B. digitalis. Niche Breadth and Overlap Numbers of Bullia digitalis and B. pura collected were generally low and fluctuated over the sampling period (Table 2), while numbers of B. rhodostoma increased steadily towards October. Bullia digitalis was absent dur- ing mid-winter (July-August) and numbers of B. pura declined over this period. B values were all under 5, the theoretical maximum that would indicate a uniform dis- tribution of snails throughout the study area. The greatest niche breadth was recorded in B. rhodostoma for October, the value reaching 78% of the maximum. Niche breadths of B. digitalis and B. pura proved variable, ranging from 0 to 3.5. Calculation of the overall B for each species showed B. rhodostoma to be the most widespread popu- lation, with B. pura confined to the narrowest zone. In- terspecific zonation is suggested by the average B values (Table 3), as these never reach 60% of the value indicative of an even distribution. The greatest species overlap occurred between Bullia rhodostoma and B. pura (Table 4) particularly during April and October. Overlap between each of the former species and B. digitalis was negligible, except for October when the overlap proved noteworthy. Bullia digitalis and B. pura populations overlapped markedly when both species were present on the shore. DISCUSSION Distribution The same basic pattern of distribution appeared on all three sandy beaches. Bullia rhodostoma occupied a broad band of the intertidal, whereas the adults of B. digitalis and B. pura restricted themselves to the subtidal, the ju- veniles remaining offshore beyond the breakers. Mc LACHLAN et al. (1979a) recorded a size-based zonation at low tide, with the smaller snails situated uppermost on the shore. The effect of shell length on distribution pat- terns was not measured in this study. On Muizenberg beach in the Western Cape, BROWN (1971) found snails of Bullia rhodostoma and B. digitalis occurring in well-defined groups of a single species within a narrow size range. No vertical pattern of zonation was noted. He attributed this distribution to a sorting process promoted by wave action. Mc GwyNn_E (1980) demon- strated that B. rhodostoma, aided by a large foot and a light shell, surfed at a faster rate than the other two species. Both B. digitalis and B. pura have smaller feet and heavier shells, which enable them to withstand current surges more effectively than B. rhodostoma and, thus, maintain their subtidal position on the shore. The physical characteristics of B. rhodostoma also promote speed and efficiency in crawling on wet sand, essential attributes for this species to reach carrion stranded in the swash. Bullia digitalis and B. pura are not equipped to compete with B. rhodostoma L. E. Mc Gwynne & A. Mc Lachlan, 1985 20h00 (5)(6) vas 17h00 jf og fac SINAN Se mae Se (14)(6)| | % ‘e 14h00 RG gee ty (16)(19) &@ Brel crcteielcleleletersiciclelersicieicye: TIME (h) 11ho0o (2)(5) UPPERMOST LINE OF SWAS 08hoo \ ae (5)(2) 4 f, VERTICAL EHWS METERS Page 33 we SURF ZONE Roh monn nanan nanan nnn. SHALLOW SURF ZONE 140 120 100 80 60 40 20 O HORIZONTAL METERS Figure 4 The migration patterns of beach populations of Bullia pura (--—) and B. digitalis (----: ) during a spring tidal cycle in March 1979. Actual numbers of each species caught in the dredge at each sampling time are shown in parentheses below the time scale on the vertical axis. The profile of Kings beach is also shown. in this pursuit and probably exploit different feeding strategies subtidally in their quest for food and, therefore, survival. In experiments correlating the distribution of Bullia digitalis with the grade of sand on west coast beaches, BROWN (1961) and BALLy (1981) showed this species to prefer no particular sediment. At Bluewaterbay beach in Algoa Bay, a series of dredge hauls from the beach to- wards and along the mouth region of the Swartkops River indicated the presence of Bullia until a point where the ripple effect of waves became negligible. This coincided with an increase in the slope of the river bank. During the hauls, there was no marked change in either sediment grade or salinity of seawater. A combination of the effects of wave action, water currents, and beach slope could act as factors limiting the distribution of the snails. Abundance The absence of juveniles of both Bullia digitalis and B. pura from the beach and their appearance in offshore dredged samples suggest that the egg-laying females mi- grate to deeper, less turbulent waters to deposit their egg capsules. This presumed offshore migration could account for the decrease in the beach populations of B. pura during January and February and B. digitalis from February to April. Eggs of B. rhodostoma are spawned in mid-summer (December/ January) (Mc LACHLAN et al., 1979a) at the same time as those of B. pura, but slightly before B. dig- italis (Mc GWYNNE, 1980). The entire population of Bulla digitalis moves offshore during the winter, while snails of B. pura, although few in number, are always found on the beach. This may be related to the apparently seasonal occurrence of carrion (Mc Gwynne, 1980) and the possible presence of a more constant food supply in deeper waters in winter. Bullia digitalis has been found at depths exceeding 20 m (BROWN, 1982) along the west coast, particularly off exposed beach- es with steep slopes. Snails of both species return in late spring to early summer to the shallows. Copulation takes place then, with the males exploiting the mobility of the swash to find mates. Females remain on the beach for about two to three Page 34 The Veliger, Vol. 28, No. 1 21h00 (1)(1)(15) 18h00 (0)(0)(9) 15h00 (1)(3)(26) TIME (h) 12h00 (0)(1)(17) O9h00 (2)(2)(3) VERTICAL METERS 60 50 40 30 20 10 O HORIZONTAL METERS Figure 5 Migrations undertaken by beach populations of Bullia pura (- - —), B. digitalis (+--+: ), and B. rhodostoma ( ) during a spring tidal cycle in November 1979. Actual numbers of each species caught in the dredge at each sampling time are shown in parentheses below the time scale on the vertical axis. The profile of Kings beach is also shown. months after copulation, with the sperm presumably stored. Tidal Migrations The second peak of abundance of Bullia digitalis and B. pura, in March and April, respectively, could indicate the return of the females to the beach after depositing their egg capsules offshore. All three Bullia species migrate with the tides, differing in the extent of their vertical penetration. Bullia rhodosto- ma, located higher on the shore than B. digitalis and B. pura, is carried with the swash and often stranded above Table 2 Numbers of Bullia rhodostoma (Br), B. digitalis (Bd), and B. pura (Bp) at five beach sites, each 4 m’, from the swash (site no. 1) to water of approximately 1 m deep (site no. 5). B = niche breadth. April June July August October Site no. Br Bd Bp Br Bd Bp Br Bd Bp Br Bd Bp Br Bd Bp 5 0 1 1 2 1 0 6 0 1 1 0 1 0 2 0 4 0 3 1 0 1 2 3 0 0 9 0 1 32 2 0 3 0 2 11 4 0 2 2 0 3 3 0 1 24 1 3 2 5 1 13 10 0 3 17 0) 0 2 0) 0) 41 0 0 1 12 0 20 19 0 1 22 0 1 50 0) 0 40 0 4 Total 17 i 46 35 2 8 50 0 5 65 0 3 137 5 7 B st Be Sil 7 AO BS) 0) OO Ar 16 O00 3.1 39 27 19 L. E. Mc Gwynne & A. Me Lachlan, 1985 Table 3 Niche breadth values, B, calculated for each species. % B of Species of Total B over theoretical Bullia all months Average B maximum B. rhodostoma 0.46 DES) 50 B. digitalis 0.87 1.6 32 B. pura 0.53 2.9 58 the tide, where it forages in search of stranded carrion. The latter two species are carried up the beach with waves on the incoming tide, and burrow rapidly into the sand as the velocity of the incoming wave diminishes. ANSELL & TREVALLION (1969), working on Bullia species on the Indian coastline, found the changing veloc- ity of the backwash of the ebbing tide to be the critical factor in the emergence and subsequent movement back to deeper water of these snails. Bullia digitalis and B. rhodostoma on west coast beaches occupy the same intertidal position (BROWN, 1971). Here, they kept pace with the tides during low and mid-water, emerging occasionally in the saturated foreshore to feed; they always remained buried at high water. On east coast beaches, B. digitalis (and B. pura) maintains a subtidal position, probably feeding on dead-animal remains wash- ing back and forth in the waves. It has not been seen foraging in the manner of the west coast species (MC GwyNNgE, 1980). The movements of the plough shells over a large part of the intertidal zone give them access to food in a wider area than if they remained fixed at one tidal level. Niche Breadth and Overlap Niche breadth indices suggest some spatial partitioning between the three Bullia species on the shore. Dredging Table 4 Niche overlap (a) between each of the three intertidal Bullia species over the period from April to June 1979. Br = B. rhodostoma. Bd = B. digitalis. Bp = B. pura. Species everlapping/ Niche overlap (a)? GOGGGS: je ee ee overlapped April June July August October Overall Br/Bd 0.14 0.06 0.00 0.00 . 0.35 0.17 Bd/Br 0.07 0.08 0.00 000 0.49 0.09 Br/Bp 1.20 072 036 0.20 0.46 £0.54 Bp/Br 0.65 0.53 0.41 0.11 O102" 1 0%47/ Bd/Bp 0.37 0.43 0.00 0.00 0.16 0.17 Bp/Bd 0.38 0.25 0.00 0.00 0.23 0.28 * The values indicate the probability of one species overlapping the other with the species being overlapped read as the denom- inator. Page 35 covered only an area from the swash line to a water depth of approximately 1 m. The niche breadth and overlap analyses, therefore, only indicate spatial niches in the shallow zone of swash sampled, and do not reveal the total dominance of the upper intertidal by B. rhodostoma or the absence of this species from deeper waters. The niche indices also give no indication of the competitive interac- tions involved in the partitioning of food between the three whelk species, the access to which is vital for their survival on the beach. The offshore migrations presumably undertaken by Bullia digitalis and B. pura would obviously reduce spatial overlap between all three species and increase their niche breadths. The species diversity and population dynamics of deep-water macrofauna have not been measured. The area beyond the breakers remains relatively unexplored. ACKNOWLEDGMENTS The authors thank members of the Zoology Department at the University of Port Elizabeth, particularly Mr. C. Cooper and Mr. E. Plumstead for their assistance during sampling and Miss M. Maree for preparing the figures. Financial assistance from the University of Port Elizabeth and the Council for Scientific and Industrial Research is gratefully acknowledged. LITERATURE CITED ANSELL, A. D. & A. Mc LACHLAN. 1980. Upper temperature tolerances of three molluscs from a South African sandy beach. J. Exp. Mar. Biol. Ecol. 48:243-251. ANSELL, A. D. & A. TREVALLION. 1969. Behavioural adap- tations of intertidal molluscs from a tropical sandy beach. J. Exp. Mar. Biol. 4:9-35. BALLy, R. 1981. The ecology of three sandy beaches on the west coast of South Africa. Doctoral Thesis, University of Cape Town. Brown, A. C. 1961. Physiological-ecological studies on two sandy beach Gastropoda from South Africa: Bullia digitalis Meuschen and Bullia laevisssma Gmelin. Z. Morph. Okol. Tiere. 49:629-657. Brown, A.C. 1964. Food relationships on the intertidal sandy beaches of the Cape Peninsula. S. Afr. J. Sci. 60(2):35-41. Brown, A. C. 1971. The ecology of the sandy beaches of the Cape Peninsula, South Africa. Part 2. The mode of life of Bullia (Gastropoda: Prosobranchiata) Trans. Roy. Soc. S. Afr. 39:281-319. Brown, A.C. 1982. The biology of sandy beach whelks of the genus Bullia (Nassariidae). Oceanogr. Mar. Biol. Ann. Rev. 20:309-361. Du Preez, H. H. 1981. The biology of the three-spotted swimming crab, Ovalipes punctatus (de Haan) (Brachyura: Portunidae) with special reference to feeding. Master’s The- sis, University of Port Elizabeth. Dye, A. H. & L. E. Mc Gwynne. 1980. The effect of tem- perature and season on the respiratory rates of three psam- molittoral gastropods. Comp. Biochem. Physiol. 66A:107- 111: HUTCHINSON, G. E. 1958. Concluding remarks. Cold Spring Harbor Symp. Quant. Biol. 22:415-427. Page 36 LEvVINS, R. 1968. Evolution in changing environments. Prince- ton University Press: Princeton, N.J. Mc Gwynne, L. E. 1980. A comparative ecophysiological study of three sandy beach gastropods in the Eastern Cape. Mas- ter’s Thesis, University of Port Elizabeth. 144 pp. Mc LacHLaNn, A. 1977. Studies on the psammolittoral meio- fauna of Algoa Bay, South Africa. I. Physical and chemical evaluation of the beaches. Zool. Afr. 12:15-32. Mc LacHuan, A. 1978. A temporal study of niche breadth and overlap in two sympatric species of Mystacocarida (Crustacea). Zool. Afr. 13(2):351-357. Mc LacHan, A. 1979. Volumes of sea water filtered through East Cape sandy beaches. S. Afr. J. Sci. 75:75-79. Mc LacHian, A. 1980. The definition of sandy beaches in relation to exposure: a simple rating system. S. Afr. J. Sci. 76:137-138. The Veliger, Vol. 28, No. 1 Mc Lacuian, A., C. Cooper & G. VAN DER Horst. 1979a. Growth and production of Bullia rhodostoma on an open sandy beach in Algoa Bay. S. Afr. J. Zool. 14(1):49-53. Mc Lacutan, A., T. Erasmus, A. H. Dye, T. WOOLDRIDGE, G. VAN DER Horst, G. Rossouw, T. A. LASIAK & L. E. Mc Gwynne. 1981. Sand beach energetics: an ecosystem approach towards a high energy interface. Estuarine, Coast- al and Shelf Science 13:11-25. Mc Lacuian, A., T. WOOLDRIDGE & G. VAN DER Horst. 1979b. Tidal movements of the macrofauna on an exposed sandy beach in South Africa. J. Zool. (Lond.) 188:433-442. Mc Lacuian, A. & N. YouNG. 1982. Effects of low temper- ature on the burrowing rates of four sandy beach molluscs. J. Exp. Mar. Biol. Ecol. 65:275-284. The Veliger 28(1):37-51 (July 1, 1985) THE VELIGER © CMS, Inc., 1985 Aspects of Reproduction, Larval Development, and Morphometrics in the Pyramidellid Boonea impressa (=Odostomia impressa) (Gastropoda: Opisthobranchia) by MARIE E. WHITE,' CHRISTOPHER L. KITTING,’? anp ERIC N. POWELL'* ‘Department of Oceanography, Texas A&M University, College Station, Texas 77843 "University of Texas at Austin, Marine Science Institute, Port Aransas, Texas 78373 Abstract. Boonea impressa is an important ectoparasite of the American oyster, Crassostrea virginica. Here, the reproductive and larval life history, intraspecific variation in certain shell characters, and the internal anatomy of the feeding apparatus are described for populations of B. impressa from the western Gulf of Mexico (Texas) and, for the latter two subjects, the western Atlantic (North Carolina). Larval development in the Pyramidellidae is reviewed. The life-span of B. impressa was approximately one year. Reproduction occurred throughout the year, but peaked in mid-summer. Eggs (182-238 um diameter) were deposited in numbers of 20-250 per egg mass. Larval development from oviposition to hatched veliger required 3.3-4.8 days. Two days after hatching, the veligers became negatively pho- totaxic. Metamorphosis occurred within one week of hatching. The developmental mode of B. impressa fits that designated as Type II-lecithotrophic, and agrees with that expected for an opisthobranch with a stable food source. The short pelagic life-span may facilitate dispersal for a species with a non-mobile, but patchy host. Recently metamorphosed B. impressa often attached near the aperture of an adult. This behavior may protect the young snail from predation and increase access to its food supply. The internal anatomy of the feeding apparatus differed from European odostomians in the absence of a well developed first buccal pump. Shell sculpture (number of cords per whorl) was most dependent on the length of the whorl. Adult snail size, whorl length, whorl width, and the number of spiral cords varied significantly between populations collected from Texas and from North Carolina. Egg size, size of the components of the feeding apparatus, whorl length-width ratio, and protoconch size differed less. These latter characters might be employed advantageously in the study of interspecific differences among odostomians where, heretofore, characters with greater intraspecific variability typically were used. INTRODUCTION PYRAMIDELLID GASTROPODS ARE important components of many shallow-water benthic communities (SANDERS, 1958; WELLS et al., 1961; FRANZ, 1976). Presumably, all are Parasitic (FRETTER & GRAHAM, 1949). As such, their impact on host population dynamics and subsequent changes in community structure may be important. Little is known, however, about pyramidellid life histories or their impact on host populations. * Person to whom reprint requests should be addressed. The pyramidellid Boonea impressa (=Odostomia im- pressa) is a frequent component of oyster reef communities on the east and Gulf coasts of the United States. ROBERTSON & MaAu-LASTOVICKA (1979) found that B. impressa can feed on 36 different gastropod and bivalve species. The predominant host, however, was the oyster Crassostrea virginica. WHITE et al. (1984) showed that the growth rate of juvenile oysters was reduced significantly at a parasite level of 10 Boonea impressa per oyster. Num- bers as high as 100 per oyster occurred on the Texas Coast (WHITE et al., 1984). WHITE et al. (1984) concluded that B. impressa may have a significant impact on oyster pop- ulations and oyster population dynamics. Page 38 The Veliger, Vol. 28, No. 1 ROBERTSON (1978) found that Boonea impressa, like all other American odostomians* studied, utilize spermato- phores for sperm transfer, whereas European species use penial copulation (FRETTER, 1951, 1953; Maas, 1964). Data of WELLS (1959), ROBERTSON (1978), and WHITE et al. (1984) indicated that reproduction occurs throughout the year, but peaks during early summer. ROBERTSON (1978) also noted that spermatophores are larger in Texas populations of B. impressa than in North Carolina pop- ulations. Additional morphometric data for North Caro- lina populations were reported by PORTER (1976) and PORTER et al. (1979). In view of the potential impact of Boonea impressa par- asitism on oyster populations and the limited data avail- able on the reproductive life history of B. umpressa, we undertook a study of its reproductive cycle and larval life history. In addition, we review the available data on other pyramidellids to elucidate whether the general trends in larval development described for other opisthobranchs are also applicable to the ectoparasitic Pyramidellidae. Taxonomic and ecologic studies on the Pyramidellidae have been hindered by a poor understanding of intraspe- cific variation within the group. Species distinctions and species descriptions tend to rely on highly variable char- acters. Species identifications often are difficult. ROBERTSON’s (1978) work is notable for the use of ana- tomical criteria beyond shell characters to confirm taxo- nomic distinctions. Intraspecific variability in anatomical characters still is documented poorly in the Pyramidelli- dae, however. The degree of variability in shell characters between populations also is little known. POWELL (1981) found whorl width to be highly variable between popu- lations in some Turbonilla, for example, thus limiting its taxonomic usefulness. Here, we also report data on some aspects of morphometrics, both of shell characters and internal anatomy, in North Carolina and Texas popula- tions of B. impressa with particular emphasis on a com- parison of the variability present in the internal anatomy of the feeding apparatus vis-a-vis that observed in shell morphology. MATERIALS ann METHODS Oyster clumps were collected from Big Slough on Harbor Island near Port Aransas, Texas, and kept in a running seawater system with adult Boonea impressa. Approxi- mately every 3 or 4 days, these clumps were examined for the presence of egg masses. Egg masses were removed with forceps and placed in small dishes of filtered seawater (24-26°C) to which penicillin G and streptomycin sulfate were added to control bacterial growth (BONAR & HAD- * The term odostomian is used here for species usually re- ferred to the genus Odostomia prior to ROBERTSON (1978) (e.g., ABBOTT, 1974; DALL & BARTSCH, 1909). FIELD, 1974). Development from oviposition through metamorphosis was studied by examining these egg masses under a microscope at hourly intervals. Additionally, daily observations were made of oyster clumps with attached egg masses that were kept in large bowls under the same conditions. Specimens of Boonea impressa were collected every oth- er month from a relatively undisturbed reef located off the south side of Mud Island near Port Aransas, Texas. Clumps of oysters were shaken vigorously in a bucket of seawater to remove all B. impressa. A careful visual check was done and the process repeated until no more snails were found. Snails were separated from debris by using a 500-um sieve and preserved in formalin. For shell mor- phometrics, the following measurements were taken using an ocular micrometer: shell length, shell width, number of whorls, length, width and number of cords of the second and sixth whorl, and the width of the larval shell. Shell length was determined by measuring the length from the apex to the abapical end of the shell. Only those shells with an intact protoconch were used. Shell width was determined by measuring the width of the largest whorl with the aperture facing upwards. These measurements also were recorded for specimens of B. zmpressa from North Carolina generously loaned by H. Porter. Collections were made at Virginia Creek near Topsail Sound and at Wil- liston Creek. PORTER (1976) and PORTER et al. (1979) gave additional information on these specimens. Specimens from Big Slough (Texas) and from North Carolina were dissected while living. The feeding appa- ratus including the proboscis, stylet apparatus, buccal pump, and salivary glands were removed and measured. Occasional staining with 0.5% toluidine blue or methylene blue during the dissection proved to be efficacious (DAVIS, (1967). The longest dimension of eggs taken from egg masses laid by both groups of snails also was measured. Boonea impressa specimens from each sample taken at Mud Island were decalcified using 0.5 M EDTA (eth- ylene diamine tetraacetic acid) and subsequently embed- ded in a paraffin medium and sectioned at 6 wm. Sections were stained for one hour with toluidine blue (method of PREECE, 1972, modified by using colloidinization during the rehydration step to ensure further that sections would remain on the slides). The sections were examined for the presence of sperm, as well as the number and size of the oocytes. RESULTS Larval Development and Population Dynamics The mean size of Boonea impressa in the population from Mud Island increased from October to May (Figure 1). The population’s size-frequency distribution changed significantly (Chi-square test, P < 0.05) between all sam- pling periods except March and May (P < 0.10 for the latter two). M. E. White e¢ al., 1985 Relative Percentage = ne) io) p oun a a a a > jee) oO 3 I a 2) N December 70 50 n= 513 30 10 0 A B Cc March 80 60 n = 238 40 20 0 A B Cc May 60 n= 49 40 20 0 A B iS July Maximum Width of Shell (mm) Page 39 Recruits (specimens 0.5-1.00 mm in width) were ob- served in all samples, but recruitment in December (22% of the population sampled were new recruits) was notice- ably higher than in October (10%), March (5%), or May (3%). The largest recruitment of juvenile snails was in July (55% of the population sampled were new recruits). Just prior to the July sampling, an extremely low tide and high temperature caused extensive mortality among the intertidal oysters. Although the population of B. zm- pressa sampled was subtidal, the population structure for July may not be indicative of the normal summer condi- tion. Mean oocyte size did not differ significantly among the October, March, and May collections (Duncan’s multiple range test, P < 0.05) (oocyte diameters in wm for October, March, and May were 17 + 2, 15 + 4 and 17 + 3 respectively). Significantly more oocytes were found in the snails collected in May than in those collected in either March or October (Kruskal-Wallis test, P < 0.05) (Table 1). The latter two months’ collections were not signifi- cantly different. No oocytes were found in any specimen collected in December, and oocytes were found in only a single specimen collected in July. Sperm were present in most or all specimens examined in every collection period. The eggs (Figure 2a) of Boonea impressa were laid in clear, irregular, gelatinous masses (Figure 3a) often de- posited in crevices near the edge of the oyster shell. The number of eggs in an egg mass varied from approximately 20 to 250 under laboratory conditions. Egg diameter (maximum dimension) ranged from 182 to 238 um. Sig- nificant differences were present in mean egg size between egg masses (Table 2). The range in egg size within an egg mass was always less than the range in egg size among all the egg masses measured. Eggs in a single egg mass tended to be of similar size so that some egg masses con- sisted almost exclusively of small eggs, whereas others consisted almost exclusively of large eggs. No difference was apparent, however, between the North Carolina and Texas populations. The early embryological development of Boonea im- pressa followed the typical pattern for opisthobranch mol- lusks by exhibiting spiral cleavage and asynchronous cell division (RAVEN, 1958, 1964). Total developmental time required from oviposition to hatched veliger was 80-114 h (Table 3). The first cleavage of eggs occurred 2 h after oviposition, the second and third divisions (8 cells) after 4-6 h (Figure 2b). After 26-30 h, a gastrula had formed (Figures 2c-e). Further development was divided into three stages: ear- Figure 1 Size-frequency distributions for the Boonea impressa population at the Mud Island reef during Fall 1981 to Summer 1982. Shell widths of A, B, and C are 0.50-1.00 mm, 1.01-1.66 mm, =1.67 mm respectively. Page 40 Table 1 The mean, standard deviation, and range of the number of oocytes observed in the stained sections of Boonea im- pressa gonad. The snails were collected from the Mud Island reef. Sampling period Mean + SD Range October 53:25) c215:38 48-60 December 0 0 March 50.25 + 20.02 33-75 May 140.00 + 27.12 102-160 July 229) 45 0-9 ly-, mid- and late-veliger. The early-veliger stage, reached 32-36 h after oviposition, was characterized by the first noticeable movement of the embryo and the beginnings of velum development; however, neither shell nor statocysts were present and a bipartite velum was not observed. The mid-veliger stage, reached 50-54 h after oviposition, was characterized by the presence of statocysts, a bipartite ve- lum, and a partially developed shell. The shell, however, did not extend down to the level of the statocysts, but covered only the upper part of the visceral mass (Figure 2f). The capability of retraction into the shell was not present nor was the velum completely developed. In par- ticular, although the velum was ciliated, the long cilia characteristic of the velum of the hatched veliger were not present. Movement within the egg was most rapid at this stage and slowed noticeably thereafter. Between this stage and the late-veliger stage, reached 56-60 h after oviposi- tion, the embryo grew rapidly from a size roughly one- half of the egg volume to a size nearly filling the entire egg volume. Prior to this, development had not markedly increased embryo size. The late-veliger stage was char- acterized by a fully developed velum and a completely developed shell extending down below the level of the statocysts. The long cilia characteristic of the velum of a planktonic larva were fully formed only at this stage. Ad- ditionally, the embryo first showed the capability of re- traction into the shell at this stage. Hatching occurred 80-114 h after oviposition (Figure 3b). After hatching, veligers frequently were caught at the air-water interface by surface tension. Strands, probably the remnants of the “mucus string” that bound the eggs together (RASMUSSEN, 1944), often connected as many as 20 floating larvae together. Trapped veligers did not seem to be capable of submerging and subsequently died at the Figure 2 The Veliger, Vol. 28, No. 1 surface, unless the surface water was actively disturbed. Submerged veligers immediately demonstrated rapid movement both horizontally and vertically. During the first two days, movement was positively phototaxic and rapid. Afterwards, the veligers became negatively photo- taxic and movement slowed considerably. For lengthy pe- riods of greater than 1 h, the veligers often remained re- tracted into the shell or were stationary on algae or the bottom. Many larvae metamorphosed in the large bowls in which oyster clumps were present: however, only a single indi- vidual was observed to metamorphose under the micro- scope. This occurred seven days after hatching. In an at- tempt to get more larvae to metamorphose, various possible metamorphosis inducing factors such as oyster shells, liv- ing oysters, algae typically found on oyster shells, and adult Boonea impressa were placed separately in bowls with apparently competent larvae (in the sense of CHIA, 1978), but without success. Larvae probably were com- petent to metamorphose, however, when negative photo- taxis was observed, about two days post-hatching. Thus, seven days is probably an overestimate of the average lar- val life span in this species. Newly metamorphosed snails were observed crawling freely, but most often they attached just outside the ap- erture on the outer lip of an adult Boonea impressa (Fig- ures 3c, d). Juvenile B. wmpressa up to two teleoconch whorls frequently were observed demonstrating this be- havior. Shell Morphometrics The mean length, width, and length-width ratio of whorls 2 and 6 of snails from Mud Island, Texas, were similar in each of the five collecting periods (Table 4). The number of spiral cords in whorl 2 was not signifi- cantly different for any of these samples (Duncan’s mul- tiple range test) either, averaging about three. The num- ber of spiral cords in whorl 6 varied considerably more. Snails collected in May had significantly fewer cords than in any other month but October (Duncan’s multiple range test, P < 0.05). The October, December, March, and July samples did not differ significantly, nor did October sam- ples differ significantly from May (P > 0.05). The width of the protoconch varied only slightly, ranging from 234 to 240 um. This variation was considerably less than noted for egg size. The two populations of North Carolina snails were not significantly different except for the number of spiral cords and length of whorl 6; thus, they were treated as one Developmental stages of Boonea impressa from the Mud Island reef, Texas. A, egg. B, 4-celled stage. C, multi- celled stage. D, blastula. E, gastrula. F, mid-veliger stage. Scale bars: A, 40 um; B, 24 um; C, 23.6 wm; D, 57.5 pm; E, 21.5 um; F, 38.8 um. M. E. White et al., 1985 Page 41 Page 42 The Veliger, Vol. 28, No. 1 Figure 3 Larval and juvenile stages of Boonea impressa collected from the Mud Island reef, Texas. A, adult B. impressa and egg mass. B, hatched veliger. C and D, juvenile on adult aperture. Scale bars: A, 2 mm; B, 48 um; C, 68.5 wm; D, 240 um. sample (Table 5). Overall, North Carolina snails were larger. The width of whorl 6, for example, was signifi- cantly larger (Duncan’s multiple range test, P < 0.05) than in any of the Texas samples, except May. The length of whorl 6 was also larger than in any of the Texas sam- ples, but the difference was significant only for the De- cember and October samples (Duncan’s multiple range test, P < 0.05). The number of spiral cords in whorl 6 in the North Carolina snails was significantly higher than in any of the Texas samples (Duncan’s multiple range test, P < 0.05). In contrast, protoconch size and the length- width ratio of whorl 6 differed little between the two populations. Mean egg size also was very similar (Table 2). A Pearson product-moment correlation test based on all shell measurements was conducted on the Texas samples (Table 6). The shell length, width, and number of whorls all were significantly correlated with each other for every month. In most samples, the whorl width and length for whorl 6 were correlated with both the total shell width and total shell length. This was not true for whorl 2. The width of whorl 2 was correlated with the width of whorl 6, but the lengths were not consistently correlated. Thus, the rate of whorl expansion was more constant than the rate of whorl translation. Consequently, although the length and width of whorl 6 were correlated in most sam- ples, the correlation coefficients were low. The length of whorl 6 and the number of spiral cords in whorl 6 were also correlated in most samples, but width of whorl 6 was correlated less well with spiral cord number than was the M. E. White ef al., 1985 Page 43 Table 2 Mean and range of the sizes of eggs in each Boonea 1m- pressa mass. Seven egg masses were measured: five from snails collected at Big Slough, Texas (No. 1, 2, 3, 4, 7) and two from North Carolina (No. 5, 6). n is the number of eggs measured. A, B, and C indicate results of Duncan’s multiple range tests (a = 0.05) where mean egg size for egg masses with the same letter are not significantly dif- ferent. Signifi- Egg Mean Range cance mass (um) (um) n (a = 0.05) 1 218 214-238 8 A 2 209 198-222 15 B 3 207 198-222 9 B 4 205 214-222 12 B 5 205 190-222 12 B 6 192 190-206 12 Cc 7 189 182-190 10 C length. The number of spiral cords in whorl 2 was poorly correlated with any other shell feature. Larval shell width was poorly correlated with any other character except the width of whorl 2. In order to determine whether the number of cords in whorl 6 was influenced by any other shell feature besides whorl length, a stepwise regression test (maximum R? improvement) was conducted. This procedure (for all samples—Texas and North Carolina) showed that the length of whorl 6 was the best one-variable model for Table 3 Development time for the embryonic stages of Boonea im- pressa. Time from oviposition to when stage was Larval stage first observed Two cells 2h Four cells 4-6h Gastrula 26-30 h Early-veliger 32-36 h Mid-veliger 50-54 h Late-veliger 56-60 h Hatched veliger 80-114 h Veliger negatively phototaxic Metamorphosis ~ 6 day (2 day post-hatch) ~11 day (7 day post-hatch*) * Probably a maximum. determining the cords in whorl 6. The R? values, however, were only 0.04, 0.13, 0.18, 0.18, and 0.24 (October, De- cember, March, May, and North Carolina respectively). For the Texas populations, addition of shell width, width of whorl 6, and the number of whorls improved the cor- relation only marginally (corresponding R? = 0.05, 0.19, 0.25, 0.20). In contrast, the same procedures improved the correlation considerably for the North Carolina popula- tion. By adding the width of whorl 6, R? increased from 0.24 to 0.45, and then to 0.54, by adding the number of whorls and shell width. Table 4 Mean and standard deviation of shell characters (in mm) of the samples of Boonea impressa from the Mud Island, Texas reef. Number measured is in parentheses. Length-width Length-width Number Width Shell Shell of Length— Width— ratio— Cords— Length— Width— ratio— Cords larval Sample length width whorls whorl 6 whorl 6 whorl 6 whorl 6 whorl 2. whorl 2 whorl 2 whorl 2 shell October 3.44 1.47 5.99 0.71 1.43 0.50 3.92 0.23 0.48 0.48 3.00 0.234 +0.96 +0.27 #1.15 +0.05 +0.09 +0.34 +0.02 +0.03 +0.03 +0.01 (293) (400) (293) (231) (231) (231) (293) (293) (293) (293) December Det 1.42 5.53 0.74 1.42 0.52 3.99 0.23 0.48 0.48 2.99 0.236 +1.40 +0.41 +1.52 +0.07 +0.09 +0.43 +0.01 +0.02 +0.08 +0.02 (331) (462) (331) (143) (143) (143) (331) (331) (331) (331) March 3.75 1.65 6.20 0.77 1.48 0.52 4.09 0.24 0.49 0.49 3.01 0.238 +#1.35 +0.32 +£1.31 +0.07 +0.12 +0.49 +0.01 +0.03 +0.18 +0.01 (280) (537) (280) (170) (170) (170) (280) (280) (280) (280) May 4.05 1.78 6.39 0.76 1.51 0.50 3.76 0.24 0.49 0.49 3.00 0.239 +1.23 a0) ail 7 +0.07 +0.13 +0.51 +0.01 +0.03 +0.00 +0.01 (133) (237) (133) (104) (104) (104) (133) (133) (133) (133) July 2.06 1.03 4.14 0.76 1.44 0.53 4.00 0.24 0.50 0.48 3.00 0.240 +0.94 +0.31 sell yall +0.07. +0.14 +0.00 +0.02 +0.04 +0.00 +0.00 (49) (49) (49) (8) (8) (8) (49) (49) (49) (49) Page 44 Mean and standard deviation of shell characters (in mm) Table 5 of samples of Boonea impressa from North Carolina. Williston Virginia Creek Creek Character (n = 22) (n = 9) Combined Number of whorls 6.98 + 0.32 7.00+0.65 6.99 + 0.43 Width of shell 1.73 +0.11 1.75+40.16 1.74+0.12 Length—whorl 6 0.80 + 0.05 0.75 +0.03 0.79 + 0.05 Width—whorl 6 1.56+0.10 1.51+0.09 1.55 +0.10 Cords—whorl 6 4.55+0.65 4.06+0.17 4.40 + 0.60 Width—larval shell 0.24 0.24 0.24 Length-width ratio— whorl 6 0.51 0.50 0.51 Feeding Apparatus The proboscis and associated feeding structures of Boo- nea impressa were similar to other odostomians described by Maas (1965). Here, we use the terminology of FRET- TER & GRAHAM (1949) and cross-reference it to that of Maas (1965) as much as possible. Rather than repeating the detailed descriptions of Maas (1965), we emphasize only the differences observed. The feeding apparatus of B. impressa consisted of a buccal pump (Pumpbulbus II of Maas, 1965) to which the esophagus and salivary glands attached at its proximal end, a long tubular structure ho- mologous to the first buccal pump (Pumpbulbus I) de- scribed by Maas (1965), the stylet and associated struc- tures, and the proboscis (Figure 4). The first buccal pump, well developed in the European odostomians described by Maas (1965), ANKEL (1949a, b), and FRETTER & GRA- HAM (1949), was poorly developed. In its place was a long tubular structure connecting the buccal pump to the stylet tube. This long tube thickened gradually but noticeably over the last 25% or so of its length at the end where it connected with the buccal pump. Possibly this thicker por- tion functions as the first buccal pump does in European odostomians. The salivary glands consisted of four sections: a prox- imal section containing about 2 or 3 small linearly ar- ranged, circular rings of cells; a larger, wider middle section with 5 or 6 linearly arranged, circular groups of cells; a second but usually narrower middle section with 15-17 circular groups of small cells; and a very narrow distal region that might function as a storage compartment for the salivary cells’ products (FRETTER & GRAHAM, 1949; Maas, 1965). Serial sections were not studied; cell num- bers were determined by staining during dissection. Thus, the variability in the number of cells observed within each group might be an artifact of preparation rather than true variability. Considerable variation was present in the width of the salivary glands so that, on occasion, the salivary glands were nearly cylindical in shape, as opposed to the The Veliger, Vol. 28, No. 1 more common appearance depicted in Figure 4. Even when cylindrical, however, the four groups of cells were readily distinguishable. The salivary glands closely resembled those described by Maas (1965) and ANKEL (1949b) from Odo- stomia plicata and by FRETTER & GRAHAM (1949) from O. unidentata, except that the proximal group of small cells is absent in O. plicata. Differences in shape observed by Maas (1965) for O. eulimoides, however, which are similar to differences described above for Boonea impressa, and the hypothesized cell cycle whereby new cells origi- nate near the middle of the gland and move proximally as they grow (ANKEL, 1949b; FRETTER & GRAHAM, 1949) suggest that salivary gland morphology may be variable from snail to snail. Thus, the significance of the similar- ities and differences noted by Maas (1965) and us as taxonomic criteria remains unclear. Approximate sizes for the various components of the feeding apparatus are given in Table 7. Except for the two cuticularized structures, the stylet (Stachel of Maas, 1965) and the stylet tooth (Stempel of Maas, 1965), the sizes varied considerably in different preparations due to relaxation or contraction and should be considered as rough estimates only. No significant difference in the sizes of any component was found between the North Carolina and Texas populations. DISCUSSION Reproduction and Growth Our results agree with those of WELLS (1959) from North Carolina that the life-span of Boonea impressa is approximately one year and that reproduction and re- cruitment to the population occurs more or less continu- ously. Reproduction and recruitment rates are not con- stant, however. Although sperm were present in all adult specimens in all months sampled, marked differences in oocyte numbers were found. In May, approximately 38% more oocytes were found than during any other sampling period. No oocytes were found in December; this could possibly be correlated with the cold water temperatures encountered at that time. WELLS & WELLS (1961) sug- gested that reproduction in B. seminuda was directly re- lated to water temperature. The absence of oocytes in most of the specimens in July probably was due to the young age of the majority of the specimens collected. Growth rates also were comparable between the Texas population studied here and the North Carolina popula- tion examined by WELLS (1959). In both populations, the large summer set reached adult size in early spring of the following calendar year. Both populations consisted of predominately juvenile individuals in mid-summer and predominately adult individuals in late spring. Thus, re- production and recruitment, although continuous, are markedly higher in early summer (May-July). This more or less coincides with the peak period of oyster reproduc- tion in the study area (GUNTER, 1941; COPELAND & HOESE, 1966). Adults of Boonea impressa were most abun- Page 45 M. E. White et al., 1985 14s LV0 [ease] * Em 67:0 €7'0 ‘€EL'0 ‘E70 cz 0 = a aaa WPIM 610 cr'0 ‘Zr €70 9 JZoyM - x ‘C70 “LZ'0 ‘9€'0 ‘€1'0 LV0‘L1'0 610 ‘6€'0 ‘670 S70 ‘1€0 820 ‘Zr'0 “6€'0 —Spi0p 80 6€'0 “6€'0 bz0 08'0 ‘Lt'0 6c 0 9 [Zoym 1 ve ‘ST * ‘VEO “€€'0 ‘€v'0 ‘LZ'0 veo ‘870 ‘€7'0 ‘090 ‘Z€"0 ‘LL'0 ‘LE'0 ‘9¢'0 ‘790 —UIPIM 1¢'0 80 L9°0 490 9 JZoyM 1 ee 6 b€ ‘7 * “S70 ‘870 820 ‘9F'0 ‘Or'0 ‘€€0 ‘8b'0 ‘770 ‘LS°0 ‘19°0 ‘050 ‘bS'0 —yisue] Z [HOUM == = = a — Z1'0 €l0 €0 €1'0 —Spi0p, ~ [Woum GL Gal ne G. | CaCl * SL'0 “610 ‘87'0 ram) 910 11-0 Z1'0 “L1'0 —UIPIM ~ [YoumM 1 € 1 €‘l S ‘eT * 910 ‘STO 910 “L1'0 ‘%'0 810 ‘810 ‘€z0 —yisueT +60 ‘760 460 ‘€6'0 sjioym = ve T Gee 1 ve Z z ZI * ‘¥6'0 ‘260 ‘760 ‘96'0 ‘L6'0 ‘960 Jo roquin Ny 460 960 yapIm = wie See G.l ie Gl Cac SG! Gi Cau * ‘S60 ‘€6'0 960 14s yi3u2] = ve ‘7 TEGm Cal aaa | € ‘7 € ‘C1 G ‘bE ST G ‘bE ‘ST * Pus 1194s 9 J1Oym 9 [ouM 9 Jyoym Z [oum Z Hoym sjuoym WIPIM TPPYS yrsuzq [24s yeas] —Sspiop —UIPIM —yisua'] —UIPIM —yisus'T jo saquinn YIPIM ‘Ajaatoadsas 6770 pure ‘7€°0 “OF'0 ‘Z9'0 = 4 HIM fF ‘€ ‘Z “T SYUOUT UT syIBUEZT [[PYs YIM payefassoo APULoytUsIS sem g [LOYM Jo UYXpIM ‘a[dwexa 10,7 ‘ouTTpTU [euoZeIp ay} Jo MOjaq pur Io] SLUT JOLITUT IY} UT UDA a1v SUONLIIIIOI JULIYTUBIS 10J syUITOYJI09 UONRIIIO|D “ATaaoeadsau A[n[{ pue ‘AvP ‘yourepy ‘Jaquiso9q, ‘¥aqoIO wos sajduies jussaidas ¢ pur ‘p ‘¢ ‘Z ‘J suaquINN, “(GQ'O = %) APURoyIUsIs payejais09 a1aM YOIYM vdssacdun vauoog jo ssajoereYyO d1nydyNos ][[IY4S Oe IGUue Page 46 dhe Veliger Voly285 Nol == Buccal Pump I Introverted Proboscis Introverted Proboscis Stylet Tube Stylet Tooth | | = Stylet — =a | Figure 4 The internal anatomy of the feeding apparatus. Bars indicate the method of measurement for dimensions reported in Table 7. Above, the entire feeding apparatus excluding the proximal portion of the proboscis. Below, an enlargement of the stylet tube. dant during the spring when oyster gonadal development and spawning occurred, and juveniles of B. 7zmpressa were most abundant in the summer and fall when oyster spat were also most common. Larval Development—Boonea impressa THOMPSON (1967) classified the larval development of opisthobranchs into three categories: Type I-plankto- trophic larvae with small ova (40-170 wm), a short em- bryonic period (2-28 days), and a free-swimming veliger stage usually of >3 days duration; Type II-lecithotrophic larvae with larger eggs (110-250 um), a longer embryonic development (4-42 days), and a free-swimming veliger stage usually of <3 days duration; and Type III-direct development with even larger ova (205-400 um), an even longer embryonic period (13-50 days), and hatching at the post-larval stage. Type II includes THORSON’s (1950) planktonic larvae with a short pelagic life-span. Devel- opment in pyramidellids fits more or less into THOMPSON’s (1967) scheme. In Table 8, we compare our data on Boo- nea impressa with other data available on pyramidellids where both egg size and embryonic development time are M. E. White e al., 1985 Page 47 Table 7 Measurements of components of the feeding apparatus of Boonea impressa. Terminology of Maas (1965) appears in parentheses. Measurements were made as shown by the bars in Figure 2. Mean Structure Population (um) Maximum width of shell NC 1920 TX 1856 Buccal pump, length NC 1497 (Pumpbulbus 11) TX 1616 Buccal pump connecting NC 3834 tube, length TX 3676 (Pumpbulbus 1) Salivary glands, length NC 861 TX 921 Stylet bulb, length NC 152 (blind sack) TX 131 Stylet length (Stachel) NC 218 TX 207 Stylet width NC 45 TX 44 Stylet tooth, length NC 199 (Stempel ) TX 196 Stylet opening, length NC 47 (Stachelofinung ) TX 40 Range Mean Range (um) n (um/um) (um /um) 1840-2000 6 1760-1920 5 1310-1736 6 0.781 0.616-0.904 1499-1767 5 0.873 0.781-0.960 2761-4418 5 1.990 1.381-2.301 3077-4181 5 1.984 1.603-2.186 757-994 9 0.447 0.395-0.540 742-1136 8 0.500 0.386-0.617 136-174 3 0.077 0.068-0.091 124-143 4 0.070 0.067-0.074 202-233 5 0.113 0.101-0.127 182-225 5 0.112 0.095-0.128 39-47 4 0.023 0.020-0.024 39-47 5 0.023 0.022-0.026 174-221 5 0.103 0.087-0.120 174-233 5 0.106 0.091-0.132 39-54 5 0.024 0.020-0.029 39-43 4 0.022 0.020-0.023 known. Boonea impressa is Type Il. Egg size and devel- opment time from oviposition to hatching are well within the range suggested by Thompson. Larval life-span is somewhat longer than Thompson’s range for other Type II larvae; however, if the advent of negative phototaxic behavior marks the initial competence for metamorphosis, then the minimum planktonic life-span would be about 3 days rather than 7. This is close to THOMPSON’s (1967) range for Type II life-spans. Furthermore, B. wmpressa does not show significant growth during the planktonic phase. Egg size and hatched veliger size are no less than 80% of size at metamorphosis. Egg sizes range up to 238 um and protoconch size as measured on the adult was also in this range. Thus, feeding, if it occurs, probably is rel- atively unimportant in the planktonic stage. Overall, development in Boonea impressa most closely resembles that described for the form of Brachystomia ris- soides with a planktonic larva (RASMUSSEN, 1944) and Odostomia eulimoides (FRETTER & GRAHAM, 1949), both of which also are Type II. Egg size is similar, as are de- Table 8 Comparison of egg size, development time, and larval life-span in the Pyramidellidae. Development time: Species Egg size oviposition to hatching Larval life-span Authority Boonea impressa 182-238 um >3 to <5 days 7 days our data (probably 3-7) Brachystomia rissoides 300-650 um 25 days none RASMUSSEN (1951) Brachystomia rissoides ~200 um 6.5 days p RASMUSSEN (1944, 1951) THORSON (1946) Eulimella nitidissima ~100 um 5 days long RASMUSSEN (1944) Odostomia eulimoides ~160 um 10-12 days 3-4 days FRETTER & GRAHAM (1949) LEBOUR (1932) Chrysallida cincta 300-340 wm 22-27 days none LAFOLLETTE (1977, 1979) Odostomia omaensis 120-150 um 8 days ? AMIO (1963) Odostomia desimana 130-160 um 14 days ? AMIO (1963) Page 48 velopment time and larval life-span (Table 8). The longer development times for O. eulimoides and Brachystomia ris- soides probably are due to a lower temperature regime (see SPIGHT, 1975). On the other hand, RASMUSSEN (1944) found that the shell and statocysts of Brachystomia rissoides with a planktonic larva developed prior to formation of a bipartite velum, and observed first movement only after 100 h. In Boonea impressa, the statocysts and a bipartite velum were present prior to complete formation of the shell. First movement was observed at 32-36 h, prior to shell formation or the development of a bipartite velum. In fact, in this regard, Boonea impressa more closely re- sembles Eulimella nitidissima for which movement was ob- served at 53 h, prior to the development of a bipartite velum or statocysts (RASMUSSEN, 1944). Additionally, the 75 h embryo is similar to our mid-veliger stage reached at 50-54 h in that shell formation is incomplete: the shell covers only the visceral mass somewhat above the level of the statocysts. Eulimella nitidissima, however, has a Type I-planktotrophic larva. Egg size is considerably smaller than in Boonea impressa and the larva more than triples in size during the planktonic phase (RASMUSSEN, 1944). Thus, although the larval development of Boonea impressa is best described as Type II overall, certain aspects of its embryonic development more closely resemble that of E. nitidissima which results in a Type I larva. Larval Development—Pyramidellidae Some information is available for a number of other pyramidellid species. Parthenia decussata, which grows considerably during its planktonic life-span and has a small egg size (90-120 wm) (LEBouR, 1936), also can be con- sidered Type I. At the other extreme, Chrysallida cincta and one form of Brachystomia rissoides have direct devel- opment (Type II of THompson, 1967) (RASMUSSEN, 1951; LAFOLLETE, 1977, 1979). ROBERTSON & ORR (1961) sug- gested that Odostomia chitonicola also may have direct de- velopment. AMIO (1963) discussed two additional Odosto- mia species with egg sizes and development similar to Boonea impressa. Thus, all three types of larval develop- ment described by THOMPSON (1967) are present in pyr- amidellids, with each larval type represented by at least two of the seven species for which some data are currently available. Apparently, ectoparasitism has produced no obvious universal modification to the opisthobranch developmental plan. This suggests that factors determining develop- mental mode in opisthobranchs generally might apply to the Pyramidellidae also. CLARK & GOETZFRIED (1978) suggested that trophic stability was an important factor. Direct development would be favored when the food source was stable or predictable, a planktonic larva when the food source was unstable or unpredictable. The pyrami- dellid species listed in Table 8 having either direct devel- opment (Type III) or a lecithotrophic larva (Type II) usually parasitize organisms with long life-spans or or- The Veliger, Vol. 28, No. 1 ganisms that are components of persistent (in the sense of BOESCH et al., 1976) communities. Chrysallida cincta has direct development and parasitizes gastropods such as Halvotis corrugata and Tegula eiseni whose life-spans prob- ably exceed 10 yr (LAFOLLETTE, 1977). Similarly, hosts for Brachystomia rissoudes and Odostomia eulimoides live 10-20 yr (FRETTER & GRAHAM, 1949; COMFoRT, 1957; ANKEL & CHRISTENSEN, 1963). The host of Boonea im- pressa is the keystone species of a particularly persistent community, the oyster reef, so that food supply and lo- cation is dependable year to year. In contrast, although the host of Eul:mella nitidissima is unknown, the plank- totrophic larva of EF. nitidissima suggests that the host’s population will be temporally less stable than in the above species. Although adult snails frequently move from one host to another (ANKEL & CHRISTENSEN, 1963; WHITE et al., 1984), movement by adults between host populations probably is rare. A short pelagic life-span of the type demonstrated by Boonea impressa might be expected, par- ticularly when the host species is immobile, even though trophic stability might favor direct development. Both gene flow and dispersal between host populations would be facilitated. Of the three species with uniformly only Type II or Type III development, both species (B. impressa and Odostomia eulimoides) which primarily parasitize immo- bile hosts (bivalves in these cases) have larvae with a short pelagic phase. In contrast, the one species with only direct development, Chrysallida cincta, parasitizes gastropods, all of which have at least some mobility that might facilitate adult dispersal. The few data available suggest that development time increases with increasing egg size in pyramidellids, as was suggested for other gastropods (e.g., SPIGHT, 1975; STRATHMANN, 1977). The shorter time for Boonea im- pressa relative to other species of the same egg size prob- ably can be attributed to the higher temperature regime of Texas bay waters. There appears to be little relation- ship between development mode and taxonomic status. Disparate modes are found in one species, Brachystomia rissoides, and very similar modes in clearly distinct genera (e.g., Boonea and European Odostomia). Juvenile Behavior The behavior of the young Boonea impressa veligers was positively phototaxic the first two days but then became negatively phototaxic. THORSON (1950) suggested that positive phototaxis allowed young larvae to stay near the surface where currents might aid their dispersal, whereas negative phototaxis in older larvae that were ready to metamorphose increased the time spent near the bottom and, thus, increased their chances of finding a suitable substrate for settlement. The frequent observations of juvenile Boonea impressa attached near or at the aperture on the outer lip of the shell of adult B. impressa are too frequent to be simply M. E. White et al., 1985 accidental, but suggest a behavioral mode that might in- crease juvenile survival. Several advantages are possible. (1) Predation might be decreased, particularly by preda- tors that are too small to attack an adult snail. Small predators, such as polychaetes and juvenile crabs, are common on oyster reefs. Movement over the host might be accomplished more safely by hitching a ride because fewer potential predators would be encountered. (2) Small B. impressa may be unable to approach the oyster’s mantle closely enough for feeding or to maintain a stable foothold on the oyster shell because the proboscis and foot are small and the edge of the oyster shell tends to be ragged. Adult B. impressa may provide a more stable substrate. (3) In fact, one cannot rule out the possibility that juveniles ac- tually might feed on the adults for a short time until a size is reached that allows feeding on the oyster host. It seems unlikely that the outer mantle fold of the oyster can be fed upon because newly formed periostracum would interfere (see GALTSOFF, 1964; WALLER, 1980), and the remainder of the mantle might be difficult to reach with the short proboscis of a juvenile. Juvenile gastropods fre- quently utilize food resources not used by adults (KITTING, 1984). Boonea impressa certainly is capable of feeding on a variety of species, some of which may be more easily utilized by juveniles than are oysters. Morphometrics—Shell Characters Lopes (1958), WHARTON (1976), PORTER (1976), Por- TER et al. (1979), POWELL (1981), and others discussed the intraspecific variability in certain shell characters often used for taxonomic identification in pyramidellids. Some, such as axial rib number and spiral cord number, are particularly variable. The North Carolina and Texas populations differed considerably in some respects. Snails from the North Carolina population were larger, and they had a greater width and length at whorl 6 than the Texas snails. Mean width of whorl 6, for example, ranged from 1.42 to 1.51 mm among the Texas samples, but was 1.55 mm in the North Carolina snails. In addition, the number of cords in whorl 6 was significantly greater in the North Carolina specimens than in any sample from the Texas population. The number of populations sampled was too few to suggest a regional difference in size or cord number. The data do indicate, however, that significant inter-pop- ulation differences are present in shell sculpture and size. POWELL (1981), PORTER (1976), and PORTER et al. (1979) described similar variability in other pyramidellid species. Unfortunately, both shell sculpture and size are often used as taxonomic characters for identification. In Boonea impressa, certain characters are much less variable. North Carolina and Texas specimens had very similar length-width ratios at whorl 6. Egg size and pro- toconch size were nearly identical. The size and shape of the feeding apparatus, including stylet, buccal pump and salivary glands, also were very similar. POWELL (1981) found that both length-width ratios and protoconch size Page 49 were less variable between populations of several 7urbo- nilla (Pyrgiscus) species than other shell characters, and suggested their taxonomic usefulness in the Pyramidelli- dae. Our data support this conclusion. The number of cords at whorl 6 was more closely cor- related to whorl 6 length than any other parameter. Cer- tainly, the larger lengths of whorl 6 in the North Carolina snails explain the larger number of cords observed. Whorl 6 length alone, however, cannot explain all of the varia- tion observed. The significant differences in cord number for whorl 6 between some collections from the Texas pop- ulation (e.g., the May and March collections), for exam- ple, cannot be explained easily by differences in whorl 6 length or in any other shell character measured. Thus, seasonal or other environmental changes also may influ- ence cord number. The size of the protoconch was correlated with only one other shell feature, the width of whorl 2. Interestingly, the widths of whorls 2 and 6 were correlated much better than the lengths of the same two whorls. POWELL (1981) pointed out that the length-width ratio and the sculpture of early whorls frequently differ considerably from those of the later adult whorls in pyramidellids. That is, both shell sculpture and growth form often change dramatically with age (see also Laws, 1937). Increased variability with age is an important taxonomic problem in the Pyrami- dellidae where species frequently are described from ju- venile individuals. Our data suggest that, for Boonea im- pressa, whorl width and the rate of whorl expansion are determined to a larger extent by factors also determining protoconch size than are the whorl length and the rate of whorl translation. Additionally, the number of cords in whorl 2 was not correlated with any other shell feature, unlike whorl 6 where a good correlation with whorl length was present. In fact, there was almost no variability in cord number in whorl! 2, and this number was less than that typically given in descriptions of the species (7.e., three rather than four cords). Morphometrics—Feeding Apparatus ROBERTSON (1978) distinguished American and Euro- pean odostomians based on several features including the method of sperm transfer. European odostomians used penial copulation, whereas spermatophores were present in American species. The feeding apparatus of Boonea impressa exhibits another striking difference between Boo- nea and European odostomians. In all European odosto- mians studied, the first buccal pump is well developed (Maas, 1965; FRETTER & GRAHAM, 1949) and attaches closely to the stylet tube. In B. impressa, the first buccal pump is very poorly developed and attaches by way of a long tube (over twice as long as the second buccal pump) to the stylet tube. This reinforces ROBERTSON’s (1978) suggestion that American odostomians are deserving of a separate generic status from their European counterparts Page 50 and suggests that anatomical studies may provide impor- tant information for species and generic determinations. Descriptions by FRETTER & GRAHAM (1949), FRETTER (1953), and Maas (1965) all suggest that stylet length and size of the salivary glands and buccal pump may be good taxonomic characters, but measurements relative to shell size for comparison to Boonea impressa are unavail- able. Nevertheless, the similarity between populations in the feeding apparatus (and in the size of the larval shell) sharply contrast to the differences present in many shell characters normally used for species distinctions. Char- acters with limited inter-population variability should be highly useful taxonomic characters when species specific differences are present. The evidence suggests that de- tailed studies of the feeding apparatus in the Pyramidel- lidae may provide useful comparative data when shell morphological criteria are too variable to provide unam- biguous results, just as internal anatomical characteristics have in other groups of small, taxonomically abstruse groups of snails (DAvis, 1967; Davis & CARNEY, 1973). ACKNOWLEDGMENTS Special thanks go to Hugh Porter who sent us snails from his collection and who collected living snails and sent them to us for the dissections. We thank B. Rogers for assistance in collecting Texas snails. We thank Drs. S. Ray, J. Brooks, and Mr. J. Parrack for helpful comments on the manuscript, and D. Lang for typing the manuscript and tables. The research was funded by a Texas A&M Uni- versity mini-grant and an institutional grant #HNA83AA- D-0061 to Texas A&M University by the National Sea Grant College Program, National Oceanic and Atmo- spheric Administration, U.S. Department of Commerce to EP, and by the University of Texas Research Institute and C. Kleberg Foundation (CK). LITERATURE CITED ABBoTtT, R. T. 1974. American Seashells. 2nd ed. Van Nos- trand Reinhold Co.: New York. Amio, M. 1963. A comparative embryology of marine gastro- pods, with ecological considerations. Shimonoseki Univ. Fisheries J. 12:15-144. ANKEL, F. & A. M. CHRISTENSEN. 1963. Non-specificity in host selection by Odostomia scalaris Macgillivray. Vidensk. Medd. Dansk. Naturh. Foren. 125:321-325. ANKEL, W. E. 1949a. Die Mundbewaffnung der Pyramidel- liden. Arch. Molluskenk. 77:79-82. ANKEL, W. E. 1949b. Die Nahrungsaufnahme der Pyramidel- liden. Verh. Dtsch. Zool. Ges. Kiel 1949:478-484. Boescu, D. F., M. L. Wass & R. W. VIRSTEIN. 1976. The dynamics of estuarine benthic communities. Pp. 177-196. In: M. L. Wiley (ed.), Estuarine processes, Vol. 1. Uses, stresses, and adaptation to the estuary. Academic Press, Inc.: New York. Bonar, D. B. & M. G. HADFIELD. 1974. Metamorphosis of the marine gastropod Phestilla sitbogae Bergh (Nudibranchia: Aeolidacea). I. Light and electron microscope analysis of larval and metamorphic stages. J. Exp. Mar. Biol. Ecol. 16: 227-255. The Veliger, Vol. 28, No. 1 CuIA, F.S. 1978. Perspectives: settlement and metamorphosis of marine invertebrate larvae. Pp. 283-285. In: F. S. Chia & M. E. Rice (eds.), Settlement and metamorphosis of ma- rine invertebrate larvae. Elsevier: New York. Cuiark, K. B. & A. GOETZFRIED. 1978. Zoogeographic influ- ence on development patterns of North Atlantic Ascoglossa and Nudibranchia, with a discussion on factors affecting egg size and number. J. Moll. Stud. 44:283-294. Comrort, A. 1957. The duration of life in molluscs. Proc. Malacol. Soc. Lond. 32:219-241. CopELAND, B. & H. HogEse. 1966. Growth and mortality of the American oyster, Crassostrea virginica, in high salinity shallow bays in central Texas. Publ. Inst. Mar. Sci. Univ. Tex. 11:140-158. DaL_, W. H. & P. BartscH. 1909. A monograph of west American pyramidellid mollusks. Bull. U.S. Natl. Mus. 68: 1-258. Davis, G.M. 1967. The systematic relationship of Pomatiopsis lapidaria and Oncomelania hupensis formosana (Prosobran- chia: Hydrobiidae). Malacologia 6:1-143. Davis, G. M. & W. P. Carney. 1973. Description of Onco- melania hupensis lindoensis, first intermediate host of Schis- tosoma japonicum in Sulawesi (Celebes). Proc. Acad. Natur. Sci. Philadelphia 125:1-34. FRANZ, D. 1976. Benthic molluscan assemblages in relation to sediment gradients in northeastern Long Island Sound, Connecticut. Malacologia 15:377-399. FRETTER, V. 1951. Turbonilla elegantissima (Montagu), a parasitic opisthobranch. J. Mar. Biol. Assoc. U.K. 30:37- 47. FRETTER, V. 1953. The transference of sperm from male to female prosobranchs with reference, also, to the pyramidel- lids. Proc. Linn. Soc. Lond. 164:217-224. FRETTER, V. & A. GRAHAM. 1949. The structure and mode of life of the Pyramidellidae, parasitic opisthobranchs. J. Mar. Biol. Assoc. U.K. 28:493-532. Ga.TsorF, P. 1964. The American oyster Crassostrea virginica Gmelin. U.S. Fish. Wildl. Serv. Fish. Bull. 64:1-480. GuNTER, G. 1941. Seasonal condition of Texas oysters. Tex. Acad. Sci. Proc. Trans. 25:89-93. Hopkins, S. 1956. Odostomia impressa parasitizing southern oysters. Science 124:628-629. KiTTING, C. L. 1984. Selectivity by dense populations of small invertebrates foraging on seagrass blade surfaces. Estuaries 7:276-288. LAFOLLETTE, P. I. 1977. Inbreeding and intraspecific varia- tion in Chrysallida Carpenter, 1857 (Gastropoda: Pyrami- dellidae). Western Soc. Malacol. Ann. Rep. 10:18-23. LAFOLLETTE, P. I. 1979. Observations on the larval develop- ment and behavior of Chrysallida cincta Carpenter, 1864 (Gastropoda: Pyramidellidae). Western Soc. Malacol. Ann. Rep. 11:31-34. Laws, C. R. 1937. Review of the Tertiary and Recent Neo- zelanic pyramidellid molluscs No. 1—The genus Turbonilla. Trans. Proc. Roy. Soc. N.Z. 66:402-422. Lesour, M. V. 1932. The eggs and early larvae of two com- mensal gastropods, Stilifer stylifer and Odostomia eulimordes. J. Mar. Biol. Assoc. U.K. 18:117-119. Lesour, M. V. 1936. Notes on the eggs and larvae of some Plymouth prosobranchs. J. Mar. Biol. Assoc. U.K. 20:547- 565. Lopes, H. DE S. 1958. Sobre “Turbonilla (Pyrgiscus) dispar” Pilsbry, 1897 (Gastropoda, Pyramidellidae). Rev. Bras. Biol. 18:17-21. Maas, D. 1964. Uber Cuticularbildungen am Penis von Pyr- amidelliden. Zool. Anz. 173:137-148. M. E. White et al., 1985 Maas, D. 1965. Anatomische und histologische Untersuch- ungen am Mundapparat der Pyramidelliden. Z. Morphol. Oekol. Tiere 54:566-642. PorTeER, H. J. 1976. Spiral cord variation of Odostomia im- pressa (Say) and O. seminuda (C. B. Adams) family Pyr- amidellidae. Bull. Amer. Malacol. Union for 1976:38-41. Porter, H. J., L. A. Howie & R. B. DeERIso. 1979. Mor- phometric character variation in Boonea impressa (Say) and B. seminuda (C. B. Adams)—family Pyramidellidae. Bull. Amer. Malacol. Union. for 1979:43-48. PowELL, E. N. 1981. Three 7urbonilla (Pyramidellidae, Gas- tropoda) of North Carolina, with comments on pyramidellid systematics. J. Elisha Mitchell Sci. Soc. 97:37-54. PREECE, A. 1972. A manual for histologic technicians. Little, Brown & Company: Boston. 428 pp. RASMUSSEN, E. 1944. Faunistic and biological notes on marine invertebrates I. The eggs and larvae of Brachystomia rissoides (Harl.), Eulimella nitidissima (Mont.), Retusa truncatula (Brug.) and Embletonia pallida (Alder & Hancock), (Gastro- poda marina). Vidd. Medd. Dansk. Naturh. Foren. 107: 207-233. RASMUSSEN, E. 1951. Faunistic and biological notes on marine invertebrates II. The eggs and larvae of some Danish ma- rine gastropods. Vidd. Medd. Dansk. Naturh. Foren. 113: 201-249. RAVEN, C. P. 1958. Morphogenesis: the analysis of molluscan development. Pergamon Press: New York. 311 pp. RAVEN, C. P. 1964. Development. Pp. 165-195. In: K. M. Wilbur & C. M. Yonge (eds.), Physiology of the Mollusca. Vol. I. Academic Press: New York. ROBERTSON, R. 1978. Spermatophores of six eastern North American pyramidellid gastropods and their systematic sig- nificance (with the new genus Boonea). Biol. Bull. 155:360- 382. ROBERTSON, R. & T. Mau-Lastovicka. 1979. The ectopar- asitism of Boonea and Fargoa (Gastropoda: Pyramidellidae). Biol. Bull. 157:320-333. ROBERTSON, R. & V. ORR. 1961. Review of pyramidellid hosts, Page 51 with notes on an Odostomia parasitic on a chiton. Nautilus 74:85-91. SANDERS, H. L. 1958. Benthic studies in Buzzards Bay. I. Animal-sediment relationships. Limnol. Oceanogr. 3:245- 258. SpIGHT, T. M. 1975. Factors extending gastropod embryonic development and their selective cost. Oecologia 21:1-16. STRATHMANN, R. R. 1977. Egg size, larval development, and juvenile size in benthic marine invertebrates. Amer. Natur. 111:373-376. TuHompson, T. E. 1967. Direct development in a nudibranch, Cadlina laevis, with a discussion of developmental processes in Opisthobranchia. J. Mar. Biol. Assoc. U.K. 47:1-22. THORSON, G. 1946. Reproduction and larval development of Danish marine bottom invertebrates with special reference to the planktonic larvae in the Sound (Oresund). Danmarks Fiskeri. og. Havundersogelser, Medd. fra. Komm. ser: Plankton 4:1-523. TuHoRSON, G. 1950. Reproductive and larval ecology of marine bottom invertebrates Biol. Rev. Camb. Philos. Soc. 25:1- 45. WALLER, T. R. 1980. Scanning electron microscopy of shell and mantle in the order Arcoida (Mollusca: Bivalvia). Smithsonian Contrib. Zool. 313:1-58. WELLS, H. 1959. Notes on Odostomia impressa (Say). Nautilus 72:140-144. WELLS, H. & M. WELLS. 1961. Three species of Odostomia from North Carolina, with description of new species. Nau- tilus 74:149-157. WELLS, H. W., M. J. WELLS & I. E. Gray. 1961. Food of the sea-star Astropecten articulatus. Biol. Bull. 120:265-271. WuHarTON, R. A. 1976. Variation in the New England pyr- amidellid gastropod, Turbonilla nivea (Stimpson). Nautilus 90:11-13. White, M. E., E. N. POWELL & C. L. KitTinc. 1984. The ectoparasitic gastropod Boonea (=Odostomia) impressa: pop- ulation ecology and the influence of parasitism on oyster growth rates. P.S.Z.N.I.: Mar. Ecol. 5:283-299. The Veliger 28(1):52-62 (July 1, 1985) THE VELIGER © CMS, Inc., 1985 On the Anatomy and Fine-Structure of a Peculiar Sense Organ in Nucula (Bivalvia, Protobranchia) by G. HASZPRUNAR Institut ftir Zoologie der Universitat Wien, A-1090 Wien, Althanstr. 14, Austria Abstract. A peculiar, tubelike sense organ, called Stempell’s Organ (StO) is found in the protobranch genus Nucula immediately dorsal to the anterior adductor muscle. The single organ forms a closed tube which is cerebrally innervated. So-called collar receptors present in the sensory portion of the StO indicate a mechanoreceptive function of the organ. Three special muscles are attached to the StO: two of them (m2 and m3) stabilize the tube, the third (m1), whose contractions are detected by the organ, is used in connection with the special mode of feeding (by palps, palp proboscides, and ctenidia) found in the Nuculidae. Comparison is made between the StO and other molluscan sense organs, likewise having collar receptors. INTRODUCTION AT THE END OF THE last century, STEMPELL (1898) de- scribed a peculiar, tubelike organ in Nucula nucleus, lo- cated dorsally to the anterior adductor muscle. Although he investigated its histology in some detail, the author could not trace any special function of the suggested sense organ. Later on, this organ was noticed by DREw (1901) in his admirable paper on the ontogeny of Nucula delphi- nodonta. According to the author this “organ of unknown function” appears during embryogenesis together with the first anlage of the ctenidium, a short time after the test of the embryo is shed. Since that time this peculiar sense organ has not been reported by scientists. In honor of its discoverer I shall call this structure Stempell’s Organ (StO). In this paper a detailed descrip- tion of the anatomy and the fine-structure of the StO will be presented with a discussion on its presumed function. MATERIAL ann METHODS Nucula nucleus (Linné, 1758) and Nucula sulcata (Bronn, 1831), both from the Atlantic (Bergen, Norway), were histologically and fine-structurally investigated with re- spect to the StO. For histological investigations serial sections were used, stained with Heidenhain’s Azan. For ultrastructural research entire specimens (3-5 mm) of Nucula nucleus were fixed in phosphate-buffered glu- taraldehyde (2.5%) and osmium (2%), decalcified with as- corbic acid (1%) after DIETRICH & FONTAINE (1975), and embedded in an epon-araldite mixture (MOLLENHAUER, 1964). Semithin sections were stained with 0.1% toluidine- blue, while ultrathin sections, made with a diamond knife, were stained with uranyl acetate and lead citrate. For observation a Zeiss EM9/S2 was used. RESULTS Anatomical Context of Stempell’s Organ Position and innervation: The StO is located dorsal to the anterior adductor (Figures 1, 2). There is some vari- ation with respect to the position of the posterior end of the StO, which may reach the first tooth of the hinge or a little into the dorsal mantle process. The organ forms a narrow, elongate tube nearly as long as the anterior ad- ductor in adult specimens 700-800 um, closed at both ends. Its diameter varies between 30 and 60 wm, depend- ing on the state of contraction of the attached muscles (see below). The anterior end of the StO is always expanded and forms a bulb (see DREw, 1901, and Figure 1b). Innervation is from the pair of anterior pallial nerves which emerge from the outside of the most dorsal/anterior parts of the cerebral ganglia (Figure 1b).! Each nerve runs ‘Tn this respect it should be stressed that the pleural ganglia of Nucula (N. nucleus, N. sulcata investigated here, N. delphino- donta after DREW, 1901) are not separated as described by PEL- SENEER (1891), but are fused as in all other bivalves. In addition, the visceral loop is not a nerve, but a neural cord over its whole length, as known from the primitive cephalopod Nautilus. Since neural cords in primitive gastropods are pedal ones, evolution of ganglia in higher conchiferous groups is clearly due to conver- gence, contradicting a (monophyletic) taxon “Ganglioneura” (LAUTERBACH, 1984). G. Haszprunar, 1985 Page 53 la Figure 1 Nucula sulcata. Position and innervation of Stempell’s Organ (StO). Figure 1a, lateral view of the left side (left mantle omitted). Figure 1b, detail view to show innervation of the StO (all tissues are shown transparent). a, anus; aa, anterior adductor; al, anterior retractor of labial palp; apr, anterior pedal retractor; at, anterior teeth of hinge; c, cerebral ganglion; cp, cerebropedal connective; ct, ctenidium; cv, cerebrovisceral connective; dg, digestive gland; f, foot; lp, labial palp; m, mantle; m1, attachment zone of muscle m1; mc, central cleft of mantle margin; mr, mantle retractors; n, nerve of StO; 0, oral opening; oe, eosophagus; pa, posterior adductor; pl, pleural ganglion; pp, palp proboscides; ppr, posterior pedal retractor; pt, posterior teeth of hinge; StO, Stempell’s Organ. Scale bars: 1a, 5 mm; 1b, 1 mm. Page 54 The Veliger, Vol. 28, No. % £ 241 Figure 2 Nucula nucleus. Cross section of the dorsal mantle at middle zone of Stempell’s Organ. aa, anterior adductor; dv, dorsal blood vessel; i, inner fold of mantle margin; Ic, longitudinal clefts; m1, 2, 3, 4, 5, 6—muscles m1, m2, m3, m4, m5, m6; n, nerve of StO; 0, outer fold of mantle margin; p, periostracum; s, sensory fold of mantle margin; sbl, specialized basal lamina (attachment zone of m2 and m3); StO, Stempell’s Organ. Scale bar: 50 um. G. Haszprunar, 1985 between the anterior adductor and the first pedal retrac- tor, and then forward beneath the StO dorsal to the an- terior adductor (Figures 1b, 2). The nerve supplies the organ, especially in the anterior region, by several very thin neural fibers passing laterally through the basal lam- ina into the epithelium (Figure 4). After passing the an- terior end of the StO the nerve runs into the anterior mantle margin. Mantle epithelium: Three folds of the mantle margin can be distinguished in the region of the StO (Figures 2, 8): The inner fold (i) forms a cleft anteriorly, but changes gradually into a strong central fold at the middle region of the StO; the sensory fold (s) is very small and there are no special sensory elements in this region; the outer fold (0) produces the periostracum at its inner side. Laterally in the mantle epithelium there appear two longitudinal clefts (Ic) limiting the insertions of muscles m2 and m3 (see below). Ventral to this zone there are no special features until the attachment area of the anterior adductor begins. Muscle system: Several special muscles are found near the StO and three of them (paired) are attached to its basal lamina: The thickest of these muscles (m1) is attached to the StO at its ventral side. The muscle runs obliquely forward and is attached to the shell immediately anterodorsally to the anterior adductor (Figure 1b). The attachment epi- thelium consists of very flat (1-2 um high) cells containing many bundles of microfilaments. It is similar to the at- tachment epithelium of the anterior adductor, which is higher (2-3 wm), but nearly lacks nuclei (Figure 16). A second pair of muscles (m2) is attached to the StO immediately dorsal to m1. The muscles m2 turn dorsally and laterally and are attached to the epithelium of the mantle (Figure 2). The attachment epithelium of these muscles (m2) is characterized by an extremely thick (2.5- 3 um) basal lamina that is divided by a very thin electron- dense membrane into two layers (Figure 17). The muscle fibers penetrate the lower layer only and are attached by an electron-dense vesicle. The epithelium itself consists of two cell types, one with electron-dense cytoplasm and few microfilaments (x), the other with a more electron-lucent cytoplasm (y). The third pair of muscles (m3) is attached dorsally to the StO. The muscles m3 cross each other (Figures 2, 3, 5) and run sideways to reach the mantle epithelium im- mediately dorsal to muscle m2. Their attachment epitheli- um is elaborated in the same way as described above for muscle m2 (Figure 17). Along the whole length of the StO, the mantle is coated on the inside by a substantial muscle (m4), extending from the outside of the inner cleft (i) of the mantle margin ventrally (Figures 2, 9). It is attached to the shell dorsal to muscle m1 (or dorsal to the anterior adductor in the posterior region). The attachment epithelium of this mus- cle looks like that of muscle m1. Page 55 Two pairs of longitudinal muscles (m5 and m6) are found near the StO, reaching into the dorsal mantle pro- cess up to the hinge, where they are fused and attached to the shell. The larger pair (m5) is located ventral to muscle m1; the smaller pair (m6) is found lateral to the StO, immediately above the nerve (n) (Figures 2, 3, 8). Structure of Stempell’s Organ General organization: Stempell’s Organ forms a narrow tube that is closed at both ends and thus lacks direct con- tact with the external water. In general the lumen of the organ is not placed centrally, but is shifted dorsally by a thickened ventral epithelium. The lumen is additionally narrowed by a high, longitudinal, ventral crest whose cilia fill it almost entirely. In the following, the structure of the StO as a whole is described at five positions (a—e) from the anterior to the posterior end (Figures 4-8). All measurements are for adult specimens. (a) A short distance behind the anterior end a cross section of the StO is circular, with a diameter of 60 um. The dorsal epithelium is very flat (1-2 um), extending 15 um ventrally, and the crest is 25 um high (Figure 4). A special central zone is not elaborated, but most of the innervation is in this region. (b) From a short distance behind the anterior bulb to the posterior quarter the StO has the following organi- zation. The diameter is smaller (50 x 30 um), and the extremely flat dorsal epithelium lacks nuclei. A special tissue, forming “longitudinal septa” (see below), separates a central zone below the crest which is narrower in this region (Figures 2, 3, 5). (c) In the last quarter of the StO its diameter increases, the dorsal epithelium is thickened to 8 um, and it contains nuclei (Figure 6). (d) The crest flattens toward the end of the organ, then disappears together with the central zone (Figure 7). (e) Finally the lumen disappears. There is no posterior bulb in Nucula nucleus and N. sulcata as figured by DREW (1901) for N. delphinodonta (Figure 8). Structure of the non-specialized epithelium: Although the height of the epithelium lining the lumen of the StO varies greatly, its structure does not change. All cells have more or less round nuclei and bear a microvillous border, but otherwise there are no special features. Anteriorly some nervous tissue is found at the bases of the cells, running from the place of innervation (always lateral) downward into the central zone, penetrating the “longi- tudinal septa.” The basal lamina of the StO is thick (2-3 wm). This seems to be necessary for the attachment of the muscles, which are fixed to the lamina by prominent toothlike pro- jections (especially m1, see Figure 3). Laterally the basal lamina is penetrated by the thin neural fibers emerging from the laterally placed nerves (Figure 4). Page 56 The Veliger, Vol. 28, No. 1 ihventittaayy & Sif NS NWN Uh Ww ZAMAN SSNS pW ANY SWS WS \\ ZINC U7 Wi Wry, YIN \) Gy WAN VANS MM SWANS Ny UU NGAP Ee a SINS SSSA THON Hi Wp My LON; ORYSSS ANY) ENR Cid eae ae ANG His) Mipe = SSA a Zi ee } Se ZAM Mig hans Ly / BAW Ny Ce ( — \ On PLS ay ih Un BCE lg \ 8 SUL HALE Figure 3 Nucula nucleus. Cross section of middle zone of Stempell’s Organ (semischematic). bl, basal lamina; cc, ciliated cell; co, connective tissue; el, elastic layer (a specialized portion of the basal lamina); k, sensory knob; m1, 2, 3, 6— muscle m1, m2, m3, m6; n, nerve of StO; se, sensory cell; sp, supporting cell; nc, neighboring cell (with a specialized portion); v, vesicle; w, wrapper. Scale bar: 5 um. G. Haszprunar, 1985 Page 57 Structure of the “longitudinal septa”: The symmetri- cally placed longitudinal septa separate a central zone in the ventral area of the lining epithelium. These septa are located exactly above the place of attachment of the thick oblique muscle m1. They consist of two portions (Figures 3, 15): (a) Ventrally a specialized region (el) of the basal lamina, conical in cross section, contains numerous thin microfilaments. (b) This region is continued upward by muscular portions of the laterally adjacent cells (nc) en- tering the crest and forming its lateral basis. The attach- ment to the lower portion of the longitudinal septa is by numerous rootlike projections which are invested with electron-dense material. The nuclei of the adjacent cells are slightly different from those of the non-specialized epithelium, having a more oval shape and bigger granules (150-200 nm) within their reduced euchromatin. Structure of the central zone and of the crest: The central zone consists of large cells with oval nuclei. The cytoplasm of these cells is granular and more electron- lucent than that of the supporting cells (Figures 3, 15). They project slender (1 um) processes up to the median surface of the crest. There, these processes form a kind of knob (2-3 wm), being somewhat higher than the sur- rounding ciliated cells (cc). The processes are arranged in transverse rows (Figure 14). At the anterior end of the StO up to seven processes are found within a row; going backward this number is gradually reduced to two. Each of the knobs bears a so-called collar receptor, consisting Explanation Nucula nucleus. Figures 4 to 7. Cross sections of Stempell’s Or- gan at different zones of the organ. Scale bars: 10 um. Figure 4. Immediately behind the anterior end (arrow: inner- vation). Figure 5. Middle zone (for details see Figure 3). Figure 6. Posterior quarter. Figure 7. Near the posterior end. Figure 8. Cross section of the dorsal mantle process at the first anterior hinge-tooth (immediately before the posterior end of Stempell’s Organ). Scale bar: 50 wm. Figure 9. Detail of Figure 2 to demonstrate the secretion of hypostracum material (decalcified). Scale bar: 10 um. of a specialized cilium that is surrounded by nine spe- cialized microvilli (=“‘stereo-cilia” of many authors) (Fig- ure 12). These cilia lack striped roots, have an thickened outer membrane, and are somewhat stouter (280-300 nm) than the cilia of the ciliated cells (200-230 nm). The structure of the collar cilia is likewise distinctive, showing a 9 x 3 pattern of outer microtubules and an electron- dense circle around the central tubules up to their tips (Figure 12). The basal body of the cilium forms a starlike plate from which several rootlets (not striped) run down- ward (Figure 13). The microvilli are triangularly shaped in cross section with amplified tips toward the central cilium. They are connected by a dense glycocalix forming a kind of fence around the central cilium. Between the rows of processes and surrounding them, ciliated cells (cc) form the bulk of the crest. The whole breadth of the crest is always occupied by a single ciliated cell which obviously lacks a nucleus. The ciliated cells bear many cilia, but only few mitochondria are found. Whereas the more dorsally placed cilia have short roots, those of the more laterally placed cilia are very long and cross each other at the center of the crest (Figures 3, 14). Since these roots are alternately arranged with the rows of processes, a striped pattern is found in oblique sections of the crest (Figures 4, 14). The cilia of the ciliated cells are connected one to another by a net of glycocalix (Fig- ures 3, 10, 11, 14) and so form a kind of matrix. In contrast to the shafts of the cilia, which are of normal structure, the microtubule pattern is progressively dis- of Figures 4 to 13 Figures 10 to 13. Specialized cilia of Stempell’s Organ; all cross sections are slightly oblique and are centripedally arranged. Scale bars: 200 nm. Figure 10. Spearlike tips of supporting cilia together with the netlike wrapper. Figure 11. Supporting cilia connected by glycocalix. Figure 12. Typical collar receptors. Figure 13. Bases of collar receptors. h, hypostracum; i, inner fold of mantle margin; lc, longitudinal cleft; m(5+6), fused muscles m5 and m6; 0, outer fold of mantle margin; p, periostracum; s, sensory fold of mantle margin; StO, Stempell’s Organ. Explanation of Figures 14 to 17 Nucula nucleus. Figure 14. Oblique section of the crest of Stem- pell’s Organ immediately behind the anterior end. Figure 15. Cross section of the central zone and the longitudinal septa of Stempell’s Organ near middle zone (see also Figure 3). Figure 16. Attachment epithelium of the anterior adductor. Figure 17. Attachment epithelium of muscle m2. at, adhesive tissue; bl, basal lamina; cc, ciliated cell; cr, collar receptor; ct, connective tissue; el, elastic layer (a specialized por- tion of the basal lamina); ibl, inner layer of basal lamina; hy, hypostracum; k, sensory knob; mf, muscle fibrils; nc, neighboring cell; obl, outer layer of basal lamina; se, sensory cell; sp, sup- porting cell; v, vesicle; w, wrapper; x and y, cell types x and y (see text). All scale bars: 2 um. The Veliger; Vola 285sNoml G. Haszprunar, 1985 Page 59 5 po Sect ee ay Foes cet 9g Fo, 000" | | 9 9. iu weet ve why P86] ‘UYVNNUdZsVH ‘P86 ‘TAZLALS Wd) 21e] -ndai 0} s}UatIND 191eM Teyedqns jo uondasas The Veliger, Vol. 28, No. 1 Jo}dasa10ueyoauI $]9]}001 YIM ‘uOT} -99S SSOJO UI pUNOL ‘auTU ainjons}s TeWIOU ‘00} [eseq pue joo1 YIM ‘(ui CG] 0} Z) afqeiszea S][90 JUa0 -e[pe ueyy soysty Apystys Aytaeo yerpedqns 24} JO 193@M YIM 39e]U0D (éjetpesydso pue) yeinayd jetpedqns ‘safoSnur [Jays ay} Jo apa Jo1Jaj}Ue ay} punoie om} esso[s000q, ‘Tqndun seunsdzsepzy ‘1861 “dSIXD ‘P961 ‘ONIVI 29 GIVNOGQOVII (sajoequa} yerpodida Y}IM UuoNs9NU0D UT) w0}}0q 94} Woy I[N -ws yono} jo uondasar Jo}dasa.10ueysoul sjap}001 But -yor] SUOTI9as ssoID UT Jepnsuers ‘aut Jo 81a Be) (0) a eho ese) Asossazoe {ainjoniys pezipersads ssay S00] [eseq pue i001 Suryory] ‘(wil Og 01 p) a[qeIseA s]]90 jusoe(pe uey) s9daap Jayem [eUsa}x9 YIM jOe1U00 [epod sopoejuay [e -tpodida jo saseq yesuaa snoJsuUINnU 0} OM} epodonsesna A, qLLol “aIOW yono} jo uondasas Jo}dasa.10ue your $13] -}001 Suryoe] {uonoas SsOJO UI punol {(¢) 1Y48Io aInjonNs [UIIOU [00] [eseq pue s}oo1 3u0] yim (uu ¢ 0} dn) qs0ys s[]20 jusoel[pe YIM JayI3 -0} eyided e Sururs0j Jovem yeuso}x9 YIM JOe1U0D [e120stA uIsIeUl spUeU snosawinu (éeap -louldag) uajvadoonjg “Sployool} pue syoues S861 ‘UVNNUdZSVH ‘C861 ‘AON.NANAS 29 NVCVHZ ‘€//6| ‘YIOW way} a3eyN -3a1 0] s}udIINd 19}3eM Teyyed jo uondasas Joidasa10ueyoour S]9]}001 YIM ‘uOT} 998 SSO1D UI puNod ‘fouTU SOUBIQUIDUI UdTJOMS sdUIMaUWIOS fainjons [eusou 00} [eseq pue $1001 Suoy] yim {(urT 00S 03 dn) Buoy AsaA s]]@9 juaoelpe se iySIay owes Ayiaeo apueul 24} JO 197M YIM 39e]U09 [BIOOSTA (eIptus}s Jo apts}no JO apIsur) 10}onppe Jotsaisod 3y} 0} [eUaA OM} IO 2Uu0 DIUOGILY -oany pue eydsowio1i3ig Jaded sty3 SJUIWIZAOUI BUIPIIy pue sjuasino 133eM a}e[NZa1 0} suonoeny -uood aposnul jo uondasa1 Jo}dasa10ue yout $19] -1001 Suryoe] Suotsas SsoJO UT Je[NSueLN ‘suru ainjon.sys pazterads joo] ye -seq pure s}0o1 3UryOR] ‘(uM ¢ 0} dn) joys s]]99 Juaoef -pe ue} saysry Apystys (2qn} pasoyd) joe}U09 OU [e1qe129 JojONppe JOII9}Ue JY} 0} [esiop 3uo (caeplpynonNy) vjnonvy qosnaz :(0861) NIMVIg-INIATVS 0} Sutpsoo0y , ee LeeeeEEeEEEeEE s20UdIIJaY uonouny Arepuosas pajsassng uonouny Areurtid pajsassng (. e119-09.1918,,) T[[IAOsoIUL pazipersadg 10jda00.1 Je][oo jo unIpIO [eUaD Jo}da0a1 Ie]]Oo jo siseg WINIPsUl YIM 19e]U0T) uOTIeAIIUUT uontsog uaurtoads sad saquinyy ul punovy Se a ae d 1s Azosuas yerjedqng Page 60 uv3Jo asuas fetpodidq sapoejua} opueUL SUO'T ueZ1O asuas [eUTWIOpqy ue3i0 s jjaduiaig uesiO ‘s10j}daoar se]]oo YIM suUeBJO asuas UvOSNI[OU payeSnsaaut [[am Jo uostredui075 ES) ESTA G. Haszprunar, 1985 solved near the tip of the cilium. The tip itself is spearlike and contains a single tubule only. Since the tips are found bent (Figure 10, 14), they appear to be flexible. Between the cilia, vesicles (v) of various diameters (0.5-2 um) are found that are probably transported up to the tips of the cilia. There the vesicles form a netlike wrapper enveloping the whole crest complex (Figures 3, 10, 14). DISCUSSION On the Structure of the Receptor Elements There can be little doubt that the so-called collar re- ceptors found in the crest are sensory structures. This is shown not only by their specialized structure, but also by the fact that the presumed sensory cells, which bear these receptors, are in synaptic contact with the fine nervous fibers emerging from the lateral nerve. Such collar receptors are found in many invertebrate groups, but appear to be often confused by authors with the so-called choanocyte-like cells (e.g., CRISP, 1981). Choanocyte-like cells are not sensory and lack the special features of the central cilium as well as of the surrounding microvilli. Within the Turbellaria collar receptors are found in the integument of nearly all groups, and struc- tural differences are used for phylogenetic suggestions (e.g., EHLERS, 1977; EHLERS & EHLERS, 1977; SOPOTT-EHLERS, 1984). Within the Mollusca, collar receptors are found so far in very different organs: they occur (a) in the subpallial sensory stripe of the Docoglossa (STUTZEL, 1984; HASZPRUNAR, 1984); (b) in the epipodial sense organs of Vetigastropoda? (Crisp, 1981; Haszprunar, unpublished); (c) at the mantle slit (or siphon) of fissurellids (Herbert, personal communication); (d) at the ventral mantle mar- gin of the pteropod Cveseis virgula (Haszprunar, unpub- lished); (e) in the abdominal sense organs of Pteriomorpha (Morr, 1977a; HASZPRUNAR, 1985); (f) at the long mantle tentacles of the scallop Placopecten magellanicus (MOIR, 1977b); and (g) in Stempell’s Organ of Nucula (this pa- per). Although the organs of the various classes and tribes, where collar receptors occur, are certainly not homolo- gous, it seems hardly likely that such a complex structure would have evolved independently in each case. However, as outlined in the cases within the Mollusca, which have been investigated in detail, there are several differences in the detailed structure of the collar receptors of the respec- tive organs (Table 1). Thus, two possibilities remain: (i) the collar receptors of different organs are the products of convergence, independently developed from a choanocyte- like cell, or (ii) there is a common basal genetic infor- mation to develop an archetype of these receptors, which have been secondarily specialized for the special function of the particular sense organ. This view is the theoretical 2 After SALVINI-PLAWEN (1980): zeugobranchs and trochoids. Page 61 basis of all phylogenetic implications based on the struc- ture of collar receptors within the Turbellaria (EHLERS, 1977). This would be a special kind of “normative” ho- mology (e.g., RIEDL, 1975) which is normally restricted to single organelles only (e.g., mitochondria, cilia), or known as “serial” homology with respect to organs (e.g., Roru, 1984). In any case this type of receptor appears to be typical for mechanoreceptors (although there are many mechano- receptors, such as statocysts, lacking collar receptors). In the case of the abdominal sense organ its suggested mecha- noreceptive function (regulating water currents, see THIELE [1889], HASZPRUNAR [1985]) has been recently confirmed by electrophysiological results (ZHADAN & SEMEN’KOV, 1982). Further, a chemo- or osmosensitive StO is very improbable, since the organ has no contact with external water, being closed. Therefore, it is very probable that the StO is a mecha- noreceptor. On the Structure of the Supporting Elements The well-developed roots found in the ciliary cells in- dicate a high mechanical stress on their cilia (especially the laterally located ones). In contrast, several structural facts indicate that these cilia do not move, but form a kind of elastic matrix covered by the netlike wrapper: (i) the presence of few mitochondria in the ciliated cells (Figure 14); (ii) the connection of the cilia among each other by a glycocalix (Figure 11); (iii) the spearlike tips of the cilia which appear to be flexible (Figures 10, 14). It follows that there is a passive mechanical stress on these cilia. In fact, a highly developed structure to transmit me- chanical forces from outside to the crest is found in the paired longitudinal septa, immediately situated above the place of attachment of muscle m1. Any contraction of this muscle is transmitted via the specialized basal lamina and via the muscular portion of the adjacent cell to the lateral basis of the crest (Figures 3, 15). Comparing the three muscles attached to the StO, the following main differences between muscles m2 and m3 and muscle m1 are found. The former muscles are sym- metrically arranged with respect to the longitudinal axis of the StO. Their mode of attachment at the lateral mantle epithelium by a thickened basal lamina (Figure 17) does not allow extreme mechanical stress. In addition, there are no special structures to transmit contractions of these muscles into the StO, and their attachment zone on the StO is not toothlike (as for muscle m1), likewise indicating a low degree of mechanical stress. Thus, it is probable that the muscles m2 and m3 are necessary for the stability of the StO, but do not act in the reception process. In contrast, muscle m1, which is the thickest, runs obliquely forward and this appears to be correlated (a) with the presence of the majority of the collar receptors at the an- terior end of the StO, and (b) with the fact that penetrat- ing neural fibers likewise are found only in the anterior Page 62 third of the organ. In addition, the muscle is directly at- tached to the shell via a special attachment tissue similar to that of the adductor muscle (Figure 16), and its con- tractions can be transmitted by the longitudinal septum. On the Function of the StO Summing up the considerations presented so far, it can be concluded that the StO is a mechanoreceptor, detecting contractions of muscle m1. To date, a StO has been found only within the genus Nucula, but is likely present also in other genera of the Nuculidae (Nuculoidea). Because the discoverer of the StO did not describe a similar structure in any of the members of the Nuculanoidea and Solemyoidea he inves- tigated (STEMPELL, 1898, 1899; DREw, 1899), the StO appears to be restricted to the Nuculidae. The Nuculidae is the sole family among all Bivalvia which has retained an anterior-posterior water current (similar conditions found in the Lucinoidea are accepted by most authors to be secondary, see e.g., ALLEN [1958], Morton [1979]). STASEK (1961) stated that feeding in nuculids is done (a) by the palp proboscides (as in all protobranchs), (b) by the outer palp lamellae (only nu- culids in such a degree), and (c) by the ctenidia (less important in nuculids). Thus, the incoming water is used not only for respiration, but also for feeding. Reflecting that the adhesive zone of muscle m1 is located immediately dorsal to the anterior adductor and thus exactly beside the inhalant opening of the water current (Figure 1b), it ap- pears probable that the StO detects movements (also lon- gitudinal) of this important region, where disturbances are essential for two main life processes. Therefore, the presence of the StO within the Nuculi- dae is additional evidence for the ideas of STASEK (1961) that nuculids are not direct forerunners of higher Bivalvia. They represent an early offshoot of the bivalve stock, spe- cialized in a considerable degree. The StO represents one example of this specialization. ACKNOWLEDGMENTS I thank Mr. W. Pekny (Univ. Vienna) for supplying me with living material of Nucula spp., on which this study is mainly based. Further I thank Dr. H. Nemeschkal (Univ. Vienna) for critical reading and commenting on the manuscript. This study was supported by the “Fonds zur Férderung der wissenschaftlichen Forschung in Os- terreich” (project number: P 4751). LITERATURE CITED ALLEN, J. A. 1958. On the basic form and adaptations to habitat in the Lucinacea (Eulamellibranchia). Phil. Trans. Roy. Soc. Lond. B 241:421-484. Crisp, M. 1981. Epithelial sensory structures of trocnids. J. Mar. Biol. Assoc. U.K. 61:95-106. Dietricu, H. F. & A. R. FONTAINE. 1975. A decalcification method for ultrastructure of echinoderm tissue. Stain Tech- nol. 50: 351-354. The Veliger, Vol. 28, No. 1 Drew, G. A. 1899. The anatomy, habits and embryology of Yoldia limulata Say. Johns Hopkins Univ. Mem. Biol. Lab. 4(3):1-37. Drew, G. A. 1901. The life history of Nucula delphinodonta (Mighels). Quart. J. Micr. Sci. 44:313-391. EHLERS, U. 1977. Vergleichende Untersuchungen tiber Col- lar-Rezeptoren bei Turbellarien. Acta Zool. Fennica 154: 137-148. EHLERS, U. & B. EHLERS. 1977. Monociliary receptors in interstitial Proseriata and Neorhabdocoela (Turbellaria, Neoophora). Zoomorphology 86:197-222. HASZPRUNAR, G. 1984. The fine morphology of the osphradial sense organs of the Mollusca. I. Gastropoda, Prosobranchia. Phil. Trans. Roy. Soc. Lond. B 307:457-496. HASZPRUNAR, G. 1985. The fine-structure of the abdominal sense organs of Pteriomorpha (Mollusca, Bivalvia). J. Moll. Stud. (in press). LAUTERBACH, K.-E. 1984. Das phylogenetische System der Mollusca. Mitt. Dtsch. Malak. Ges. (Frankfurt A.M.) 37: 66-81. MacDonaLp, J. & C. B. Maino. 1964. Observations on the epipodium, digestive tract, coelomic derivates, and nervous system of the trochid gastropod Tegula funebralis. Veliger 6(Suppl.):50-55. Mor, A. J.G. 1977a. On the ultrastructure of the abdominal sense organ of the giant scallop Placopecten magellanicus (Gmelin). Cell. Tiss. Res. 184:359-366. Morr, A. J. G. 1977b. Ultrastructural studies on the ciliated receptors of the long tentacles of the giant scallop Placopec- ten magellanicus (Gmelin). Cell. Tiss. Res. 184:367-380. Morton, B. 1979. The biology and functional morphology of the coral-sand bivalve Fimbria fimbriata (Linnaeus, 1758). Rec. Austral. Mus. 32(11):371-387. MOLLENHAUER, H. H. 1964. Plastic embedding mixtures for use in electron microscopy. Stain Technol. 39:111. PELSENEER, P. 1891. Contribution a l’étude des Lamelli- branches. Arch. Biol. 11:147-312. RIEDL, R. 1975. Die Ordnung des Lebendigen. Parey Verlag: Hamburg. 372 pp. Rotn, L. V. 1984. On homology. Biol. J. Linn. Soc. 22:13- 29: SALVINI-PLAWEN, L. V. 1980. A reconsideration of systematics in th the Mollusca (phylogeny and higher classification). Malacologia 19:249-278. SOPOTT-EHLERS, B. 1984. Epidermale Collar-Rezeptoren der Nematoplanidae und Polystyliphoridae (Plathelminthes, Unguiphora). Zoomorphology 104:226-230. STASEK, C. R. 1961. The ciliation and function of the labial palps of Acila castrensis (Protobranchia, Nuculidae), with an evaluation of the role of the protobranch organs of feed- ing in the evolution of the Bivalvia. Proc. Zool. Soc. Lond. 137:511-538. STEMPELL, W. 1898. Beitrage zur Kenntnis der Nuculiden. Zool. Jb. Suppl. 4:339-430. STEMPELL, W. 1899. Zur Anatomie von Solemya togata Poli. Zool. Jb. Syst. 13:89-170. STUTZEL, R. 1984. Anatomische und ultrastrukturelle Unter- suchungen an der Napfschnecke Patella L. unter besonderer Beriicksichtigung der Anpassung an den Lebensraum. Zoo- logica (Stuttgart) 135:1-54, 36 pl. THIELE, J. 1889. Die abdominalen Sinnesorgane der Lamel- libranchiaten. Z. Wiss. Zool. 48:47-59. ZHADAN, P. M. & P. G. SEMEN’KOV. 1982. Function of the abdominal organ in Patinopecten yessoensis. Dokl. Biol. Sci. 262(1-6):87-90. The Veliger 28(1):63-79 (July 1, 1985) THE VELIGER © CMS, Inc., 1985 The Anatomy and Histology of Phyllidia pulitzern Pruvot-Fol, 1962, with Remarks on the Three Mediterranean Species of Phyllidia (Nudibranchia, Doridacea) by HEIKE WAGELE Fachbereich 7, Arbeitsgruppe Zoomorphologie, Universitat Oldenburg, Postfach 2503, 2900 Oldenburg, West Germany Abstract. In this study is given the first extensive examination of the anatomy and the histology of the organ systems of a phyllidiid: the Mediterranean species Phyllidia pulitzeri Pruvot-Fol, 1962. The digestive tract of Phyllidia pulitzeri differs from that of other species of the genus by the lack of the typical oral glands lying on the outside of the oral tube. Oral glands can be found only at the inside wall of the oral tube. The genital system corresponds to the triaulic scheme of all other Doridacea. The penis has no armament. Gastro-esophageal ganglia, as described by other authors, are not present. Special features of the excretory system are the very long renopericardial duct and a glandular mass that lies in the posterior part of the nephridium and continues in the ureter. Basophilic subepithelial glands that are scattered over the mantle, foot, and gills seem to be responsible for the intensive secretion of mucus. A comparison of the three Mediterranean species of Phyllidia (P. rolandiae, P. aurata, and P. pulitzert) led to the conclusion that P. rolandiae is a nomen dubium, because its description does not allow a clear distinction from the two other species. INTRODUCTION “THE PHYLLIDIIDAE ARE the longest known of all nudi- branchs (Bergh, 1876) and were figured as early as 1735” (Marcus, 1962:477). Nevertheless, little is known about their anatomy and even less about the phylogeny of this family or the life histories of its species. It is occasionally maintained in the literature that dissection of phyllidian species is unnecessary for identification (e.g., PRU- VOT-FOL, 1962). Therefore, several species are known only by their external features. That dissection is an absolute necessity is demonstrated by several studies (EALES, 1938; ELIOT, 1903, 1904; Marcus, 1962) which show that there are both interspecific and intraspecific differences. The phyllidiids have an oval, flat form similar to the cryptobranchiate Doridacea, but they can be easily distin- guished from those by their gills lying ventrally between the notum and foot, being interrupted only by the mouth and the genital papilla. The anus lies dorso-median (Phyl- lidia, Phyllidiella, Phyllidiopsis, Ceratophyllidia) or ventro- median between the gill leaflets (Fryeria). The phyllidiids have no radula or mandible, a feature that they have in common with the Dendrodorididae and on which was based their association with the latter family into the group Porostomata (BERGH, 1876, after BERGH, 1889). The most important anatomical feature for systematics is the foregut, as it distinguishes the two genera Phyllidia and Phyllidiopsis. Yet, the first description of Phyllidia pulitzert Pruvot-Fol, 1962, is based on a single specimen only, and its internal anatomy is completely unknown. As a consequence its generic status is not certain. Does it belong to the genus Phyllidia or to Phyllidiopsis? Having a few specimens in well-fixed condition at my disposal, I have been able to answer this question and to give a detailed description of the anatomy and the histol- ogy of some of the organs. Moreover, this examination is regarded as a first step toward a complete systematic re- Page 64 The Veliger, Vol. 28, No. 1 H. Wagele, 1985 vision of the family Phyllidiidae and toward a determi- nation of its correct phylogenetic position. MATERIAL anD METHODS In September 1980, four specimens of Phyllidia pulitzeri were collected by SCUBA diving in Khalkidhiki (North- ern Aegean Sea/Greece). The animals were found on a sponge of the genus Axznella in depths between 10 and 20 m (Figure 1a). In May 1983, additional specimens were discovered in Gozo/Malta in depths between 5 and 30 m. They were collected from crevices, and a group of five animals was found on rock material at the back of a large cave. The material from Greece (length of specimens: 22-32 mm; breadth: 12-22 mm) was fixed in 4% formaldehyde/ seawater, that from Malta (length: 12.5-22 mm; breadth: 6-13 mm) in Bouin’s fluid. Previous narcosis with MS 222 prevented deformation of the animals during fixation. For anatomical and histological examinations, three specimens were embedded in paraplast. Serial longitudi- nal sections (8 um) were made of two small animals (spec- imen L1: 12.5 mm; specimen L2: 22 mm) and serial cross sections (8 um) of a larger specimen (C1: 31 mm). Stain- ing of the sections was carried out according to ROMEIS (1968) with May-Grtinwald/Giemsa or trichrome (azan or hemalaun/lightgreen). Four specimens were dissected under the stereomicroscope. As the fixatives did not penetrate the digestive gland, histological examination of this organ was not possible. The following descriptions apply to all specimens ex- amined, except where special reference is made to partic- ular specimens. One specimen is deposited as a neotypus in the Mu- seum National d’Histoire Naturelle, Paris. RESULTS External Morphology In the living animal the notum above the visceral hump is covered with white tubercles arranged in five longitu- dinal lines. Between them are smaller, orange-colored tu- bercles. On the margin of the notum the white and or- ange-colored warts are scattered irregularly (Figure 1b). Immediately behind the yellow rhinophores a small white tubercle is located on each rhinophore sheath (Figure 2a: Page 65 rhst), the so-called ‘“‘Rhinophorenscheidentuberkel”’ (SCHMEKEL & PORTMANN, 1982:140). Fixed animals are ivory-colored and the arrangement of the white tubercles in longitudinal lines is hardly visi- ble. The tubercles of different animals vary in size and form. In some they are higher than broad and stand close together (Figure 2d). In others the warts are flat, widely separate, and protrude only slightly. This great variability in the form of the tubercles is due to fixation, because it was not observed in living animals. The rhinophores, each with approximately 12-14 la- mellae, are situated in the anterior ¥, to % of the body (Figure 1b). The rhinophore sheaths are very small. The anal papilla lies in an anal tubercle of varying size located medio-dorsally in the posterior % of the body (Figure 2b). The ventral side of the living animal is uniformly gold to light yellow in color. The margin of the notum is slight- ly transparent and the radial and net-shaped arrangement of the spiculae is visible. Beneath the slitlike oral aperture, which is surrounded by two short, triangular, marginally grooved, oral tentacles, the foot has a short, longitudinal notch (Figure 2e). The 100-130 gill leaflets, lying in a groove between the foot and mantle margin, are triangu- lar-shaped and are fused to the ventral notum along their broad base. Large and small gill leaflets alternate more or less regularly (Figure 2c). The genital papilla lies on the right side in the anterior third of the body between the gill leaflets. The vestibulum (v), with vagina and vas deferens, opens on the distal part of the papilla and the oviduct (ov, Figure 2c) opens on the proximal part. All the specimens fixed in formaldehyde are covered with small crystalline globules (Figure le). Some of them could easily be detached, but the greater part of them could not be removed without the surrounding tissue. Digestive Tract Anatomy: The general anatomical outline can be seen in Figure 2f. The vertical slit of the oral aperture leads into a vestibulum, from where the oral tube rises. After enter- ing the perivisceral cavity (pc) the oral tube widens into a club-shaped chamber; the highly folded walls of this chamber consist of a glandular epithelium (ogl). There are no glands on the outside of the oral tube. Two thick retractor muscles arise laterally, one on each Figure 1 a. Living animal in natural surrounding, on a sponge of the genus Axinella. Photograph was taken in Corfou/ Greece in August 1974. Specimen was not collected; depth 10 m. b. Living animal from Gozo/Malta; note the transparency of the mantle margin. c. Longitudinal section through statocyst; arrows, neurilemma; May-Griin- wald/Giemsa; scale 10 wm. d. Longitudinal section of the anterior digestive tract; note the protruding parts of the oral gland; azan; scale 50 um. e. “Crystallized” mucus on the ventral side of the mantle margin; scale 10 um. Key: f, foot; gnc, giant nerve cell; m, mouth; 0, oral tentacle; ogl, “oral gland”’; ot, oral tube; st, statolith. Page 66 The Veliger, Vol. 28, No. 1 Vaan a L els HAY HY Ue . f TILA ogl dgl odgl Figure 2 a-e. External morphology, details; scales 1 mm. a. Openings of the rhinophoral cavities. b. Anal papilla. c. Ventral side; genital papilla and gill leaflets (after WAGELE, 1984). d. Tubercles of the fixed animal; they are high and stand very close together. e. Ventral side; oral tentacles (0). f. General outline of the digestive tract, longitudinal H. Wagele, 1985 side of the oral tube. The muscles run caudo-medially, inserting in the dorsal body wall. The pharynx arises between these two retractor muscles, narrows after two small bends into the esophagus and then passes through the central nerve ring. The esophagus leads directly backward and opens into the stomach (Figure 3a). A distinct separation between the stomach and a central lumen of the digestive gland is not present: several tubes of the digestive gland open laterally into a central cavity of the digestive tract (Figure 2f: odgl), where also the esophagus ends and the intestine starts dorsally. The digestive gland occupies nearly *4 of the whole visceral hump. The ventral side is covered with a wide- spread network of vessels, which open into four visceral sinuses (see WAGELE, 1984). The intestine originates in the posterior third of the stomach, runs forward, and turns to the left, making an arch in front of the heart (Figure 3a: i). It then runs posteriorly on the right side under the pericardium and ends in the anal papilla. Histology: Around the vestibulum and along the sides of the mouth lies a thick glandular mass; these glands open into the vestibulum or externally at the base of the oral tentacles. The anterior oral tube is surrounded by connective tis- sue containing longitudinal muscular fibers laterally, and mainly transverse muscular fibers dorsally and ventrally. Cilia could not be detected by light microscopy. Goblet cells filled with basophilic grana are interspersed between the epithelial cells. The anterior oral tube seems to be very distensible: in one specimen, the oral glandular folds of the posterior oral tube (see below) were protruded out of the mouth, and the anterior oral tube was several times wider than in other specimens (Figure 1d). The oral glands are situated inside of the posterior, club-shaped part of the oral tube (Figure 2f), the ecto- dermal epithelium of which projects inward, forming the folded and finger-shaped glands. The glandular folds (Figure 3b) contain, in particular, transversal and longi- tudinal muscle fibers. The ectodermal epithelium consists mainly of tall columnar cells, with the nuclei lying basally. Goblet cells similar to those of the oral tube are inter- spersed between them. Toward the apex of the glandular folds these mucus-secreting cells replace the epithelial cells. Subepithelial basophilic glands (bsgl) were recognized along the ectodermal epithelium. In the apex, additional granular glands with acidophilic contents (agl) could be seen in the connective tissue. However, it was not possible Page 67 to decide whether these glands were single glandular cells or multicellular glands. Most fibers of each of the great retractor muscles insert in one glandular septum of the corresponding side. Other branches envelop the posterior part of the oral tube and the anterior part of the pharynx. The pharynx consists (from outside to inside) of a thin layer of connective tissue (ct), a thin layer of circular muscles (cm), a layer of fewer radial muscle fibers (rm), and a thin layer of longitudinal muscle fibers (Im). An apocrine-secreting epithelium lines the pharyngeal lumen, which in cross section has the shape of a triangle. The inner layer of longitudinal muscle fibers is well developed along the sides of the triangle (Figure 3c). The transition into the esophagus is abrupt. The esoph- agus is enclosed by a thin outer layer of connective tissue, followed by a layer of circular and a thick layer of lon- gitudinal muscle fibers. The lumen is lined by smooth columnar epithelium (Figure 3d). Cilia could not be found by light microscopy. In the layer of circular muscles two nerves were observed (Figure 3d: nv). The entrance of the esophagus into the stomach is char- acterized by regular folds of the, here, ciliated cuboidal epithelium. These folds continue in the dorsal wall of the stomach. Glandular cells were not detected. Ciliated epi- thelial linings exist only in dorsal areas around the open- ings of the esophagus and intestine, and sometimes around the large entrances into the digestive gland. The ventral epithelium is similar to that of the digestive gland. The intestine is characterized by a cuboidal ciliated ep- ithelium (Figure 7c), which in the proximal part is highly folded and invested by muscle fibers. Genital Apparatus Anatomy: The structure of the genital system of Phyllidia pulitzert corresponds to that of all Doridacea: it is triaulic and bears two vesicles (bursa copulatrix and receptaculum seminis) on the vaginal duct (Figure 4a). The hermaphroditic gonad is a more or less flat organ surrounded by the kidney dorsally and laterally and the digestive gland ventrally. Running initially for a short distance along the esophagus, the hermaphroditic duct (gonoduct gd) opens into a yellow-whitish oval ampulla (Figures 4b, c: a). The postampullar gonoduct (pogd) leads ventrally along the female gland, then divides into the vas deferens and oviduct. After a short distance the vas deferens enlarges into the brownish prostate (pr) which, after a few bends, leads above the female gland directly to the right side of the section: dgl, digestive gland; e, esophagus; f, foot; fm, foot margin; gr, groove of oral tentacle (0); i, intestine; lg/ sg, large/small gill leaflets; oa, oral aperture; odgl, opening into the digestive gland; ogl, oral gland; ot, oral tube; ov, oviduct; pc, perivisceral cavity; ph, pharynx; rhs, rhinophore sheath; rhst, ““Rhinophorenscheidentuberkel’’; v, vestibulum with vagina and vas deferens. The Veliger, Vol. 28, No. 1 Page 68 7 = 1 ne 1 \) h “% Atty L\de een us Oa SOS ee 7 ne! Vee RWS eat) > Seo —. aie > a aa oS ¥ 100p H. WAagele, 1985 body. Shortly before entering the thick notum tissue, the prostate narrows and forms the ductus ejaculatorius (dej), then passes into the penis (p) without being separated from it. The penis is not armed, and accessory glands were not observed. The vas deferens opens into the vestibulum (Figures 4a, b). The middle-sized proximal oviduct (Figure 4b: prov) leads ventrally of the female gland to the right side, where it disappears into the gland. Therefore, the bifurcation of the proximal oviduct into the glandular oviduct and the vaginal duct could not be observed macroscopically. The vaginal duct with its two vesicles lies dorsal to the female gland. The white, oval receptaculum seminis (r) and the yellow-brownish, round bursa copulatrix (b) are both stalked. In the smaller specimens the receptacula were smaller than the bursae. From the bursa the vaginal duct leads directly into the vestibulum, thus having a com- mon opening with the vas deferens (Figure 4a). Histology: In specimen C1, the largest of the three his- tologically examined animals, the oocytes of the gonad were in the final stage of development. They measured 85 um in diameter and were full of granular yolk. The nu- tritional cells were deformed by the large oocytes into elongate structures. The few regions of sperm develop- ment were at the margin of the gonad. The epithelium of the thin preampullar gonoduct (gd) consists of cuboidal cells. Sometimes sperm were found in the lumen of the duct. The ampulla is coated with a pavement epithelium, except near the openings of the pre- and postampullar duct, where cuboidal cells predominate. The ampullae of all examined specimens were filled with autosperm, which were distributed irregularly, not lined up (Figure 5d). The muscular postampullar duct (pogd) is lined with a cuboidal epithelium with long cilia. The duct arises at the distal end of the ampulla and is separated from it by a small septum. At the bifurcation into the vas deferens and proximal oviduct there is a chamberlike enlargement, which is characterized by folds with long cilia. At the beginning of the vas deferens of C1, some sperm were observed in the lumen. The transition of the proxi- mal vas deferens (which is lined by cuboidal ciliated cells) into the glandular prostate is abrupt. The glandular cells of the prostate are high and rather slender. They are full of acidophilic grana (azan: red). Small apical nuclei in- Page 69 dicate that there are supporting cells between the secretory cells (according to SCHMEKEL, 1971). Toward the distal end of the prostate the secretory cells become smaller and disappear at the transition into the ductus ejaculatorius, where cuboidal ciliated cells, thickly underlaid by muscle fibers, dominate. The penis is a thin, muscular tube forming the end of the ductus ejaculatorius. The proximal oviduct and its bifurcation in the glan- dular oviduct and vaginal duct are lined by a cuboidal ciliated epithelium. In the proximal oviduct, spermatozoa were found, single or arranged in groups (Figure 5b: ar- rows). The high columnar secretory cells of the membrane gland are densely filled with grana, whereas the contents of the columnar cells of the ripe mucous gland stain homo- geneously. The distal part of the mucous gland is formed by a highly folded epithelium of lower, columnar cells, with basophilic granular contents. This seems to be the “adhesive region.” In specimen C1, an area was observed with cells that had large, optically empty vacuoles within the cytoplasm (Figure 5a). These seem to be “old”? mucous cells that had secreted their contents, indicating that the animal must have had an oviposition. The vaginal duct is coated with a cuboidal, ciliated epithelium and covered with a layer of muscle fibers at the distal part in particular. The receptaculum is filled with sperm, the heads of which face the coating epithelium. The tails of the sperm are arranged in lines toward the stalk of the receptaculum (Figure 5e). The coating of the bursa copulatrix in specimen C1 is a pavement epithelium with large but flat nuclei. Only around the stalk does the epithelium consist of cuboidal cells, which still might have secretory function (see SCHMEKEL, 1971). The lumen is filled with aggregates of more or less dissolved sperm and prostatic secretion gran- ula (Figure 5g). Occasionally oocyte-like cells were found (Figure 5f). The vestibulum is lined with a columnar ciliated epi- thelium with mucus-secreting goblet cells between them. The surrounding tissue is interwoven with transverse and longitudinal muscle fibers. Only spermiogenesis could be observed in the smaller specimens. The development of the prostate secretion was not so advanced as in specimen C1: in smaller specimens, Figure 3 Anatomy and histology of the digestive tract. a. Digestive tract seen from the dorsal side; arrow, position of the central nerve ring; the internal septa of the oral glands are indicated. b. Cross section of a septum of the oral gland; position of the section see Figure 2f. c. Cross section of the pharynx. d. Cross section of the esophagus. Key: agl, acidophilic glands; bsgl, basophilic subepithelial glands; gc, goblet cells; ct, connective tissue; e, esophagus; i, intestine; lot, lumen of oral tube; (c,l,r)m, (circular, longitudinal, radial) muscle fibers; nv, nerve; odgl, opening of the digestive gland; ogl, “oral gland”’; ot, oral tube; pc, perivisceral cavity; ph, pharynx; rm, retractor muscle. Page 70 The Veliger, Vol. 28, No. 1mm Figure 4 External morphology of genital system. a. General outline. b. Jn situ, seen from the front. c. Same as Figure 4b, dorsal side. Key: a, ampulla; b, bursa copulatrix; dej, ductus ejaculatorius; e, esophagus; fgl, female gland; g, gonad; gd, gonoduct; p, penis; pogd, postampullar gonoduct; pr, prostate; prov, proximal oviduct; r, receptaculum seminis; vad, vaginal duct; vd, vas deferens. H. Wagele, 1985 the contents of the cells did not stain red, but instead a bluish-gray. Also, no spermatozoa were found in the prox- imal oviduct. Contrary to that of the large specimen C1, the mucous gland in the small specimens contained a great number of immature mucous cells, their contents not being of ho- mogeneous but of granular consistency. No sperm were found in the receptacula seminis. The epithelium of the bursa consisted mainly of apocrine secretory cells (Figure 5c). A pavement epithelium, as found in C1, was not seen; this feature probably depends on the functional phase of the bursa. Nervous System and Sensory Organs Morphology: The central nerve ring is placed at the proximal part of the esophagus (Figures 3a: arrow; 6a). The cerebropleural ganglia are close to each other dorsally and connect by the visceral loop (vl) ventrally. All ganglia of this visceral loop are fused with the brain. The pedal ganglia are placed close to the cerebropleural complex. The thick pedal commissure (pc) lies between the buccal commissure (bc) and the visceral loop. The two buccal ganglia are placed slightly asymmet- rically at the left side of the buccal commissure, which is as short as the other commissures. In the following list, all nerves originating from the cerebropleural complex are designated with a C, those originating from the pedal ganglia with a P and those branching from the visceral loop with a V. Nerves lying symmetrically on both sides and innervating the same or- gans are marked by a +. Page 71 C1+ Rhinophoral nerve; the proximal part of it forms the bulbous rhinophoral ganglion. C2+ This nerve runs along the oral tube to the oral ten- tacles. C3+ Optic nerve; it is very thin and shows no sign of pigmentation in Phyllidia pulitzert. C4+ This nerve runs to the rhinophores, and branches several times in the notum; it possibly innervates the rhinophore sheaths and the anterior dorsal part of the notum. C5 This branch of the right C4 runs ventrally into the notum and could not further be observed. C6+ This stout nerve leads into the retractor muscles, ramifies, and among other organs innervates the muscle fibers of the oral gland. C7+ Nervus pallialis posterior (HOFFMANN, 1939); it runs backward along the lateral side of the mantle, oc- casionally giving off nerves leading to the kidney and heart. C8 This thin, unpaired nerve lies in the connective tissue of the kidney, innervating the latter and leading to the genital apparatus. C9 This nerve leads ventrally under the oral glandular mass and disappears into the foot. P1+ This single pedal nerve is very thick, runs along the posterior part of the club-shaped oral tube to the ventral side, and leads caudally under the visceral mass. It dis- appears into the tight tissue of the foot in the posterior third of the body. P2, P3 These two nerves are branches of the right pedal nerve (P1), and innervate the genital organs from the ventral side. V1 Visceral nerve; branching off from the visceral loop. Figure 5 Histology of the genital apparatus. a. “Old”? mucous cells of the mucous gland; May-Gritinwald/Giemsa; scale 50 um. b. Cross section of proximal oviduct: for position of the section see Figure 4b; arrows, sperm—note that their heads are turned toward the wall; May-Griinwald/Giemsa; scale 5 wm. c. Apocrine secretory epithelium of bursa copulatrix of small specimen L1; azan; scale 5 um. d. Section of ampulla, with cuboidal epithelium; May-Griin- wald/Giemsa; scale 10 um. e. Specimen C1, section of the epithelium of the receptaculum seminis; the heads of the sperm are oriented toward the wall; May-Griinwald/Giemsa; scale 5 um. f. Oocyte-like cell (00) in the lumen of the bursa; arrow, epithelium of bursa; May-Griinwald/Giemsa; scale 5 um. g. An aggregation of sperm and prostatic secretion granula; arrows, epithelium of bursa; May-Grtinwald/Giemsa; scale 5 um. Key: b, bursa copulatrix; dgl, digestive gland; lb, lumen of bursa; mgl, membrane gland; mugl, mucous gland; ogl, oral gland; 00, oocyte-like cells; r, receptaculum seminis; vad, vaginal duct. Figure 6 a. Central nerve ring; dorsal side; at the transition of pharynx (ph), coming from ventrad, into esophagus (e); the esophagus is cut before it passes the nerve ring from ventrad. Key: bg, buccal ganglion; cpg, cerebropleural ganglion; pc, pedal commissure; pg, pedal ganglion; rhg, rhinophoral ganglion; vl, visceral loop. b. General outline of the excretory system. Key: ne, nephridium; ngl, gland in the nephridium; orpd, opening of the renopericardial duct into the nephridium; peep, pericardial epithelium; rpd, renopericardial duct; sy, syrinx; u, ureter; v, digestive gland and gonad covered by the nephridium. c. Epithelia and their position. The four types of glands: agl, acidophilic subepithelial glands (4); bsgl, basophilic subepithelial glands which stain granularly (3); gc, goblet cells (1); hgl, basophilic subepithelial glands which stain homogeneously (2); mu, “crystallized” mucus. Page 72 The Veliger, Vol. 28, No. 1 See |e ore whiny oe Sion 4 rae phe . ba vw s ‘ - ’ » | Ze e ge j - =a H. Wagele, 1985 P Page 74 The Veliger, Vol. 28, No. 1 H. Wagele, 1985 Histology: As mentioned above, two nerves run along the esophagus in the layer of the longitudinal muscle fibers (Figure 3d: nv). These nerves originate in the buccal gan- glia and end before the esophagus passes into the stomach. Shortly after their origin in the buccal ganglia, a branch of these nerves runs in the opposite direction to the phar- ynx, where it shows on each side a ganglionic enlargement in the layer of the longitudinal muscles, exactly at the transition of the esophagus into the pharynx. The epithelium of the rhinophores is composed of high columnar ciliated cells with basally lying nuclei. Gland cells were not detected. The interior part of the rhino- phore contains hyaline tissue (HOFFMANN, 1939), spicu- lae, and small groups of longitudinal muscle fibers. Spi- culae were also observed in the lamellae. The oral tentacles have an epithelium similar to the ventral side of the notum. In the lateral grooves of the tentacles, larger aggregations of perikarya with well stained nuclei, but with little cytoplasm, were found. These ag- gregations always lie in hollow spaces of the hyaline tis- sue. They are probably nuclei of sensory cells, although they could not be recognized as such in this preparation (see HOFFMANN, 1939). The statocyst of C1 has a diameter of approximately 90 um. It is placed between the pedal and cerebropleural ganglia (Figure 1c). The inner side is lined by a pavement epithelium. Above this epithelium, toward the cerebro- pleural ganglion, lies the thin neurilemma (Figure Ic: arrows). The statocyst is filled with 15-20 statoliths of oval to globular form. The average size of the statoliths was 6-8 um. Cilia were not detected by light microscopy. Excretory System Anatomy: The excretory system (Figure 6b) communi- cates with the pericardium by the so-called syrinx (sy), which lies ventral to the pericardium at the right side. From there, the renopericardial duct (rpd) leads foward under the pericardium to the region of the blood gland, where it enters the nephridium (ne) from the dorsal side. The nephridium covers the gonad laterally and dorsally and also reaches anteriorly between the viscera. The rami- fied appearance of the nephridium is caused by the branched vessels in the wall, starting with the aorta pos- terior. Page 75 The ureter originates from the caudal end of the ne- phridium and opens to the outside on the right side at the base of the anal papilla. Histology: The wall of the syrinx, a thick layer of lon- gitudinally plicated tissue, is covered at the inner side with a cuboidal epithelium with long cilia. At the transition from syrinx to the renopericardial duct, the cuboidal cells are replaced by apocrine secreting cells (Figure 7b: ar- rows). In the posterior part of the body, the renopericar- dial duct is submerged into the nephridium, but in the anterior part it lies dorsal of the nephridium. The lamellae and folds of the nephridial tissue are mainly formed by the ventral wall of the nephridium. In the posterior part they divide the nephridium into cham- berlike areas (Figure 7e). In the folds a thin layer of tissue, with blood vessels in between, separates the cells from each side. The nephrocytes are large cells with large, nonstaining spaces (Figure 7f). At the posterior part of the specimen a glandular com- plex that lies mainly dorsally and laterally on the visceral mass was observed (Figure 6b: ngl). From there, tubuli with a unistriate glandular epithelium and lying in the nephridial tissue reach anteriorly. In specimen L1 these ducts were missing. The gland is connected with the proximal ureter, where the same large secreting cells of the glandular mass are present between small epithelial cells (Figure 7c: u). The submerged, secreting cells have basal nuclei, whereas the nuclei of the epithelial cells lie apically. Near the end of the ureter the secreting cells are replaced by small ciliated, cuboidal cells (Figure 7d). Pericardial glands (Figure 7a) which project into the lumen of the pericardium are attached to the anterior dorsal part of the pericardial wall. Histology of the Epidermis Figure 6c shows a general outline of the epidermis. Dorsal notum epithelium: The epithelium is formed by large but flat cells (pavement epithelium) with submerged nuclei. Cilia were not observed. Some basophilic, sub- merged secretory cells (bsgl) are scattered all over the dorsal side. The contents of these secreting cells have a Figure 7 Anatomy and histology of excretory system. a. Section of the dorsal pericardial epithelium with pericardial glands (arrows); hemalaun/lightgreen; scale 5 um. b. Cross section of renopericardial duct; note the apocrine secretory cells (arrow); hemalaun/lightgreen; scale 10 um. c. Proximal ureter with secretory cells; for position of section see Figure 6b; May-Grtinwald/Giemsa; scale 10 um. d. Distal ureter; May-Griinwald/Giemsa; scale 5 wm. e. Glan- dular ducts lying in the nephridial tissue (arrows); for position of the section see Figure 6b; May-Grtinwald/ Giemsa; scale 25 um. f. Nephrocytes (nc) of a septum; hemalaun/lightgreen; scale 5 wm. Key: g, gonad; i, intestine; Ine, lumen of the nephridium; Irpd, lumen of the renopericardial duct (rpd); n, nucleus; nc, nephrocytes; ne, nephridial tissue; pc, perivisceral cavity; pe, perciardium; u, ureter. Page 76 granular appearance. Sometimes the contents evidently had been secreted (mu), and only the lining of the epithelium was stained (dark violet: May-Grtinwald/Giemsa). Sub- merged acidophilic glands (agl) were occasionally found. Ventral notum epithelium: This epithelium consists of cuboidal ciliated cells. The submerged basophilic glands, mentioned above, are abundant here. Acidophilic glands are not present. Dorsal foot epithelium: This appears to be the same as the ventral notum epithelium; only basophilic glands are represented. Ventral foot epithelium: Columnar ciliated cells with basal nuclei dominate. Four different types of glands are to be found in the ventral foot, and, of these, Types 1 and 2 are glands con- fined to the foot epithelium. Type 1: goblet cells (gc) with basophilic granular con- tents interspersed in the epithelium (see also digestive tract: oral tube and oral gland); Type 2: subepithelial basophilic glands, with contents stained homogeneously (hgl); Type 3: subepithelial, granularly stained, basophilic glands; Type 4: subepithelial acidophilic glands, especially at the margin of the foot. DISCUSSION External Morphology The external appearance of the specimens found in Khalkidhiki and Malta agrees with that of the animals photographed (1974a; animals from Portofino) and de- scribed by BARLETTA (1974a, b) and by SCHMEKEL & PORTMANN (1982, animals from Capo Miseno and Pon- Za). Digestive System In accordance with HOFFMANN (1939) I regard the glandular complex at the opening of the mouth as the foot gland and not as the labial gland, because it lies mainly ventrally and laterally of the vestibulum, not dorsally. BERGH (1868-69) shows drawings of cross sections through the distal oral tube of Phyllidia pustulosa Cuvier, 1804, and P. varicosa Lamarck, 1801. These clearly show that the oral tube has inner folds in addition to the lobed or papillate glands at the outside of the posterior oral tube. These outer glands are missing in P. pulitzeri. Bergh ob- served these inner folds in other species too, but did not describe them in detail. BERGH (1889) also mentioned in the first description of Phyllidiopsis striata foldlike features in the enlarged part of the oral tube. Because of the lack of anatomical examinations of the outer and inner oral glands and the total absence of histological examinations, possible homologies among different genera of the family Phyllidiidae cannot be discussed. The Veliger, Vol. 28, No. 1 An “intrabulbous” part of the pharynx, such as ob- served by BERGH (1868-69) in several species (Phyllidia varicosa, P. nobilis Bergh, 1868-69), is not present. The pharynx shows some pecularities in lacking features typ- ical for other groups of the Opisthobranchia: no labial disc or other armament is present at the transition from the oral tube to the pharynx, and radula, mandible, and salivary glands are absent. A stomach completely separate from the digestive gland, as possessed by other Doridacea, does not exist within the Phyllidiidae. The dorsal part of the central lumen of the digestive tract, where the openings of esophagus and in- testine are located, is part of the stomach, whereas the ventral part, with the openings into the digestive gland, seems to represent the central collecting cavity of the digestive gland (see HOFFMANN, 1939). The digestive gland corresponds to the holohepatic type. A caecum is absent. No cuticular structures could be found in the entire diges- tive system. Genital Apparatus The anatomy of the genital apparatus of the specimens examined in this study agrees with the description given by SCHMEKEL & PORTMANN (1982). All examined spec- imens were sexually mature. The largest of the three his- tologically examined specimens (C1) seemed to be on the verge of oviposition. This is indicated by the ripe eggs of the gonad, the spermiogenesis restricted to the glandular margin, and the stout membrane gland. No eggs were found in the ampulla or in the distal parts of the genital ducts. This was to be expected, because eggs are trans- ported through the gonoduct and oviduct only during ovi- position (SCHMEKEL, 1971). The smallest of the specimens examined histologically (L1) was in the protandrous phase. Eggs were not yet developed. Whether the sperm in the proximal oviduct of the larg- er specimen were autosperm or not, could not be deter- mined. The structure and position of the membrane gland and the mucous gland are as in other Doridacea (see SCHME- KEL, 1970, 1971). The large mucous gland envelops the tightly coiled membrane gland. The size of the receptaculum seminis in relation to the size of the bursa probably depends on the quantity of sperm present in the receptaculum. In accordance with this, the empty receptaculum of L1 is small in relation to the bursa, whereas the receptaculum of C1, which was filled with sperm, was larger than the bursa. In the bursa, a dissolution of sperm and of the prostate secretion takes place (SCHMEKEL, 1971). Nervous System The enumeration of the nerves and the records of the innervations of organs are still incomplete. To gain a more H. Wagele, 1985 Page 77 accurate conception of innervation more material is nec- essary. In some phyllidiids the connectives of the buccal ganglia are very long (see IHERING [1877] for Phyllidia varicosa; RISBEC [1956] for P. honloni Risbec, 1956; BOUCHET [1977] for Phyllidiopsis gynenopla Bouchet, 1977) and the buccal ganglia lie near the stomach. In others they are short (see RISBEC [1956] for Phyllidia sereni Risbec, 1956; MARCUS & Marcus [1970b] for Phyllidia tula Marcus & Marcus, 1970) as in Phyllidia pulitzert. Several authors describe gastro-esophageal ganglia lying near the buccal ganglia (IHERING, 1877; RIsBEC, 1928, 1956; Marcus & Marcus, 1970a; EDMUNDs, 1972). It may be possible that the ganglionic enlargements of the nerves in the pharyngeal layer of the longitudinal muscle fibers represent the gastro-esophageal ganglia in Phyllidia pulitzerr. This would mean that in this species these gan- glia are only detectable by histological examinations. Excretory System A sphincter muscle around the opening of the syrinx in the pericardium, as described by Hancock (1864) for some dorids, is absent in Phyllidia pulitzer. The renopericardial duct in other dorids is described as lying ventral to the nephridium (see SCHMEKEL & PORTMANN, 1982). In Phyllidia pulitzeri it is submerged into the nephridium and is very long as compared with the renopericardial ducts of other Doridacea. The function of the great glandular complex in the posterior part of the visceral mass is not clear. Whether this gland also exists in other dorids has yet to be exam- ined. BABA (1937) describes in Okadaza an accessory renal gland that is separated from the nephridium. This acces- sory gland opens with a duct into the middle of the ureter. Further investigations must be made before possible ho- tnologies can be discussed. Epidermis Whether the glands are single glandular cells or mul- ticellular glands could not be determined in the present study. The basophilic granular type of gland (type 3) seems to be responsible for the secretion of the mucus that “‘crys- tallizes” to small globules during fixation with formal- dehyde. Sometimes these globular structures could be found even within the glands. Strangely enough, this phenome- non is only mentioned once in the literature (ELIOT, 1910, for Phyllidiopsis carinata Eliot, 1910), although Schme- kel’s specimen from Ponza and also that described by PRUVOT-FOL (1962) have this “crystallized” mucus. The existence of the crystallized mucus apparently depends on the use of certain fixatives. In the presence of acetic acid, which is part of Bouin’s fluid and nearly all staining fluids, the globules dissolve. Whether this mucus is identical with the mucus in Phyllidia varicosa, described by JOHANNES (1963), is not clear. According to him this mucus, having a pH of approximately 7, can be secreted in great quan- tities within a few seconds, and has a poisonous effect on many invertebrates and vertebrates. Comparing the histological features of the oral gland and the ventral foot epithelium it is remarkable that, ex- cept for the basophilic granular type which is absent in the oral gland, both epithelia have the same gland types. Taxonomic Remarks on the Mediterranean Species of Phyllidia Three species of the genus Phyllidia are known from the Mediterranean at present: P. rolandiae Pruvot-Fol, 1951; P. aurata Pruvot-Fol, 1952; and P. (Phyllidiopsis ?) pulitzerr Pruvot-Fol, 1962. PRUVOT-FOL (1952) was in doubt as to whether Phyl- lidia rolandiae and P. aurata should be placed within the genus Phyllidia. Although they differed in some features (e.g., they lack the black color so typical for almost all species of the family Phyllidiidae), she did not want to erect a new genus based on only two specimens. As she did not dissect P. pulitzeri, she was uncertain as to which genus it belonged. The present study settles this problem: similar to the descriptions of the type species of Phyllidia (P. varicosa Lamarck, 1801, described by BERGH, 1868- 69), P. pulitzer: has a posterior club-shaped oral tube. Furthermore, there is no oral gland lying free ventral to the oral tube; therefore, it cannot belong to the genus Phyllidiopsis. When the descriptions of Phyllidia rolandiae and P. pu- litzert. are compared, no differences of taxonomic value can be detected. Unfortunately, information given on the same specimens differs in subsequent publications, and the first descriptions of these species were inadequate. Many features of P. rolandiae can only be inferred from PRUVOT-FOL’s (1952) description of P. aurata. The gills of P. aurata are said to be “moins saillantes’’; hence, those of P. rolandiae have to be bigger, similar to those of P. pulitzert. Information on features such as size is not suf- ficiently reliable, especially in organs where different kinds of fixation may cause variability. The same applies to the form and size of the tubercles as demonstrated for P. pul- itzert in the present study. The only differences between Phyllidia rolandiae and P. pulitzeri seem to be the shape of the surface of the tubercles (“‘bosselés” in P. rolandiae |PRUVOT-FOL, 1951: 37], smooth in P. pulitzeri) and the isolated gland near the posterior part of the oral tube in P. rolandiae, which is lacking in P. pulitzert. In the first description of P. pulitzeri, PRUVOT-FOL (1962:568-569) differentiates be- tween large tubercles, composed of five or six translucent “spherules” and smaller tubercles, composed of one, two, or three ‘“‘spherules.” She further states ‘Elles tiennent solidement au tégument; et ceci, en plus de la disposition plus ou moins réguliére sur le manteau, exclut l’idée d'un artefact, qu’il fallut tout d’abord écarter, 4 cause de leur forme sphérique insolite.” What she describes here very Page 78 likely is the mucus that, indeed, can hardly be removed from the surface, being partly crystallized in the epithe- lium. This assumption is supported by the fact that, in Pruvot-Fol’s figure of P. pulitzeri, some globules are drawn on the ventral side of the notum. Unfortunately a re-examination of the holotypes of the three Mediterranean species is not possible, as the holo- types of Phyllidia rolandiae and P. pulitzer: are considered to be lost (personal communication from Dr. P. Bouchet, Paris) and the holotype of P. aurata, located at the Mu- séum National d’Histoire Naturelle in Paris, is totally dissected, so that no organs are left except the notum. The examination of this notum revealed that Phyllidia aurata is clearly distinguishable by the granular appear- ance of its tubercles. This granular surface is caused by the spiculae with their outwardly pointing ends. The tu- bercles of P. aurata are arranged in lines (not mentioned by PRUVOT-FOL, 1952), as can be seen in the fixed holo- type, although this is difficult to make out, as it is in the fixed specimens of P. pulitzert. The possibility cannot be excluded that Phyllidia rolan- diae is identical to one of the other two Mediterranean species of the genus. Most of the few features known (col- or not known, tubercles “‘bosselés,” gills larger than in P. aurata) cannot be used to characterize a species. As to the only feature of importance, the “couche glandulaire,” the possibility of confusion with the blood gland must be con- sidered. Phyllidia rolandiae also has a small anal tubercle, which is thought to be absent in P. aurata but is present in P. pulitzert. In my specimens of P. pulitzeri, however, this feature shows great variability. Therefore, at present P. rolandiae cannot be distinguished from the other species. Certainly, the “P. rolandiae” mentioned by BARASH & DANIN (1982) is a P. pulitzerz. Thus, all evidence seems to indicate that P. rolandiae has to be regarded as a nomen dubium. ACKNOWLEDGMENTS I wish to thank Prof. Dr. L. Schmekel (Miinster) and Prof. Dr. H.-K. Schminke (Oldenburg) for their advice during the preparation of this study. I am grateful to my father, G. Stanjek, for the help in taking the pictures of the living animals, and Dr. habil. S. Perry, Dr. B. San- ders-Esser, and B. Jutsch for their assistance in the his- tological preparations. Dr. P. Bouchet (Paris) and Prof. Al. Barash (Tel Aviv) kindly sent me material for re- examination. My sincere thanks are due to Prof. Schminke who carefully re-read the manuscript and helped me with my English. LITERATURE CITED BaBA, K. 1937. 8. Contribution to the knowledge of a nudi- branch Okadaza elegans Baba. Japanese Journal Zoology 7: 147-190. BARASH, A. & Z. DANIN. 1982. Mediterranean Mollusca of The Veliger, Vol. 28, No. 1 Israel and Sinai: composition and distribution. Israel Jour- nal of Zoology 31(3—4):86-118. BARLETTA, G. 1974a. Genus: Phyllidia Cuvier, 1798. Phyllidia pulitzer: Pruvot-Fol, 1962. Schede Malacol. Mediterr. No. 43:1-4. BARLETTA, G. 1974b. Secondo reperto di Phyllidia pulitzeri Pruvot-Fol, 1962. Natura - Soc. Ital. Sci. Nat., Museo Civ. Ster. Nat. e Acquario Civ., Milano 65(1-2):25-32. BeRGH, R. 1868-69. Bidrag til kundskab om Phyllidierne, en anatomisk Undersegelse. Naturhist. Tidsskr. 5:357-542. BERGH, R. 1873. Neue Nacktschnecken der Stidsee, malaco- logische Untersuchungen. J. Mus. Godeffroy (1)2:65-96. BERGH, R. 1875. Neue Nacktschnecken der Siidsee, malaco- logische Untersuchungen. J. Mus. Godeffroy 8:91-116. BERGH, R. 1876. Neue Beitrage zur Kenntnis der Phyllidi- aden. Verh. Zool.-Bot. Ges. 25:659-674. BeRGH, R. 1889. Malacologische Untersuchungen, 16. Nu- dibranchien vom Meere der Insel Mauritius. Jn: C. Semper, Reisen im Archipel der Philippinen, 2. Teil, 2. Bd.:815- 872. BoucHET, P. 1977. Opisthobranches de profondeur de locéan Atlantique: 2—Notaspidea et Nudibranchiata. J. Mollus- can Stud. 43(1):28-66. Eaues, N. B. 1938. A systematic and anatomical account of the Opisthobranchia. John Murray Exped. (Brit. Mus. Natur. Hist.) Sci. Rep. 5, no. 4, London: 77-122. EpMuNDs, M. 1972. Opisthobranchiate Mollusca from the Seychelles, Tanzania, and the Congo, now in the Tervuren Museum. Revue Zool. Bot. Afr. 85(1-2):67-92. ELIoT, C. N. E. 1903. Nudibranchiata, with some remarks on the families and genera and description of a new genus, Doridomorpha. Pp. 540-573. In: J. S. Gardiner (ed.), The fauna and geography of the Maladive and Laccadive Ar- chipelagoes 2. Cambridge. Euiot, C. N. E. 1904. On some nudibranchs from East Africa and Zanzibar. 6. Proc. Zool. Soc. Lond. for 1904:268-298. EuioT, C. N. E. 1910. Nudibranchs collected by Mr. Stanley Gardiner from the Indian Ocean in H. M. S. Sealark. Trans. Linn. Soc. Lond. 13:411-438. Hancock, A. 1864. On the structure and homologies of the renal organ in the nudibranchiate Mollusca. Trans. Linn. Soc. Lond. 24:511-530. HecntT, E. 1895. Contributions a |’étude des nudibranches. Mem. Soc. Zool. France 8:539-711. HOFFMANN, H. 1939. I. Opisthobranchia. Jn: H. G. Bronn (ed.), Klassen und Ordnungen des Tierreiches. Bd. 3, Abt. 2, Buch 3. 1247 pp. IHERING, H. von. 1877. Vergleichende Anatomie des Nerven- systems und Phylogenie der Mollusken. Verlag Wilhelm Engelmann: Leipzig. 290 pp. JOHANNES, R. E. 1963. A poison-secreting nudibranch (Mol- lusca: Opisthobranchia). Veliger 5:104-105. Marcus, E. 1962. Opisthobranchs from Florida and the Vir- gin Islands. Bull. Mar. Sci. 12(3):450-488. Marcus, E. & E. Marcus. 1970a. Some gastropods from Madagascar and west Mexico. Malacologia 10(1):181-223. Marcus, E. & E. Marcus. 1970b. Opisthobranch mollusks from the southern tropical Pacific. Pacific Sci. 24:155-179. Pruvot-FoL, A. 1951. Etude des nudibranches de la Médi- terranée. Arch. Zool. Exp. Gén. 88(1):1-80. Pruvot-FoL, A. 1952. Un nouveau nudibranch de la Médi- terranée: Phyllidia aurata n. sp. Bull. Soc. Zool. Fr. 77(5- 6):408-411. PRuvoTt-FoL, A. 1962. Deux trés rares nudibranches de la Méditerranée. Bull. Soc. Zool. Fr. 87(5-6):566-569. H. Wagele, 1985 Page 79 RisBEc, J. 1928. Contribution a l'étude des Nudibranches Néo- Calédoniens. Faune Colon. Fr. 2. 328 pp. RIsBEC, J. 1956. Nudibranches du Viet-Nam. Archs. Mus. Natur. Hist. Natl. Paris (6)4:5-78. RoMEIs, B. 1968. Mikroskopische Technik. R. Oldenbourg Verlag: Miinchen, Wien, 16. Auflage. 757 pp. SCHMEKEL, L. 1970. Anatomie der Genitalorgane von Nudi- branchiern (Gastropoda, Euthyneura). Pubbl. Staz. Zool. Nap. 38:120-217. SCHMEKEL, L. 1971. Histologie und Feinstruktur der Geni- talorgane von Nudibranchiern (Gastropoda, Euthyneura). Z. Morph. Tiere 69:115-183. SCHMEKEL, L. & A. PORTMANN. 1982. Opisthobranchia des Mittelmeeres. Nudibranchia und Saccoglossa. Springer Verlag: Berlin, Heidelberg, New York. 410 pp. WAGELE, H. 1984. Kiemen und Hamolymphkreislauf von Phyllidia pulitzeri (Gastropoda, Opisthobranchia, Dorida- cea). Zoomorphology 104:246-251. The Veliger 28(1):80-93 (July 1, 1985) THE VELIGER © CMS, Inc., 1985 The Nudibranch Genera Onchidoris and Diaphorodoris (Mollusca, Opisthobranchia) in the Northeastern Pacific by SANDRA V. MILLEN Department of Zoology, University of British Columbia, Vancouver, B.C., Canada V6T 2A9 Abstract. An anatomical review was conducted on the genus Onchidoris in the northeastern Pacific. Comparisons were based on the published literature and specimens obtained from both the North Atlantic and North Pacific oceans. Onchidoris muricata (Miller, 1776) occurs in both the Atlantic and Pacific oceans, and O. varians (Bergh, 1878) and O. hystricina (Bergh, 1878) are junior synonyms of this species. The nudibranch commonly considered in California to be O. hystricina was an unnamed species belonging to the genus Diaphorodoris. This new species, D. lirulatocauda, is described and compared with other species in the genus, including Diaphorodoris mitsuti (Baba, 1938) comb. nov. The relationships of the genus Diaphorodoris with other genera in the family Onchidorididae are discussed. INTRODUCTION FOUR SPECIES OF Onchidoris have been reported from the northeastern Pacific. The species Onchidoris bilamellata Linnaeus, 1767, and O. muricata Miller, 1776, have also been recorded from both sides of the Atlantic (THOMPSON & BROWN, 1976). Onchidoris hystricina (Bergh, 1878) and O. varians (Bergh, 1878) were described from the north- eastern Pacific as cognate species of O. muricata. ABBOTT (1974) suggested that both these species are synonyms of O. muricata. Unfortunately, the type material is lost (Zoo- logisk Museum, Copenhagen, personal communication), and comparisons must therefore be made on the basis of the literature. Three species of Onchidoris are found in British Colum- bian waters. Onchidorvs bilamellata is easily recognized due to its mottled brown (rarely white) coloration and its unique habit of preying on barnacles. Another species is identical to the Californian species commonly referred to as O. hystricina. The third species is of the type referred to as O. muricata by BEHRENS (1980:67). An anatomical investigation of the latter two species was undertaken and comparisons made with O. muricata of the Atlantic and the literature. Onchidoris muricata (Miller, 1776) Doris aspera Alder & Hancock, 1842. Doris diaphana Alder & Hancock, 1845. Doris pallida Agassiz, 1850. Doris ulidiana Thompson, 1845. Lamellidoris varians BERGH, 1878:613-614; BERGH, 1879: 365; BERGH, 1880a:216-219, pl. 11 (figs. 13, 14); BERGH, 1880b:67-70, pl. 11 (figs. 13, 14), pl. 13 (fig. 1); BERGH, 1890:985; BERGH, 1892:1153 (161); syn. nov. Lamellidoris hystricrna BERGH, 1878:605, 614, pl. 68 (figs. 17-23); BERGH, 1879:365; BERGH, 1880a:219-221; BERGH, 1880b:70-72; BERGH, 1890:985; BERGH, 1892: 1153 (161); syn. nov. Onchidoris hystricina (Bergh, 1878): Marcus, 1961:28, 57, pl. 5 (figs. 89-91). External morphology: I examined preserved specimens from Norway which ranged in length from 4 to 11 mm, from the Atlantic coast of the United States with lengths of 4 to 10 mm, and from British Columbia, Canada, from 1 to 10 mm in length. Most animals were 5-7 mm long. The body shape is oval, slightly wider and more truncate in front, with a low arch (Figure 1A). A small mantle margin overhangs the sides and foot. The mantle becomes disproportionately larger as the animal’s size increases. The notum is covered with rounded tubercles, flattened and uneven on top, constricted at their bases, giving them a mushroom shape. The tubercles are large, with a few, scattered, small tubercles. Towards the mantle edge all of the tubercles are small. The tubercles of specimens from the Canadian Pacific, Atlantic, and Norwegian Sea had the same aver- age tubercle size. The larger tubercles of specimens from S. V. Millen, 1985 Page 81 these three areas were 0.51-0.56 mm high and 0.59-0.65 mm wide at the top. Spicules run lengthwise in the tubercles (Figures 1B, 2A, B) and are capable of being protruded through open- ings in grooves along the flattened top (KREss, 1981, figs. 5E, F). In a relaxed state the spicules do not protrude and the tubercle is an inflated mushroom shape. When con- tracted, the tubercles appear cylindrical with flattened spiculose tops. Short spicules radiate in a star-like pattern in the notum at the tubercle bases. In the notum there is a dense spicule arrangement that shines through the in- tegument in a cross, transverse, circular, radiating pattern as diagrammed by ALDER & HANCOCK (1855, pl. 48 [fig. 2]). The margins of the rhinophores bear 2 (sometimes 3) tubercles, and there are 3 to 8 tubercles inside the bran- chial circlet, which is located on the posterior midline. The simply pinnate, contractile gills are separate, ar- ranged in a nearly complete, transverse oval, broken mid- posteriorly by the post-anal tubercle. Gill number varies between 6 and 18 in Pacific specimens, 8 to 14 in Atlantic specimens. The anterior-most gills are the largest, de- creasing gradually in size toward the posterior. The rhinophores are long and slender with a rectan- gular, flat-topped tip. The stalk is short, and most of the clavus has long sloping lamellae. The lamellae, except for the most distal one, are attached along the anterior line. Posteriorly only the first 3 or 4 are complete, with an ever-widening bare space proximally. Atlantic specimens had 9 to 20 lamellae, Pacific specimens had 6 to 10. The rhinophore margin is not raised and is smooth except for 2 (sometimes 3) tubercles which are positioned on either side of the anterior border. The head has a semicircular velum, usually with folds to mark the triangular tentacles that are attached poste- riorly. The large foot is truncate anteriorly, thickened but not bilabiate. The foot is wider in front than behind and ends in a bluntly rounded tail which is covered by the mantle margin. Living Pacific specimens were usually white, occasion- ally creamy-white, light yellow, or light orange. The no- tum is semitranslucent. Through it can be seen the bright red digestive gland, which in mature specimens becomes obscured by creamy gonads. In mature animals, a dark brown spot, the sperm-filled bursa copulatrix, can be seen through the anterior right side of the dorsum. Ventrally, the red digestive gland shows clearly for % of the body length. It extends farther forward on the left side. The leaves of the rhinophores are dusky yellow or orange. The branchiae are lighter than the body, white or dusky yellow with an opaque white base. A color photograph appears in BEHRENS (1980:fig. 72). Atlantic specimens are either white or pale yellow, the latter color being more common at the northern end of its range (THOMPSON & Brown, 1976). Digestive tract and radula: The internally folded buccal tube is short, broad, and flaccid. The buccal bulb has a dorsal rounded sucking crop with a broad median mus- cular band and a short stalk. The radular sac projects posteriorly. It is long, cylindrical, and usually bent to one side. The lip disk has been described as having a thick yellowish cuticle (BERGH, 1880).! With the aid of the scanning electron microscope, I found the lip disks of At- lantic and Pacific specimens to be finely papillate toward the central area. The opening is guarded by two ventral flaps (Figure 1C). The lip papillae were illustrated by BERGH (1878, pl. 68 [fig. 17]) for Onchidoris hystricina, and reported by BERGH (1880) in O. varians. Atlantic specimens of Onchidoris muricata have been reported to have radulae ranging in length from 29 to 44 rows. I found that specimens from the Atlantic have from 27 to 34 rows and specimens from the Pacific have 20 to 33 rows. The radula is narrow, with the formula 1.1.1.1.1 (Fig- ure 1D). The central (rachidian) tooth is an elongate rect- angular shape with thickened sides. In Atlantic specimens of Onchidoris muricata its length was 0.05 mm (BERGH, 1880). Specimens that I examined from the Atlantic had central-tooth lengths of 0.03-0.04 mm (X = 0.04 mm; n = 9) and from the Pacific 0.02-0.06 mm (X = 0.04 mm; n = 25). Each large lateral tooth (Figure 1D) has a triangular- shaped base with a denticulate hook. At the base of the denticulations is a knoblike projection from which a small wing extends down the inner side of the tooth. Reported tooth height for Onchidoris muricata is 0.075-0.12 mm (BERGH, 1880; MEYER, 1971). My specimens from the Atlantic had a tooth height of 0.07-0.10 mm (X = 0.08 mm; n = 8) and from the Pacific 0.04-0.10 mm (X = 0.08 mm; n=16). Bergh did not measure the lateral tooth height of O. hystricina but reported them to be smaller than the 0.12-0.17 mm he found in O. varians, although of the same shape (see Figures 9C, D). The number of denticles varied in Atlantic O. muricata from 9 to 16 and in Pacific specimens from 8 to 18. There was substantial variation in the strength of denticulation and the numbers of denticles. Older, worn teeth had the tips of the hooks ground away and the denticles extended almost to the tip as in O. varians. Younger teeth had a longer, straighter, smooth cusp with denticles only near the base as in O. hystricina. The marginal teeth have a triangular base with a single strong recurved hook facing posteriorly (Figure 1D). Onchidoris muricata from the Atlantic had a marginal tooth height of 0.04 mm (BERGH, 1880). Atlantic specimens that I examined had a height of 0.03-0.04 mm (X = 0.03 mm; n = 9), while those from the Pacific had a tooth height of 0.02-0.05 mm (X = 0.03 mm; n = 19). ' Text references to BERGH (1880) refer both to BERGH (1880a) and (1880b) listed separately in the Literature Cited. They are the same paper published in two different journals. (hii ae ji y, Wy Hj i; ee The Veliger, Vol. 28, No. 1 S. V. Millen, 1985 Page 83 0.1mm Figure 2 Tubercles of Onchidoris muricata showing the arrangement of the spicules. A. Contracted state. B. Relaxed state. At the posterior end of the buccal bulb is a narrow tubular esophagus. The salivary glands are attached on either side of the base of the esophagus. BERGH (1880) described them as 2 or 3 thick, white coils in Onchidoris muricata, and as a large, whitish mass in O. varians. I found them to be small and U-shaped in both Pacific and Atlantic specimens. The stomach is buried in the digestive glands except at the junction of the intestine, where a small, round, stalked caecum is given off. The digestive glands appear as one oval reddish mass hollowed on the anterior right due to the reproductive organs. The narrow tubular intestine loops to the right around or over the caecum and runs straight to the anus, located at the pos- terior of the branchial circlet at the base of a large tuber- cle. The anal opening is simple and not raised. The in- conspicuous renal pore is located within the circlet to the right of center, surrounded by tubercles. Circulatory system: The pericardial sac contains a pos- terior, thin-walled, triangular auricle and a ventricle. The aorta ends in a large, granular, white, blood gland situated above the central nervous system. Central nervous system: The central nervous system has been well described for Onchidoris muricata by BERGH (1880). He also described and illustrated this system for O. varians (pl. 13 [fig. 1]) and described it for O. hystricina. In all three species the cerebral and pleural ganglia are fused, ovate or rounded, and connected by a short com- missure. The almost separate pedal ganglia are rounded and only slightly smaller than the cerebro-pleurals. The eyes are on moderately long, fine stalks. There were no discernible differences in this system in specimens from the Atlantic or the Pacific. Reproductive system (Figure 3): The ovotestis consists of creamy-yellow lobules on the dorsal surface, sides, and part of the ventral surface of the digestive glands. Its his- tology and maturation have been studied by BEHRENTZ (1931) and Topp (1978a). The branched gonoducts of the ovotestis merge forming a thin pre-ampullary duct. This duct widens into a U-shaped ampulla, which is at- tached to the inner, lower curvature of the albumen gland. It narrows to form a thin post-ampullary duct, which ends at a triple junction. One branch becomes the vas deferens, another leads to the buried receptaculum seminis (fertil- ization chamber), and a third, the short oviduct, enters the female gland mass. The vas deferens is narrow and prostatic for a short distance. It becomes non-prostatic, looping dorsally, then enlarging into a muscular penial sac. Inside, the vas deferens coils and then straightens, ending in an unarmed, simple, bifurcate or trifurcate pe- nis. The vagina, which is short and muscular, has a sepa- rate opening posterior to the penial sac. The vagina leads to a bluntly rounded, blind sac, where on one side the moderately long duct to the large round bursa copulatrix is given off. The fertilization duct is long, muscular and convoluted, terminating in a buried, oval receptaculum seminis. The arrangement of the ducts is semiserial. The female gland mass has a separate nidamental duct ventral to the vagina. This mass has an anterior, yellow- ish, albumen gland and a posterior, inner, mucous gland. The receptaculum seminis is buried in the albumen gland. The reproductive openings are located on the right side a short distance posterior to the anterior margin of the foot. Figure 1 A. Onchidoris muricata, 7 mm. Photograph of a live Pacific specimen. B. Tubercle of O. muricata, SEM micrograph of a Pacific specimen. Scale = 100 um. C. Lip disk of O. muricata, SEM micrograph of an Atlantic specimen. Scale = 100 wm. D. Radula of O. muricata, SEM micrograph of a Pacific specimen. Scale = 40 um. E. Diaphorodoris lirulatocauda, 12 mm. Photograph of a live specimen. F. Posterior half of D. lirulatocauda showing tubercles and gills. SEM micrograph. Scale = 1 mm. G. Lip disk of D. lirulatocauda, SEM micrograph. Scale = 50 um. H. Radula of D. lirulatocauda, SEM micrograph of medial area showing connecting membranous wing of laterals. Scale = 10 um. I. Radula of D. lirulatocauda, SEM micrograph of lateral and first marginal teeth. Scale = 20 um. J. Radula of D. lirulatocauda, SEM micrograph of inner marginal teeth. Scale = 10 wm. K. Radula of D. lirulatocauda, light microscope photograph of one half row including reduced outer platelet. Page 84 The Veliger, Vol. 28, No. 1 Figure 3 Onchidoris muricata reproductive system, drawn using a camera lucida. Key: al, albumen gland; am, ampulla; bc, bursa copulatrix; bs, blind sac; mu, mucous gland; ov, oviduct; p, penis; pd, preampullary duct; pr, prostate; ps, penial sac; rs, receptaculum seminis; ud, uterine duct; v, vagina; vd, vas deferens. This system was found to be essentially the same for Atlantic and Pacific specimens. It agrees with the descrip- tion given for Onchidoris muricata by BERGH (1880) and with his partial descriptions for O. varians and O. hystni- cina. It should be noted that Bergh did not consider the vas deferens (spermatoduct) of O. hystricina to be very long. The reproductive system was reconstructed by BEH- RENTZ (1931, figs. 6-9) using serial sectioning. Behrentz noted that the penis has three “horns.” On specimens of O. muricata that I examined, the penis was often a simple cylinder or showed two lobes. Ecology: Onchidoris muricata occurs in the low intertidal zone and shallow subtidal to 20 m (THOMPSON & BROWN, 1976). It ranges from the White Sea to Cape Finisterre, Greenland, from Nova Scotia to Rhode Island in the At- lantic, and from Kyska, Alaska, to California in the Pa- cific. The southern range limit is uncertain due to con- fusion with the next species and an undescribed Adalaria species. I examined specimens from as far south as Aba- lone Beach, Humboldt County, California and Marcus’s (1961) specimen came from Dillon Beach, California. In Britain this species eats a variety of encrusting bryo- zoans, especially Electra pilosa, Membranipora membra- nacea, and Alcyonidium polyoum (THOMPSON & BROWN, 1976; Topp, 1978b, 1979a, 1981). In the Pacific it also eats a variety of encrusting bryozoans, most of which are cheilostomate. Specimens have been reported feeding on Reginella mucronata, Eurystomella bilabiata, and Muicro- porella cribosa (MCDONALD & NYBAKKEN, 1978; GODDARD, 1984). In British Columbia the species feeds most often on Schizoporella unicornis but has also been found on Hippodiplosia insculpta, Cheliopora praelonga, Lagenipora sp., Microporina borealis, and Membranipora serrilamella (personal observations). It feeds by sucking the bryozoan polypides out of their skeletons, approxi- mately 19-32 polypides being eaten per adult per day (Topp, 1979a, 1981) at a rate of 0.4-5.2 polypides per hour (TopD, 1979b, 1981). The nudibranchs are nor- mally found on rocks and under boulders. Topp (1978b) found that their distribution on the undersurface of boul- ders showed aggregation, which was particularly pro- nounced during breeding season. The life cycle has been studied in Sweden (BEHRENTZ, 1931) and in Britain (THOMPSON, 1961b; MILLER, 1962; Topp, 1979a, b, 1981). In both places this species was an annual, settling in the summer, growing until the early spring, when animals over 3 mm spawn. Spawning ani- mals die in June, leaving a brief interval between gener- S. V. Millen, 1985 Page 85 X X Jee iViea Md 3) Ass © ND Figure 4 Onchidoris muricata. Annual cycle of Pacific animals. Preserved length in mm versus month collected. Spawn present in months marked “x.” n= 91. ations. Specimens collected from British Columbia (Fig- ure 4) had a similar annual cycle. The spawn mass is white or pale orange in 12-2 whorls. The eggs are in a string folded up and down in the whorl, although this pattern is not obvious. The egg mass is quite thick and leans inward. The eggs are usually found one to a double-walled capsule. The eggs are 75.0-77.3 um (GODDARD, 1984), 73-100 um (THompPsON, 1967), or 80- 100 um (Topp, 1979b) in diameter, with a capsule size of 99-117 um (personal observation). There are approx- imately 2500 eggs/ribbon (THOMPSON, 1967; Topp, 1978a). Each animal lays approximately 15,000-34,000 eggs in its lifetime (TODD, 1979b). Egg masses are pro- duced in a regular 4-5 day cycle, with mating necessary prior to each spawning (TODD, 1978b). The eggs hatch in 7-20 days (MILLER, 1958; THOMPSON, 1967; Hurst, 1967; GODDARD, 1984; per- sonal observations) into Type-1 veligers of “THOMPSON (1967), with shell Type 1 (THOMPsoN, 1961a) and having a length of 117-136 wm (GODDARD, 1984; personal ob- servations). The larvae are long-term planktotrophs with a larval life estimated at 12 wk (BEHRENTZ, 1931) or 7- 8 wk (Topp, 1979a, b). I maintained larvae, fed on phy- toplankton, in the laboratory at 8°C for 7 weeks, when they suddenly died without metamorphosing. This cor- roborates Todd’s estimate. Animals in Britain preferen- tially settle on the bryozoan Callopora aurita even though as adults they prefer Electra pilosa (TODD, 1979a, b, 1981). Adult animals are quite active compared to other dorid nudibranchs and have the disconcerting habit of crawling out of their dish and drying up on the sides when kept in the laboratory. Diaphorodoris lirulatocauda Millen, spec. nov. Onchidoris hystricina (Bergh, 1878): BEEMAN & WILLIAMS, 1980:328 (section 14.42), pl. 105 (fig. 14.42) (non 0.1mm Figure 5 Diaphorodoris lirulatocauda tubercle showing the arrangement of the spicules. Bergh); BEHRENS, 1980:66-67 (fig. 71) (non Bergh); McDOonaLpD & NYBAKKEN, 1981:16, 31, 44-45 (fig. 18) (non Bergh); NyBAKKEN & MCDONALD, 1981:440, 442 (fig. 1H) (non Bergh); MCDONALD, 1983:198-199 (fig. 36) (non Bergh); JAECKLE, 1984:209 (non Bergh). Onchidoris sp. (cf. Onchidoris hystricina): GODDARD, 1984: 143-163. Material: Holotype: British Columbia Provincial Mu- seum, BCOPM-984-347-1, 5 June 1979, Earls Cove, Brit- ish Columbia, Canada (49°45'N, 124°01’W), 20 m depth, rocky substrate on bryozoans growing on Rhabdocalyptus dawsonu, spawning, coll. S. Millen. Paratypes: In the British Columbia Provincial Mu- seum three lots: BCPM-976-1037-6, 27 Mar. 1976, Juan Perez Sound, Queen Charlotte Islands, British Columbia, Canada (52°35.8'N, 131°25.2’W), 10-20 m, rock with coralline algae, 3 specimens, coll. P. Lambert; BCPM- 976-1057-5, 18 June 1976, Arbutus Island, British Co- lumbia, Canada (48°42.4'N, 123°26.1'W), <13 m, rocky substrate, 4 specimens, coll. P. Lambert; BCPM-976- 1073-10, 2 Aug. 1976, Discovery Passage, British Colum- bia, Canada (50°19.7'N, 125°26.4'W), <25 m, rock with hydroids, 6 specimens, coll. P. Lambert. In the California Academy of Sciences two lots: CASIZ 031680, 8 Aug. 1968, Hazard Canyon, San Luis Obispo Co., California, 1 specimen, coll. D. Roller; CASIZ 031682, 8 Aug. 1964, Moss Beach, California, 1 specimen, coll. L. Andrews. Etymology: The name lirulatocauda is derived from the Latin lirulatus, meaning “ridged,” and cawda, meaning “tail,” and refers to the mid-dorsal ridge on the tail. This feature distinguishes this species from similar small white dorids in the family Onchidorididae. Page 86 1mm Figure 6 Head of Diaphorodoris lirulatocauda. External morphology: I examined 44 specimens from British Columbia and California ranging in length from 3 to 12 mm. The body shape is elongate-oval (Figure 1E), wider in front than behind, with a trailing, keeled tail. The mantle margin is not wide, but covers the high sides and is slightly longer in front, covering the head. The notum bears elongate, slender, cylindrical tubercles with blunt but not inflated ends. The tubercles taper slightly from their bases. They are soft and capable of slight con- traction. The tubercles show little variance in size, al- though smaller tubercles predominate toward the mantle edges. They are spaced fairly far apart, not crowded. In some specimens, the tubercles appear to form longitudinal rows. Larger tubercles are from 0.34 to 0.64 mm high and 0.13 to 0.28 mm wide. Spicules are found in the tubercles, but they do not protrude, even when the tubercle contracts (Figure 1F). The spicules are densely packed in the central core of the tubercles and have trifurcate ends (Figure 5). At the bases of the tubercles the spicules extend in a radial, star-like pattern through the notum. In the notum there are large, curved spicules with side prongs and slightly smaller S-shaped spicules scattered in the connective tissue. These spicules do not form a definite pattern, nor are they visible in living animals. In the foot, the spicules form a crisscross pattern. The margins of the rhinophores bear three tu- bercles, two anterior and one posterior. There are no tu- bercles within the branchial circlet. The branchial margin is smooth except for one large posterior tubercle, but sev- eral slightly smaller tubercles are sometimes present (Fig- ure 1F). The simply pinnate, contractile, branchiae are non-re- tractable, enclosed in a common sheath, and joined at their bases. There are 4-9 gills, the most anterior being the longest and the most posterior two being very small. The rhinophores are long and slender, with a long, blunt tip. The stalk is long and most of the clavus bears sloping lamellae. The 6-10 lamellae are attached along a vertical, anterior line, except for the most distal one or The Veliger, Vol. 28, No. 1 two. The lamellae slope ventrally and meet posteriorly forming a chevron, except for the most proximal two or three which are incomplete. The head (Figure 6) is rounded, not extended into a large veil, but appearing as a double, rounded mound, separated by the vertical mouth opening. There are small, longitudinal, lateral tentacle grooves. The foot is narrow and elongate, wider and truncate anteriorly. The anterior foot edge is not bilabiate. Dor- sally the protruding tail, which ends in a sharp point, has a medial ridge. Living specimens are white or creamy-white, with opaque white, granular flecks in the notum, the top of the foot, and head, but not on the tubercles. In mature ani- mals, the mid-dorsal region appears darker yellow due to the creamy gonads underneath. Sometimes there is a brown spot on the anterior right, indicating the location of the sperm-filled bursa copulatrix. Ventrally the dark-brown digestive gland is visible through the foot, although it is often obscured by the creamy-yellow gonads. The rhino- phores are creamy-yellow, as are the gills. The gills may have white granulations near their bases and an opaque white line up the central shaft. Color photographs appear in BEEMAN & WILLIAMS (1980: pl. 105 [fig. 14.42]), BEH- RENS (1980:fig. 71), and MCDONALD & NYBAKKEN (1981: fig. 18). Digestive tract and radula: The soft buccal tube is sur- rounded by glands made up of large granules. The buccal bulb has a dorsal, rounded sucking crop which is sessile. The crop has a broad muscular band dorsally, but only a thin muscular strip posteriorly. On the ventral surface a small radular sac projects posteriorly. The lip disk is smooth, except in the central passageway, where it has small, oval papillae that are 7-8 wm in diameter (Figure 1G). The radula has 29-33 rows. The radular formula is 2.1.0.1.2, with no central (rachidian) tooth. A membra- nous wing runs from the inner posterior corner of each lateral tooth to join just inside the inner base of the fol- lowing tooth (Figure 1H). The large lateral teeth (Figure 1I) have a wide triangular base with a small, needlelike recurved hook. At the inner base of the hook extends a comblike row of 11-13 denticles. Above and inside the row of denticles is a prominent knob. The lateral teeth range in height from 0.05 to 0.06 mm. The innermost marginal tooth has an oval base with a posterior-facing middle hook (Figure 1J). The height is 0.02 mm. On the outside of this is a small, insubstantial, oval plate (5-10 um long), representing a rudimentary second lateral tooth (Figure 1K). At the posterior end of the buccal bulb is the long, thin esophagus. The salivary glands insert at its base. These are long, thin straps running down the sides of the diges- tive gland for half its length before bending ventrally. The small stomach is buried in the digestive gland, but a round, short, stalked caecum extends from it to the surface. In S. V. Millen, 1985 Page 87 Figure 7 A. Diaphorodoris lirulatocauda reproductive system, drawn using a camera lucida. Key: al, albumen gland; am, ampulla; bc, bursa copulatrix; fe, fertilization chamber; me, membrane gland; mu, mucous gland; ov, oviduct; p, penis; pd, preampullary duct; pr, prostate; rs, receptaculum seminis; ud, uterine duct; v, vagina; vd, vas deferens. B. Conical penis with everted, armed, central core. mature animals the digestive gland is covered by the go- nads. The oval, brown digestive gland appears as one mass. The intestine emerges beside the caecum, curves to the right, and runs as a long, thin tube to the anus, located in the center of the branchial circlet. The renal pore is to the right and slightly anterior to the anal opening. Circulatory system: The auricle is large, triangular, and thin. The ventricle is small, muscular, and rounded. The muscular aorta ends in fluffy, white, blood glands located just posterior to and slightly over the central nervous sys- tem. Central nervous system: The cerebro-pleural ganglia are fused, large, and elongate oval in shape. The smaller, rounded pedal ganglia are ventrally located and are con- nected by a short circumesophageal commissure. The ol- factory bulbs have a short stalk. The eyes are connected to the cerebro-pleural ganglia by long optic nerves with small bulbs at their bases. The paired buccal ganglia are separated by a short commissure and each has a gastro- esophageal ganglion attached by a short stalk. Reproductive system (Figure 7): The ovotestes are creamy-yellow lobules entirely covering the digestive gland, including the ventral side. The gonoducts are broad, shiny white and conspicuous, uniting to form a central pre- ampullar duct, which widens into the U-shaped ampulla. This ampulla is attached to the inner side of the female Page 88 Figure 8 Diaphorodoris lirulatocauda. Annual cycle. Preserved length in mm versus month collected. Spawn present in months marked “x. n = 40. gland mass. The post-ampullar duct bifurcates into a short, wide fertilization duct and an extremely long, coiled vas deferens. The inner portion of the vas deferens widens into a soft granular prostatic section, then narrows to a coiling, muscular portion. Near the outer body wall the vas deferens widens slightly to join a common atrium with the vagina. This atrium has a plicate edge. The conical penis is located at the anterior of the atrium, and has a central protrusible core bearing spines. The core, which can extend 300-400 um, bears approximately 8 irregular rows of 15-25 spines. The spines are simple rods with pointed, downward-tipped ends. They vary in length from 8 to 23 um, the shortest being most proximal. The vagina is long and cylindrical, wider near its pos- terior opening in the atrium and gradually narrowing into a long duct leading to the sessile bursa copulatrix. Just after it narrows, the uterine duct branches off. The vagi- nal duct beyond this point has a double-partitioned inte- rior. The bursa copulatrix is a large, round thin-walled sac, dark-brown when filled with sperm. At its junction with the vagina, the moderately long duct of the club- shaped receptaculum seminis bends anteriorly. The ar- rangement of the ducts is vaginal. The uterine duct is combined with the vagina for one-half of its length. It then separates from the vaginal duct and crosses the pos- terior portion of the female gland mass. It terminates in a slightly swollen fertilization chamber next to the short oviduct, which in turn enters the female gland mass. The female gland mass consists of a membrane gland, mucous gland, and albumen gland. Its exit is located just The Veliger, Vol. 28, No. 1 ventral to the vaginal opening. A short duct widens into an interior, white membrane gland and then widens fur- ther into a more dorsal, yellow, highly convoluted albu- men gland. On top of the albumen gland is the soft, gran- ular, white coil of the mucous gland. The oviduct enters the mucous gland at its junction with the albumen gland. The genital openings are located on the right side, a short distance behind the anterior margin of the foot. Ecology: The habitat and life cycle of this species has been frequently confused with those of Onchidoris mun- cata. I will therefore restrict my observations to specimens that I have examined. This species occurs in the low in- tertidal down to 126 m subtidally. It ranges from Juan Perez Sound, Queen Charlotte Islands, British Columbia, to Point Loma, California. This species has been observed eating the ctenostome bryozoan Nolella stipitata. The nudibranchs are usually found under rocks intertidally or crawling on rocky sur- faces and sponges subtidally. They have been found from February to October, reaching their largest size in the summer (Figure 8). Spawning has been observed in June, July, and September (GODDARD, 1984; personal obser- vations). The spawn mass and development time have been described by GODDARD (1984). He found the spawn mass to be white, in a sausage-shaped cord laid in a dis- orderly spiral of 1-4 turns. The single egg per capsule had a diameter of 62.6-64.0 wm and hatched in 9-11 days (at 12-16°C) into Type-1, eyeless veligers of ‘THOMPSON (1967). The veligers have shell Type 1 of THOMPSON (1961a) and a length of 113.3-116.6 um. The duration of the larval stage is unknown. DISCUSSION Synonyms of Onchidoris muricata Onchidoris muricata from the Atlantic has the same in- ternal and external morphology as the animals from the Pacific that are described as O. varians. I therefore consid- er them synonymous. BERGH (1880) distinguished O. var- zans on the basis of its bluish color as opposed to the light yellowish, white, or yellowish-white colors of O. muricata. Nevertheless, he conceded that a variety of O. varians is yellowish-white or yellowish. The radula formula for O. varians is 30-41 xX 1.1.1.1.1, with 15-20 denticles reach- ing to the end of the hook (Figure 9C). This is within the range for O. muricata from Norway (Table 1, Figure 9A). Live specimens from the eastern Atlantic are reported to reach a length of 17 mm (BEHRENTZ, 1931), whereas the largest live specimen found on the Pacific coast was 12 mm. When Pacific animals from Vancouver Island were compared with Norwegian animals (Table 1, Figure 9B), the following differences were found. Identically sized specimens were alike, but some larger animals had been collected in Norway. These latter specimens bore dispro- portionately large tubercles and a similarly oversized mantle margin. They had been labeled Adalaria loveni S. V. Millen, 1985 Page 89 Table 1 Morphological features of the species of Onchidoris and Diaphorodoris examined. O. muricata O. muricata Location Atlantic Pacific Color white, yellowish white, yellowish Body oval oval Head veliform veliform Foot short short Branchiae disk large disk large separate pits separate pits 8-14 6-18 Tubercles knobbed club knobbed club spicules project spicules project Skin spicules show spicules show Radula 27-44 (1.1.1.1.1) 20-33 (1.1.1.1.1) Denticles 9-16 fine 8-18 fine Vas deferens short short Penis large large Receptaculum buried buried (Alder & Hancock, 1862) on the basis of their external anatomy. However, when compared with equal-sized bona fide A. loveni, it could be seen that the tubercles of the large Norwegian specimens of O. muricata were not as large and were more constricted at their bases. Internally, the single marginal tooth per half row, as opposed to the 8-12 found in A. loveni, provided positive identification of these mislabeled animals. The only unexplained differ- ence noted by Bergh for O. varians is the lack of spicula- tion. This was probably an artifact of preservation. The species Onchidoris hystricina has been the object of confusion on the Pacific coast. BERGH (1878, 1880) ob- tained one specimen that Dall found in Alaska. He sep- arated it from O. muricata by its color (bluish rather than yellowish-white). He separated it from O. varians appar- ently because of slight differences in the nervous system, a thinner belt of denticles on the lip cuticle, and smaller lateral plates (0.075 versus 0.12 mm in height). The den- ticulation on the lateral teeth was weaker, there were few- er denticles (8 versus 20), and the denticles did not extend as far out toward the tip. Bergh did not consider these differences to be great, and he concluded that “... the possibility cannot be denied that further investigations may show both the Pacific ‘species’ to be merely varieties of the old Lamellidoris muricata of the Atlantic.” Onchidoris hystricina is compared with the other two species (Table 1, Figure 9D). The differences noted by Bergh fall within the range of variability of O. muricata. Onchidoris muricata can be an almost translucent bluish color, opaque white, or pale yellow. MEYER (1971) reported teeth with a height of 0.075 mm in O. muricata, the same height that Bergh found in O. hystricina. I found O. muricata from Norway with 9-16 denticulations, usually ending before the tip, but at times continuing to the end. Tooth heights varied from 0.057 to 0.090 mm. Younger teeth had longer, O. hystricina O. varians D. lirulatocauda Pacific Pacific Pacific bluish, yellowish bluish yellowish oval oval elongate veliform veliform knobbed short short elongate disk large disk large disk small separate pits separate pits common pit 12 12-20 4-9 clubbed clubbed cylindrical spicules project R not projecting spicules show no spicules spicules buried 40 (1.1.1.1.1) 30-41 (1.1.1.1.1) 29-33 (2.1.0.1.2) 6-8 fine 15-20 stronger 11-13 very strong short short long large large small ? ? free straighter hooks; older, worn teeth had shorter, blunter, more curved hooks, much as shown by THOMPSON (1958: 51, fig. 2) for Adalaria proxima (Alder & Hancock, 1854). BERGH’s (1879:pl. 68 [figs. 18-23]) drawings of the teeth of O. hystricina are consistent with newer, unworn teeth (Figure 9D). I therefore consider O. hystricina to be a junior synonym of O. muricata. Misidentification of Diaphorodoris lirulatocauda and Onchidoris muricata Confusion has arisen in the literature due to the mis- taken association of the distinctive new species Diaphoro- doris lirulatocauda with the name Onchidoris hystricina. Three factors probably led to this error. Firstly, O. mur- icata had its known range extended to California by Marcus (1961) under the misnomer of O. hystricina. Be- cause later researchers realized two species occurred in California, one was correctly identified as O. muricata; the other (D. lirulatocauda) was given the name O. hystricina, as this was the only other name reported from California for a similar appearing animal. Secondly, both O. hystri- cina and D. lirulatocauda have tooth denticles that do not extend to the end of their strongly hooked laterals. This reinforced the misidentification even though the lateral teeth differ in shape (Figures 9D, E). Thirdly, the tuber- cles of O. hystricina are mistakenly reported by BERGH (1880) as being 1.2 mm high, which is much higher than the 0.51-0.56 mm actually found in O. muricata. Diapho- rodoris lirulatocauda has slightly longer tubercles than O. muricata, and this reinforced its identification with the name O. hystricina. However, the longest tubercles of D. lirulatocauda are only 0.64 mm long, which is not nearly as long as in Bergh’s report. When the features of D. lirulatocauda are compared closely with O. hystricina as described by Bergh, many important differences emerge E Figure 9 Radular teeth. A. Onchidoris muricata (Atlantic) from BERGH, 1880:pl. 11 (fig. 10). B. Onchidoris muricata (Pacific). C. Onchi- doris varians from BERGH, 1880:pl. 11 (figs. 13, 14). D. Onchi- doris hystricina from BERGH, 1878:pl. 68 (figs. 18, 21). E. Dia- phorodoris lirulatocauda. Not drawn to scale. (Table 1). These differences, particularly those of the head shape, gill arrangement, radula, and length of vas defer- ens, show that Bergh’s O. hystricina belongs with the species O. muricata rather than the animal we have been com- monly calling O. hystricina. This latter animal is in fact a new species, which I have described in this paper. Discussion of Diaphorodoris This new species, Diaphorodoris lirulatocauda, has been placed in the genus Diaphorodoris Ireland & O’ Donoghue, 1923, because it has broadly based, triangular, denticulate teeth with no central tooth, an elongate body with a trail- ing keeled tail, a double-knobbed head, and a reproductive system similar to that of the type species D. luteocincta (M. Sars, 1870). Another important feature distinctive to this species is branchiae that are enclosed in a common sheath much as in the cryptobranch dorids, although in Diaphorodoris the branchiae are nonretractable as the sheath does not close over the branchiae. The genus Dia- phorodoris was first created by IREDALE & O’ DONOGHUE (1923) as a genus of Onchidorididae for the species D. luteocincta, although no distinctive characters were given. PORTMANN & SANDMEIER (1960) provide a history of the genus, redescription of the type species, and describe a new species, D. papillata, which varies only in color and tubercle shape from the type. Since then, no new species have been added and the generic status of Diaphorodoris The Veliger, Vol. 28, No. 1 dablerZ Anatomical characters separating Onchidoris and Diaphorodoris. Onchidoris Diaphorodoris Shape oval elongate Head veliform lobiform Tail not extending trailing Branchiae separate pits common sheath enclosing tubercles no tubercles enclosed Radula central present no central or absent Reproductive vagina short vagina elongate bursa stalked bursa sessile receptaculum buried receptaculum free semiserial vaginal is usually ignored. FRANC (1968) considers Diaphorodoris to be a subgenus, although he does not state of which genus. He places it in the family Lamellidoridae A. Pru- vot-Fol, 1954, although the name Onchidorididae Alder & Hancock, 1845, has priority. THOMPSON & BROWN (1976) and THOMPSON (1976) place the species /uteocincta in the genus Onchidoris. It is clear that in spite of the early arguments of PRUVOT-FOL (1932), the completeness of PORTMANN & SANDMEIER’s (1960) description, and its recent use by SCHMEKEL & PORTMANN (1982), the estab- lishment of Diaphorodoris as a distinct genus has not been universally accepted. I believe Diaphorodoris should retain its generic status. In support of this, I have summarized the important differences between the two genera in Ta- ble 2. Comparison of Species in the Genus Diaphorodoris Diaphorodoris lirulatocauda conforms closely to the morphology of D. /uteocincta. It differs externally in hav- ing slimmer rhinophores, longer dorsal tubercles (0.6 ver- sus 0.2 mm) and more branchial gills (7-9 versus 5-7). Diaphorodoris luteocincta normally has a yellow marginal ring and dorsal crimson blotching, but the red color is missing in the variety alba although the yellow ring is present. Diaphorodoris papillata can be distinguished from D. lirulatocauda by its red-colored, soft, inflated tuber- cles, which reach up to 0.8 mm in length. Internally D. lirulatocauda differs from the others by having an extra, reduced, outer external plate giving it the formula 2.1.0.1.2. In addition, the vas deferens of Diaphorodoris lirulatocau- da is longer, with more coils, and the penis is armed with spines. Diaphorodoris mitsui (Baba, 1938) comb. nov. The species Lamellidoris mitsuti (for which BaBA, 1938, created a new subgenus Lamellidorella because it has an S. V. Millen, 1985 Page 91 vS Figure 10 Reproductive systems—female portion. A. Acanthodoris from BERGH, 1880:pl. 13 (fig. 5). B. Actodoris from BERGH, 1880:pl. 6 (figs. 18, 19). C. Calyctdoris from ROGINSKAYA, 1972:pl. 1 (fig. 16). D. Diaphorodoris. E. Onchidoris. Not drawn to scale. armed lip cuticle) must also be compared with the genus Diaphorodoris. The teeth have the same triangular, den- ticulate shape and lack a central plate. The body is also elongate with cylindrical tubercles and a trailing tail. Most importantly the branchiae are enclosed by a common cav- ity into which, according to BABA (1938, 1949), they can retract. This species has a rounded, double-lobed head like other Diaphorodoris species. The orange-yellow marginal ring is reminiscent of D. luteocincta var. alba. The small scales on the lip disk, which BaBa regarded as distinctive (1938:131 [fig. 1B]), are similar to the small papillae on the lip cuticle of D. rulatocauda. Unfortunately, the re- productive system of this species is not known, but exter- nally, and according to its radular morphology, it appears to be conspecific with Diaphorodoris. I therefore designate it Diaphorodoris mitsuii (Baba, 1938) comb. nov. Discussion of Generic Relationships in the Family Onchidorididae One of the major characteristics separating Dzaphoro- doris and Onchidoris is that the branchiae do not possess separate pits, but emerge from a common cavity. This is similar to the branchial arrangement of cryptobranch do- rids, although the gill pocket does not close over to protect the gills and thus the gill system can still be classified as nonretractile. Of the genera in the family Onchidorididae, Aciodoris, Adalaria, Arctadalaria, Doridunculus, Onchidoris, and Prodoridunculus have nonretractile branchiae that contract and are arranged in separate cavities. The genus Acanthodoris has gills that connect at their bases. How- ever, the branchial margin of the acanthodorids indents between each gill and the gills usually enclose a tuber- culated portion of the notum. Calycidoris and Diaphoro- doris both have a single cavity containing gills that join at their bases. ROGINSKAYA (1972) proposed a new family, Calycidorididae, for the monotypic genus Calycidoris be- cause its gills retract into a common sheath. However, in specimens that I examined it appears that the gill margin does not close over the gills and, thus, even when the gills are maximally contracted the system can still be consid- ered nonretractile. The gills of Diaphorodoris mitsuii are probably similarly contracted, rather than retracted as claimed by Basa (1938, 1949). This branchial arrange- ment is very close to that of true cryptobranchs (which is termed retractile) differing only in that the gill margin does not close itself over the gills. Diaphorodoris can be distinguished from Calycidoris by its broad, triangular-based, denticulate lateral teeth and its elongate body shape with a trailing, keeled tail. It is separated from other genera in the family Onchidorididae because its branchial gills are arranged in a common sheath. Diaphorodoris is more closely allied to the genera Calycidoris, Acanthodoris, and Aciodoris on the basis of the vaginal arrangement of the uterine duct than to Onchi- doris, which has a semiserial arrangement (Figure 10). As in Aciodoris, the penis can be armed with spines. The radular teeth are most similar to some of the species in the genus Onchidoris. The elongate body shape, knobbed head, trailing keeled tail, common branchial pit, and vag- inal arrangement of the copulatory bursa are all charac- teristics that validate the generic separation of Onchidoris and Diaphorodoris. ACKNOWLEDGMENTS I would like to thank the following people and institutions for providing specimens: Dave Behrens, the British Co- lumbia Provincial Museum, the California Academy of Sciences, Jeff Goddard, Terry Gosliner, Will Jaeckle, the U.S. National Museum, the University of Alaska Mu- seum—Fairbanks, and the Zoologisk Museum—Bergen, Page 92 Norway. Thanks also to the submersible Pisces IV and her crew for its use in obtaining depth distributions. Ron Long provided photographs for Figures 1A, E. Annette DeHault helped with German translations. Sven Donald- son and Terry Gosliner gave me many helpful comments on the manuscript. This research was jointly funded by the Department of Zoology, U.B.C., and NSERC grant # 6759061. LITERATURE CITED ABBoTT, R. T. 1974. American seashells, the marine Mollusca of the Atlantic and Pacific coasts of North America. 2nd ed. Van Nostrand Reinhold: New York. 633 pp. ALDER, J. & A. Hancock. 1855. Monograph of the British Nudibranchiate Mollusca 7. Ray Society: London. 54 pp. Basa, K. 1938. Three new nudibranchs from Izu, Middle Japan. Annot. Zool. Japon. 17(2):130-133. Bapa, K. 1949. Opisthobranchia of Sagami Bay. Iwanami Shoten: Tokyo. 194 pp. BEEMAN, R. D. & G. C. WILLIAMS. 1980. Opisthobranchia and Pulmonata. Pp. 308-354. Jn: R. H. Morris, D. P. Ab- bott & E. C. Haderlie (eds.), Intertidal invertebrates of Cal- ifornia. Stanford Univ. Press: Stanford, Calif. BEHRENS, D. W. 1980. Pacific coast nudibranchs: a guide to the opisthobranchs of the northeastern Pacific. Sea Chal- lengers Inc.: Los Osos, California. 112 pp. BEHRENTZ, A. 1931. Trekk av Lamellidoris muricata’s biologi og av dens generasjonsoraganers bygning. Nyt mag. Naturv. Oslo 70:1-26. BERGH, L. S. R. 1878. Malacologische Untersuchungen. Pp. 603-645, pl. 66-68. In: C. G. Semper, Reisen im Archipel der Philippinen von Dr. Carl Gottfried Semper. Zweiter Theil. Wissenschaftliche Resultate. 2,2(14). BerGuH, L. S. R. 1879. Gattungen Nordischer Doriden. Ar- chiv. fiir Natur. 45(1):340-3069, pl. 19. BERGH, L. S. R. 1880a. On the nudibranchiate gasteropod Mol- lusca of the North Pacific Ocean, with special reference to those of Alaska. Sci. Results Explor. Alaska 1(6) 2:189- 276, pl. 1-16. BERGH L.S.R. 1880b. On the nudibranchiate gasteropod Mol- lusca of the North Pacific Ocean, with special reference to those of Alaska. 2. Proc. Acad. Natur. Sci. Philadelphia. 32: 40-127, pl. 9-16. BerGH, L. R. S. 1890. Malacologische Untersuchungen. Die nudibranchien des “Sunda-Meeres.” Pp. 873-991, pl. 85- 89. In: C. G. Semper, Reisen im Archipel der Philippinen von Dr. Carl Gottfried Semper. Zweiter Theil. Wissen- schaftliche Resultate. 2,3(17). BeRGH, L. R. S. 1892. Malacologische Untersuchungen. Sys- tem der nudibranchiaten Gastropoden. Pp. 995-1165 (3- 173). In: C. G. Semper, Reisen im Archipel der Philippinen von Dr. Carl Gottfried Semper. Zweiter Theil. Wissen- schaftliche Resultate. 2,3(18). FRANC, A. 1968. Sous-Classe des Opisthobranches. Pp. 608- 893. In: P. Grassé (ed.), Traité de Zoologie 5:3. Mollusques Gasteropodes et Scaphopodes. Masson et Cie: Paris. GODDARD, J. 1984. The opisthobranchs of Cape Arago, Or- egon, with notes on their biology and a summary of benthic opisthobranchs known from Oregon. Veliger 27(2):143-163. Hurst, A. 1967. The egg masses and veligers of thirty north- east Pacific opisthobranchs. Veliger 9(3):255-288. IREDALE, T. & C. H. O'DONOGHUE. 1923. List of British The Veliger, Vol. 28, No. 1 nudibranchiate Mollusca. Proc. Malacol. Soc. Lond. 15(4- 5):195-233. JAECKLE, W. B. 1984. Opisthobranch mollusks of Humboldt County, California. Veliger 26(3):207-213. Kress, A. 1981. A scanning electron microscope study of no- tum structures in some dorid nudibranchs (Gastropoda: Op- isthobranchia). J. Mar. Biol. Assoc. U.K. 61:177-191. Marcus, E. 1961. Opisthobranch mollusks from California. Veliger 3(Suppl.):1-85. McDona_p, G. R. 1983. A review of the nudibranchs of the California coast. Malacologia 24(1-2):114-276. McDonaLp, G. R. & J. W. NYBAKKEN. 1978. Additional notes on the food of some California nudibranchs with a summary of known food habits of California species. Veliger 21(1):110-119. McDonaLp, G. R. & J. W. NYBAKKEN. 1981. Guide to the nudibranchs of California. American Malacologists Inc.: Melbourne, Florida. 72 pp. Meyer, K. B. 1971. Distribution and zoogeography of four- teen species of nudibranchs of northern New England and Nova Scotia. Veliger 14(2):137-152. MILLER, M.C. 1958. Studies on the nudibranchiate Mollusca of the Isle of Man. Doctoral Thesis, University of Liverpool [not seen]. MILLER, M. C. 1962. Annual cycles of some Manx nudi- branchs, with a discussion of the problem of migration. J. Anim. Ecol. 31(3):545-569. NYBAKKEN, J. W. & G. R. McDonaLp. 1981. Feeding mech- anisms of west American nudibranchs feeding on Bryozoa, Cnidaria and Ascidiacea, with special respect to the radula. Malacologia 20(2):439-449. PORTMANN, A. & E. SANDMEIER. 1960. Zur kenntnis von Dia- phorodoris (Gastr., Nudibranchia) und ihrer mediterranen formen. Verh. Nat. Gesel. 71:174-183. PRuvoT-FoL, A. 1932. Notes de systématique sur les Opis- thobranches. Bull. Mus. Nation. Hist. Nat. 2(4):322-331. Roainskaya, I. S. 1972. Calycidoris guenther: (Gastropoda, Nudibranchia). Taxonomy and distribution. Akad. Nauk S.S.S.R. Zool. Zh. 51(6):913-918. SCHMEKEL, L. & A. PORTMANN. 1982. Opisthobranchia des Mittelmeeres: Nudibranchia und Saccoglossa. Springer- Verlag: New York. 410 pp. THOMPSON, T. E. 1958. Observations on the radula of Adalaria proxima (A. & H.) (Gastropoda Opisthobranchia). Proc. Malacol. Soc. Lond. 33:49-56. TuHompson, T. E. 1961a. The importance of the larval shell in the classification of the Sacoglossa and the Acoela (Gas- tropoda Opisthobranchia). Proc. Malacol. Soc. Lond. 34(5): 233-238. THompson, T. E. 1961b. Observations on the life history of the nudibranch Onchidoris muricata (Muller). Proc. Mala- col. Soc. Lond. 34(5):239-242. TuHompson, T. E. 1967. Direct development in a nudibranch, Cadlina laevis, with a discussion of developmental processes in Opisthobranchia. J. Mar. Biol. Assoc. U.K. 47(1):1-22. TuHompson, T. E. 1976. Nudibranchs. T. F. H. Publications Inc.: Neptune, New Jersey. 96 pp. TuHompson, T. E. & G. H. Brown. 1976. British opistho- branch molluscs. Mollusca: Gastropoda. Synopsis of the British fauna (new series) 8. Linnean Soc. London. Aca- demic Press: New York. 203 pp. Topp, C. D. 1978a. Gonad development of Onchidoris muri- cata (Miiller) in relation to size, age and spawning (Gas- tropoda: Opisthobranchia). J. Moll. Stud. 44(2):190-199. Topp, C.D. 1978b. Changes in spatial pattern of an intertidal S. V. Millen, 1985 population of the nudibranch mollusc Onchidoris muricata in relation to life-cycle, mortality and environmental het- erogeneity. J. Anim. Ecol. 47:189-203. Topp, C.D. 1979a. The annual cycles of two species of Onchi- doris (Opisthobranchia: Nudibranchia). Pp. 65-72. In: E. Naylor & R. G. Hartnoll (eds.), Cyclic phenomena in ma- rine plants & animals. Proc. 13th Eur. Mar. Biol. Symp. Pergamon Press: Toronto. Page 93 Topp, C. D. 1979b. Reproductive energetics of two species of dorid nudibranchs with planktotrophic and lecithotrophic larval strategies. Mar. Biol. 53:57-68. Topp, C. D. 1981. The ecology of nudibranch molluscs. Oceanogr. Mar. Biol. Ann. Rev. 19:141-234. The Veliger 28(1):94-98 (July 1, 1985) THE VELIGER © CMS, Inc., 1985 A New Subgenus of Helminthoglypta (Gastropoda: Pulmonata: Helminthoglyptidae) WALTER B. MILLER Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, Arizona 85721 Abstract. A new subgenus Rothelix is described for the land snail genus Helminthoglypta in southern California. It is distinguished from the nominate subgenus by its long, sausage-shaped, lower sac of the penis and by its atrial sac in which the vagina enters just below the dart sac instead of near the genital pore. INTRODUCTION THE SPECIOSE GENUS Helminthoglypta comprises over 100 species and subspecies of western North American land snails. Classification within the genus has consisted main- ly of allocating species into series whose characteristics are primarily based on shell sculpture, size, shape, and color. Subgeneric classification has been limited to the desig- nation of two subgenera, namely the nominate subgenus, whose type species is Helminthoglypta tudiculata (Binney, 1843), and Charodotes Pilsbry, 1939, whose type species is H. trasku (Newcomb, 1861). Helminthoglypta s.s. is characterized by a double-walled penis with an outer tube (i.e., an eversible inner tube with an outer tube) and Cha- rodotes, as reported by PILsBRyY, 1939, by a single, thick, muscular tube. PILSBRY (1939:68) also prepared a key in which Charodotes was listed as having a large dart sac with a short common duct of the mucus glands, while Helminthoglypta s.s. was divided into two main groups of species, one group having the dart sac and common duct as in Charodotes and the other having the dart sac small and much shorter than the common duct. Between 1956 and 1964, the late Wendell O. Gregg and I undertook to conduct a careful examination of the anatomy of nearly every known (and many yet unde- scribed) species and subspecies of Helminthoglypta in or- der to find characteristics that could be used to establish additional subgeneric categories. Publication of our find- ings was delayed by Gregg’s declining health and eventual death. Foremost among our determinations was the fact that all species examined, including H. trasku, had a dou- ble-tubed penis, of varying length, thereby synonymizing Charodotes with Helminthoglypta s.s.; this information has been reported by MILLER (1981), and the name Charo- dotes, while available, is invalid because it is a junior syn- onym. Furthermore, we found that the comparative mea- surements of dart sac and common duct of the mucus glands were generally useless diagnostically, because we could find specimens with a large sac and short duct in the same populations with specimens with a small sac and long duct. On the other hand, the studies revealed that the lower sac of the penis, which PILsBRyY (1939:69) refers to as ‘‘a short neck with a simple wall,” differed considerably and consistently in one group of species which includes Hi. lowe: (Bartsch, 1918), H. cuyamacensis cuyamacensis (Bartsch, 1916), and several additional species yet to be described. The reproductive anatomy of this group of species differs so markedly from that of all other species of the genus, including the type species, H. tudiculata, that it warrants classification in a separate, new subgenus, de- scribed below. It is imperative, in comparing reproductive system anatomies, that the terms used in describing the accessory organs be unequivocally defined. BERRY (1953) described and compared the reproductive systems of H. lowe: and H. thermimontis Berry, 1953, in minute detail. In H. lowei, however, he considered the epiphallus to in- clude the double-tubed, eversible portion, and he then identified the lower sac of the penis, which is long and saccular in this species, as the entire penis. In this report, in order to be consistent with Pilsbry’s earlier definitions, the epiphallus (Figure 1, ep) is considered to be the mus- cular, single-tubed duct which begins at the junction of the vas deferens and the epiphallic caecum, and ends at the beginning of the double-tubed, eversible duct. The penis, then, consists of an upper part (Figure 3, upe) W. B. Miller, 1985 . Page 95 Depth, very generalized; the depth indicated is considered the ‘optimum,’ presently available. > or where the species is most abundant, based on data * Special habitat: the special habitat attributed to the taxa shown does not mean that the species are limited to that situation; no special habitat is known for species of some taxa. > Salinity: very generalized. is present in C. gigas); the accessory heart is a thick, mus- cular, cylindrical tube without tributaries. Four species are recognized, each with a long list of junior synonyms. Crassostrea angulata (Lamarck, 1819) occurs in the eastern Atlantic, from the equator northward to the Mediterranean and Atlantic coast of the Iberian Peninsula; it has been introduced into southwestern France. Crassostrea virginica (Gmelin, 1791) is found in the west- ern Atlantic, from Brazil northward through the Carib- bean and Gulf of Mexico, including the Antilles, to the Page 154 Family Subfamily S a | 5 = 2 Hyotissini a £3 oe) mls 1S} oO] »> Ay % Lophini & iS a. | Myrakeenini Ostreini o 3 & v/s S| 2 isi. Cryptostreini 5 YP’ e) Undulostreini Pustulostreini vo a & : ates 2 Striostreini Je] n 3 n a S Crassostreini be present in some Lophinae. Neopycnodontini Neopycnodonte Table 2 Se aussie 8 § §| 3 OB ols e/g N Nn| Qa Q'S. ‘a alo of 2 Genus or ats & SSsye subgenus IPs Can (ee Hyotissa x BOE) 5 Parahyotissa s.s. x FFF P. (Pliohyotissa) x RR P. (Numismoida) R xX Lopha Alectryonella Dendostrea Myrakeena xX Anomuostrea xX Ostreola xX} xX Ostrea s.s. xX 4 O. (Eostrea) xX xX Nanostrea X| xX Planostrea xX xX Cryptostrea Teskeyostrea Booneostrea Pustulostrea xX Saccostrea™ Striostrea s.s. S. (Parastriostrea) xX y+ | Erect lamellae* ? xX ? ieee ol se Jes) es The Veliger, Vol. 28, No. 2 Comparison of shell traits of genera and subgenera of living oysters. No chalk deposits° Non-vesic. chalk dep.° Musc. scar reniform® Lophine chomata? Ostreine chomata? Chomata absent always’ Chomata may disappear in larger shells’ Muse. scar distinctive Surface pustulose" Organic deposits® ~ XM =| Vesicular chalk dep.® Appressed lamellae* <> | No lamellae‘* » | Commissural shelf’ <> > > «| Musc. scar circuJar® me ae x > > | Vermicular chomata’ *% *% a ~ Ae | MMM *% x x x x x x x Xx x x x x x x x x x x x xX|X xX x x x xX xX x x x x x x x x | x x x x x x x x x x x x x x x x x x x x [owen fe Pe = F = frequently; R = rarely; X = generally as indicated. ' Size: indicates the size usually attained. * Plications: these vary in shape, spacing and prominence in different species. The lamellar striations of Striostrea and Teskeyostrea are not considered to be plications. > Hyote spines: these vary in form and occurrence in different species. * Lamellae of the right valve only. * Organic deposits: may be absent in some specimens of species in which they generally occur. ° Chalk deposits: variable within most species. ’ Commissural shelf: considered to be present if the inner margin is usually sharply defined. * Shape of the adductor muscle scar. * Chomata: in Neopycnodonte the chomata are intermediate between the two types indicated; both lophine and ostreine chomata may '° Muscle scar distinctively colored from the rest of the shell’s interior; the color may not be limited to the area within the scar’s margin. '' Pustules of the outer surface of the shell vary in size between species, and in position and prominence within a species. '? Undulostrea: the undulations of the shell margin are very different from the plications of other oysters. '3 Saccostrea: hyote spines are present in juveniles only; chomata may be absent in some ecomorphs. Table 3 Comparison of characters of the flesh and reproductive habits of genera and subgenera of living oysters. Genus and Tribe subgenus Hyotissa aay SS. Hyotissini P. ( Pliohyotissa) P. a Neopycnodontini Neop | Neopycnodonte | te Lopha Lophini Alectryonella® Dendostrea see Myrakeena Ostreola Ostrea s.s. Ostreini O. (Eostrea) Nanostrea Planostrea Subfamily .-kid.-rec. complex gryphaeid? Ht.-kid.-rec. complex ostreid? Anal appen. a thin circular collar® Accessory heart lacks tributaries® ~~ XM | Labial palp fusion long* Accessory heart has tributaries® Anal appen. digitiform$ Anal appen. cardiform? R promyal passage narrow? Labial palp fusion short* Mantle lobes thick’ L promyal passage open? promyal passages closed? No fusion of palp* * x XM | Anal appen. a bulbous collar® No anal appendage*® Broods larvae’ ~ ~*~ | «xX x MX |R promyal passage wide? we w ~ mK XK eV XK ~ ~*~ & | No accessory heart® x > > «| Mantle lobes thin! ~ | Mm mM | Ht Gryphaeidae Pycnodonteinae aw *% amex | XM ~ ~ xm xX ~ ~ xx uK x ~ | x Cryptostrea Cryptostreini Teskeyostrea Booneostrea Pustulostreini Pustulostrea Saccostrea Striostrea s.s. S. (Parastriostrea) Mmm | MMMM OM mmm | MMM MM axa MM Ostreidae o iss} g aS i a cS) = o 3S de o =I JB n Oo Crassostreinae ‘Mantle: thickness may be determined in part by age and nutritional state. ? Promyal passage: the left one is closed in all species but Hyotissa hyotis. > Heart-kidney-rectum complex: in the gryphaeid condition the heart auricles are outpocketed, fused to the pericardium wall by their ventral surfaces, with very abundant brown granulocytes internally; the rectum distinctly penetrates the heart ventricle; the kidney is large, covering the dorsal curvature of the adductor muscle (which is circular in profile), with posterior caecum, and little outpocketing; renopores are at the dorsal ends of the anterior horns. In the ostreid condition, the auricles are not outpocketed, nor fused ventrally to the pericardium, and the amount of brown granulocytes is variable, but usually small; the rectum passes immediately behind the ventricle, the kidney is much reduced, with no large sac over the dorsal surface of the adductor muscle (which is reniform in profile), not posterior caecum, and the whole is much outpocketed as short, dendritic alveoli, with the renopores at the ventral ends of the anterior horns. * Outer labial palp: the extent of fusion in the Ostreidae presents almost a continuous variable, but is here reduced to three alternatives. > Anal appendage: this was difficult to determine in the limited amount of material available for some species (see text). ° Accessory heart: this was often difficult to determine precisely in the limited amount of preserved material available for some species. 7 Reproductive habits: data are partly from the literature. ® Data on the flesh of Alectryonella are entirely from ToRIGOE (1981). ° The flesh of Anomiostrea has not been available for study. 10 Crassostrea: the labial palp is unfused, and there is no anal appendage in C. virginica, the genotype; but there is slight fusion of the palp, and a distinct circular flange on the anus in C. sivas. Page 156 St. Lawrence River estuary in eastern Canada (O. rhizo- phorae Guilding, 1827, is a junior synonym); in the east- ern Pacific, C. columbiensis (Hanley, 1846) occurs from Ecuador northward to the Gulf of California (O. corte- ziensis Hertlein, 1951, is a junior synonym). Crassostrea gigas (Thunberg, 1793) occurs in the Indo-West Pacific from Pakistan to Japan and Korea, and in the Philippine Islands, Borneo, and Sumatra, but possibly not in the rest of Indonesia, or southward. SUMMARY TABLES Some of the more important characters which vary among Recent oysters are summarized in the preceding three comparative tables (pp. 153-155). The data refer to all known species of each genus and subgenus, not merely to the type species of the taxa cited, and they include vari- ation found within the species. Therefore some characters which would seem to be mutually exclusive, as type of chomata in Neopycnodonte, Alectryonella and Dendostrea, and shape of the anal appendage and fusion of the labial palps in Crassostrea, appear in more than one column. For more precise information one must refer to the text. LITERATURE CITED ADAMS, A. & L. REEVE. 1850. Zoology of the voyage of H.M.5S. Samarang. Reeve and Benham: London. Mollusca. (1848- 1850) Born, I. von. 1778. Index rerum naturalium musei Caesarei Vindobonensis, Pars I, Testacea. Verzeichniss der nattir- lichen Setenheiten des K.K. naturalien Kabinet zu Wien, Erster Theil, Schalthiere. Officina Krausiana: Vienna. 458 plus several unnumbered pages. CARPENTER, P. P. 1857. Catalogue of the collection of Ma- zatlan shells in the British Museum collected by Frederick Reigen. London: British Museum. Pp. i-iv + 552. Reprint- ed, Paleontological Research Institute, 1957. CARPENTER, P. P. 1864. Supplementary report on the present state of our knowledge with regard to the Mollusca of the West Coast of North America. Rept. British Assoc. Adv. Sci. for 1863:517-686. CARRIKER, M. R., R. E. PALMER & R. S. PREZANT. 1980. Functional ultramorphology of the dissoconch valves of the oyster Crassostrea virginica. National Shellfisheries Associ- ation, Proc. 70(2):139-183. CHANLEY, P. & P. DINAMANI. 1980. Comparative descriptions of some oyster larvae from New Zealand and Chile, and a description of a new genus of oysters, Tiostrea. New Zealand J. Mar. Freshw. Res. 14(2):103-120. CHILDREN, J. G. 1822-1824. Lamarck’s genera of shells, translated from the French, with plates from original draw- ings by Miss Anna Children. Quarterly J. Sci., Lit. Arts, 14:64-86 (Oct. 1822); 14:298-322 (Jan. 1823); 15:23-52, 2 pls. (Apr. 1823); 15:216-258, 2 pls. (July 1823); 16:49- 79, 2 pls. (Oct. 1823); 16:241-264 (Jan. 1824). Designation of the type of Ostrea Linné is 1823, 15:44, pl. 2, fig. 94. CrosszE, H. 1869. Diagnoses molluscorum novorum. J. de Conchyl, Ser. 3, 9:183-188 (and pl. 2 of vol. 10, 1870). Crowson, R. A. 1970. Classification and biology. Heinemann Educational Books, Ltd.: London. 350 pp. DaLL, W. H. 1898. Contributions to the Tertiary fauna of The Veliger, Vol. 28, No. 2 Florida with especial reference to the Silex Beds of Tampa and the Pliocene Beds of the Caloosahatchie River including in many cases a complete revision of the generic groups treated of and their American Tertiary species. Wagner Free Inst. Sci., Philadelphia, Trans., 3(4):viii + 571-947, pls. 23- 35. DALL, W. H. 1914. Notes on West American oysters. Nautilus 28:1-3. DALL, W. H., P. BARTSCH & H. A. REHDER. 1938. A manual of Recent and fossil marine pelecypod mollusks of the Hawaiian Islands. Bernice P. Bishop Mus. Bull. 153:233 pp., 57 pls. DESHAYES, G. P. 1863. Catalogue des mollusques de I’Ile de la Réunion (Bourbon). Pp. 1-144, pls. 28-41. In: L. Mail- lard, Notes sur |’Ile de la Réunion. DUJARDIN, F. 1835. Mémoire sur les couches du sol en Tou- raine et descriptions des coquilles de la craie et des faluns. Soc. Geol. France, Mém. 2(2):211-311, pls. 15-20. FINLAY, H. J. 1928. The Recent mollusca of the Chatham Islands. New Zealand Inst., Trans. and Proc. 59(2):232- 286, pls. 38-43. Forbes, M. L. 1964. Distribution of the commensal oyster, Ostrea permollis, and its host sponge. Bull. Mar. Sci. Gulf Caribb. (Univ. Miami) 14(3):453-464. Forses, M. L. 1966. Life cycle of Ostrea permollis and its relationship to the host sponge, Stellata grubi. Bull. Mar. Sci. Gulf Caribb. (Univ. Miami) 16(2):273-301. GaLtTsorF, P. S. 1964. The American oyster Crassostrea vir- ginica Gmelin. U.S. Bureau of Commercial Fisheries, Fish. Bull. 64:1-480, 400 figs. GLENN, L. C. 1904. Pelecypoda. Vol. 1:274-401. In: Mary- land Geological Survey, Miocene. Text, pp. i-xix, 1-543; atlas of pls., 10-135. GMELIN, J. F. 1791. Caroli a Linné Systema naturae per regna tria naturae. Ed. 13, vol. 1, pt. 6, pp. 3021-3910. G. E. Beer: Leipzig. GouLp, A. A. 1861. Descriptions of shells collected in the North Pacific Exploring Expedition under Captains Ring- gold and Rodgers. Boston Soc. Natur. Hist., Proc. 8:33-40. Gray, J. E. 1825. A list and description of some species of shells not taken notice of by Lamarck. Ann. Philos. 25:134- 140, 407-415 (sometimes cited as Ser. 2, vol. 9). GROBBEN, C. 1888. Die Pericardialdrusse der Lamellibran- chiaten. Ein Beitrag zur Kenntniss der Anatomie dieser Molluskenclasse. Arbeit des Zoolog. Institutes zu Wien 7(3): 355-444, 6 pls. GUILDING, L. 1828. Observations on the zoology of the Carib- bean Islands. Zool. Journal Vol. 3, no. 12, art. 61, pp. 527- 544. Hase, T. 1957. Descriptions of four new bivalves from Japan. Venus (Fukuyama) 19(3/4):177-183. HaseE, T. & S. KosuGe. 1966. New genera and species of the tropical and subtropical Pacific molluscs. Venus 24(2):312- 341, pl. 29. HANLEy, S. C. T. 1846. A description of new species of Os- treae, in the collection of H. Cuming, Esq. Zool. Soc. Lon- don, Proc., for 1845:105-107. Harry, H. W. 1968. An alternate view on the phylogeny of the Mollusca. Proceedings of the Symposium on Mollusca, Pt. 1. Marine Biol. Assoc. of India, Ramanathapuram Dis- trict, Madras, India. Pp. 170-187. Harry, H.W. 1971. Reconciling biological nomenclature and the phylogeny of organisms. The Echo (Abstracts and Pro- ceedings of the Third Annual Meeting of the Western So- ciety of Malacologists) 3:41-53. H. W. Harry, 1985 Harry, H. W. 1981. Nominal species of living oysters pro- posed during the last fifty years. Veliger 24(1):39-45. Harry, H.W. 1981a. Newly discovered anatomical characters useful in classifying oysters (Ostreacea, Gryphaeidae and Ostreidae). Amer. Malacol. Bull. for 1981:34. Harry, H. W. 1983. Homology of the five types of marginal denticles (chomata) of living oysters (Gryphaeidae and Os- treidae). Amer. Malacol. Bull. 1:90. Harry, H. W. & D. T. DocKERY. 1983. Notes on the Lower Oligocene oysters of Mississippi. Mississippi Geology 4(2): 7-14. HERRMANNSEN, A. N. 1846-1852. Incicis generum malaco- zoorum primordia. 2 vols. Theodor Fischer: Cassel. HERTLEIN, L. G. 1951. Descriptions of two new species of marine pelecypods from West Mexico. So. Calif. Acad. Sci., Bull. 50(2):68-75, pls. 24-26. Hutton, F. W. 1873. Catalogue of the Tertiary Mollusca and Echinodermata of New Zealand, in the collection of the Colonial Museum. New Zealand Colonial Mus. and Geol. Surv. Dept., xvi + 48 pp. (Wellington). IHERING, H. von. 1907. Les mollusques fossiles du Tertiaire et du Crétacée supérieur de l’Argentine. Museo. Nac. Bue- nos Aires, Ann., ser. 3, 7:xiii + 611 pp., 16 figs., 8 pls. TREDALE, T. 1939. Mollusca, Pt. I. Brit. Mus. Nat. Hist., Rec. 19:267-340, pls. 20-24. KEEN, A. M. 1971. Sea shells of tropical West America. 2nd ed. Stanford Univ. Press: Stanford, Calif. 1064 pp. LAMARCK, J. B. P. DE M. 1802-1809. Sur les fossiles des environs de Paris, comprenant la determinations des espéces qui appartiennent aux animaux marins sans vertébres, et dont da plupart sont figures dans la collection des velins du Museum. Annales du Museum National d’Histoire Natu- relle (Paris). Various places in Vols. 1-14; reprinted, 1978, by the Paleontological Research Institution, Ithaca, New York. LaMaRCK, J. B. P. DEM. 1804. Une nouvelle espéce de Tri- gonie, et sur une nouvelle espéce d’huitre, découvertes dans le voyage du capitaine Baudin. Annales du Museum Na- tional d’Histoire Naturelle (Paris) 4:351-359, pl. 67. Lamarck, J. B. P. DE M. 1815-1822. Histoire naturelle des animaux sans vertébres. Paris. 7 vols. (Vol. 6, 232 pp., ap- peared 1819; oysters are in part 1 of it, pp. 195-220.) Lamy, E. 1929-1930. Revision des Ostrea vivants du Museum National d’Histoire Naturelle de Paris. J. de Conchyl. 73 (ser. 4, vol. 27), No. 1 (30 April 1929):1-46, 3 figs.; No. 2 (20 July 1929):71-108; No. 3 (30 October 1929):133-168; No. 4 (28 February 1930):233-257, pl. 1. LINNE, C. 1758. Systema naturae per tria regna naturae. Ed. 10, vol. 1. Stockholm. 823 pp. LIsCHKE, C. E. 1869-1874. Japanische Meeres-Conchylien, mit besonderer Riicksicht aus die geographische Verbrei- tung derselben. Novitates Conchologicae. W. Dunker (ed.), Suppl. 4, 3 pts. in 1. McLean, R. A. 1941. The oysters of the western Atlantic. Notulae Naturae, (Acad. Natur. Sci. Philadelphia) No. 67: 1-14, 4 pls. MELVILL, J. C. 1898. Further investigations into the mollus- can fauna of the Arabian Sea, Persian Gulf and Gulf of Oman, with descriptions of forty species. Memoirs and Pro- ceedings of the Manchester Literary and Philosophical So- ciety 42(2):1-40 (reprint), 2 pls. MonTeErosaTo, T. A., MARCHESE DI. 1884. Nomenclatura generica e specifica di algune conchiglie mediterranee. Virzi: Palermo. 152 pp. Page 157 O.tsson, A. A. 1951. New Floridian species of Ostrea and Vermicularia. Nautilus 65(1):6-8, pl. 1. OrsIGNY, A. D’. 1835-1847. Voyage dans l’Amérique Méri- dionale ... executé pendant les années 1826-1833. 7 vols. of text, 2 vols. of atlas. Paris (Mollusques, 1847, in vol. 5, pt. 3, 83 pls.). PAYRAUDEAU, B. C. 1826. Catalogue déscriptif et methodique des annelides et de mollusques de I’Isle de Corse. Paris. 218 pp., 8 pls. Puiippl, R. A. 1845-1847. Abbildungen und Beschreibung neuer oder wenig bekannter Conchylien. Vol. 2 (in parts issued at intervals, pages and plates not consecutively num- bered). Theodore Fischer: Cassel. (The part on Ostrea, pp. 81-82, is dated Feb. 1846.) Piussry, H. A. 1904. New Japanese marine Mollusca: Pele- cypoda. Proc. Acad. Natur. Sci. Philadelphia 56:550-560, pl. 40. Poul, J. X. 1791-1827. Testacea utriusque Siliciae eorumque historia et anatomie tabulis aeneis illustrata. 3 vols. Parma, Italy. RAFINESQUE, C. S. 1815. Analyse de la nature ou tableau de Punivers et des corps organisés. Palermo. 224 pp. RANSON, G. 1967. Les espéces d’huitres vivant actuellement dans le monde définies par leurs coquilles larvaires ou pro- dissoconques. Revue des Travaux de l'Institut des Peches Maritimes, Paris 31 (part 2, June):127-192, 25 figs.; (part 3, September):205-247, figs. 26-55. ROCHEBRUNE, A.-T. DE. 1895. Diagnoses du mollusques nou- veaux, provenant du voyage de M. Diguet en Basse-Cali- fornie. Bull. Mus. d’Hist. Nat. (Paris) 1:239-243. ROpDING, P. F. 1798. Museum Boltenianum sive catalogus ci- meliorum. Pars secunda continens conchylia. C. J. Trapii: Hamburg. viii + 199 pp. Sacco, F. 1897. Pelecypoda (Ostreidae, Anomiidae e Dimyi- dae) of L. Bellardi & Federico Sacco, 1872-1904, I mol- luschi dei terreni Terziarii de Piemonte e della Liguria (30 pts. separately paged, pt. 23, 66 pp., 11 pls.). Carlo Clausen: Torino. Say, T. 1834. American conchology, or descriptions of the shells of North America illustrated by coloured figures from original drawings executed from nature. Vol. 1, No. 6. School Press: New Harmony, Indiana. 42 pages (unnumbered), pls. 51-60. ScHENCK, E. T. & J. H. McMasrers. 1948. Procedure in taxonomy (revised ed.). Stanford Univ. Press: Stanford, Cal- if. 93 pp. Sowerby, G., JR. 1870-1871. Monograph of the genus Os- traea. In: L. Reeve, 1843-1878, Conchologia Iconica; or il- lustrations of the shells of molluscous animals (20 vols.) vol. 18, 33 pls. and index (2 pp.) (Oct. 1870-Nov. 1871) L. Reeve & Co.: London. STENZEL, H. B. 1959. Cretaceous oysters of southwestern North America. Congr. Geol. Internac., XXa session, Mexico City, 1956. El sistema Cretacico. 1:15-37, 19 figs. STENZEL, H. B. 1971. Oysters. Pp. i-iv, N953-N1224, 153 figs. In: R. C. Moore (ed.), Treatise on invertebrate paleon- tology. Part N, vol. 3, Mollusca 6, Bivalvia. Geol. Soc. America. SUTTER, H. 1917. Descriptions of new Tertiary Mollusca oc- curring in New Zealand, accompanied by a few notes on necessary changes in nomenclature. Pt. 1. New Zealand Geol. Survey Paleont. Bull. 5:vii + 93 pp., 13 pls. SwaInson, W. 1835. The elements of modern conchology brief- ly and plainly stated, for the use of students and travelers. London. viii + 62 pp. Page 158 THIELE, J. 1889. Die abdominalen Sinnesorgane der Lamel- libranchier. Zeitschrift fiir Wissenschaftliche Zoologie 48(1): 47-59, pl. 4. THUNBERG, C. P. 1793. Techning och beskrifning pa en stor Ostronsort ifran Japan. K. Svenska Vetensk. Akad., Hand- lingar. 14:140-142. TorIGOE, K. 1981. Oysters in Japan. J. Sci. of Hiroshima Univ., Ser. B, Div. 1 (Zool.) 29(2):291-419, 36 pls. TorIGOE, K. 1983. Systematic position of Ostrea sedea setoensis Habe, 1957. Venus 41(4):291-295, pls. 1-2. The Veliger, Vol. 28, No. 2 VyaLov, O. S. 1936. Sur la classification des huitres. Acad. Sci. USSR, Comptes Rendus (Doklady) New Ser., vol. 4 (13), No. 1:17-20. Wapa, S. 1953. Larviparous oysters from the tropical West Pacific. Rec. Oceanogr. Works Japan, New Ser. 1(2):66- 72. [not seen] Waite, K. M. 1942. The pericardial cavity and the pericar- dial gland of the Lamellibranchia. Proc. Malacol. Soc. Lond. 25(2):37-88. NOTE ADDED IN PROOF While this paper was in press a paper was published with extensive descrip- tions, good photographs and many locality records of Parahyotissa (Pliohyotissa) quercinus (Sowerby, 1871), from the Gulf of California and southward to Man- zanillo, Colima, Mexico: Gemmell, J.. C. M. Hertz & B. W. Myers, 1985. A problem oyster in the Gulf of California (“Ostrea” quercinus Sowerby, 1871 rediscovered). Festivus (Publication of the San Diego, Calif., Shell Club) 17(5): 43-48. The Veliger 28(2):159-174 (October 1, 1985) THE VELIGER © CMS, Inc., 1985 Gonatus ursabrunae and Gonatus oregonensis, Two New Species of Squids from the Northeastern Pacific Ocean (Cephalopoda: Oegopsida: Gonatidae) by KATHARINE JEFFERTS College of Oceanography, Oregon State University, Corvallis, Oregon 97331 Abstract. Two new species of gonatid squids are described from the northeastern Pacific. Gonatus ursabrunae spec. nov. is distinguished by the presence of greatly enlarged suckers in two locations: in the lateral rows of the middle portions of arms I-III, and in the proximal portion of the dactylus. This species has been taken off Oregon and west to the central Aleutian Islands. Gonatus oregonesis spec. nov. is characterized primarily by the number of club suckers, greater than in any other Gonatus (s.s.) species. This species has been taken only off Oregon. The status of systematics in the family is discussed and the species compared. Based on existing collections, up to three additional species of Gonatidae may remain undescribed in the North Pacific, and one in the Antarctic. Gonatus phoebetriae Imber, 1978, is shown to be a nomen dubium. INTRODUCTION THE FAMILY Gonatidae is of major importance to the ecol- ogy of the Subarctic Pacific; species of this group dominate the pelagic cephalopod fauna in this area (JEFFERTS, 1983; KUBODERA & JEFFERTS, 1984), and are important in the diets of seabirds, fishes, and marine mammals (LEBRAS- SEUR, 1966; SANGER & BAIRD, 1977; Fiscus, 1982). Greater knowledge of the taxonomy of gonatids is vital to an understanding of broader ecologic questions in the Subarctic Pacific. Three genera are currently recognized in the family: Gonatopsis Sasaki, 1920, Berryteuthis Naef, 1921, and Gonatus Gray, 1849. Gonatopsis is distinguished by the loss of tentacles in the adults. The following species are recognized: Gonatopsis octopedatus Sasaki, 1920, G. bo- realis Sasaki, 1923, G. makko Okutani & Nemoto, 1964, G. japonicus Okiyama, 1969, and G. okutanu Nesis, 1972. Another form of Gonatopsis, type A of KUBODERA (1978) has been described, but not named. Berryteuthis is char- acterized by the absence of club hooks, a septemdentate radula, and a carpal-locking zone that extends up onto the manus and dactylus as a “fixing apparatus” (BERRY, 1913). Two species are presently recognized: Berryteuthis magister (Berry, 1913) and B. anonychus (Pearcy & Voss, 1963). The taxonomy of Gonatus has recently been in a state of flux. The genus is characterized by a quinquedentate radula and a carpal-locking zone consisting of alternating ridges with large suckers medially and grooves with fleshy knobs medially. Prior to 1972 only two species were rec- ognized: G. fabric (Lichtenstein, 1818) and G. antarcticus Loénnberg, 1898. Gonatus berryi Naef, 1923, had been forgotten until YOUNG (1972) revived usage of the name and redescribed the species. Gonatus kamtschatica Mid- dendorff, 1849, was originally inadequately described on the basis of a specimen no longer extant, and has been declared a nomen dubium (KUBODERA & OKUTANI, 1981a). Since 1972 seven new species have been described: G. onyx Young, 1972, G. pyros Young, 1972, G. californiensis Young, 1972, G. tenro Nesis, 1972, G. madokai Kubodera & Okutani, 1977, G. middendorffi Kubodera & Okutani, 1981a, and G. steenstrupi Kristensen, 1981. Two others have been described, but not named: Gonatus type C of KUBODERA, 1978 (synonym, Gonatus type A of KUBODERA & OKUTANI, 1981b) and Gonatus sp. of BUBLITZ, 1980. A form that probably represents an additional species oc- curs in Antarctic waters (YOUNG, 1972). Gonatus phoe- betriae Imber, 1978, was described on the basis of a single lower beak. Because variation in the form of beaks within species is frequently broad (Fiscus, 1983, personal com- munication), and complete specimens are absent, I con- sider G. phoebetriae to be a nomen dubium, as does KRIs- TENSEN (1981). Page 160 MATERIALS anp METHODS The material examined (Table 1) was collected by two separate research programs: one conducted by the Uni- versity of Washington aboard the R/V Brown Bear (ARON, 1958, 1962) and the other conducted by the Oregon State University (O.S.U.) Nekton group aboard the R/V’s Ya- quina and Cayuse (e.g., PEARCY, 1964). Both sampling programs employed Isaacs-Kidd Midwater Trawls (Isaacs & Kipp, 1953) of various sizes (1.83, 2.44, 3.05 m de- pressor width) and configurations. Several of the O.S.U. Nekton samples were taken with Isaacs-Kidd Midwater Trawls (IKMT) that had multiple plankton samplers (MPS) as closing cod ends. The MPS was developed by BE£ (1962) and modified by PEARCY & HUBBARD (1964), PEARCY & MESECAR (1971), and PEARCY et al. (1977) to fish three or five nets at discrete subsurface depth hori- zons. Mesh size in the Brown Bear IKMT was 7.6 cm, with a 1.3 cm liner in the aft portion; the O.S.U. Nekton sampling program used 5 mm mesh in all but the cod end, which was 0.571 mm Nitex. The samples were preserved in 10% buffered formalin- seawater solution at sea and transferred to fresh 5% buff- ered formalin in the laboratory before examination. Samples were subsequently transferred to 50% isopropyl alcohol, although often as long as 24 years after collection. The specimens were examined, enumerated, and mea- sured to the nearest mm (or 0.5 mm, depending on di- mension of the structure in question). Initial drawings were made with a camera lucida. The following counts and measurements were made, although not all measure- ments were always possible on all specimens. Measure- ments not further defined here correspond to those of Voss (1956). DML, dorsal mantle length MW, mantle width FL, fin length FW, fin width HW, head width ED, eye diameter, maximum diameter of bulbus AL I, length of arm I, measured from the base between arms I to the tip AL II, length of arm II, measured from the base between arms II and III to the tip AL III, length of arm III, measured from the base be- tween arms III and IV to the tip AL IV, length of arm IV, measured from the base between arms IV to the tip TL, tentacle length, total length of tentacle stalk and club CL, club length, measured from basal sucker of carpus to tip of dactylus AH, arm hooks (present/absent) CH, largest (central) club hook (present/absent) OCH, other (than central) club hooks (number present/ absent) DH, hook distal to large central club hook (present/ab- sent) The Veliger, Vol. 28, No. 2 PH, hooks proximal to large central club hook (number present/absent) CS, total number of club suckers, counted from basal sucker of carpus to tip of dactylus HAC [-IV, half arm count (number of suckers or suck- ers/hooks on the proximal half of arms I-IV Other abbreviations: MWI, mantle width index = MW/DML FLI, fin length index = FL/DML FWI, fin width index = FW/DML EDI, eye diameter index = ED/DML TLI, tentacle length index = TL/DML CLI, club length index = CL/DML CAS, California Academy of Sciences, Department of In- vertebrate Zoology, Golden Gate Park, San Francisco OSUI, Oregon State University Invertebrate reference collection; College of Oceanography, Corvallis USNM, National Museum of Natural History, Smith- sonian Institution, Washington, D.C. Family GONATIDAE Hoyle, 1886 Characterized by a simple, straight funnel-locking car- tilage; buccal connectives that attach ventrally to arms IV; tetraserial brachial armature, including two medical rows of hooks (except male Berryteuthis anonychus, which lack hooks). Genus Gonatus Gray, 1849 Radula with five teeth in a transverse row; tentacles well developed, club with carpal-locking zone consisting of alternating ridges with large suckers medially and grooves with fleshy knobs medially. Gonatus ursabrunae Jefferts, spec. nov. (Figures 1, 2) ? Gonatus fabricit SASAKI, 1929 (pars):269-290, pl. 22, fig. 14; text fig. 128C. Gonatus sp. A, JEFFERTS, 1983:88-93, including table 5 and fig. 31. Material examined: Holotype: a juvenile of 24 mm DML; R/V Brown Bear cruise 235, haul 46; W. Aron and P. McCrery; south of Alaska Peninsula; 53°57'N, 157°39'W, 1.8 m IKMT fished open 0-225 m; 25 July 1959, 0129-0222 h; CAS 040163. Paratype: 1 juvenile, 20 mm DML; R/V Brown Bear cruise 235, haul 46; W. Aron and P. McCrery; south of Alaska Peninsula; 53°57'N, 157°39'W, 1.8 m IKMT fished open 0-225 m; 25 July 1959, 0129-0222 h; OSUI 701. Paratype: 1 juvenile, 19 mm DL; R/V Yaquina cruise YALOC 66, haul 849, south of Alaska Peninsula; 52°58.5'N, 162°48’W, 3.0 m IKMT- MPS fished open 0-2400 m; 6 July 1966, 0707-1330 h; USNM 816326. Paratype: 1 juvenile, 18 mm DML; R/V Yaquina cruise YALOC 66, haul 845; south of Alaska Peninsula; 54°58.2'N, 166°02'W, 1.8 m IKMT fished open K. Jefferts, 1985 Gas Page 161 i a Se a SY Figure 1 Gonatus ursabrunae spec. nov. A, lower and upper mandibles, CAS 057606, 17 mm DML. B-D, USNM 816236, 19 mm DML: B, funnel organ; C, funnel cartilage; D, nuchal cartilage. E, whole animal, dorsal aspect, OSUI 701, 20 mm DML. 0-200 m; 4 July 1966, 0240-0321 h; OSUI 696. Para- type: 1 juvenile. 17 mm DML; R/V Yaquina cruise YALOC 66, haul 849; south of Alaska Peninsula; 52°58.5'N, 162°48’W, 3.0 m IKMT-MPS fished open 0- 2400 m; 6 July 1966, 0707-1330 h; CAS 057606. Para- type: 1 juvenile, 15 mm DML; R/V Yaquina cruise YALOC 66, haul 837; J. Donaldson; south of Adak Is- land; 50°32.3’N, 176°04.5’'W, 1.8 m IKMT fished open 0-160 m; 22 June 1966, 0031-0107 h; CAS 057607. Paratype: 1 juvenile, 12 mm DML; R/V Yaquina cruise YALOC 66, haul 842; J. Donaldson; southeast of Adak Island; 51°43.8'N, 175°20’W, 1.8 m IKMT fished open 0-200 m; 1 July 1966, 0309-0347 h; USNM 816325. Additional material (all in the collections of Oregon State University): 2 juveniles, mantles missing; R/V Ya- quina cruise YALOC 66, haul 843; J. Donaldson; south of central Aleutian Islands; 51°01.0'N, 171°32.0’W, 1.8 m IKMT fished open 0-200 m; 2 July 1966, 0305-0345 h; 1 juvenile, mantle missing; R/V Yaquina cruise YALOC 66, haul 850; J. Donaldson; south of Alaska Peninsula; 53°33.8'N, 160°08.0'W, 1.8 m IKMT fished open 0-160 m; 7 July 1966, 0125-0205 h; 3 juveniles, 23 mm DML, Page 162 2 missing mantles; R/V Brown Bear cruise 176, haul 34; Allen and P. McCrery; south of Alaska Peninsula; 52°29'N, 160°59’W, 1.8 m IKMT fished open 0-60 m; 1 August 1957, 0146-0222 h; 1 juvenile, 18 mm DML; R/V Brown Bear cruise 176, haul 85; P. McCrery, Se- mon, and Linger; south of Aleutian Islands; 51°26’N, 174°10’W, 1.8 m IKMT fished open 0-225 m; 24 August 1957, 0032-0124 h; 1 juvenile, 15 mm DML; R/V Brown Bear cruise 235, haul 23; W. Aron and P. McCrery; Gulf of Alaska; 52°49'N, 142°45.5'W, 1.8 m IKMT fished open 0-225 m; 20-21 July 1959, 2334-0028 h; 1 juvenile, 18 mm DML; R/V Brown Bear cruise 235, haul 44; W. Aron and P. McCrery; south of Alaska Peninsula; 53°55.5'N, 153°17'W, 1.8 m IKMT fished open 0-400 m; 24 July 1959; 0337-0454 h; 1 juvenile, 23 mm DML; R/V Brown Bear cruise 235, haul 45; W. Aron and P. McCrery; south of Alaska Peninsula; 53°56.5'N, 157°57.5'W, 1.8 m IKMT fished open 0-400 m; 25 July 1959, 0004-0119 h; 2 juveniles, 21, 25 mm DML; R/V Brown Bear cruise 235, haul 47; W. Aron and P. Mc- Crery; south of Alaska Peninsula; 53°57’N, 157°49'W, 1.8 m IKMT fished open 0-60 m; 25 July 1959, 0223-0300 h; 1 juvenile, 22 m DML; R/V Yaquina sta. NH-65; off Oregon coast; 44°43.3'N, 125°41.1'W, 1.8 m IKMT fished open 0-200 m; 14 February 1967, 0418-0500 h. Description: Mantle plump, widest at anterior margin (MWI = 30-53, widest in small individuals; meristic in- dices summarized in Table 2), narrowing to pointed tip and adhering to gladius (Figure le). Mantle of soft con- sistency. Eyes large, occupying entire lateral surface of head (EDI = 18-21); anterior sinus small and broad. Ventral surface of mantle slightly emarginate at anterior edge. Fins relatively small, FWI = 41-58, FLI = 26-50, very thin, posteriorly attached just dorsal and anterior to the posterior tip of the gladius; posterior edge of fins united at midline, and projecting slightly posteriorly (Figure 1e). Funnel not extending as far as midpoint of eyes; in most specimens anterior tip of funnel not, or only just, visible at ventral margin of mantle (perhaps owing to contraction on preservation). Mantle-locking cartilage straight, slight- ly expanded posteriorly (Figure 1c). Dorsal component of funnel organ a broad inverted V, with expanded posterior lobes (Figure 1b). Ventral element of funnel organ a pair of small ovoid pads each about half the length of each branch of dorsal element. Funnel valve large and broad. Nuchal folds low and indistinct, no more than two folds observed on one side; the low and short olfactory papilla is just posterior to the eyes, in line with and anterior to the funnel-mantle locking cartilages. Arm formula generally HI = II > I = IV. Arms rel- atively short: ALI for longest arms (III or II) 42-56, ALI for shortest arms (IV) 25-44. Aboral keels well developed on arms IV. Trabeculate protective membranes very well developed on arms I-III, especially in larger individuals. Brachial armature quadriserial; suckers of the two medial rows small (0.18-0.20 mm diameter) in all individuals examined (largest individual examined, the holotype, was The Veligers Vola 28-sNowZ 24 mm DML). These medial suckers have about nine long, slender, blunt teeth on the inner distal margin (Fig- ure 2c), those on the largest suckers reach about 0.024 mm in length, or approximately one-fifth of the diameter of the sucker opening. The two lateral rows of suckers are borne on trabeculae and consist of suckers that are greatly enlarged along the middle third of arms I-III (Figure 2f). The largest of these suckers are 0.50 mm in diameter and have 9-16 short, blunt teeth (much shorter than on suck- ers from the medial rows) on the distal inner margin (Fig- ure 2d). Arms IV bear four rows of equally sized suckers (0.10-0.12 mm in diameter in holotype) which are smaller than the medial suckers of arms I-III. Half-arm counts for the two largest specimens are given in Table 3, but are not very consistent between the two specimens. It is extremely difficult to make accurate counts on smaller specimens. These counts of only two specimens are of little use by themselves; when a larger body of data becomes available such counts may show consistent differences be- tween species. Tentacles are of moderate length, TLI = 53-79, and the clubs are moderately short, CLI = 13-25 (Figure 2a). A dorsoaboral keel is present on the club from the level of the central hook (or enlarged sucker) to the tip of the dactylus. The medial zone of the manus contains a central hook in the holotype, and an enlarged central sucker in specimens of 19 and 20 mm DML, with three or four proximal suckers in all three specimens (Figure 2a). A carpal-locking zone consists of approximately five alter- nating ridges and suckers. The dorsal marginal zone con- tains suckers in four rows, and the ventral marginal zone bears four to five rows. The dactylus suckers are disposed in about six rows just distal to the central hook (or en- larged sucker), but these rapidly decrease to four regular rows which continue out the length of the dactylus to a circlet of small suckers at the tip. The suckers distal to the central hook number approximately 110 (full club sucker counts are impossible, as no mature specimens are available, and many of the proximal suckers remain as buds even in the larger specimens). The dactylus suckers just distal to the central hook reach a maximum diameter of 0.30 mm in the holotype, and decrease in size distally (Figure 2b). In other species of Gonatus (specimens of similar size were used where possible), dactylus suckers never approach this maximum size: Sucker Species DML diameter californiensis 112 0.25 oregonensis 46 0.20 madokai 40 0.14 pyros 35 0.14 berryt 30 0.13 onyx 26 0.08 middendorffi 35 buds madokai 22 buds sp. C of Kubodera 15 buds K. Jefferts, 1985 Page 163 Figure 2 Gonatus ursabrunae. A-F, USNM 816326, 19 mm DML: A, tentacle; B, enlarged dactylus sucker; C, normal brachial sucker of medial row, A III; D, enlarged brachial sucker of lateral row, A III; E, radula; F, right arm III. Scale (B, C, D), 1 mm. Page 164 The Veliger, Vol. 28, No. 2 Table 1 Station data for the material examined, including type number (OSUI: 687-701. CAS: 040162-057608. USNM: 816325- 816328. * indicates holotypes); station or haul number; date; latitude (N); longitude (W); gear depressor width of IKMT (m), or IKMT + MPS (hauls 2057#4, 2107#5, 2110#1 and #5); depth sampled (m); local time of sampling; and the vessel from which the sample was collected (Y, R/V Yaquina; B, R/V Brown Bear; C, R/V Cayuse). Type no. Haul no. Latitude Longitude Gear Date Time Depth Vessel Gonatus ursabrunae 696 845 54°58.2' 166°02' 1.8 4 Jul 66 0240-0321 0-200 Y 700, 701 235-46 SSro V5 193.92 1.8 25 Jul 59 0129-0222 0-225 B 040163* 235-46 53°57 157/739! 1.8 25 Jul 59 0129-0222 0-225 B 057606 849 52°58.5' 162°48’ 3.0 6 Jul 66 0707-1330 0-2400 Y 057607 837 50°32.3' 176°04.5' 1.8 22 Jun 66 0031-0107 0-160 Ye 816325 842 51°43.8' 175°20' 1.8 1 Jul 66 0309-0347 0-200 Ww 816326 849 52°58.5' 162°48' 3.0 6 Jul 66 0707-1330 0-2400 Y — 843 51°01.0' 171°32.0' 1.8 2 Jul 66 0305-0345 0-200 Y — 850 5373318) 160°08.0' 1.8 7 Jul 66 0125-0205 0-160 Y — 1016 44°44. 4! 125°44.6' 1.8 14 Feb 67 0418-0500 0-200 Ww — 176-34 52729: 160°59' 1.8 1 Aug 57 0146-0222 0-60 B _ 176-85 51°26' 174°10' 1.8 24 Aug 57 0032-0124 0-225 B — 235-23 52°49! 142°46' 1.8 20 Jul 59 2334-0028 0-225 B — 235-44 53°55:51 U5 3217 1.8 24 Jul 59 0337-0454 0-400 B — 235-45 53°56.5' SSeS) 1.8 25 Jul 59 0004-0119 0-400 B — 235-47 5325s 157°49' 1.8 25 Jul 59 0223-0300 0-60 B Gonatus oregonensis 687 1011 44°46.2' 125°52.0’ 1.8 13 Feb 67 1347-1728 0-1500 Y 690 2057#4 44°35.1' 125°32.5' 2.4 21 Jul 71 0314-0348 300-400 Y 692 2110#5 44°37.4! 125°41.3' 2.4 29 Nov 72 0327-0335 surface Y 040162* 1692 44°39.1' 128°21.8' 1.8 21 Aug 69 0327-0414 0-240 C 057608 1563 44°40.2' 127°49.1' 1.8 30 Jun 69 2330-0020 0-220 Y 057609 2110#1 44°33.9' 125°39.2' 2.4 29 Nov 72 0105-0216 0-200 NY. 816327 1091 44°40.9' 127°56.2' 1.8 3 Jun 67 2300-2343 0-185 Y 816328 2107#5 44°37.2' 125°42.3' 2.4 28 Nov 72 0641-0715 200-300 Y _- 884 44°54.2' 125°25' 1.8 25 Aug 66 0100-0540 0-2000+ Y — 953 44°39.0' 125°41.4' 1.8 18 Dec 66 0042-0415 0-950 Y Tentacle stalk suckers are small (about 0.04 mm diame- ter) and numerous. In a 19 mm individual, there are 25 suckers in the ventral row, 28 in the dorsal row, and 57 on the oral face between the two rows. In smaller indi- viduals, the stalk suckers appear to be arranged in roughly six alternating rows. Measurements of the holotype and paratypes are given in Table 3. Buccal connectives are attached dorsally to arms I and II and ventrally to arms III and IV. Seven short buccal lappets are present. A spindle-shaped liver is present in smaller individuals, oriented obliquely to the body axis. Complete hook de- velopment is unknown, but the central hook develops at 20-24 mm DML; arm hooks and other club hooks may develop at sizes greater than 24 mm DML. No trace of chromatophores remains on these speci- mens, most likely due to preservation. No photophores are present. The radula (Figure 2e) is of the normal Gonatus type, with five rows of teeth: a tricuspid rhachidian, and simple admedian and lateral teeth on each side. No ridges are visible on the teeth. The central tooth of the rhachidian is off-center, alternating sides with each row, v.e., the teeth of the second and fourth rows are aligned, as are the teeth of the first and third rows. The upper mandible is slightly curved and acutely pointed; both the upper and lower are pigmented only at the tips in a specimen of 19 mm DML (Figure 1a). Type designation: The holotype is a juvenile of 24 mm DML. R/V Brown Bear cruise 235, station 46; W. Aron and P. McCrery; south of Alaska Peninsula, northeast Pacific; 53°57'N, 157°39'W;; collected with a 1.8m IKMT fished open 0-225 m; 25 July 1959, 0129-0222 h. Location of type: California Academy of Sciences, De- partment of Invertebrate Zoology, Golden Gate Park, San Francisco. Catalogue number: CAS 040163. Etymology: ursabrunae, after the vessel R/V Brown Bear, from which the holotype was collected. Distribution: The known distribution is limited to the northeastern Pacific, but may extend into the northwest- K. Jefferts, 1985 Page 165 180° 150° W 120° 60° = = 30° N 180° ISO°W 120° N Figure 3 Location of hauls capturing Gonatus ursabrunae (x) and Gonatus oregonensis (0). For clarity, not all hauls are shown: three hauls captured G. ursabrunae in the area 53°56-57'N, 157°39-58'W; seven hauls captured G. oregonensis in the area 44°34-54'N, 125°25-52'W. ern Pacific, considering Sasaki’s specimen (see Discussion below). Twenty individuals have been collected in 14 mid- water hauls (all open; mostly 0-200 m; two hauls 0-400 m, one 0-2400 m; see Table 1) from the northern Cali- fornia Current and the Alaska Current as far west as 176°W (Figure 3). Okutani (zn Jitt., 1982) has seen three similar individuals in the collections of the University of Alaska. One was collected at Seward (60°N, 149° W), and the others in southeastern Alaska (56°N, 134°W; 58°N, 135°W). Discussion: Gonatus ursabrunae clearly belongs in the genus Gonatus because of the structure of the radula and the development of a central hook on the tentacular club. Comparison of similarly sized specimens shows that it is not G. pyros, as it lacks an optic photophore, nor is it G. berry, as no arm hooks are yet evident, as they are in juvenile berry:; neither can it be G. tinro, as it does have a club hook. Comparably sized individuals are known for G. onyx, G. madokai, G. californiensis, G. middendorffi, and Gonatus oregonensis (spec. nov., described and discussed below); none of these demonstrates the enlarged brachial and club suckers characteristic of this species. BUBLITZ (1980:76) stated that some of his specimens of Gonatus type A (which was described as G. middendorffi, KUBODERA & OKUTANI, 1981a) showed slightly enlarged suckers in the lateral arm rows (“‘1.5-2 times as large as the corre- sponding median sucker’); however, his figure (pl. 30) clearly shows just the reverse, that the median sucker (transforming into a hook) is larger than the lateral suck- er. These specimens otherwise agree with the description of G. middendorffi, which is separable from Gonatus ur- sabrunae by its MWI and the size at which club hooks develop. There are other differences, especially in club armature, which serve equally well to differentiate all of these species (Tables 4, 5). Gonatus type C of KUBODERA, 1978, is known from individuals as large as 16 mm DML; there is no indication of enlarged suckers in these, and this type is further characterized by a separated epidermis, which does not occur in Gonatus ursabrunae. BUBLITZ’s (1980) new species also has no indication of enlarged suckers in the lateral rows: “each sucker of the median two rows is about 1.2 times as large as the corresponding lateral sucker” (BUBLITZ, 1980:61), and has five rows of sucker buds on the tentacular stalk as opposed to six in larvae of G. ursabrunae. The tentacles of Bublitz’s species are shorter (TLI = 37-49) but have clubs of about the same size (CLI = 18-24); in addition, the ventral mar- ginal zone comprises 3 or 4 rows of suckers in Bublitz’s species, vs. 4 or 5 in G. ursabrunae. Several other differ- Page 166 The Veliger, Vol. 28, No. 2 Table 2 Meristic indices for Gonatus ursabrunae. Abbreviations as in methods section, with additions: ALIM, arm length index for longest arm (length of arm over DML x 100); X, mean; SD, standard deviation. Type no. Index 040163 701 816326 696 057606 DML 24 20 19 18 AW MWI 33 30 42 44 53 FLI 21 20 21 22 18 FWI 46 45 53 56 41 HWI 29 36 26 33 29 EDI 21 20 21 19 18 ALIM 42 45 47 56 53 TLI 79 60 63 — Bl CLI 21 25 21 — 18 ences in tentacle sucker counts and disposition are evident: suckers distal to the central hook number approximately 85 in Bublitz’s species (22 mm DML), but about 110 in G. ursabrunae (19 mm DML); suckers of the dorsal row, ventral row, and oral face of the tentacular stalk number about 15, 10, and O in Bublitz’s species (22 mm DML), but 28, 25, and 57, respectively, in G. ursabrunae (19 mm DML). SASAKI (1929) included one larva (pl. 22, fig. 14; text fig. 128C) in the description of Gonatus fabricit which appears to correspond to Gonatus ursabrunae. Measure- ments of this individual have been included here, in Table 3. SASAKI (1929:269) noted: ‘““The suckers of the first three 057607 816325 Range xX SD 15 12 12-24 17.9 3.80 40 42 30-53 40.6 7.52 20 17 17-22 19.9 Ea 53 58 41-58 50.3 6.32 33 33 26-35 31.1 3.18 20 21 18-21 20.0 1.15 47 50 42-56 48.6 4.79 53 67 53-79 65.5 9.03 13 _ 13-25 19.6 4.45 pairs of arms, uniform, except in the largest larva referred to, where the suckers of the outer two series on these arms are much larger than those of the inner two series.” In addition, he noted that the proximal suckers on arms IV were also enlarged, and numbered from two to seven, in the larvae of G. fabrici, although it is not clear from the description to which specimen(s) he was referring. No such condition has been noted for Gonatus ursabrunae. The geographical origin of Sasaki’s specimen is unknown; it apparently came from collections of the Albatross, and SASAKI (1929:270) listed the following localities from which the Albatross collected G. fabrici: “Milne Bay, Simushir I., Kurile group; Bowers Bank, Bering Sea; near Near Table 3 Measurements (in mm) of selected individuals of Gonatus ursabrunae. “Sasaki” refers to the specimen described by SASAKI (1929) which is discussed in the text. Type no.: 040163, 057606, 057607 are CAS; 696, 701 are OSUI; 816325, 816326 are USNM. ES, enlarged sucker. Type no. Index 040163 701 816326 696 DML 24 20 19 18 MW 8 6 8 8 FL 5 4 4 4 FW 11 9 10 10 HW i 7 5 6 ED 5 4 4 3.5 AL I 7 6 vi 8 AL I 9 9 8 10 AL III 10 9 9 10 AL IV 8 6 6 8 TL 19 12 12 — CL 5 5 4 — AH — — = CH — ES ES — OCH -— — — HAC I 18 23 -- — HAC II 25 19 — — HAC III 22 20 — — HAC IV 30 27 = — 057606 057607 816325 Sasaki 17 15 12 14 9 6 5 6.5 3 3 2 — 7 8 i — 5 5 4 — 3 3 2s) — 6 6 5 3.5 8 7 6 4 9 7 6 4 6 5 3 DES) 12 8 8 7 3 2 2 ? Kewjetterts, 1935 Page 167 Figure 4 Gonatus oregonensis spec. nov. A-E, USNM 816327, 46 mm DML: A and B, hook of right arm III, with hood removed and with hood intact, front and lateral views; C, funnel organ; D, funnel cartilage; E, nuchal cartilage. F, whole animal, ventral aspect, CAS 057608, 31 mm DML. is., Aleutians; east of Kamchatka; south of Alaska; and near Commander Is.” Gonatus oregonensis Jefferts, spec. nov. (Figures 4, 5) Gonatus sp., E. JEFFERTS, 1983:94-98, including table 6 and fig. 31. Material examined: Holotype: a juvenile of 39 mm DML; R/V Cayuse haul 1692; R. Findley; off the coast of Oregon; 44°39.1'N, 128°21.8’W, collected with a 1.8 m IKMT fished open 0-240 m; 21 August 1969, 0327-0414 h; CAS 040162. Paratype: 1 juvenile, 46 mm DML; R/V Yaquina haul 1091; R. Eagle; off Oregon coast; 44°40.9’N, 127°56.2'W, 1.8 m IKMT fished open 0-185 m; 3 June 1967, 2300-2343 h; USNM 816327. Para- type: 1 juvenile, 35 mm DML; R/V Yaquina haul 1011; off Oregon coast; 44°46.2'N, 125°52.0’W, 1.8 m IKMT fished open 0-1500 m; 13 February 1967, 1347-1728 h; OSUI 687. Paratype: 1 juvenile, 31 mm DML; R/V Page 168 The Veliger) Vol Z233NowZ Table 4 Comparison of species of the family Gonatidae from the North Pacific. Adult characters (indices for animals over 40 mm DML), including known size range (DML in mm), number of teeth in transverse row of radula, shape of nuchal cartilage, photophores, mantle width index, maximum arm length index, size at which arm hooks develop (DML in mm), hook and sucker pattern on club (e.g., hHhhh is one distal hook, a large central hook, and three proximal hooks), size (DML in mm) at which the central, distal, and proximal hooks develop, rows of suckers on the dactylus (from just distal to the hooks toward the end—in Berryteuthis, the manus is included; Irr, irregular), number of suckers on the club, Teeth Known size on Nuchal Photo- Species range radula cartilage phores MwWI ALIM Arm hooks Gonatus ursabrunae 12-30 5 rectangular none — = >24 pyros 7-66 5 rectangular optic 26 64 17-22 berryt 6-119 5 rectangular none 25-35 64-72 7-9 tinro 7-89 5 rectangular none 20 64 19-21 onyx 2-98 5 rectangular none 21-27 50-59 26-28 or 18-20 madokai 6-329 5 rectangular none 23-30 90-103 16-19 middendorffi 6-296 5) rectangular none 18-22 40-52 20-30 or 26-30 oregonensis 24-46 5 rectangular none 26 63 24-30 californiensis 24-112 5 rectangular none 19-33 46-55 26-29 sp. C of Kubodera 4-16 5 ? rectangular none — — >16 sp. of Bublitz 7-80 5 rectangular none 33-35 63-65 21-38 Gonatopsis borealis 4-290 7 panduriform none 22-39 44-67 30-35 Berryteuthis anonychus 5-99 V panduriform none 21-29 30-33 ? >30 magister 6-320 i, panduriform none 27-29 62-68 >16 or 55-60 Yaquina haul 1563; P. Kalk and D. Stein; off Oregon coast; 44°40.2'N, 127°49.1'W, 1.8 m IKMT fished open 0-220 m; 30 June-1 July 1969, 2330-0220 h; CAS 057608. Paratype: 1 juvenile, 30 mm DML; R/V Yaquina haul 2057#4; W. Pearcy; off Oregon coast; 44°42.4'N, 125°32.5'W, 1.8 m IKMT fished open 0-600 m; 21 July 1971, 0314-0348 h; OSUI 690. Paratype: 1 juvenile, 30 mm DML; R/V Yaquina haul 2107#5; off Oregon coast; 44°37.2’N, 125°42.3’W, 3.0 m IKMT + MPS fished 300-200 m; 28 November 1972, 0641-0715 h; USNM 816328. Paratype: 1 juvenile, 30 m; R/V Ya- quina haul 2110#1; off Oregon coast; 44°33.9'N, 125°39.2'W, 3.0 m IKMT + MPS fished 0-200-150 m; 29 November 1972, 0105-0216 h; CAS 057609. Para- type: 1 juvenile, 24 mm; R/V Yaquina haul 2110#5; off Oregon coast; 44°37.4'N, 125°41.3’W, 3.0 m IKMT + MPS fished at surface; 29 November 1972, 0327-0335 h; OSUI 692. Additional material (all in the collections of Oregon State University): 1 juvenile, 26 mm DML; R/V Yaquina haul 953; station NH-50; Coleman; off Oregon coast; 44°38.8'N, 125°20.7’W, 1.8 m IKMT fished open 0-950 m; 18 December 1966, 0042-0415 h; 1 juvenile, 19 mm DML; R/V Yaquina haul 884; station WG-16; Coleman and Wyandt; off Oregon coast; 44°54.2'N, 125°25'W, 1.8 m IKMT fished open 0-2000 m; 25 August 1966, 0100- 0540 h. Description: Mantle plump, widest in the midsection (MWI = 29-43; meristic indices are summarized in Ta- ble 6). Ventral anterior margin of mantle emarginate (Figure 4f). The corners of this emargination project at the anterior ends of the mantle-locking cartilages. Head less wide than mantle, with at least two nuchal folds. Eyes are large, occupying the entire lateral surfaces of head (EDI = 15-23); an optic sinus is at the anterior end, between the bases of the tentacle and arm III. Fins broad but relatively short: FWI = 80-90 for an- imals over 30 mm DML; FLI = 25-45. Fins united pos- teriorly, extending beyond the tip of the gladius. A car- tilaginous end cone extends to the posterior limit of the fins. Posterior margin of fins essentially straight, anterior margin convex. Margins quite thin, fragile, especially an- teriorly. K. Jefferts, 1985 Page 169 Table 4 Continued. and the sucker distribution pattern on the tentacular stalk (e.g., 1V, 1-2, 1D represents one row of suckers along ventral margin, 1-2 suckers on medial face, and 1 row along dorsal margin of the stalk). Abbreviations as in Table 3. From original data and NeEsIs (1972), YOUNG (1972), KUBODERA & OKUTANI (1977, 1981a, b), BUBLITZ (1980), BUBLITz & NISHIYAMA (MS). Club formula C Hook D Hook P Hooks ?H?? 20-24 ? (>24) ? (>24) hHhhhh 15-18 18-23 21-26 hHsshhhh 12-17 19-28 25=32 no hooks —_ —_ — sHsssss or 17-24 — — hHsssss hHhhhhh >72 >72 >72 hHsssss or > 60 > 60 >250 hHshhss hHhhhhss 24-30 24-30 35-39 hHhhhs 1V=23 24-30 35-41 -S-- —_ ? (> 16) ? (> 16) hHhhhhh 13-15 >22 22-38 no hooks = — — no hooks — — — Funnel reaching only slightly past the posterior extent of the eye. Funnel-locking cartilage slightly curved later- ally, with a shallow medial groove which widens caudad, and with a distinct anterior fold, corresponding to a pro- jection on the ventral surface of the mantle (Figure 4d). Funnel valve small and broad. Dorsal pad of funnel organ very broad, with an anterior papilla and narrow ovoid pads at the posterior ends of the arms (Figure 4c). General shape that of an inverted V, but with posterior portions of arms laterally offset from anterior portions. Ventral component of funnel organ consists of two broadly ovoid pads each nearly as long as the arms of the dorsal pad. Nuchal cartilage only slightly clavate, and slightly wider at anterior end. The cartilage has a narrow medial ridge with a medial groove, and broad lateral grooves (Figure 4e). Arms of moderate length, ALI = 59-63 in 46 mm DML individual, 43-53 in 30 mm DML individual. Arm for- mula generally III = II > IV = I. Aboral keels are strong and nearly always evident on arms IV; they are occasion- ally discernible on arms I-III. Trabeculate protective membranes are exceedingly well developed on arms I-III; Rows on dactylus Club suckers Stalk pattern 6->4 194+ 1V,57,1D Irr->4 159-181 2V, 50-125, 1D 4 162-178 1V, 1-2,1D 5-6->18 576-600 — >12->4-5 5-6->4 165-194 1V, <10,1D 5-6->4 215+ 2V, few, 1D 7-8->4 340 1V, few, 1D 7-8->5-6 295-370 1V, 63-74, 1D 7-8->4 217-269 1V, 40-80, 1D — buds — 5-6->4 183 1V, none, 1D — 55 max _ 13->4 650-738 —_— 16->4 1106-1273 — the marginal rows of suckers are borne on the trabeculae. Arms I-III bear hooks in the medial rows (Figures 4a, b, 5); these develop at a mantle length of 24-30 mm. Arms IV bear four rows of suckers. Lateral suckers of arms I to III relatively small, with about eight closely set, elon- gate, blunt teeth (Figure 5d). Half-arm counts for two of the larger individuals are given in Table 7. The tentacle is long, TLI = 60-105 (depending on preservational state), and bears a fairly large club (CLI = 21-20). A swimming keel is present on the dorsal surface of the dactylus, extending from the level of the distal hook to the tip of the dactylus (Figure 5g). Dorsal and ventral protective membranes are also present, but are very short and ill-developed. They originate on the stalk and extend along the club to its tip. The club bears a large central hook, a distal hook about half the size of the central one, and several proximal hooks. The central and distal hooks develop at a DML of about 24-30 mm, but the proximal hooks are not evident until a length of 35-39 mm is at- tained. In the proximal series (Figure 5f), the suckers next to the central hook are the first to transform into hooks, so that an animal of 39 mm may have two hooks proximal Page 170 The Veliger, Vol. 28, No. 2 Table 5 Comparison of early life history stages of species of the family Gonatidae. Characters for individuals under 40 mm DML, including size range of specimens included (mm DML), mantle width index, maximum arm length index, rows of suckers on tentacular stalk in larval forms (these suckers are lost as the club begins to develop), tentacle length index, and club length index. Information from original data and Nesis (1972), YOUNG (1972), KUBODERA & OKUTANI (1977, 1981a, b), BuBLitz (1980), and BuBLITz & NISHIYAMA (MS). (BUBLITZ, 1981, measured stretched mantle width). Rows of Species Size range MWI ALIM stalk suckers TLI CLI Gonatus ursabrunae 12-24 24-53 42-56 6 50-79 13-25 pyros 13-25 42 38-48 5-6 52 21 berryi 13-30 30-33 48 5-6 35-100 25 tinro 10-28 35-47 34-74 5-6 40 10-15 onyx 6-26 35-40 35-40 5 25-55 20-25 madokai 10-40 30-35 30-80 4-5 30-90 Z=15) middendorffi 6-40 24-40 25-45 4 30 15 oregonensis 24-39 29-43 42-67 6 60-105 21-30 californiensis 29-38 29-32 41-47 ? 66-82 21-24 sp. C of Kubodera 6-16 40-50 30 5-6 50-75 4-8 sp. of Bublitz 11-13 35-61 23-55 5 41-75 18-34 16-22 42-45 37-59 5 37-49 18-24 Gonatopsis borealis 5-30 30-40 25-40 4-5 25-30 — Berryteuthis anonychus 5-30 25-45 33-40 3-4 ~50 8 magister 7-16 40-45 35-40 5-6 50 5-13 to the central, and three to four suckers, and an animal of 46 mm may have four hooks proximal to the central, and two suckers. The carpal-locking zone consists of four to five ridges with accompanying suckers, alternating with five to six knobs. This series extends onto the stalk. The ventral marginal zone contains four rows of suckers and the dorsal marginal zone five. The tentacular stalk bears single rows of suckers on both the ventral and dorsal margin of its inner face. The space between the rows is beset with many small suckers. The number of suckers in the ventral row is at least 74 in the 46 mm specimen, in the dorsal row, at least 63, and on the medial face, at least 70. In a 24 mm specimen, the stalk suckers appear to be arranged in six, somewhat irregular, alternating rows. The dactylus bears many small but roughly equal-sized suckers (0.20 to 0.25 mm at DML 46 mm). These have narrow openings and four to six long, slender, peglike teeth on the distal border of the inner ring (Figure 5e) and are disposed in seven or eight rows just distal to the Table 6 Meristic indices for Gonatus oregonensis. Abbreviations as in Table 2. Type no. Index 816327 040162 687 057608 690 816328 057609 692 Range x SD DML 46 39 35 31 30 30 30 24 24-46 33.1 6.77 MWI 35 31 29 42 33 43 33 38 29-43 35.5 5.07 FLI 48 44 34 32 40 40 40 33 32-48 38.9 5.59 FWI 89 82 80 81 83 83 83 58 58-89 79.9 9.23 HWI 22 28 31 29 30 23 30 25 22-31 Dilee) 3.45 EDI 16 22 — 19 23 1S 20 17 15-23 18.9 3.02 ALIM 63 64 54 58 53 60 67 42 42-67 57.6 7.95 TLI 96 69 80 81 63 107 103 1S 63-107 84.2 16.1 CLI 26 23 26 23 27 30 30 2A 21-30 25.8 3.28 K. Jefferts, 1985 Page 171 ooo coo oe? S,9,9 FS O22 9 SASCOGSS oO 0 SS SS eo CKO) CXex BOOLIGIENS OKO OISICIS) ) Figure 5 Gonatus oregonensis. A, right arm III], USNM 816327, 46 mm DML. B, C, OSUI 687, 35 mm DML: B, mandibles; C. radula. D-G, USNM 816327, 46 mm DML: D, brachial sucker, right arm III; E, dactylus sucker; F, proximal series of club; G, tentacle. Scale (D and E), 1 mm. hooks, decreasing to five or six rows near the tip (Figure Buccal connectives are connected to dorsal borders of 5g). A circlet of small suckers occupies the tip of the dac- arms I and II and to the ventral borders of arms III and tylus. The total number of suckers on the dactylus, ventral IV. marginal zone, and dorsal marginal zone is 320 in the 46 The radula is of the normal quinquedentate Gonatus mm specimen, and shows a range of 295 to 370 in the type, with tricuspid rhachidian, unicuspid admedian, and other specimens. unicuspid lateral. No ridges are apparent on the radular Page 172 The Veliger, Vol. 28, No. 2 Table 7 Measurements (in mm) and counts for selected individuals of Gonatus oregonensis. Type no.: 040162, 057608, 057609, CAS; 687, 690, 692, OSUI; 816327, 816328, USNM. D, damaged; +, present; —, absent or not applicable. Haul: 1091 1692 1011 type no.: 816327 040162 687 Index DML 46 39 35 MW 16 12 10 FL 22 9/ 12 FW 41 32 28 HW 10 11 11 ED ED 8.5 D ALI Dil, 22 16 AL II 28 25 19 AL Ill 29 24 19 AL IV 28 18 17 TL 44 Dif 28 CL 12 9 9 AH oF ae + CH + =. + DH oF ote =P PH 4 D, - CS 320 347 339 HAC I 20/14 — HAS) HAC II 22/15 - 19/9 HAC III 19/15 = 17/9 HAC IV 47 — 35 teeth, and the central teeth of the rhachidian are aligned in each row (Figure 5c). The upper mandible is long and acute; both upper and lower mandibles are darkly colored only on the tips in a specimen of 46 mm (Figure 5b). Type designation: The holotype is a juvenile of 39 mm DML. R/V Cayuse, haul 1692; Findley; off the coast of Oregon; 44°39.1’N, 128°21.8'W; collected with a 1.8 m IKMT fished open 0-240 m; 21 August 1969, 0327- 0414 h. Location of type: California Academy of Sciences, De- partment of Invertebrate Zoology, Golden Gate Park, San Francisco. Catalogue number: CAS 040162. Etymology: oregonensis, after the type locality; to em- phasize the morphological similarity to another species localized in the California Current, G. californiensis. Distribution: This species is currently known only from waters off Oregon. Ten individuals were collected in ten midwater hauls (all open, mostly 0-400 m; one 0-1500 m, one 0-2000 + m) in the northern portion of the Cali- fornia Current system (Figure 3). Measurements for eight of these are given in ‘Table 7. Discussion: This species is easily separable from all but one of the described species of Gonatus (Table 4, 5). The distribution of hooks on the club separates it from G. 1563 2057#4 2107#5 2110#1 2110#5 057608 690 816328 057609 692 31 30 30 30 24 13 10 13 10 9 10 12 12 12 8 25 25 z5 25 14 9 9 q 9 6 6 Y/ 4.5 6 4 15 13 13 15 8 18 15 16 19 10 17 16 18 20 10 14 13 14 18 9 25 19 32 31 18 7 8 9 9 5 + + + + — + + a + = + + + + - 355 370 295 320 300 berry: (in which the proximal hooks are separated from the central hook by one or two suckers), from G. onyx (no proximal hooks), and from G. tinro (no club hooks). Go- natus pyros has an optic photophore, and G. madokaz has only eight to ten minute suckers on the oral face of the tentacular stalk (several other characters also serve to sep- arate these species). Gonatus middendorffi develops all club hooks at a much larger size (over 60 mm DML), and has a more slender body. Gonatus sp. of BUBLITZ (1980) has fins which are somewhat less broad (FWI = 43-87 us. 80-89 in Gonatus oregonensis), and shows significant dif- ferences in the number and disposition of club suckers (probably fewer than 100 club suckers in Gonatus sp. of BUBLITZ, 1980, arranged in four rows on the dactylus). This species is less easily separable from Gonatus cali- forniensis. The distribution of hooks on the club is the same in the two species, and the size at which all hooks develop is similar. There are, however, consistent differ- ences in fin dimensions, in sucker counts on the club, and in distribution of suckers on the dactylus. My present collection does not contain mature individuals; these dif- ferences may be better characterized on examination of larger individuals. YOUNG’s (1972) specimens of Gonatus californiensis (29-112 mm DML) showed a FWI = 54- 70. Specimens of G. oregonensis 30 mm and over had a FWI = 80-89. The clubs are also somewhat larger in G. oregonensis: CLI = 21-30 vs. 17-24 in G. californiensis. Club sucker counts show no overlap in the two species: Kew jetterts) 1985 G. oregonensis ranges from 295 to 370, and Young’s G. californiensis from 217 to 269. In Gonatus oregonensis, suckers are arranged in seven to eight rows at the base of the dactylus, and decrease to five to six rows at the tip. In G. californiensis, the dactylus suckers are disposed in eight rows basally and “decrease to four rows about halfway out on the dactylus” (YOUNG, 1972:52). The arms are also noticeably longer in G. oregonensis than in G. cali- jorniensis: at 46 mm DML, the longest arms (III and II) are 23-24 mm in G. californiensis and 28-29 mm in G. oregonensis. Further comparison supports this difference (the data for G. californiensis are from YOUNG, 1972): Arm length index for longest arms (II, III) DML californiensis oregonensis 46 mm 50-52 61-63 38-39 47 62-64 34-35 47 54 29-30 41 50-67 This new form thus represents an intermediate condi- tion between Gonatus tinro, which has a Berryteuthis-like club with no hooks but many (>400) suckers, and G. californiensis, G. pyros, and G. madokai, which have cen- tral, distal, and proximal hooks, but fewer (<270) suckers on the club. I believe that this form represents a distinct species, as several characters show no overlap with G. californiensis: fin dimensions, arm length, and sucker num- ber and distribution on the club. ACKNOWLEDGMENTS The author is grateful to T. Okutani and T. Kubodera for many enlightening discussions on gonatid systematics; to W. Aron, C. H. Fiscus, T. Okutani, W. G. Pearcy, and T. Kubodera for making specimens available; to B. Vinter for most of the illustrations; to two anonymous reviewers for comments which improved the manuscript; and to W. G. Pearcy for much constructive advice and assistance during the course of the research. The work was supported by contracts with the National Marine Fisheries Service, Northwest and Alaska Fisheries Center (03-7-208-35070, Phases I-IV), and completed while in residence at the Northwest and Alaska Fisheries Center. LITERATURE CITED ARON, W. 1958. Preliminary report of midwater trawling studies in the north Pacific Ocean. Tech. Rept. 58. Univ. Wash. Dept. Oceanogr., Seattle. ARON, W. 1962. The distribution of animals in the eastern North Pacific and its relationship to physical and chemical conditions. J. Fish. Res. Bd. Canada 19(2):271-314. BE, A.W. H. 1962. Quantitative multiple opening and closing plankton samplers. Deep-Sea Res. 9:144-151. Berry, 8S. S. 1913. Notes on some west American cephalopods. Proc. Acad. Natur. Sci. Philadelphia 1913:72-77. BUBLITZ, C. 1980. Systematics of the cephalopod family Go- Page 173 natidae from the southeastern Bering Sea. Master’s Thesis, Univ. Alaska, Fairbanks. 177 pp. BUBLITZ, C. & T. NISHIYAMA. Manuscript. Developmental mor- phology of the gonatid cephalopods, with special reference to Gonatus tinro. Fiscus, E. H. 1982. Predation by marine mammals on squids of the eastern North Pacific Ocean and the Bering Sea. Mar. Fish. Rev. 44(2):1-10: Gray, J. E. 1849. Catalogue of the Mollusca in the collection of the British Museum: I, Cephalopoda Antepedia. London. 164 pp. Hoy_e, W. E., 1886. Report on the Cephalopoda collected by HMS Challenger during the years 1873-76. Rept. Voy. Challenger, Zool. 16(44):1-246. ImBER, M. J 1978. The squid families Cranchiidae and Go- natidae (Cephalopoda: Teuthoidae) in the New Zealand region. New Zealand J. Zool. 5:445-484. Isaacs, J. D. & L. W. Kipp. 1953. Isaacs-Kidd midwater trawl. Scripps Inst. Oceanogr. Ref. 53-3. 21 pp. JEFFERTS, K. 1983. Zoogeography and systematics of cepha- lopods of the northeastern Pacific Ocean. Doctoral Thesis, Oregon State Univ., Corvallis. 291 pp. KRISTENSEN, T. K. 1981. The genus Gonatus Gray, 1849 (Mollusca: Cephalopoda) in the North Atlantic. A revision of the North Atlantic species and description of Gonatus steenstrupi n. sp. Steenstrupia, Zool. Mus. Univ. Copen- hagen 7(4):61-99. KuBoDERA, T. 1978. Systematics and morphological changes with growth in the early life stages of pelagic squids of the family Gonatidae in the Subarctic Pacific region. Master’s Thesis, Fac. Fish., Hokkaido Univ. 107 pp. KusoperA, T. & K. JEFFERTS. 1984. Distribution and abun- dance of the early life stages of squid, primarily Gonatidae (Cephalopoda, Oegopsida), in the northern North Pacific. Bull. Nat. Sci. Mus. Tokyo 10(3):91-106 et seq. in press. Kusopera, T. & T. OKUTANI. 1977. Description of a new species of gonatid squid, Gonatus madokai n. sp., from the northwest Pacific, with notes on morphological changes with growth and distribution in immature stages (Cephalopoda: Oegopsida). Jap. J. Malacol. (Venus) 36(3):123-151. KusBoperA, T. & T. OKUTANI. 1981a. A new species of go- natid squid, Gonatus middendorffi n. sp., from the northern North Pacific, with notes on morphological changes with growth and distribution in immature stages (Cephalopoda: Oegopsida). Bull. Nat. Sci. Mus., Tokyo, Ser. A 7(1):7-26. Kusoperé, T. & T. OKUTANI. 1981b. The systematics and identification of larval cephalopods from the northern North Pacific. Res. Inst. N. Pac. Fish. Hokkaido Univ., Spec. Vol.: 131-159. LEBRASSEUR, R. J. 1966. Stomach contents of salmon and steelhead trout in the northeastern Pacific Ocean. J. Fish. Res. Bd. Canada 23(1):85-100. LICHTENSTEIN, K. M.H. 1818. Onychoteuthis, sepien mit kral- len. Isis 1818:1591-1592. LONNBERG, E. 1898. On the cephalopods collected by the Swedish expedition to Tierra del Fuego, 1895-96. Svenska Expeditionen till Magellanslanderna 2(4):49-64. MIpDDENDoR FF, A. T. 1849. Beitrage zu einer Malacozoologia Rossica. I]. Aufzahlung und Beschreibung der zur Meer- esfauna Russlands gehGrigen Einschaler. Mém. Acad. Imp. Sci. St. Pétersbourg, 6me Sér., 6:329-516. Naer, A. 1921. Das System der dibranchiaten Cephalopoden und die mediterranean Arten derselben. Mitt. Zool. Stn. Neapel 22:527-542. Naer, A. 1923. Fauna und Flora des Golfes von Neapel und Page 174 The Veliger, Vol. 28, No. 2 der angrenzenden Meeres-Abschnitte. Monograph 35. Cephalopoda. 2 vols. 863 pp. (Translated by Israel Progr. Sci. Trans., 1972, Jerusalem, 917 pp.) Nests, K. N. 1972. Two new species of gonatid squids from the North Pacific. Zool. Zh. 51(9):1300-1307. OxtyaAMaA, M. 1969. A new species of Gonatopsis from the Japan Sea, with the record of a specimen referrable to Go- natopsis sp. Okutani, 1967 (Cephalopoda: Oegopsida, Go- natidae). Publ. Seto Mar. Biol. Lab. 17(1):19-32. OKUTANI, T. & T. NEMOTO. 1964. Squids as the food of sperm whales in the Bering Sea and Alaskan Gulf. Scient. Rep. Whales Res. Inst. Tokyo 18:111-122. Pearcy, W. G. 1964. Some distributional features of meso- pelagic fishes off Oregon. J. Marine Res. 22(1):83-102. Pearcy, W. G. & L. HuBBARD. 1964. A modification of the Isaacs-Kidd midwater trawl for sampling at different depth intervals. Deep-Sea Res. 11(2):263-264. Pearcy, W. G., E. E. KRyYGIER, R. MESECAR & F. RAMSEY. 1977. Vertical distribution and migration of oceanic mi- cronekton off Oregon. Deep-Sea Res. 24:223-245. Pearcy, W.G. & R.S. MESECAR. 1971. Scattering layers and vertical distribution of animals off Oregon. Proc. Int. Symp. Biol. Sound Scatt. Ocean. U.S. Naval Oc. Office MC Rept 005:381-394. Pearcy, W. G. & G. L. Voss. 1963. A new species of gonatid squid from the northeastern Pacific. Proc. Biol. Soc. Wash. 76:105-112. SANGER, G. A. & P. A. BaiRD. 1977. The trophic relationships of marine birds in the Gulf of Alaska and the southern Bering Sea. Part 14. Jn: J. C. Bartonek, C. J. Lensink, P. J. Gould, R. E. Gill & G. A. Sanger (co-principal investi- gators), Annual Report, OCSEAP RU-341. U.S. Fish. Wildl. Serv. OBS-CE. Anchorage, Alaska, 1 April 1977. SasaAkI, M. 1920. Report on the cephalopods collected during 1906 by the United States Bureau of Fisheries steamer “Al- batross” in the northwestern Pacific. Proc. U.S. Natl. Mus. 57(2310):163-203. SASAKI, M. 1923. On a new eight-armed squid from Hokkai- do, Gonatopsis borealis n. sp. Annot. Zool. Japon. 10:203- 207. SASAKI, M. 1929. A monograph of the dibranchiate cephalo- pods of the Japanese and adjacent waters. J. Fac. Agric. Hokkaido Imp. Univ. 20 Suppl. 10:1-357. Voss, G. L. 1956. A review of the cephalopods of the Gulf of Mexico. Bull. Mar. Sci. Gulf Carrib. 6(2):85-178. YounGc, R. E. 1972. The systematics and areal distribution of pelagic cephalopods from the seas off southern California. Smithsonian Contrib. Zool. 97:1-150. THE VELIGER © CMS, Inc., 1985 The Veliger 28(2):175-178 (October 1, 1985) A New Species of Hubranchus Forbes, 1838, from the Sea of Cortez, Mexico by DAVID W. BEHRENS Pacific Gas & Electric Co., Biological Research Laboratory, P.O. Box 117, Avila Beach, California 93424 Abstract. The nudibranch Eubranchus cucullus spec. nov. from the Sea of Cortez, Mexico is de- scribed. This description represents the second occurrence of the genus Eubranchus in the Sea of Cortez. THE GENUS Eubranchus forms a group composed pri- Family EUBRANCHIDAE Odhner, 1934 marily of temperate species (EDMUNDS & KREss, 1969). Eub ng laa eS During the collection of opisthobranch mollusks in the Bo ohana Sea of Cortez a new species of aeolid nudibranch belong- Eubranchus cucullus Behrens, spec. nov. ing to the genus Eubranchus Forbes, 1838, was discovered. To date the only other eubranchid nudibranch reported (Bigunes sino?) from the Sea of Cortez is Eubranchus rustyus (Marcus, Material examined: (1) Holotype: One specimen ap- 1961) (McDONALD, 1983:186). The description of a new proximately 5 mm long (preserved) collected in 10 m of species is presented here. water at Puerto Refugio, Isla Angel de La Guarda, Baja Figure 1 Dorsal view of Eubranchus cucullus spec. nov. Puerto Penasco, Sonora, Mexico. April 21, 1978. Drawn from color transparency. Page 176 The Veliger, Vol. 28, No. 2 Figure 2 Eubranchus cucullus spec. nov. Puerto Refugio, Isla Angel de La Guarda, Baja California, Mexico. Approximately 10 mm. Photograph by Jeff Hamann. Figure 3 a. Diagrammatic right lateral view of body of Eubranchus cu- cullus spec. nov.; an = anus, go = genital oriface. b. Detail of ceras of Eubranchus cucullus spec. nov. California, Mexico (Lat. 29°32’50”N; Long. 113°35’55”W) in August 1982 by Jeff Hamann. This specimen is de- posited in the collection of the California Academy of Sci- ences, Department of Invertebrate Zoology and Geology (CAS), San Francisco, California, CAS Catalogue No. 055515. (2) Paratypes: One specimen, 4 mm long (preserved) collected with the holotype is deposited in the CAS col- lection, Catalogue No. 055516. (3) One specimen, 3 mm long (preserved) collected with the holotype is also deposited in the CAS collection (Cat- alogue No. 055517). A color transparency of a living spec- imen of Eubranchus cucullus is on file at CAS. Description: The living animals were up to 10 mm long. The body is typically aeolidiform (Figures 1, 2). The foot is narrow, linear, and tapering posteriorly into a short blunt tail. The foot corners are triangular but not elon- gate. The cephalic tentacles are cylindrical with a blunt tip and slightly less than one-half the length of the rhi- nophores (Figures 1, 3a). The rhinophores are long, smooth, and tapering to a blunt tip (Figures 1, 3a). The cerata are cylindrical and irregularly inflated (Figure 3b). The liver diverticulum is nodular within each ceras. The cerata are arranged in 6 oblique rows dorsolaterally on either side of the dorsum. An example of the branchial half formula is I 5-8, II 6-7, III 6, IV 3-4, V 3, VI 2. D. W. Behrens, 1985 oe » x Page 177 a Figure 4 Radula and jaw of Eubranchus cucullus spec. nov. a. Rachidian tooth. b. Lateral tooth. c. Jaw plate. d. Masticatory edge of jaw. The first two rows are anterior to the pericardial elevation (Figure 3a). The largest cerata are closest to the midline, those at the margins being smaller. The anal pore is an- terior to the medial ceras of the third row and ventral to the pericardial elevation (Figure 3a). The genital orifice lies posteroventrally to the first ceratal row on the right side (Figure 3a). Except for the head region, rhinophores, and anterior margins of the foot, the entire body is encrusted with an opaque white pigmentation (Figure 2). The head, ce- phalic tentacles, rhinophores, and the anterior half of the foot margins are deep rust-brown. On the rhinophores this pigmentation diminishes, leaving them transparent. In some specimens, opaque white marks occur on the sides of the head and cephalic tentacles. Variable numbers of rust-brown specks and spots occur dorsomedially on the notum and subapically on the cerata. In some specimens the cnidosac appears cream colored, while in others it is transparent. The radular formula is 82 x 1.1.1. The central cusp of the rachidian tooth projects above the 3 or 4 large lateral denticles per side (Figure 4a). The lateral teeth, thin rectangular plates with a single triangular cusp di- rected toward the rachidian (Figure 4b), are typical of Eubranchus. The basal leg of the lateral tooth is extremely long and tapering, measuring from 4-5 times the height of the cusp. The jaws are narrow, tapering posteriorly (Figure 4c). The masticatory border bears about 25 con- ical denticles (Figure 4d). The penis is conical and armed with a stylet (Figure 5a). The egg mass is a white-cream colored coil of about 1% whorls attached to the substrate at the center of the whorl Page 178 The Veliger, Vol. 28, No. 2 Figure 5 a. Penis of Eubranchus cucullus spec. nov. b. Egg mass of Eu- branchus cucullus spec. nov. (All eggs not shown.) (Figure 5a). This mass consists of more whorls than that described by Hurst (1967) for Eubranchus olivaceus. The outer edge of the coil is free of egg capsules. The egg capsules are closely arranged, each containing a single egg. The egg masses collected in August 1982 averaged 2 mm in diameter and less than 0.5 mm in height. One coil was 2-3 eggs thick and 8-10 eggs wide. Egg ribbons were encountered attached to the perisarc of an unidentified plumularid hydroid. Eubranchus cucullus is known intertidally and subtid- ally to depths of 10 m. Specimens are collected most com- monly on plumularid hydroids. Localities within the northern and central Gulf of California where this species has been collected include Puerto Penasco, Sonora, Mex- ico and Puerto Refugio, Isla Angel de La Guarda, and Loreto, Baja California, Mexico. Discussion: The characteristics delineating the genus Eu- branchus are concise and well defined (EDMUNDS & KRESS, 1969). In their review of the genus, EDMUNDs & KRESS (1969) listed 24 species. ROLLER (1972) added Eubranchus sanjuanensis from the northeastern Pacific fauna. BABA (1975) described two new species from the northwest Pa- cific. and ORTEA (1979) added the most recent species from the Canary Islands. Although some taxonomic prob- lems have existed among European species, as chronicled by Edmunds & Kress, the northeastern Pacific members of the genus are clearly distinguishable. Of the 28 species known worldwide, none exhibit the striking white en- crustation over the body or the dark rust-brown head. This characteristic alone establishes Eubranchus cucullus as a distinct species. Concerning the four west American species, the greater number of rows of teeth in the radula of E. cucullus (82) is distinctive. Eubranchus misakiensis Baba, 1960, has 40-46; EF. olivaceus (O’ Donoghue, 1922), has 32-35; E. rustyus (Marcus, 1961) has 50-60; and E. sanjuanensis has 50 (ROLLER, 1972; MCDONALD, 1983). The shape of the lateral teeth and the denticulation of the masticatory edge of the jaw also separate this species as distinct. Eubranchus cucullus has a very long tapering tooth, and 25 denticles on the jaw edge, twice as many as occur in the four other west American species. The specific name cucullus, from the Latin word for “hood” or “‘cowl,” is chosen to call attention to its dark rust-brown cephalic hood. ACKNOWLEDGMENTS I am grateful to Jeff Hamann for providing me with the type material and distributional data on this species, and to Terry Gosliner for his critical review of the manuscript. LITERATURE CITED BaBA, K. 1975. On two new species of Eubranchus from Ayu- kawa, Echizen coast, Japan Sea side of middle Japan (Nu- dibranchia; Eolidoidea; Eubranchidae). Jap. J. Malacol. (Venus) 34(3-4):65-72. Epmunps, M. & A. Kress. 1969. On the European species of Eubranchus (Mollusca: Opisthobranchia). J. Mar. Biol. As- soc. U.K. 49:879-912. Hurst, A. 1967. The egg masses and veligers of thirty north- east Pacific Opisthobranchs. Veliger 9(3):255-288. McDonaLp, G. 1983. A review of the nudibranchs of the California coast. Malacologia 24(1-2):114-276. OrTEA, J. 1979. Una nueva especie de Eubranchus (Mollusca: Opisthobranchia) de Tenerife, Isla Canarias. Rev. Fac. Cienc. Univ. Oviedo (Ser. Biologia) 20-21(1979-80):169- 176. Ro.ier, R. A. 1972. Three new species of eolid Nudibranch from the west coast of North America. Veliger 14(4):416- 423. The Veliger 28(2):179-185 (October 1, 1985) THE VELIGER © CMS, Inc., 1985 The Role of Shell Geometry as a Deterrent to Predation in Terebrid Gastropods by PHILIP W. SIGNOR III Department of Geology, University of California, Davis, California 95616 Abstract. High-spired gastropod shells vary in their susceptibility to attack by durophagous (shell- destroying) predators. Slender terebrid gastropods with small apertures are damaged significantly less often than less slender species when exposed to calappid crabs. When damaged, slender terebrids are killed significantly less frequently than less slender species. The actual cause of slender terebrids’ lesser vulnerability is uncertain, but small aperture size, a geometric correlate of shell geometry, may be responsible for the observed differences. High-spired shells are probably less likely than other shell shapes to be attacked successfully by other durophagous predators. Frequencies of repair marks have been interpreted by some authors as an index of a shell’s effec- tiveness in deterring potential predators. In experiments conducted for this study, nonlethal attacks were as common on robust terebrids as on slender forms, despite the fact that a higher proportion of attacks on robust forms were successful. Comparison of repair mark frequencies obtained from labo- ratory experiments to those observed in local populations showed slender terebrids bore fewer repair marks than robust forms, although the differences were not statistically significant. These observations suggest that repair marks are not an adequate index of a turritelliform shell’s vulnerability to duro- phagous predators. INTRODUCTION PREDATION by durophagous (shell-destroying) predators is a major source of mortality among temperate and, es- pecially, tropical marine mollusks (VERMEIJ, 1977a, b, 1978, 1982a, 1983). Durophagous predators employ two fundamentally different tactics in attacking gastropod prey. One tactic is crushing and the second is peeling, where the shell is pared away leaving the snail exposed. Crush- ing is believed to predominate on hard substrates while peeling is understood to be more common on soft substrata (VERMEIJ, 1978, 1982a, 1983; PALMER, 1979), although both modes of predation are found in both types of habi- tats. Marine gastropods have evolved a variety of morpho- logical features which serve to deter durophagous preda- tors, including spines (PALMER, 1979), axial ribs (BERTNESS & CUNNINGHAM, 1981; VERMEIJ, 1982a), shortened spires (KITCHING e¢ al., 1966), thickened aper- tural lips (VERMEIJ, 1977a, 1978, 1982a), and apertures constricted by teeth or very narrow apertures (VERMEIJ, 1977a, 1978, 1979, 1982a; HUGHES & ELNER, 1979; BERTNESS & CUNNINGHAM, 1981). This paper will ex- plore further one possible consideration: the role of shell geometry as a defense against durophagous predators. More specifically, this study will focus on the interaction between terebrid gastropods and a predatory decapod crustacean, Calappa hepatica (Linnaeus). The Terebridae (Coniacea) are a family of common sand-dwelling marine neogastropods. They are limited to tropical and temperate regions and are abundant in shal- low waters throughout the Indo-Pacific region. The fam- ily first appeared in the Late Cretaceous (TAYLOR et al., 1980) and now includes approximately 150 extant species (Boss, 1971). The interrelationships of terebrid genera are not well understood but the family is believed to be derived from the Conidae or an intermediate between the Conidae and Turridae, from which the Conidae were also derived (RUDMAN, 1969). The evolution and diversifica- tion of this family are both part of a general Cenozoic radiation of predatory marine neogastropods (TAYLOR et al., 1980) and a part of the Mesozoic Marine Revolution (VERMEIJ, 1977a), wherein a variety of predation-resis- tant taxa replaced older, less resistant taxa that dominated the Paleozoic and early Mesozoic faunas. Page 180 The Veliger, Vol. 28, No. 2 Figure 1 Calappa hepatica. Note the well-developed master cheliped and the large tooth on the dactyl that is used for peeling snails. Specimen was collected at Motupore Island, Papua New Guinea; carapace length is 65 mm. The natural history of terebrids is poorly known; only three species—Hastula cinerea (Marcus & Marcus, 1960), Terebra gould: (MILLER, 1975), and Hastula incon- stans (MILLER, 1979)—have been studied in any detail. Based on this limited information, it appears that duro- phagous predators, especially the sand crab Calappa (Fig- ure 1) are important natural predators of small (<4 cm) terebrids. Calappa is distributed throughout the tropical oceans of the world and commonly co-occurs with Terebra. It at- tacks its gastropod prey by peeling (Figure 2) (SHOUP, 1968; VERMEIJ, 1982a). The crab holds the prey with the small cheliped and uses its highly modified master che- liped (Figure 1) to peel away the shell. Unlike crabs from other families (e.g., Cancer productus Randall, Eriphia se- bana [Shaw and Nodder] [ZIPSER & VERMEIJ, 1978]; Ozius verreauxit Saussure [BERTNESS & CUNNINGHAM, 1981)), Calappa does not employ peeling as a mode of attack com- plementary to others, such as clipping the spire. Instead, Calappa is exclusively a peeler. Morphological features that inhibit peeling, such as a thickened outer lip or varices, have been shown to reduce the likelihood of a successful attack by Calappa (VERMEIJ, 1982a). Another possibility is that shell geometry or a correlate of shell geometry may confer resistance to peel- ing. A high-spired shell is actually a very long, thin, tight- ly coiled and slowly expanding calcium carbonate tube. A calappid crab would need to attack the shell through a small aperture and peel a high-spired shell further than what would be necessary for shells with a more trochiform geometry. Gastropods with very slender shells can often retract deeply into their shells, as much as several whorls into the shell (VERMEIJ et al., 1980; VERMEIJ, 1982a). The small aperture of turritelliform snails may prevent Calappa from inserting its dactyl process to begin the peel- ing process. VERMEIJ et al. (1980) tested the hypothesis that slender gastropods are more resistant to peeling. They predicted that if frequencies of attack were constant among terebrid gastropods, then those shells most resistant to predation should have a higher frequency of repair marks on their shells. Surprisingly, they found that slender terebrids ac- P. W. Signor III, 1985 Page 181 Explanation of Figures 2 to 11 Figure 2. Duplicaria baileyi peeled by Calappa hepatica. The shell exterior has been peeled away leaving the columella intact. Shell length (to end of columella) is 15 mm. Figure 3. Terebra kilburni. This and specimens in Figures 4-11 are from Motupore Island, Papua New Guinea. Shell length is 24 mm. Figure 4. Terebra laevigata. Shell length is 26 mm. Figure 5. Terebra subulata. Shell length is 28 mm. tually had a lower average number of repair marks, im- plying that slender shells were more, not less, susceptible to peeling. This conclusion was supported in part by VER- MEIJ (1982a), who found that a low frequency of repair marks implied that durophages were not present in the local habitat or that most attacks were successful. The objective of this project was to provide a more direct test of the hypothesis that shell geometry or a cor- relate of shell geometry inhibits predation by peeling crabs. Also, the experiments allow comparison of actual success, Figure 6. Terebra affinis. Shell length is 22 mm. Figure 7. Terebra areolata. Shell length is 23 mm. Figure 8. Terebra columellaris. Shell length is 22 mm. Figure 9. Terebra conspersa. Shell length is 21 mm. Figure 10. Terebra dimidiata. Shell length is 47 mm. Figure 11. Terebra undulata. Shell length is 23 mm. under experimental conditions, in avoiding predation by a durophagous decapod to repair frequencies in natural populations, thus permitting direct evaluation of the ac- curacy of repair frequencies as an estimate of a shell’s effectiveness in providing protection against shell-destroy- ing predators. MATERIALS anD METHODS Specimens of Terebra were collected from shallow (3-15 m) subtidal sand patches on the northwestern side of Mo- Page 182 The Veliger, Vol. 28, No. 2 Table 1 Results of offering slender and robust species of terebrids to crabs (Calappa hepatica). n is the total number of individuals of a given terebrid species offered to the crabs. Attacks is the number of those individuals damaged by the crabs. P, is the proportion of individuals damaged by the crabs (attacks/n). Successes is the number of snails killed by the crabs. P is the proportion of successful attacks (successes/attacks). s n Attacks ey, Successes P, Slender species: Terebra kilburni 55 4 0.08 0) 0.00 Terebra laevigata 69 5 0.07 1 0.20 Terebra subulata 45 1 0.02 0 0.00 Subtotal 169 10 0.06 1 0.01 Robust species: Terebra affinis 22 3 0.14 1 0.33 Terebra areolata 15 1 0.07 1 1.00 Terebra columellaris 53 10 0.19 6 0.60 Terebra conspersa 53 10 0.19 7 0.70 Terebra dimidiata 12 3 0.25 1 0.33 Terebra undulata 46 10 0.22 5 0.50 Subtotal 201 37 0.18 21 0.57 Total 370 47 0.13 22 0.46 tupore Island (9°32’S, 147°16’E), in Bootless Bay, on the southern coast of Papua New Guinea. Individuals of over 20 species could easily be collected by sieving sand through a 3 mm mesh screen. The specimens were returned to the Motupore Island Research Center and maintained in a large water-table. Sixty-eight percent of the snails were carefully exam- ined under low-power magnification and the number of repair marks and shell length were recorded for each an- imal. Only well-defined repaired breaks that crosscut growth lines were counted. Only data for individuals ranging in size from 10 to 39 mm were used for this study (several of the species reach sizes in excess of 100 mm), in order to minimize the size bias in the numbers of re- pairs expected on any one snail (VERMEIJ et al., 1980). Calappa hepatica (Linnaeus, 1758) is abundant on sand flats at the north end of Motupore Island. The crabs are active at high tide and can be collected by hand as the tide recedes. Six crabs ranging from 41 to 65 mm in carapace width were collected and each was transferred to a run- ning seawater aquarium. Enough coarse sand was pro- vided to cover the bottom of the tank and allow the crab to burrow completely. The terebrids were divided into two shape classes: slen- der and robust. Species were assigned to the slender class according to the criteria suggested by VERMEIJ et al. (1980); slender species are those with 18 or more whorls when the shell is 20 to 29 mm in length, 20 or more whorls when the shell is 30 to 49 mm in length, or more than 21 whorls when the shell length exceeds 50 mm (Figures 3- 5). Robust shells are those with fewer whorls than re- quired for assignment to the slender class at specified lengths (Figures 6-11). The slender species offered to Ca- lappa were Terebra kilburni (Burch, 1965), T. laevigata (Gray, 1834), 7. subulata (Linnaeus, 1767), and the ro- bust species were 7. affinis (Gray, 1834), T. areolata (Link, 1807), 7. columellaris (Hinds, 1844), 7. conspersa (Hinds, 1844), 7. dimidiata (Linnaeus, 1758), and 7. undulata (Gray, 1834) (Figures 3-11). Each crab was simultaneously offered four or five (two slender and two or three robust) terebrids as potential prey. Rejected prey were not re-offered to the same crab, but undamaged individuals were occasionally re-offered to another crab. The prey were presented in the morning and remained in the crab’s tank for 24 h. Twenty-four hours was selected as the test period because it was a substantially longer time than VERMEIJ (1982a) found necessary for crab attacks but minimized disturbance within the tanks. At the end of each trial the snails were recovered and the results recorded for each individual. The snails were classified as either not damaged, damaged but not killed (shell broken but animal not harmed), or killed (shell broken and some portion of the snail eaten). Representatives of each shape class were offered each day, but only one individual of each species was placed in the tank at a time. VERMEIJ (1977b) has emphasized that there is an im- portant size component to the outcome of any encounter between gastropods and predatory crabs. When the gas- tropod reaches a relatively large size in comparison to the crab it tends to become invulnerable to the crab’s attack. Vermeij referred to this as the critical size. MILLER (1975) observed this effect in his study of Terebra gouldi, noting that individuals larger than a few centimeters were no P. W. Signor III, 1985 longer preyed upon by Calappa. To avoid a possible size bias in this experiment all the snails offered to the crabs were small in size, generally less than 35 mm. There was no apparent increase in susceptibility for small shells in the data. Calappa attacks on the terebrids occurred infrequently, sometimes raising concerns that individual crabs might not be healthy. If a particular crab failed to damage any shells for several consecutive days, a juvenile Strombus gibberulus Linnaeus was added to that crab’s tank. Healthy crabs attack §. gibberulus quickly and relatively easily (VERMEIJ, 1982a), and the snail’s demise was taken as evidence of the crab’s good condition. During the course of experiments reported here, none of the crabs became unresponsive. RESULTS Terebrids, 201 robust and 169 slender, were offered to the crabs (Table 1). Thirty-seven of the robust terebrids were damaged when exposed to Calappa and, of that num- ber, 21 were killed. Likewise, 10 of the slender terebrids were damaged but only a single individual was killed. These data can be employed to answer two separate ques- tions; first, are robust terebrids more likely than slender ones to be damaged and, secondly, once damage occurs is there a greater likelihood that damage to a robust terebrid will be fatal? More than 18% of the robust species were damaged while only 6% of the slender species had the aperture lip peeled. The frequencies of damage differ strongly from those expected if the two groups were equally likely to be attacked and damaged (x? = 10.13, P < 0.002). Slender terebrids are much less likely than robust species to be damaged when exposed to Calappa. Fewer than half of the terebrids damaged by Calappa were killed. Among the robust species, 57% of the attacks were successful. In contrast, only one of the ten (10%) slender individuals that were attacked was killed. Again, the two frequencies differ significantly from those expect- ed if the two shapes were equally likely to be killed during an attack (P < 0.02, Fisher’s Exact Test [SOKAL & ROHLF, 1981)). The sum result of these two effects is that slender te- rebrids are not very likely to be attacked successfully by Calappa. Indeed, only one individual of the 169 slender terebrids offered was killed for a frequency of kills well below 1%. This frequency is far lower than that of robust terebrids or of other gastropod species reported by VER- MEIJ (1982a). Slender terebrids collected for this project averaged 0.64 repairs per individual while robust species averaged 0.78 (Table 2). This difference was not significant (Mann- Whitney test), nor was there any significant difference between the two groups in the proportions of individuals lacking repairs altogether (chi-square test). Page 183 Table 2 Repair mark frequencies on terebrids from the Motupore Island Research Center. All data are for snails between 10 and 39 mm in length. n Repairs Frequency Slender species: Terebra kilburni 36 29 0.80 Terebra laevigata 92 60 0.65 Terebra subulata 26 10 0.38 Subtotal 154 99 0.64 Robust species: Terebra affinis 12 16 1.33 Terebra areolata 9 3 0.33 Terebra columellaris 45 31 0.76 Terebra conspersa 21 23 1.10 Terebra dimidiata 4 0 0.00 Terebra undulata 24 14 0.58 Subtotal 115 87 0.78 Total 269 186 0.69 DISCUSSION While a pattern of increased probabilities of lethal and nonlethal damage for robust terebrids is evident, the un- derlying mechanism remains unknown. The disparity in frequencies of attack among slender and robust terebrids cannot be interpreted unambiguously. The differences could arise from the way Calappa locates its prey, from prey selection by the predator once potential prey are de- tected, or from how the snails are attacked once found. Calappa searches for prey by moving over the sediment and probing for buried prey with its chelipeds (SHOUP, 1968; MILLER, 1975; personal observations). Prey with a smaller profile might be less likely to be detected. Because terebrids burrow with the long axis of the shell subpar- allel to the sediment-water interface (anterior end down), slender terebrids, by definition, will have a consistently smaller profile (see Figures 3-11). Nevertheless, this ef- fect is too small to account for the tremendous difference observed in frequencies of attack. The slender species have a cross-sectional area parallel to the shell axis at least 70% as great as the most robust species used in the experiment. This difference is not sufficient to account for the threefold difference in frequencies of attack, although it might have some effect on the overall results. Furthermore, X-radio- graphic studies of burrowing terebrids show slender te- rebrids burrow less deeply than robust forms, and should be more readily detected by the crabs (Signor, unpub- lished data). (For example, the anterior end of an 8-cm Terebra dimidiata will be 2.4 cm below the sediment sur- face, whereas the anterior end of a similar size 7. subulata will be only 1.4 cm below the surface. In both cases, the apex will lie just below the sediment surface.) Page 184 Another hypothesis is that slender terebrids are some- how less desirable food items, and are not attacked if dis- covered. This seems unlikely, as many durophagous crabs, including Calappa, are surprisingly unselective when at- tacking potential prey items, and freely attack gastropods they have virtually no chance of killing (VERME]J, 1982a, b). Carcinus maenas, while selective in its attacks on mus- sels (ELNER & HUGHES, 1978), attacks all Nucella lapillus it encounters regardless of size (HUGHES & ELNER, 1979). Crabs have even attacked plastic models of long-extinct bivalves introduced into shallow marine habitats (La- BARBERA, 1981). Calappa will eat the soft parts of slender terebrids when removed from the shell (personal obser- vations), so there is no reason to suspect the crabs find slender terebrids inedible. However, the possibility of se- lective feeding behavior by Calappa should not be dis- counted as no data are presently available to demonstrate that prey selection does or does not occur. Aperture size is a likely cause of slender terebrids’ rel- ative invulnerability to attack. Other researchers have ob- served that narrow or occluded apertures inhibit attack by predatory crabs (e.g., VERMEIJ, 1977a, 1978, 1982a; HuGHES & ELNER, 1979; BERTNESS & CUNNINGHAM, 1981). The results presented here are consistent with, but do not demand, that hypothesis. Calappa possesses a large tooth on the dactyl of the master cheliped that meshes with two protuberances on the propodus (SHOUP, 1968). Calappa uses this tooth to peel away portions of the shell lip. The small apertures of terebrids, especially the more slender species, probably restrict the insertion of the calappid peeling tooth, in much the same way that a narrow or occluded aperture restricts access. Interestingly, the only slender terebrid killed was one attacked by the smallest calappid used in the experi- ment. Regardless of the mechanism, slender terebrids are damaged or killed significantly less often when exposed to Calappa. This pattern seems to extend to comparisons between terebrids and forms with low spires. Overall, the frequencies of successful (fatal) attacks on Terebra by Ca- lappa were far below those reported for other species by VERMEIJ (1982a). Only 10% of the robust terebrids of- fered to Calappa were attacked successfully, a result close to that observed by Vermeij for Terebra affinis (9.1%) and well below frequencies obtained for Rhinoclavis aspera (16%), R. fasciata (25%) and Strombus gibberulus (33%). Terebra is also susceptible to shell crushing, the other mode of durophagous predation. Small rays are common in many of the shallow subtidal sand patches around Mo- tupore Island and prey on mollusks and crustaceans dug out of the sediments with jets of water. But shell geometry might also serve to inhibit predation by shell crushers. Shell crushing fish must take the gastropod into their mouth in order to crush the snail with their jaws or pha- ryngeal mill. A long, slender shell could likely be more difficult than other geometries for the fish to manipulate The Veliger, Vol. 28, No. 2 into the mouth for crushing. The effect of shell geometry as a defense against shell crushers has not been examined and would provide a useful complement to the results presented here. VERMEI (1982a, b) has argued forcefully that repair marks on gastropods can be employed as an index of a shell’s effectiveness as a deterrent to predation. The results presented above indicate that this generalization does not extend to high-spired snails. Robust terebrids sustained by far the higher frequency of successful attacks but also had a (not significantly) higher frequency of unsuccessful attacks (7.9% us. 5.3% for slender species). Likewise, ro- bust species have a higher frequency of repair marks than slender species in local populations, although the differ- ences are less marked than in the experimental results. (The discrepancy might reflect the activity of other du- rophagous predators in the natural environment.) The shells of slender terebrids exposed to Calappa usually es- cape damage. This phenomenon may explain why VER- MEIJ et al. (1980) were unable to use repair marks to substantiate their hypothesis that slender terebrids are less susceptible to peeling predators. CONCLUSION When exposed to the durophagous predator Calappa, slender, many whorled terebrids are damaged significant- ly less frequently than robust species. When Calappa at- tacks and damages a terebrid, the damage is fatal signif- icantly less often in slender species. Robust species also suffer non-lethal damage as often as slender terebrids. These results support the hypothesis that shell geometry, or a correlate of shell geometry, can be an effective escape from predators. Repair marks are not a valid index of shell vulnerability in terebrid gastropods. ACKNOWLEDGMENTS I thank Dr. G. J. Vermeij, J. Pandolfi, and B. Onken for comments on previous versions of this paper. Two critical reviewers also contributed suggestions. Dr. N. Polunin and K. Severin provided invaluable support in the field. J. Fay assisted in collecting repair mark data, and E. Sears and Dr. B. Steene assisted in conducting the pre- dation experiments. This work was funded in part by the University Research Expeditions Program of the Univer- sity of California and a University of California Faculty Research Grant. Acknowledgment is made to the donors of the Petroleum Research Fund, administered by the American Chemical Society, for partial support of this research. LITERATURE CITED BERTNESS, M.D. & C. CUNNINGHAM. 1981. Crab shell-crush- ing predation and gastropod architectural defense. J. Exp. Mar. Biol. Ecol. 50:213-230. P. W. Signor III, 1985 Boss, K. J. 1971. Critical estimate of the number of Recent Mollusca. Occ. Pap. Mol. 3:81-135. ELNER, R. W. & R. N. HUGHES. 1978. Energy maximization in the diet of the shore crab, Carcinus maenas. J. Animal Ecol. 47:103-116. HuGuHeEs, R. N. & R. W. ELNER. 1979. Tactics of a predator, Carcinus maenas, and morphological responses of the prey, Nucella lapillus. J. Anim. Ecol. 48:65-78. KITCHING, J. A., L. Muntz & F. J. EBLING. 1966. The ecol- ogy of Lough Ine XV. The ecological significance of shell and body form in Nucella. J. Anim. Ecol. 35:113-126. LaBarBERA, M. 1981. The ecology of Mesozoic Gryphea, Ex- ogyra, and Ilymatogyra. Paleobiology 7:510-526. Marcus, E. & E. Marcus. 1960. On Hastula cinerea. Bol. Fac. Fil. Cien. Letr. Univ. S. Paulo (Zool.) 23:25-66. MILLER, B. A. 1975. The biology of Terebra gould: Deshayes, 1859, with a discussion of life history similarities among other terebrids of similar proboscis type. Pacific Sci. 29:227- 241. MILLER, B. A. 1979. The biology of Hastula inconstans (Hinds, 1844) and a discussion of life history similarities among other hastulas of similar proboscis type. Pacific Sci. 33:289- 306. PALMER, A. R. 1979. Fish predation and the evolution of gas- tropod shell sculpture: experimental and geographic evi- dence. Evolution 33:697-713. RupMaANn, W. D. 1969. Observations of Pervicacia tristis (De- shayes, 1859) and a comparison with other toxoglossan gas- tropods. Veliger 12:53-64. SHouP, J.B. 1968. Shell opening by crabs of the genus Calap- pa. Science 160:887-888. Page 185 SOKAL, R. R. & F. J. ROHLF. 1981. Biometry. 2nd ed. W. H. Freeman and Co.: San Francisco. 859 pp. Tayor, J. D., J. N. Morris & C. N. Taytor. 1980. Food specialization and the evolution of predatory prosobranch gastropods. Palaeontology 23:375-409. VERMEIJ, G. H. 1977a. The Mesozoic marine revolution: evi- dence from snails, predators and grazers. Paleobiology 3: 245-258. VERMEIJ, G. J. 1977b. Interoceanic differences in vulnerabil- ity of shelled prey to crab predation. Nature 260:135-136. VERMEIJ, G. J. 1978. Biogeography and adaptation: patterns of marine life. Harvard Univ. Press: Cambridge. 332 pp. VERMEI, G. J. 1979. Shell architecture and causes of death of micronesian reef snails. Evolution 33:686-696. VERMEIJ, G. J. 1982a. Gastropod shell form, breakage, and repair in relation to predation by the crab Calappa. Mala- cologia 23:1-12. VERMEY, G. J. 1982b. Unsuccessful predation and evolution. Amer. Natur. 120:701-720. VERMEIJ, G. J. 1983. Shell-breaking predation through time. Pp. 649-669. In: M. J. S. Tevesz & P. L. McCall. Biotic interactions in Recent and fossil benthic communities. Ple- num. Publ. Co.: New York. VERMEI, G. J., E. ZIPSER & E. C. DUDLEY. 1980. Predation in time and space: peeling and drilling in terebrid gastro- pods. Paleobiology 6:352-364. ZIPSER, E. & G. J. VERMEIJ. 1978. Crushing behavior of tropical and temperate crabs. J. Exp. Mar. Biol. Ecol. 31: 155-172. The Veliger 28(2):186-194 (October 1, 1985) THE VELIGER © CMS, Inc., 1985 Gametogenesis in a Population of the Hard Clam, Mercenaria mercenaria (Linnaeus), in North Santee Bay, South Carolina! JOHN J. MANZI, M. YVONNE BOBO, ano VICTOR G. BURRELL, Jr. Marine Resources Research Institute, P.O. Box 12559, Charleston, South Carolina 29412 Abstract. Adult hard clams, Mercenaria mercenaria (Linnaeus, 1758), were sampled monthly be- tween December 1977 and February 1979 and semi-monthly from March to June 1981, from subtidal populations in North Santee Bay, South Carolina. Gonad development was monitored using standard histological methods and resulting slides were examined with light microscopy at 100 and 400~x. Observed gametogenic progression was best categorized by five stages or phases of development: inactive, ripe, spawning, partially spent, and spent. Both male and female clams displayed a complex progression of gametogenesis. Gonadal tissue was not uniformly dominated by clearly defined, distinct stages. Instead, gonads routinely exhibited several stages simultaneously and progressed through slow shifts in domination of stages in gonad tissue. Spawning in the population occurred continuously for six months (May to October) with at least two apparent peaks of spawning activity in the summer months. INTRODUCTION HARD CLAM, Mercenaria mercenaria (Linnaeus, 1758), landings in South Carolina have increased substantially in recent years. An estimated 6809 acres (2756 ha), or roughly 1% of South Carolina marsh-estuarine area of 746,445 acres (302,086 ha), contain clams in various com- mercial densities (ANDERSON et al., 1978). As the demand on the fishery and subsequent pressure on the resource continue to grow, it becomes important to determine re- cruitment potential of the stocks. The documentation of gametogenesis in a fishery resource is the first logical step in estimations of population recruitment. Although data are scant on the reproductive cycle in hard clam popula- tions of South Carolina, a number of studies on the go- nadal development of clams from other areas have been described. LOOSANOFF (1937a, b) determined the seasonal gonadal changes of M. mercenaria from Long Island Sound and showed that temperature is an environmental factor regulating the gonadal cycle in hard clams. PORTER (1964) studied clams from Core Sound, North Carolina, and sug- gested that gonadal differences in populations could be caused by racial differences or by phenotypic responses to ‘Contribution No. 197, South Carolina Marine Resources Center. variable environmental factors. KECK et al. (1975) com- pared the gonadal cycles of M. mercenaria from Delaware Bay and Cape Henlopen for evidence of physiological races. They found that divergent developmental patterns did ex- ist between the two areas. EVERSOLE et al. (1980) docu- mented the gametogenic cycle of M. mercenaria seed from North Carolina planted in an estuary near Clark Sound, South Carolina. PLINE (1984) compared gametogenesis in two size classes of M. mercenaria in Georgia. Differences in gonadal cycles between geographically separated populations are not limited to the hard clam (Mercenaria mercenaria) as shown by PFITZENMEYER (1962), Ropes & STICKNEY (1965), and SHAW (1962, 1965) for the soft clam, Mya arenaria (Linnaeus, 1758); HOLLAND & CHEW (1974) for the manila clam, Venerupis japonica (Deshayes); and JONES (1981) and THOMPSON e¢ al. (1980) for the ocean quahog, Arctica islandica (Linnaeus, 1767). Hard clam populations in South Carolina are subjected to environmental conditions that significantly differ from those characterizing the northeastern and Middle Atlantic states. Water temperatures are relatively moderate in win- ter (normally =10°C monthly means) and very warm in summer (228°C monthly means). This study was initi- ated to determine the natural cycle of gonadal develop- ment in native populations of South Carolina clams and to compare the results with those of previous studies on hard clam populations from other areas. J. J. Manzi et al., 1985 SANTEE RIVERS MURPHY ISLAND Page 187 CAT ISLAND Figure 1 Sampling location in North Santee Bay, South Carolina. MATERIALS ann METHODS Twelve hard clams were collected monthly from Decem- ber 1977 to February 1979 and biweekly from March to June 1981. All animals were from North Santee Bay, South Carolina (Figure 1). The bay is characterized by substrates of soft mud mixed with shell and an average depth of approximately 1-2 m at mean low water. The North Santee Bay was chosen as a study site because of its dense beds of hard clams and oysters (Cvassostrea vir- ginica Gmelin, 1791). ANDERSON et al. (1978) found the highest density hard clam populations in South Carolina in the North Santee delta system. Hard clams for this study were obtained with a 19-cm x 53-cm boxed-type oyster dredge. Hydrographic data were collected coinci- dentally with sampling and included measurements of air and water temperatures, and salinity. All clams were re- turned to the laboratory and stored under refrigeration at approximately 5°C. Tissue samples were always removed within 24 h of collection. Before dissection, shell lengths (anterior-posterior axis) were measured to the nearest millimeter with vernier calipers. Whole clams were fixed in FAA (formalin-acetic acid- alcohol) for two to four weeks. Either a cross- or longi- tudinal section was cut through the mantle, gonad, and underlying digestive gland. Dissected tissues were then placed in FAA for 24 h and washed in running tap water for approximately 4 h. Tissues were prepared for section- ing by dehydration in alcohol, clearing in toluene, and infiltration in 57°C paraffin (PREECE, 1972). After proper embedding in paraffin, the tissues were sectioned at 7 um on a rotary microtome. Sections were made at three areas of each tissue block, approximately 20 um apart. All sec- tions were stained with Gill’s hematoxylin and counter- stained with alcoholic eosin. The sections were examined with a light microscope and photomicrographs of various stages of gametogenesis were taken at 40, 75, and 185x. Examination of slides made it quite evident that a ga- metogenic index had to be devised to organize or catego- rize the developmental stages in this study. The slides were examined first under low power (100) to scan the entire gonadal area and under high power (400) to as- sess each follicle. Often, two or more stages occurred si- multaneously within each section; therefore, stage criteria decisions were based upon the condition of the majority of the section. In most cases 80% or more of the section represented no more than one stage. This technique was Page 188 particularly useful in staging the spawning and partially spent categories which were segregated on the basis of percent lumen filled with ovocytes and spermatocytes (representing spawning activity) and the number of ma- ture gametes remaining. RESULTS Initial surveys of collected gonadal tissues indicated that only five readily identifiable stages of gametogenesis were apparent in the populations of South Carolina Mercenaria mercenaria sampled. PORTER (1964) and KECK et al. (1975) used 14 and 10 developmental stages, respectively, to de- scribe the gametogenic cycle of M. mercenaria. EVERSOLE et al. (1980), however, also classified the gonads of M. mercenaria into five developmental stages. In contrast to the latter, the degree of spawning that we observed in this study made classifying the gonads into the single stage of “ripe and spawning” difficult. Staging gonads as early active and active was also difficult. This necessitated the formulation of an index suitable for the prolonged ripe and spawning period and reduced inactive period char- acterized by these hard clams. The five developmental stages (spent, inactive, ripe, spawning, and partially spent) established were distinguished by the following character- istics. Spent (Figures 2, 3): Follicles are nearly empty, and a thin band of spermatocytes or few small ovocytes may occur along the follicle wall. Few large undischarged ovo- cytes or spermatozoa and some spermatids are found free in the lumen of some follicles. Usually follicles are ex- tended but may occasionally be compressed in shape or size. Undifferentiated cells may be present in a few fol- licles. Inactive: Gonads are in a state of quiescence, and follicles are either absent or very few in number. A few follicles contained numerous undifferentiated cells, but no recog- nizable primary or secondary gametogonia are present. Ripe (Figures 4, 5): Follicles are extended and full of darkly staining ova or spermatozoa with their tails point- ing toward the center of the lumen. Large ovocytes have been freed from the follicle wall into the lumen, with the nucleus apparent in most cells and the nucleolus apparent in some. Some attached spermatocytes and ovocytes are located along the follicle wall. Spermatids, staining lighter than the spermatocytes or spermatozoa, lie in a layer or in groups outside the central core of the spermatozoa. Spawning (Figures 6, 7): Generally large expanded fol- licles still contain many mature ova in the lumen or dense bands of ripe spermatozoa surrounding a partially empty lumen. Around the periphery of the follicles may be many spermatids, spermatocytes, or ovocytes still attached to the follicle wall. Partially spent (Figures 8, 9): Most follicles are empty The Veliger, Vol. 28, No. 2 of spermatozoa or of large ovocytes but generaly have a thin band of spermatocytes or small ovocytes along the follicle wall. Other follicles are still dense with sperma- tozoa or ripe ova scattered in the lumen. The progression of gametogenesis in female clams as reflected by this study was not characterized by clearly defined, distinct stages. Instead, the gonads often exhibited several stages simultaneously and progression of the sex- ual cycle occurred in a gradual but complex manner. Fig- ure 11 illustrates this progression in oogenesis throughout the study period. Approximately 80% of the female gonads examined from late December 1977 samples were spent. Gonads with partly discharged ovocytes and others with follicles filled with ripe ova in the largest part of the lumen appeared in February and continued until May. By the end of May, as the water temperature increased to 22°C, all females examined were ripe and normal oval-shaped ovocytes had reached a size of 65-70 um. Indications of spawning were evident in late June (water temperature, 26°C) with a distinct decrease in the number of ripe ova and an increase of gonads with partially spent lumen. This spawning condition continued through Oc- tober showing a gradual shift from a ripe appearance to a partially spent condition. LOOSANOFF (1937b) suggested that the spawning of an individual hard clam is not com- pleted in one attempt but is extended for a certain period of time, depending upon the individual and ecological pe- culiarities. Some follicles with few undischarged ovocytes free in the lumen were characteristic of all of the females examined in November. This completely spent condition was not evident in December when ripe ova reappeared in half of the gonads examined. The percentage increase in ripe females in December, January, and February could indicate that active gametogenesis occurred immediately after the spawning. The new ovocytes at this time varied considerably in size and averaged 55-60 um. In most cases, the ova were oval in shape with the nucleus clearly ap- parent. Samples taken biweekly, March through June 1981, are combined into monthly means in Figures 10-12. Most female gonads (Figure 11) were marked with a spawning or partially spent condition. The number of large un- spawned ova found in the ovaries of these clams varied greatly with individuals. In some cases, almost all the ova were discharged with only a few ripe ovocytes remaining (Figure 8). In most, however, a comparatively large num- ber of ova were retained (Figure 6). Ovocytes at this time averaged 55-60 wm in size. Some small ovocytes were observed in March but these were relatively sparse. Ovo- cytes greater than 68 um in size had increased in number by May and June, but the majority of the ova were still between 55 and 60 um. Male clams reiterated the complex progression of ga- metogenesis exhibited by the females. Figure 12 illus- trates, by percent of sample population in various sexual stages, this progression of the male sexual cycle. In Feb- J. J. Manzi et al., 1985 Explanation of Figures 2 to 5 Figure 2. Spent stage of oogenesis (x 40). Figure 3. Spent stage of spermatogenesis (x75). ruary, gonads in 85% of the males examined were ripe, with most follicles containing mature spermatozoa in the lumen. A decline in the percentage of ripe males occurred in March when approximately half of the gonads exam- ined were ripe while all of the others were spawning. In April, all male gonads were ripe with follicles packed with darkly stained spermatozoa with their tails oriented in- wards toward the center of the lumen. A mature condition continued in the majority of the males through October with only a small percentage in a spawning or partially spent condition. By November when water temperatures Figure 4. Ripe female (x75). Figure 5. Ripe male (x40). were approximately 18°C, all male gonads examined had discharged their gametes and were in a generally spent condition. Males exhibiting extended follicles filled with ripe spermatozoa were apparent again in December, and a high percentage of ripe males remained through Feb- ruary. Water temperatures recorded during the study were not significantly different from typical South Carolina coastal conditions. Water temperatures ranged from a low of 5.3°C in February 1978 to a high of 30.2°C in August 1978. Ambient water temperatures recorded at sample collection Page 190 Jo oe fay a oo 4G » “ The Veliger, Vol. 28, No. 2 * (i 5 te Explanation of Figures 6 to 9 Figure 6. Spawning female, with many ova retained (x75). Figure 7. Spawning male, with many sperm retained (x40). are presented in Figure 10. Available salinity records in- dicate a salinity range of 24.0 to 34.0%o over the collection period. The seasonal variation of the gonadal cycle is presented in Table 1. Data are presented as percentage of individ- uals in each developmental stage per season. The table indicates that during the winter of 1977-1978 clams were undergoing all stages of gametogenesis, although a high percentage (36%) were ripe. The largest percentage (27%) of inactive clams were observed during this season. By the summer, ripe gonads were present in 69% of the popu- lation examined. All other clams were spawning or par- Figure 8. Partially spent female (x75). Figure 9. Partially spent male (x 185). tially spent. The fall season of 1978 was characterized by the greatest percentage of spent individuals. Follicles con- taining ripe gametes were apparent in most clams exam- ined during the winter of 1978-1979. The spring and early summer of 1981 were represented by a high pro- portion of ripe clams, reiterating the observations of 1978. DISCUSSION Temperature is a major environmental factor regulating gametogenesis in a variety of marine bivalves (LOOSANOFF, 1937a, b; GrESE, 1959; ANSELL, 1961; GALTSOFF, 1964; J. J. Manzi et al., 1985 PERCENT MONTHS D [? WATER TEMP.(°C) 6.6 MALL Uta Page 191 LL L] spent G4 inactive Bg RIPE SS spawnine ES) part. spent M A J J Ss (eo) N D J F M A M J 11.0 18.6 21.6 266 281 29.0 26.9 20.9 183 14.2 9.2 11.9 17.6 208 28.2 Figure 10 Stages of gonad development in Mercenaria mercenaria from North Santee Bay, South Carolina. Shaded areas represent percent frequency of clams in each stage from December 1977 to February 1979 (see legend). March to June 1981 data represent composites of biweekly samples. Temperatures are ambient recorded water temperatures at time of sampling. PORTER, 1964; KEcK et al., 1975; EVERSOLE et al., 1980; PLINE, 1984). The seasonal temperature differences that exist between Long Island Sound, Delaware Bay, Geor- gia, and North and South Carolina coastal waters would Table naturally lead to differences in their respective bivalve population reproductive cycles. Local environmental con- ditions can also influence and confound gametogenesis in bivalves. CARRIKER (1961) showed that depth of water il Seasonal variation (as percentiles) in developmental stages of gonads from a hard clam, Mercenaria mercenaria, population in North Santee Bay, South Carolina. Seasons Winter (1977-1978) Dec.-Feb. Ze Spring Mar.-May 34 Summer Jun.-Aug. 35 Fall Sep.-Nov. 33 Winter (1978-1979) Dec.—Feb. 36 Spring/Summer (1981) Mar.-Jun. 90 Number examined Spent 18 33 Stage Partially Inactive Ripe Spawning spent 27 36 5 14 74 21 6 69 26 6 12 30 9 15 75 19 6 56 29 14 Page 192 1977 1978 1004 rr 804 60 FEMALES PERCENT dhe Veligers Vola2s Now, = © NX oO = {75 mm SL). KECK et al. (1975) found no case where it was impossible to determine sex, indicating a paucity of in- active specimens in their survey. LOOSANOFF (1937b) did not report any undifferentiated gonads for Long Island male clams and mentions the presence of undifferentiated cells along the inner walls of the ovarian follicles only immediately after spawning. PLINE (1984), comparing lit- tlenecks (38-68 mm SL) to chowders (>78 mm SL), in- dicated that there was evidence that chowders ripen more quickly than littlenecks. He also observed that chowders had longer and more extensive spawning periods and a shorter redevelopment period than littlenecks. Active gametogenesis also appears to continue after spawning since so many ripe gonads were found in De- cember and January. LOOSANOFF (1937b) reported major redevelopment immediately after spawning in Long Is- land waters, and observed many ripe gonads in December and January. PORTER (1964) reported that at least 50% of his clams retained a ripe appearance through the fall and winter. EVERSOLE et al. (1980) reported an increase in the number of ripe and spawning clams in December, March, and April collections, and suggested that this in- dicated that regeneration occurred after fall spawning and continued into spring. EVERSOLE et al. (1980) observed differences in shell length between sexes in a young population of Mercenaria mercenaria. Females appeared larger than males and males were larger than undifferentiated clams. However, they speculated that as clams in a cohort continued to grow and entered subsequent breeding periods, these size dif- ferences should become less apparent. ANSELL (1961) found no significant size difference between male and female clams. There was no evidence of a size-sex relationship among the North Santee Bay chowders examined. Shell lengths ranged from 54 to 109 mm with approximately equal numbers of males and females in all size classes. Ova, approximately 55-60 wm in diameter, were ob- served in the follicles of female clams during most of the sampling period. Although this dimension appeared to be the average size of the ovocytes, smaller and less numerous ovocytes (25-30 wm) appeared in small numbers during late winter and early spring. Occasionally, during the summer, large ova reaching the previously reported (LOOSANOFF & DAVIS, 1963) maximum size of 70-73 um were seen. LOOSANOFF (1937b) found large ovocytes mea- suring 55-60 um in almost every gonad studied. He in- dicated that they represented cells of previously developed gonad tissue. PLINE (1984) also noticed large ovocytes (50- 60 wm), which he suggested were residual egg cells from previous ovogenic cycles, in gonads that were in an early Page 194 active phase. BRICELJ & MALOUF (1980) indicated that mature ovocytes can vary widely in size, from about 50 to 97 wm. Hard clams from North Santee Bay, South Carolina exhibited no progression of well-defined stages of gonad development. Instead, there were gradual shifts in game- togenesis with two or more stages usually present in the same gonad simultaneously. PLINE (1984) found many male littlenecks showing a considerable overlap of devel- opmental phases among follicles within a single gonad. There also appeared to be no time at which the hard clams in the North Santee Bay population were truly “inactive.” Samples contained some clams in various stages of ga- metogenesis throughout the year, although the percent of the population undergoing active gametogenesis changed significantly. This condition is not unique to South Car- olina adult clams. LOOSANOFF (1937b) found no undiffer- entiated gonads in hard clams from Long Island Sound, and KECK et al. (1975) and PLINE (1984) found only small percentages of hard clams in a recuperative phase in Del- aware Bay and Wassaw Sound, Georgia, respectively. The occurrence of morphologically ripe sperm and ova throughout most of the year in North Santee Bay is an interesting feature of the sexual behavior of local Merce- naria mercenaria. LOOSANOFF (1937b) commented that the sexual cycle of the hard clam was not in phase with other bivalve mollusks in Long Island Sound. Results from this study substantiate the suggestion that M. mercenaria ex- hibits an unusual cycle of gonadal development. In South Carolina the cycle is further confounded by an extremely long spawning season and abbreviated periods of early active gametogenesis. ACKNOWLEDGMENTS We thank the following members of the Marine Re- sources Research Institute for their assistance: Ms. Nancy Beaumont for typing the manuscript; Ms. Karen Swanson for her preparation of the figures; Mr. George Steele for his assistance in preparing the photographs; Mr. Joe Car- son and Ms. Josie Williams for their assistance in the field collections; and Ms. Nancy Hadley for her critical review of the manuscript. We also thank Dr. Pete El- dridge of the NMFS-Charleston Laboratory, for his con- structive criticism. LITERATURE CITED ANDERSON, W. D., W. J. KeirH, F. H. Miuus, M. E. BaAILEy & J. L. STEINMEYER. 1978. A survey of South Carolina’s hard clam resources. South Carolina Wildlife and Marine Resources Dept., S.C. Marine Resources Center Tech. Rep. #32. Charleston, S.C. 32 pp. ANSELL, A. D. 1961. Reproduction, growth and mortality of Venus striatula (da Costa) in Kames Bay, Millport. J. Mar. Biol. Assoc. U.K. 41:191-215. ihe Veligers Volk 233NoeZ ANSELL, A. D., K. F. LANDER, J. COUGHLAN & F. A. LOOSMORE. 1964. Studies on the hard shell clam, Venus mercenaria, in British waters. Growth and reproduction in natural and experimental colonies. J. Appl. Ecol. 1(1):63-82. BRICELJ, V.M. & R. E. MALour. 1980. Aspects of reproduc- tion of hard clams (Mercenaria mercenaria) in Great South Bay, New York. Proc. Natl. Shellfish. Assoc. 70:216-229. CaRRIKER, M. R. 1961. Interrelation of functional morphol- ogy, behavior, and autecology in early stages of the bivalve, Mercenaria mercenaria. J. Elisha Mitchell Sci. Soc. 77(2): 168-241. EVERSOLE, A. G., W. K. MICHENER & P. J. ELDRIDGE. 1980. Reproductive cycle of Mercenaria mercenaria in a South Car- olina estuary. Proc. Natl. Shellfish. Assoc. 70:22-30. GALTSOFF, P. S. 1964. The American oyster Crassostrea vir- ginica Gmelin. U.S. Fish & Wildl. Serv. Fish. Bull. 64:1- 480. GiEsE, A. C. 1959. Comparative physiology: annual repro- ductive cycles of marine invertebrates. Ann. Rev. Physiol. 21:547-576. V. E. Hall (ed.). Ann. Rev., Inc.: Palo Alto, Calif. HOLLanD, D. A. & K. K. CHEW. 1974. Reproductive cycle of the manila clam (Venerupis japonica), from Hood Canal, Washington. Proc. Natl. Shellfish. Assoc. 64:53-58. Jones, D. S. 1981. Reproductive cycles of the Atlantic surf clam Spisula solidissuma, and the ocean quahog Arctica islan- dica off New Jersey. J. Shellfish Res. 1:23-32. Keck, R. T., D. MAURER & H. LinD. 1975. A comparative study of the hard clam gonad developmental cycle. Biol. Bull. (Woods Hole) 148:243-258. Loosanorr, V. L. 1937a. Development of the primary gonad and sexual phase in Venus mercenaria Linnaeus. Biol. Bull. (Woods Hole) 72:389-405. LoosanorF, V. L. 1937b. Seasonal gonadal changes of adult clams, Venus mercenaria (L.). Biol. Bull. (Woods Hole) 72: 406-416. Loosanorr, V. L. & H. C. Davis. 1963. Rearing of bivalve molluscs. Pp. 1-236. In: F. S. Russell (ed.), Advances in marine biology, Vol. 1. Academic Press: New York. PFITZENMEYER, H. T. 1962. Periods of spawning and setting of the soft-shelled clam Mya arenaria, at Solomons, Mary- land. Chesapeake Sci. 3:114-120. PLINE, M. 1984. Reproductive cycle and low salinity stress in adult Mercenaria mercenaria L. of Wassaw Sound, Georgia. Master’s Thesis, Georgia Institute of Technology. PorTER, H. J. 1964. Seasonal gonadal changes of adult clams, Mercenaria mercenaria (L.), in North Carolina. Proc. Natl. Shellfish. Assoc. 55:35-52. PREECE, A. 1972. A manual for histologic technicians. Little, Brown and Co.: Boston. 428 pp. Ropes, J. W. & A. P. STICKNEY. 1965. Reproductive cycle of Mya arenaria in New England. Biol. Bull. (Woods Hole) 128(2):315-317. SHAW, W.N. 1962. Seasonal changes in female soft-shell clams, Mya arenaria, in the Tred Avon River, Maryland. Proc. Natl. Shellfish. Assoc. 53:121-132. SHaw, W. N. 1965. Seasonal gonadal cycle of the male soft- shell clam, Mya arenaria, in Maryland. U.S. Fish & Wildl. Serv., Spec. Sci. Rept., Fish. No. 508:1-5. TuHompson, I., D. S. JONEs & J. W. Ropes. 1980. Advanced age for sexual maturity in the ocean quahog Arctica islandica (Mollusca: Bivalvia). Mar. Biol. 57:35-39. The Veliger 28(2):195-199 (October 1, 1985) THE VELIGER © CMS, Inc., 1985 Surficial Shell Resorption in Nautilus macromphalus Sowerby, 1849 PHILIP W. SIGNOR III Department of Geology, University of California, Davis, California 95616 Abstract. Like many prosobranch gastropods, Nautilus macromphalus Sowerby, 1849, resorbs a thin layer of shell material from the surface of its penultimate whorl prior to extending the edge of the black layer during growth. The resorption occurs along a millimeter wide band skirting the edge of the black layer and apparently is accomplished by the mantle edge. The depth of resorption is uneven within and between individuals. This is the first report of shell resorption among the extant Cephalo- poda. Surficial resorption in cephalopods must have evolved independently from the gastropods but is postulated to serve a common function in both groups: to provide a fresh surface to which new shell material may be attached. INTRODUCTION RESORPTION IS a normal component of shell growth in many marine gastropods. In addition to the well-known examples of resorption to remove obstacles to further growth, such as varices on muricid gastropods (CARRIKER, 1972), or to enlarge the shell interior, as occurs in Conus (KOHN et al., 1979), more subtle patterns of resorption are common. Many prosobranch gastropods resorb a thin sheet of shell material from the surface of the penultimate whorl (Gray, 1833; SIGNOR, 1982). The resorption is lim- ited to an area in the parietal region as wide as, or slightly wider than, the shell surface the body whorl will cover after further growth. Surficial resorption removes very little shell material; in Terebra dimidiata (Linnaeus, 1758) (Neogastropoda: Terebridae) the total thickness of the re- sorbed material is only a few micrometers. One hypoth- esized function of surficial resorption is to provide a fresh surface upon which new shell material can be deposited (SIGNoR, 1982). If the foregoing hypothesis is correct, one should expect surficial resorption to occur in the other extant group of multiwhorled mollusks with conjoined whorls, the Nau- tiloida. In secreting its shell, Nautilus must contend with the same constructional problems encountered by gastro- pods. Both must securely attach new shell material to the surface of the conch without detracting from the shell’s structural integrity. Furthermore, because the first nau- tiloids were cyrtoconic, surficial resorption must, if pres- ent, have evolved independently from the Gastropoda or any common ancestor. In this perspective, Nautilus is an ideal comparison group for testing the structural integrity hypothesis for the functional significance of surficial re- sorption. Shell resorption has not been described previously in Nautilus, and surficial resorption has not, to my knowl- edge, been observed among extant Cephalopoda. Shell re- sorption in the Nautiloida has been postulated previously (TASNADI-KUBACSKA, 1962) but only in the context of decollation of primitive nautiloids. The Nautilus shell is planispiral, involute, and consists of about three whorls in mature specimens (Figure 1). Microstructurally, the outer shell wall is comparatively simple, consisting of three aragonitic layers (BOGGILD, 1930; GREGOIRE, 1962; STENZEL, 1964; ERBEN et al., 1969; MutTVvEI, 1972). The outer shell layer (often referred to as the porcellaneous layer) is composed of irregular pris- matic crystals while the thicker, middle layer is nacreous. The thin inner prismatic layer consists of small prisms oriented perpendicular to the shell surface (ERBEN e¢ ai., 1969). The septal microstructure is more complex but is also primarily nacreous (B@GGILD, 1930; GREGOIRE, 1962; STENZEL, 1964; ERBEN ef al., 1969; MuTVEI, 1972). Growth is determinate (COWEN ef al., 1973); the animal ceases to deposit the characteristic irregular color bands in the last half whorl of growth and the final septa are closely approximated. Nautilus deposits a dull black organic film above the dorsal region of the aperture (Figure 1). The origin of this layer is uncertain but it is apparently deposited by Page 196 The Veliger, Vol. 28, No. 2 P. W. Signor III, 1985 the mantle edge (JOUBIN, 1892; STENZEL, 1964). Once deposited, the black layer remains unbroken until it is covered by nacre. Shell resorption, if present, must occur along the dorsal perimeter of the black layer, where the mantle edge rests during life. MATERIALS anp OBSERVATIONS Live specimens of Nautilus macromphalus Sowerby, 1849, were collected by Peter D. Ward near Noumea, New Caledonia. The animals were removed from the shells and the clean shells returned to the University of California, Davis, for further study. Only immature specimens were examined in the scanning electron microscopy phase of this study. If surficial resorption exists in Nautilus and occurs in conjunction with growth, as in gastropods, there would be no reason to expect evidence of resorption in adults, where growth has ceased. Immature individuals were initially identified by size and color pattern; this characterization was later confirmed by a lack of approx- imated terminal septae when specimens were sectioned. The specimens were prepared for examination by scan- ning electron microscopy (SEM) by cutting free centi- meter-square size pieces of the shell exterior. Most pieces were cut so as to center the boundary between the black layer and the unmodified surface of the penultimate whorl but other areas of the shell surface were also examined. The specimens were cleaned with ethanol and an ultra- sonic bath, mounted on SEM stubs, and then sputter- coated with gold/palladium and examined under SEM. Visual examination of Nautilus macromphalus reveals that the edge of the black layer is usually slightly lower than the unmodified shell surface. (This is often most easily detected by running a fingernail over the boundary between the black layer and the unmodified shell surface.) The degree of offset varies from individual to individual, in some cases appearing flush and in others having the black layer visibly below the unmodified shell surface. The offset cannot reflect the presence of a low growth line, because the edge of the black layer is not congruent with growth lines. The only possible explanation for the black layer lying below the level of the shell surface over Explanation Figure 1. Immature Nautilus macromphalus from New Caledo- nia. Photo courtesy of P. Ward. Scale bar is 3.4 cm. Figure 2. Shell surface of Nautilus macromphalus. Illustrated area is from flank of body chamber. Arrow indicates position of small, irregular cuesta interpreted here as growth line. Scale bar is 5 ym. Figure 3. Boundary between unmodified shell surface and re- sorbed area on flank of penultimate whorl. Arrow indicates po- Page 197 which it is extending is that shell resorption occurs before the advance of the black layer. Examination of specimens cut perpendicular to the boundary of the black layer shows that the thin outer shell layer, the porcellaneous layer, extends under the black layer. Therefore, resorption cannot remove more than about 0.18 mm of shell material, the approximate thick- ness of the porcellaneous layer. Measurement of the thick- ness of the outer shell layer, using an ocular micrometer, shows the portion under the black layer is approximately 0.12 mm, or averages about two-thirds the thickness of the uncovered portion of the outer shell layer. Resorption prior to deposition of the black layer is the only plausible explanation for this reduction in the thickness of the por- cellaneous layer. Under low-power optical magnification the normal shell surface appears vitreous. Along the edge of the black layer the shell surface has a hazy luster, suggesting that some modification of the shell surface has occurred. Under scanning electron microscopy, the shell surface of Nautilus macromphalus has an irregular, pitted appear- ance (Figure 2). Fine growth lines are visible as uneven cuestas, apparently formed when new growth extends the shell from beneath the previous shell edge. Small pits are scattered unevenly over the surface and are densely con- centrated in some areas. These concentrations usually fall along growth lines or where the shell apparently was damaged. The pits can approach one micrometer in size but most are less than half that diameter. The origin of the pits is unknown; one possibility is that they are pro- duced by an endolithic organism, perhaps a boring fungus. At the edge of the black layer the shell is irregularly eroded to a depth of several micrometers (Figure 3). The erosion occurs only along a band, about 1 mm in width, between the black layer and the unmodified shell surface. The depth of resorption is extremely variable, from a few micrometers up to several hundred micrometers. The eroded area is rough in appearance, with irregular hum- mocks of shell material separated by more deeply eroded areas (Figure 4). The advancing black layer eventually covers the eroded region and is itself later overgrown by further deposits of nacreous shell material. of Figures 1 to 6 sition of boundary. Direction of growth is to top. Scale bar is 50 pm. Figure 4. Resorbed area at boundary of black layer. Note un- even, hummocky appearance. Scale bar is 5 um. Figures 5 and 6. Stereo pair of the boundary between the re- sorbed area and the unmodified shell surface. Direction of growth is to the right. Scale bars are 4 um. Page 198 The Veliger, Vol. 28, No. 2 The morphology of the resorbed area is shown in Fig- ures 5 and 6, a stereoscopic pair of micrographs taken of a single specimen. (The two pictures show the same region but are taken from two different angles, 6 degrees apart.) Perceived depth-of-field in SEM images can often be de- ceiving, because electron shadowing in SEM micrographs is quite different from patterns of illumination normally encountered in the human environment. In Figures 5 and 6, the resorbed area to the right of the micrographs some- times appears to overlie the unmodified shell surface shown at the left of the picture. When examined through a ste- reoscopic viewer, it is immediately obvious that resorption has cut down into the shell surface shown at right. Shell resorption in Nautilus macromphalus occurs along the entire margin of the black layer, from umbilicus to umbilicus. No portion of the shell’s surface is covered by the black layer before the surface is modified by resorp- tion. DISCUSSION Surficial resorption in Nautilus macromphalus is generally similar in form to that observed in many prosobranch gastropods, although the precise pattern of resorption dif- fers somewhat in detail. In N. macromphalus the resorp- tion is relatively deep and irregular, whereas the shallow, even resorption in the prosobranch Terebra dimidiata pro- duces a smooth, flat surface (SIGNOR, 1982). Unlike most prosobranch gastropods, N. macromphalus alters the entire surface of the penultimate whorl, less narrow bands at the umbilici, but this reflects differences in shell geometry and not function. Despite these small differences, the ef- fects of the resorption are identical: to remove the surface of the penultimate whorl as growth proceeds. Surficial resorption is so widespread among different taxa of prosobranch gastropods that it is difficult to imag- ine resorption serving a function relating to the specific ecology of each given species. The convergent evolution of surficial resorption in Nautilus macromphalus greatly strengthens this argument. The ecology of Nautilus is very different from any prosobranch gastropod; what Nautilus and prosobranch gastropods have in common is a coiled shell where fresh growth surfaces contact and overlie older shell. The function of surficial resorption most likely lies in the few commonalities shared by prosobranch gastro- pods and Nautilus. The hypothesis that the function of surficial resorption is constructional, and that the mantle edge prepares a suitable surface to which the black layer and new shell material can be attached, meets the fore- going criterion. Alternatively, the function, if any, of sur- ficial resorption could be to remove small epibionts or boring micro-organisms which might infest the shell’s sur- face. Relatively large calcareous epibionts are demonstrably too large to be removed by surficial resorption. Serpulid tubes not removed by the Nautilus during growth are sim- ply plastered over by the black layer and, later, by na- creous deposits (JOUBIN, 1892; STENZEL, 1964). LANDMAN (1983) documents the occurrence of a barnacle that grew on a live, juvenile Nautilus. The side of the barnacle was plastered with alternating layers of black organics and aragonite. Surficial resorption in Nautilus macromphalus must be a convergently evolved character, because the most prim- itive and earliest orders of nautiloids, the Plectronocerida, Ellesmerocerida and others, consist of orthoconic and breviconic forms (SWEET et al., 1964; YOCHELSON et al., 1973; Dzik, 1981; CHEN & TEICHERT, 1983). Since growth in these straight or slightly curved forms would not involve overgrowth or extension of the mantle over previously deposited shell, surficial resorption would not have occurred. Surficial resorption must have appeared later in the evolution of the nautiloids, along with or after the evolution of coiled conchs where successive whorls were in contact with each other. It would be interesting to determine whether the other great clade of fossil cephalopods, the Ammonoidea, exhib- ited surficial resorption. Answering this question would require extremely well-preserved fossil specimens. Thus far, I have not been able to obtain sufficiently well-pre- served material to allow detection of surficial resorption, if present, in this group. While resorption of shell material is commonplace among the Gastropoda, to my knowledge this is the first report of shell resorption among the extant Cephalopoda. Resorption may have occurred among extinct cephalo- pods, however. For example, the decollate nautiloids (e.g., Sphooceras truncatum [Barrande, 1868]) might have re- sorbed a portion of the conch, allowing separation of the deciduous portion (TASNADI-KuBACSKA, 1962), in much the same way as the gastropods Caecum (BERNER, 1942) or Rumina decollata (KAT, 1981) weaken their shells by resorption prior to shedding the deciduous portion. The mechanism by which this resorption, if present, would have occurred is uncertain. Resorption requires direct ap- plication of the mantle to the area where shell material is to be removed. Although authors have postulated the pres- ence of “‘cameral mantle” to account for the formation of cameral deposits in some nautiloids (TEICHERT, 1933, in FISCHER & TEICHERT, 1969; FLOWER, 1939; KOLEBABA, 1974), there is no compelling evidence for the presence of living tissues within the camerae, except the siphuncle, of any nautiloid (for recent reviews of the debate over for- mation of cameral deposits, see FISCHER, in FISCHER & TEICHERT, 1969; Dzik, 1981; Crick, 1982). Thus, it is uncertain if resorption did occur in conjunction with de- collation in primitive nautiloids and, if so, how that re- sorption might have occurred. An important and unresolved question is how gastro- pods and Nautilus accomplish shell resorption, and what happens to shell material secondarily removed by the an- imal. It is not certain that uptake of ions removed from the shell occurs in the mantle, nor is it certain what part of the mantle might be responsible for the resorption. P. W. Signor III, 1985 Lacking this information, the term “resorption,” while well established in the literature for this process, must be applied with caution. Based on current knowledge, it can only be used in the sense of “localized secondary disso- lution.” The experiments necessary to demonstrate uptake of secondarily dissolved ions through the mantle are tract- able, however, and would permit resolution of this ques- tion. ACKNOWLEDGMENTS Peter D. Ward kindly supplied specimens of Nautilus for this study. I thank R. Cowen, J. Haggart, C. Hickman, N. Landman, C. Teichert, P. D. Ward, and an anony- mous reviewer for helpful discussions, suggestions, and comments which greatly improved earlier drafts of this paper. SEM facilities were provided by the Department of Geology at the University of California, Davis. Early stages of this project were funded by a Collo Murphy Scholarship from the National Capitol Shell Club, and the final phases were supported in part by a grant from the donors of the Petroleum Research Fund, administered by the American Chemical Society. LITERATURE CITED BERNER, L. 1942. La croissance de la coquille chez les Gas- tropodes. Bull. Inst. Ocean. Monaco 816, 16 pp., 1 pl. BocGILD, O. B. 1930. The shell structure of the mollusks. K. Danske Vidensk. Selsk. Skrifter, Naturvidensk og Mathem. 2:232-326, 15 pls. CARRIKER, M. R. 1972. Observations on removal of spines by muricid gastropods during shell growth. Veliger 15:69-74, 1 pl. CHEN, J. & C. TEICHERT. 1983. Cambrian cephalopods. Ge- ology 11:647-650. Cowen, R., R. GERTMAN & G. WIGGETT. 1973. Camouflage patterns in Nautilus, and their implications for cephalopod paleobiology. Lethaia 6:201-214. Crick, R. E. 1982. The mode and tempo of cameral deposit formation: evidence of orthoconic nautioloid physiology and ecology. Proc. Third North Amer. Paleont. Conf. 1:113- 118. Dzik, J. 1981. Origin of the Cephalopoda. Acta Palaeonto- logica Polonica 26:161-191. ERBEN, H. K., G. Fiajs & A. SIEHL. 1969. Die Fruhonto- genetische Entwicklung der Schalenstruktur ectocochleater Cephalopoden. Palaeontographica Abh. A. 132:1-54. Page 199 FIscHER, A. G. & C. TEICHERT. 1969. Cameral deposits in cephalopod shells. Univ. Kansas Paleont. Contr. Paper 37, 30 pp. FLoweEr, R. H. 1939. Study of the Pseudoorthoceratidae. Pa- laeontographica Americana 2:1-219. Gray, J. E. 1833. Some observations on the economy of mol- luscous animals, and on the structure of their shells. Phil. Trans. R. Soc. Lond. 123:771-819. GREGOIRE, C. 1962. On submicroscopic structure of the Nau- tilus shell. Bull. Inst. Roy. Sci. Nat. Belg. 38:1-71. Jousin, L. 1892. Recherches sur la coloration du tégument chez les cephalopodes. 4me partie. Gland sécrétant le vernis noir chez le Nautile. Arch. Zool. Exper. Gen. Ser. 2, 10: 319-324. Kat, P. W. 1981. Shell shape changes in the Gastropoda: shell decollation in Rumina decollata. Veliger 24:115-119, 1 pl. Koun, A. J., E. R. MEYERS & V. R. MEENAKSHI. 1979. In- ternal remodeling of the shell by a gastropod mollusc. Proc. Natl. Acad. Sci. U.S.A. 76:3406-3410. KOLEBABA, I. 1974. A new orthocerid with a cameral mantle. Vest. Usti. Usat. Geol. 49:293-297. LANDMAN, N. H. 1983. Barnacle attachment on live Nautilus: implications for Nautilus growth rate. Veliger 26:124-127. Mutvel, H. 1972. Ultrastructural studies on cephalopod shells. Part I. The septa and siphonal tube of Nautilus. Bull. Geol. Inst. Univ. Upsala. N.S. 3, 8:237-261. SCHINDEWOLF, O. H. 1967. Analyse eines Ammoniten-Ge- hauses. Akad. Wiss. und Literatur (Mainz), Math.-Natur- wiss. K]., Abh. Jahrg. 1967, no. 8:137-188, pls. 1-16. S1GNorR, PHiLip W., III. 1982. Growth-related surficial re- sorption of the penultimate whorl in Terebra dimidiata (Lin- naeus, 1758) and other marine prosobranch gastropods. Ve- liger 25:79-82, 1 pl. STENZEL, H. B. 1964. Living Nautilus. Pp. K59-K93. In: R. C. Moore (ed.), Treatise on invertebrate paleontology, Part K, Mollusca 3. The Geological Society of America and the University of Kansas Press. SwEET, W. C., C. TEICHERT & B. KUMMEL. 1964. Phylogeny and evolution. Pp. K106-K114. Jn: R. C. Moore (ed.), Treatise on invertebrate paleontology, Part K, Mollusca 3. The Geological Society of America and the University of Kansas Press. TasNADI-KusacskA, A. 1962. Pathologie der vorzeitlichen tiere. Paladopathologie 1:1-269. TEICHERT, C. 1933. Der Bau der actinoceroiden Cephalopo- den. Palaeontographica Abt. A 78:111-130, 8 pls. YOCHELSON, E. L., R. H. FLOWER & G. C. WEBERS. 1973. The bearing of the new Late Cambrian monoplacophoran genus Knightoconus upon the origin of the Cephalopoda. Lethaia 6:275-310. The Veliger 28(2):200-203 (October 1, 1985) THE VELIGER © CMS, Inc., 1985 Egg Capsules and Veligers of the Whelk Bullia digitalis (Gastropoda: Nassariidae) F. M. pa SILVA anp A. C. BROWN Department of Zoology, University of Cape Town, South Africa Abstract. The sandy beach whelk Bullia digitalis can package its eggs in two different ways. Clumps of eggs may be contained in a single large sheath and deposited in the sand, or each clump of 150 eggs or more may be contained in its own capsule and held on the ventral surface of the maternal foot. In the latter case up to 40,000 eggs may be produced at one time. The eggs and capsules are described for the first time, as is the veliger stage, which is passed within the egg. The reproductive strategy of B. digitalis is contrasted with that of B. tenuis. INTRODUCTION Bullia digitalis (Dillwyn) is a nassarid whelk which is abundant on medium to high energy sandy beaches along the west and south coasts of southern Africa. Its biology, together with that of other species of the genus, has been reviewed by BROWN (1982). On beaches in the Eastern Cape Province of South Africa, gametogenesis occurs be- tween March and May, vitellogenesis and egg storage taking place from June to December or January, after which the females spawn (MCGwyNNE, 1980). We be- lieve that the timing of events on the west coast may be both different and more variable than in the Eastern Cape (BRowNn, 1971) and in recent years we have discovered females with eggs from early July to late January at Van Riebeeck Strand (Ou Skip), just north of Table Bay. Females of several intertidal species of Bullia tend to migrate offshore before producing their egg capsules (BROWN, 1982). Bullia digitalis appears to be no exception (McGwynng, 1980), although this migration of females is more marked in some areas than in others. The gravid females found at Van Riebeeck Strand were all buried just below low water mark, the migratory tendency thus being poorly developed. Egg cases of Bullia digitalis were first described by Pro- fessor J. Omer-Cooper in a letter to one of us (A.C.B.), this description being subsequently confirmed by BROWN (1971). A case measured about 2 cm in length and 1.2 cm in width and contained more than 1500 eggs arranged in clumps of 50 to 100 or more. Such egg cases were found 4 to 12 cm below the surface of the sand, usually in the presence of an adult female. The present work was undertaken due to the discovery of Bullia digitalis eggs, from Van Riebeeck Strand, that were packaged differently, being held in numerous small capsules under the maternal foot, and also to the acqui- sition for the first time of eggs containing veligers. MATERIALS anp METHODS Of the several females of Bullia digitalis discovered car- rying egg capsules beneath their feet, four were returned to the laboratory from Van Riebeeck Strand. The capsules and the eggs within them were counted and measurements made using a graduated eyepiece in a binocular micro- scope. In addition we had on loan from the South African Museum a female with capsules collected on Fish Hoek beach, False Bay, by Mrs. C. M. Connolly on 5 January 1961; in these capsules all the eggs had hatched or were on the point of hatching, as miniature adults. More recently, a whelk collected at Van Riebeeck Strand produced a full batch of egg capsules in the laboratory. These were discovered on 19 January 1984, well over a month after the animal had been captured. It is almost certain that this whelk had copulated in the field, the sperm being stored in the spermatheca. A number of cap- sules shed from the parental foot were held over sand in flowing seawater at 14°C. In each egg a veliger larva could be observed, which swam actively in the water if mechan- ically released from the egg. Several such veligers were examined and photographed under light microscopy, us- ing various types of illumination, first while they were swimming freely and later while held immobile under a coverslip. F. M. da Silva & A. C. Brown, 1985 Page 201 Figure 1 Egg capsules of Bullia digitalis removed from the foot of a gravid female (x12). The cases are typical except for one near the center of the picture, which contains few eggs. RESULTS Eggs and Egg Capsules The number of capsules per female varied from 150 to 203, each capsule typically containing 150 to 200 eggs, although an occasional capsule had only 30 or 40. Each capsule measured 3.00 + 0.05 x 1.5 + 0.15 mm, had an extremely thin, transparent, membranous wall, and pos- sessed an attachment thread at either end, one thread being more coiled than the other. The capsules were attached loosely to the undersurface of the maternal foot and to one another by a sparse but viscous mucous secretion and were further anchored to one another by their coiled attachment threads. A group of such egg capsules, removed from the foot and comprising about a quarter of those present, is shown in Figure 1. Each egg was about 220 um in di- ameter, as were the eggs and newly hatched young col- lected by Mrs. Connolly on Fish Hoek beach. Gravid whelks in the laboratory protected their cap- sules by curling the foot over them to form a tubular brood pouch, in the manner described from Bullia melanoides by ANSELL & TREVALLION (1970), while during crawling only the margins of the foot were used. These protective behavior patterns did not appear to be entirely adequate, however, as the whelks tended to shed capsules. The Veligers Each veliger carried a very thin, transparent protoconch consisting of 1% whorls. The veligers measured 205 + 25 um between the apex of the protoconch and the leading edge of the head. At its widest the shell diameter was 98 + 7 um. The head was bordered by a bilobed velum, which was heavily ciliated with cilia of two types; the longer (10 um in length) exhibited metachronal rhythm, while the shorter cilia, only about half that length, showed a more random pattern of movement. Two well-defined tentacles Page 202 The Veliger, Vol. 28, No. 2 SIPHON ( : CILIA TENTACLES HEART STATOCYST vISC. MASS PROTOCHONCH EYES PIGMENT YOLK SACK DIGESTIVE GLAND OPERCULUM VELUM 4 CILIA PROTOCHONCH STATOCYSTS PIGMENT VISC. MASS YOLK SACK DIGESTIVE GLAND FOOT Figure 2 Veliger larva of Bullia digitalis. Above, dorsal view. Below, lateral view. F. M. da Silva & A. C. Brown, 1985 and a siphon were present, as was a ridge of dark orange pigment on the velum that probably represents the res- piratory complex of the adult. A pair of eyes was appar- ent, despite the fact that the adults lack eyes (BROWN, 1982). Laterally and slightly anterior to the eyes, a pair of statocysts lay close to the actively pumping heart. Tor- sion had already occurred but it could not be determined whether torsion was complete. A small foot and opercu- lum were present. A sac lying on the outside of the body and attached near the base of the visceral mass was ten- tatively identified as a yolk sac, as its contents dissolved rapidly on contact with acetone. indicating the presence of lipids. A veliger of Bullia digitalis is shown in Fig- ure 2. Attempts to rear the eggs failed, the veligers becoming lethargic and darkly pigmented; within five days they had become infested by larvae of a digenic trematode and died soon thereafter. Eggs in capsules that remained attached to the feet of other individuals also failed to develop. DISCUSSION Although the literature on planktonic prosobranch larvae is voluminous, descriptions of veliger stages passed within the egg are rare and no Bullia veliger has previously been described. THIRIOT-QUIEVREUX (1980) described the planktonic veligers of Nassarius, a genus closely related to Bullia, but these differ considerably from the veligers de- scribed here. On the other hand, veligers of Littorina lit- torea are quite similar to those of Bullia digitalis, both in size and at least superficially in morphology (FISH & FIsH, 1977), with the exceptions that the cilia on the velum of Littorina are 3 to 5 times longer and no tentacles are visible. The eggs of all species of Bullia so far investigated pro- duce crawling young, the larval stages being passed within the egg (BROWN, 1982). Bullia digitalis is no exception and the small size of the eggs of this species may thus be remarked upon, as one might have expected eggs of little more than 0.2 mm in diameter to hatch at a much earlier stage. It is also of interest that every egg we examined had within it a living veliger and that every egg in the capsule collected by Mrs. Connolly contained a miniature adult; there are thus no nurse eggs in this species, despite ten- tative previous suggestions (BROWN, 1971, 1982). The numbers and size of young Bulla digitalis contrast with those of B. tenuis, a subtidal species whose egg cases and young have recently been described (BROWN, 1985). The adults of these two species are of similar size and appearance, but B. tenuis produces only about 60 egg cap- sules at a time and each capsule contains only one devel- oping egg, although nurse eggs are also apparently pres- Page 203 ent. By contrast, B. digitalis appears capable of producing up to 40,000 young at one time, but these are minute compared with the young of B. tenuis, which may attain a shell length of 5.3 mm before emerging from their cap- sules (BARNARD, 1959; BROWN, 1985). It is clear that these extremes represent very different strategies and it must be supposed that juvenile mortality is high in B. digitalis as compared with B. tenuis. Finally, Bullia digitalis can package its eggs in two dif- ferent ways—either with each clump of eggs contained in its own capsule, as reported here, or with all the clumps in a single all-embracing case or sheath, as described by Professor Omer-Cooper and subsequently by BROWN (1971). In the former circumstance, the tiny capsules are loosely attached to the undersurface of the foot, while if contained in a single large case they are deposited in the sand. It is clear that such a large case must be formed outside the body of the parent and it is logical to suppose that it is molded by the foot after the eggs have been extruded; its size and shape certainly support this expla- nation. Differences in egg packaging according to circum- stances of food availability are not unknown among the Nassariidae (McKILLUP & BUTLER, 1979) but the pres- ent example would appear to be the most extreme so far reported for any of the Prosobranchiata. LITERATURE CITED ANSELL, A. D. & A. TREVALLION. 1970. Brood protection in the stenoglossan gastropod Bullia melanoides (Deshayes). J. Natur. Hist. 4:369-374. BARNARD, K. H. 1959. Contributions to the knowledge of South African marine Mollusca. Part II: Gastropoda: Prosobran- chiata: Rhachiglossa. Ann. S. Afr. Mus. 45:1-237. Brown, A. C. 1971. The ecology of the sandy beaches of the Cape Peninsula, South Africa. Part 2: the mode of life of Bullia (Gastropoda: Prosobranchiata). Trans. Roy. Soc. S. Afr. 39:281-320. Brown, A.C. 1982. The biology of whelks of the genus Bullia (Nassariidae). Oceanogr. Mar. Biol. Ann. Rev. 20:309-361. Brown, A. C. 1985. Egg capsules and young of Bullia tenuis (Nassariidae). J. Moll. Stud. (in press). Fiso, J. D. & S. FisH. 1977. The veliger larva of Hydrobia ulvae with observations on the veliger of Littorina littorea (Mollusca: Prosobranchiata). J. Zool. (Lond.) 182:495-503. McGwynng, L. E. 1980. A comparative ecophysiological study of three sandy beach gastropods in the Eastern Cape. Mas- ter’s Thesis, University of Port Elizabeth. 144 pp. MckKI.uvup, S. C. & A. J. BUTLER. 1979. Modification of egg production and packaging in response to food availability in Nassarius pauperatus. Oecologia 43:221-231. THIRIOT-QUIEVREUX, C. 1980. Identification of some plank- tonic prosobranch larvae present off Beaufort, North Car- olina. Veliger 23:1-9. The Veliger 28(2):204-210 (October 1, 1985) THE VELIGER © CMS, Inc., 1985 ‘The Ecology and Local Distribution of Non-marine Aquatic Gastropods in Viti Levu, Fiji by A. HAYNES School of Natural Resources, University of the South Pacific, P.O. Box 1168, Suva, Fiji Abstract. Freshwater habitats throughout the island of Viti Levu, Fiji were investigated for gastro- pods, water conductivity, water hardness, temperature, substrate, and current speed from August 1982 to February 1984. In general the values of conductivity, hardness, and temperature increased toward the sea; but this was not true of all river sytems and these factors were not as important in influencing the distribution of the 32 gastropod species as were physical factors, specifically distance from the sea, substrate, and current speed. Using these physical factors the running water gastropods were classified into five groups. Gastropods were absent from long stretches of all rivers where the water was deep and turbid and the bottom unstable. INTRODUCTION DuRING 1971 STARMUHLNER (1976) sampled the gastro- pods at stations near the town of Suva and near the for- estry station of Nadarivatu on the island of Viti Levu, Fiji. However, no further sampling of gastropods in the remaining extensive network of rivers and streams on Viti Levu has been reported. The aim of this study was to find the distribution of the freshwater gastropods on the island of Viti Levu, Fiji and to establish the factors that were most important in influencing the distribution of the various species. STUDY AREA Viti Levu is an oval-shaped island, reaching about 150 km long and 100 km wide (Figure 1). The interior is mountainous and the highest peak, Mt. Victoria (Toma- nivi), is 1312 m high. Because Viti Levu is in the path of the southeast trade winds, the southeastern side and the interior receive heavy rainfall and are covered in rain forest, while the northwestern side is comparatively dry and much of it is used for growing sugar cane. The mean annual temperature is 29-30°C and the annual rainfall is about 3000 mm. The two longest rivers are the Rewa and Sigatoka, which rise in the central high country and flow southward. The Rewa River system drains nearly one-third of the island. Recently two artificial lakes have been formed. The con- struction of a hydroelectric dam on the upper Rewa River resulted in Lake Monasavu, and a dam to impound water to supply the towns of Lautoka and Nadi has produced Lake Vaturu (Figure 1). MATERIALS anp METHODS Gastropods were collected from rivers, streams, and lakes from July 1982 to February 1984. The collecting stations 1-47 are shown on the map of Viti Levu (Figure 1). They were chosen to be as representative as possible while being accessible by road or track. The river bed and plants at each station were searched for 30 min. The upper and lower surfaces of stones and boulders were searched, leaf litter and water-weed were inspected, and sand and gravel were sieved. Representa- tives of all gastropod species were collected and taken to the laboratory for identification. Shell, operculum, radula, and reproductive organs were used in the identification of the snails following several authors: Mousson (1870), RIECH (1937), BENTHEM- JUTTING (1956), FRANC (1956), STARMUHLNER (1970, 1976). Water speed was estimated by timing a float between two fixed points, the water temperature was taken to the nearest 0.5°C, and in some cases a water sample was col- lected from the station. At the Institute of Natural Re- sources, University of the South Pacific, water samples were analyzed for conductivity (us), which indicates the total ion concentration, and for hardness (mg CaCO,/L) by titration with EDTA (ethylene-diaminetetraacetic acid). A. Haynes, 1985 Page 205 1) ae VIT!I Levu 6 34 ':500000 = a Xe @35 SBA 36 O NADARIVATU \ 42 LAUTOKRO 11 43 e cs Ws> eS abet? Ne TES 3/@ 47® Vaturu L \ “Oo 16 Monasavu ° 2 NA dj : i Oy, %15 © 45 4, ‘ny 19@ ‘ njvod) %, - O 8 ts \S NaDl 1a f Warmar 17O KO/Rovoy 4) 6 4 é x z Ae” y? 33 sig? 5 7. as? 14 wat y Rewa 15 ae ONA was ie Camm esor 2 TONG fj 29 13 1 SIGATOKA : 26 2 ° SUVAKD 23 =e 4 2 O28 33 XQ S32 e31 7 3 0 Figure 1 A map of the main river sytems of Viti Levu, Fiji showing the localities of sampling stations 1-47. RESULTS flowing river (station 17), and two fast flowing inland Species Found rivers (stations 11, 19). This suggests that it is also wide- spread but often overlooked because of its small size (<5.0 mm). Gyraulus montrouziert was found on water-weed in still and slowly flowing water at two stations (23, 30). The remaining gastropod species lived in running water. Using the parameters of distance from the sea, current speed, and bottom substrate of the river or stream in which they lived, these gastropods can be divided into 5 groups: Thirty-two species were found and identified (Table 1). The nomenclature of STARMUHLNER (1976) has been used where possible. The gastropods found in still water were the pulmo- nates Planorbarius corneus (Linnaeus, 1758), Physastra na- suta (Morelet, 1856), Ferrissia noumeensis (Crosse, 1871), Gyraulus montrouzieri (Gassies, 1863), and the proso- branch Melanoides tuberculata (O. F. Miller, 1774). The (1) 200 m-2 km from the sea in currents from 0 to 10 European snail Planorbarius corneus was probably intro- cm/s, substrate of mud or sand with some rocks and duced into station 24 from a freshwater aquarium. Mel- water plants: Asszminea crosseana (Gassies, 1858) on anoides tuberculata was the most widespread species; it plants, Clithon oualaniensis (Lesson, 1831) on sand or was found at 22 of the 47 stations. Physastra nasuta was mud, Neritina turrita (Gmelin, 1790) on mud, Clithon the next most widespread species, being found at 14 sta- spinosus (Budgin, 1845), and Neritina auriculata La- tions. Both species lived in ditches and dalo patches on marck, 1816, on rocks. gravel and mud as well as on stones in water currents as (2) 300 m-8 km from the sea in a current up to 40 cm/s, fast as 80 cm/s (Table 1). Ferrissia noumeensis was present substrate of stones and rocks: Clithon corona (Linné, on stones, gravel, and water plants at seven stations—two 1758), Clithon diadema souleyetana (Récluz, 1841), and ponds, two small streams (stations 21, 23), one slowly Melanoides arthuri (Brot, 1871). The Veliger, Vol. 28, No. 2 Page 206 piasyjnd vuiiany “vjnsou pisvskyg “vyvjns1aqny “JA ‘vsojn] saprouvjaywy ‘vapiodnd vdndoianjy ‘sisuaaunou vississaq DjDINILaqN} SaproUv]ayyy Dpigns DYIYIIaNT ‘SIsUaaUNOU DISSILLAY DID] -NdUUND DUIjLaNy “DuDJakajnos DUapDIp UoY;)/) pypjnz41aqny ‘W ‘vsojn] saprouvjayy ‘vapiodnd vdndoianjy pinsou viysoskyg DjDINILaqNy Sapiouv]ayyy SISUIBULNOU DISStLLaY “DLQVIS DADLY J, ‘vinspu vuysoskyg “Djpjno1aqny sapioun -jay ‘vapiodnd vdndoianpy ‘vjojnovu vwopiligy Dssalq -ap DuvjjaoL0d visnjdagy “vjvjno1aqn]} sapiounjayyy pynspu viysoskyg Dinsou vAysvsKYg “vIDjINILagGN} sapiouD -jay ‘vapiodnd vdndoinpy “vjojnovu vwopiliy pinsou vuysoskyg “oyojnavu vuopilry (wep JayemyYsad & “Da00/010 DsS11Dg) pjoidiuonbs ‘py “1Uu0jiny DUtBLLaAT sup.idsp sapiouvnjayyy “vypau -4) visvjdas ‘vjoidiuonbs “ay “1uojiny DUIzLLaNy DUOLOD U0YI1)/) “DSOI -14Jaq viviy yf ‘oj21diwonbs “py “u0jiny DUIyILaAT DUDvassOLI pauiussp ‘psorijjaq viDlYy J ‘vssasdap vunp}a9 -L0d visvjdag “njnidiwonbs “ay “uojzing Durpi4any juasaid spodonsesy Toe a4 88e Sol ve Ge 99 8¢ aN GE I¢ GSC 6€ aN 14 8L Sut) S68 66 re 4 LEL 08 TIL Oot (1/00eD (s7) von -P1IJUIDUOD ssoupiepyY uol jejoy, -sodwiay, 8c G'6¢ CC €c Sé 6c Sc Gc €C VC G9C G9C Lc Lé 9¢ SZ (De) ounye 0¢-0 O£-02 OS-O¢ OS-O€ O¢-0¢ O1-0 O1-0 OL-O (s/w) poods JOIe MA SI Ol c8 cv ov é (wry) eas WO] soue\siq] sJaprnoq 2g sauojs ‘Jaaes3 Sauo}s 29 [9AvIB poom -19}eM 2 pnul sauo}s 2 pues ‘pnur SJap[nog 2 sau0}s sauo}s SJap[Noq 29 sauo0}s S9Uu0}s SIOp[Nnoq 7 sauojs sJap[noq sauojs 29 pnul sJop -[Noq 2 2]3urys sJop -[noq 7 2]Surys aysurys a[surys 29 pnu poom 29 poam ‘pnul siueyd 29 pues ‘pnur pnur sjuey|d 2 syoos ‘pues ayesqng SLESOSAX “YT ISOATUTe AA O}UT URIS S7COCIAX “SU VILWILAA WY ISIATUIAA JO UOTOUN[ SCCOCOAX NOAOIOY 3 “Y VIEUTIEAA SI¢899 AX ‘peor JO pus ye “Y eIeUTIE AA SSOCIZAX ‘OWN 1e “Y CUIPIE AA SSTOLPAX ‘peor naeseuoyy “yy CUIPIeE AA LO6VLTAX “peor ISOWBN “Y NULUIE AA OUT UIeII1G 6SO0CSSHX ‘asnoyduind “y nueuire A oLPOPEAX “exeNAOY “OY JO YINos wy gy “Y eynqrure 6LPOPEAX “MeIOTIE AA YINOS wy [| “WY eyNqiure~A OUI WeINS LLESIIAX “Navseuopy 24e'T 6€COGTAX “Nestureyey aaoqe Wy CG] “WY PTPWIUIe AA OJUT WIeanS LOETOTAX “NA -a[aseT MOfaq WY Z “YW P[EWIUIE AA S9ZOBEAX “Pas9g We “Y PpeUTUTE A SELOESAX “Nagineg 3 “Y eMoy OLIZSSAX “A eMoy TrOlS9AX ‘aspliq Wosnen “Y eMIY 090¢C9AX Ssodire wosneyy (yep) “Yy eMay €86rb9EX ‘Surpury eryory “(jepn) “Y eMoy (nAa"T BIA 000'0S:1) aouasajor deur 2 JOA 6l 81 LI oT SI vl cl cl N I Jaquinu uone}s surjd -weg ‘pautuiajap jou = CN ‘suoneis Surjdures ay} ye Juasoid spodosses pure ‘stsAyeue 193eM JO s}[Nsas ‘suoNTpuod [eotsAyd ay], VS IGE Ns Page 207 A. Haynes, 1985 6LS67OAX “MeAUepeN vapiognd vdndownpy 9°9¢ Ss cc 0S-0 81 Sdop[noq ye “AY BQnyzeqnyle’ AA OFUT ULI 9¢ Si9p 0268861IM vinsou vuysoskyq “vypjnosagny saprouvjayy cr9 co SEL SC 09-0F vi -[noq 7 2]Surys ‘aseTTIA ye “YY eqnyzeqnylE A G€ pssasqap vuvjjao10g vrivjdag “pjaidiwonbs * Ay “tuojin} vurpiiany “Duvjakajnos DULapoIp UoYst)/) 6 LEC 109 O€ OT-0 ¢c SIUC Sey POUL OZLSZ6AM ‘23Plq UM} “Y eNAL TL, ve os 8L808LA.M “dins -sasdap Dunjja2140g vi10jgag ‘o1adyjnd vuriszany dN dN 8Z Ov-O€ Z sauo}s -11e WoJJ weansdn “ID nAsjo10y €€ DjDjNI1aqNy sJap[noq saprounjay “uuad durjisany “Du0109 UoYsi/-) aN aN 6c O¢-OL £0 2g Jeavsd ‘pues SS8S78AM “ISB09 [e1O-) UO WRaIIS ce SIALI] Lysrc lax pinsou vLysvskyg “vjvjNILaqQN} saplouvjay dN dN 8Z 0 Z peop 2 pnur ‘moqiep oyleg ‘durems o8es I€ DIDI 2G ie SOP GUuni2 yy, 6l88elax “‘ualzno4quoU snjnoidyy ‘sisuaawnou DISsiLay dN dN 8Z 0 €0 sjued 29 pnur ‘moqiep oyweg ‘spuod Ajry O€ S19p 0666c1AX vaasynd vuipisany aN aN 4 0S—O¢ 8é -[Nog 2g 2]SuTYs ‘enuenue Ny 3e “Y eNaen] 6¢ ra SE S89 S¢ 0c SI Sauoys 29 JaAvIs 888S81AX ‘Nruedien ye “Y enaeyy 82 sjueyd ruojiny “Ny “Djaigiuonbs “ny ‘vpiqns vurpisany €€ 8°89 SC 01-0 v a syoor “pnur 9S8P77AX ‘Spluq ye “Y enaeN Le SZOCIEAX ‘Pros DIDINILaGN] “Ay “WsnyjLD sapiouvjayy S61 9'6P 97 Or-O€ + a[surys 2 Jaaess ISOUI® NJ-ISAABYNGEN Iptsaq YIoI") 92 Djnsvup * J ‘asoyoisq1aq piviy J “ossasdap vuvjjasL0g “¢ “1uasf{ns v1.40} -dag “idosaynsovu ‘ny “yuyad "xy ‘bsadyjnd SYyIO1 29 vuiiiany ‘suvsidsn “Wy “Djvjnr12qn} sapiouvjayy Te 8°c9 VC 08-OF S S1aprnog ‘sapsurys LOOGTSAX “MINTTEM “ID 08k A SC snauLod snisvgsouDnjg “vinsou vsjsoskyg CCOCESAX ‘pyojnz1aqn} Sapiounjayyy ‘Sisuaauinou vIssiLay dN dN 8Z 0 €0 sjue|d 29 pnur ‘jooysg seuUTeLy eAng ‘puod Ajry 74 DINSDU DL] -spskyg “Ninyj1v “Al “Djoj]nIL1aqN} sapioudjayy paom 8Z6TLSAX ‘uaiznosuou sninoiksy ‘sisuaauinou DISsiLLaq dN dN 62 OL €0 2 jaaeis ‘pnur ‘wreans ‘sndures gsQ “rp tunimnyp €t (DJDJNpUN DULLO}IVT ‘snavgoLvIS Diyjh ‘snjoisys snduvjayy :UILeUL) 0714107 9COVeSAX ‘N “DJDINILND DUIILLaNy ‘SIsuatuUD]ONO UOYII’) dN dN 62 0 Z0 SOABI] 29 pNul ‘eang ‘saaoisuewl “ID TunyIN] 144 DIDINILaqN] SaplouvjayW] ‘SisuaaUNoOU DISSILLAay dN dN O€ 072-0 + P22M 29 YOOL OLISE9AX ‘Asjsos0j ye wvANS 1Z Dyoaur OSCSP9AX piuvjgas ‘ojoidiuonbs vurjisany “DuoL02 Uu0YyyI) dN dN 6Z OI-0 G souo}s 29 pnul ‘peor JO pus We “Y oI[eprern 02 quasaid spodonses (T/00rD (s7) won (De) (s/wid) (uy) aIeLISQNS (nA2T BIA 000‘0S'1) Jaquinu Sul) -esjUI0U0D «= auNye poads eas WO} souasajar deur 2 JOATY uorje}s ssoupseFy uor[ejoy, -Jadway sajeAA =: 90 SIGT suryd -uIes *ponunuor) T 91981 The Veliger, Vol. 28, No. 2 Page 208 pinsou vaysoskyg “DjvjnILaqny saplouDjayy Dinsou viysoskyg ‘vyvjns4aqn} ‘W ‘vsojn) sapiounjayy “vapiodnd vdndoianjy aLoyous -d4a} paviy J “ossasdap puvjjasL0g *s “D2 -1u0zn) niunjdag “njaidiwonbs vurpisany ‘v1409 -yd sapiounjayy ‘nuvjakajnos nwapoip uoyjtyy) quaf[ns °S ‘vssasdap vunjja2L0g viivjdas ‘psasyjng vurjisany ‘DjJDjINILaqQN} SapiouDjayy tuaif{ns pisvjdasy “bjnspu piysoskyg “vypjnz1aqny "Jl ‘sunsidsp sapiounjayy pInspu Diysvskyg “vypjns41aqn} “JAI ‘vsojn] Sapiounjayy DjojnILaqn} Sapiouvjayy ‘vaprodnd vdndoinpyq DLQDIS DADIY J “DIDINILAGN] Saplouvjayyy pou0zm piLoj gag “DPILIN] “AJ “DI -pynaiino “Ny “Wuoj4N} DUIRILaNy “DsoUuIds UoYIY)’) Ds -sasdap puvjjazL10g viivjdag ‘suvsidsv sapioun -/2JW “Duo4109 +7 “puvjakaynos vuapoip uoYyjt’) juasoid spodonsey ce 66 06 aN aN c 1d IZ Tel col 16 aN G's Te? 67 dN aN ley LI91 L9C oO IST r9l aN (T/ODFD (s7) uo su) -e.1]U99U09 ssoupieFY uol [ej}oy, -sodway, 8c Sé ce Le 8c CC 0:92 S'9¢ 8¢ 8¢ Le (De) oinye 0 8¢ 09-0€ G€ 01-0 14 08-05 9 OF —-02 8 Or —-0¢ Oll 09-05 99 09-0 OS OS-O€ c 01-0 c Ov—-O¢ S‘0 (s/w) (wy) paods eas WO] JeAA —-0URSIGG ‘ponunuor) b SI9e%.L SQUO}S 29 PNUI sJap[noq W® squojs ‘acess souo}s 29 pnur SYIOI 2 S9U0}S S9UuO0}S siap -[Noq 7 2]3urys S9Uu0]S SyI01 2 sauo}s ‘JaavsB sauo}s sauo}s 29 poom ‘pnul sauo}s ayeysqns GLESOLAM ‘TINIE A 24k] 6PELILAIM ‘eMPIEN 12 “Y IPEN CCESPPAM ‘g3pliq UMO} 18 “Y IPEN OIPSCSHM ‘uoneys 1amod seau “yY oaqes S7SS6SAM ‘HNqeyeA ie ureaS 99SCCOAX “NIeATIep -ON 1e “Y eyoyeSig OUI “ID eyepeN OETSEBAM ‘SPAT wos weansdn wy Z oY eyoesig VIZOGLAM “AN, Jeau “YY eyoeSIg OUT WRITS O€1SCOAM ‘?SPliq woJj wesnsdn wy oz “Yy eyoesISg 8rOSPSAM ‘adpliq UMO} Je “YW eyoieSICg OrP9OLOAX AORN 2Q TWOpo'T usaMJaq WeINS (naa'T NIA 000°0S"1) aouasajor deur 2 JOATY LY ov SP vy CV (a4 Ip OF o£ 8e Le Joquinu uone}s suryd -wies A. Haynes, 1985 Page 209 (3) 2-11 km from the sea in currents from 0 to 10 cm/s, substrate of mud and sand with some rocks and water plants: Melanoides plicaria (Born, 1780) on mud and sand, Neritilia rubida (Pease, 1867) on plants, Septaria lineata (Lamarck, 1816) and Neritina squamipicta Ré- cluz, 1843, on plants and rocks, Neritina turtoni (Ré- cluz, 1843) on mud and rocks, and Thiara bellicosa (Hinds, 1844) on mud. Neritina squamipicta and N. turtont were found 22 km from the sea (station 4). (4) 300 m-11 km from the sea in currents from 20 to 80 cm/s, substrate of rocks and boulders with patches of stones and gravel: Melanoides aspirans (Hinds, 1847), Thiara amarula (Linné, 1758), and Thiara terpsichore (Gould, 1847) on gravel and stones in slower currents, and Neritina macgillivray: (Reeve, 1855), Neritina pe- titi Récluz, 1843, Neritina pulligera (Linné, 1767), Septaria porcellana depressa (Linné, 1758), and Sep- taria suffreni (Récluz, 1841) on rocks. Septaria porceil- lana depressa was found 77 km (station 10) and N. pulligera 28 km (station 29) from the sea. 15-110 km from the sea in currents from 30 to 80 cm/s, substrate of stones, boulders, and rocks: Fiji- doma maculata (Mousson, 1865) on stones, Fluviopupa pupoidea Pilsbry, 1911, Melanoides lutosa (Gould, 1847), and Thiara scabra (O. F. Miller, 1774) on stones, boulders, and rocks (Table 1). (5 WY Characteristics of the River Systems When the physical data from all the river systems were pooled, an inverse correlation was found between distance from the sea and temperature (P < 0.05), hardness (P < 0.05), and total ions (P < 0.01) (Spearman’s rank cor- relation coefficient r,, ELLIOTT, 1977). Generally, however, the number of gastropod species did not follow this inverse correlation. More often there were more species near the mouth of the rivers and in the headwaters than in the middle reaches. The pattern of decreasing temperature, total ions, and hardness with increasing distance from the sea was not apparent in all rivers. However, this was the trend in the short, steep Waikubakuba-Tavua river system (Figure 1, stations 34, 35, 36) where temperature (30, 25, 22°C), total ions (601, 135.2, 55.1 ws), and hardness (237.9, 66.2, 26.6 mg CaCO,/L) decreased as distance (2, 14, 36 km) increased. The lowland Wainivesi-Waimara system illustrated the lack of correlation between distance from the sea and the number of species found. Here, there were two gastropod species 2 km from the mouth, two other species 5 km from the sea at station 17, one species 10 km inland at station 18, and six species in the headwaters at station 19 (Figure 1). In this river system there was little variation in water hardness (55, 38.8, 44.9, 39.1 mg CaCO,/L), or temper- ature (28, 28.5, 29.5, 28°C) from mouth to headwater (Table 1). All of the rivers and streams studied contained sufficient dissolved ions to support a gastropod population. The lowest conductivity, 43.1 us, was obtained at Nadala Creek, station 42 (Figure 1), where the three gastropods Mela- noides tuberculata, M. lutosa, and Physastra nasuta were frequently found. The total ion concentration here was low compared with that in the water of the English Lake district where the main ions are sodium and calcium in equal proportions and the conductivity is 112 us, or with those in a limestone stream such as the river Avon (Wilt- shire, England) which has a conductivity of about 450 us, or with those in water of salinity 3%o which has a con- ductivity of 6000 ws (MACKERETH et al., 1978; HAYNES, 1982). The concentration of calcium ions necessary for the presence of mollusks varies but if the quantity of water is great enough they will tolerate low concentrations. For example, Planorbis carinatus is common in Lake Win- dermere, England where the calcium concentration is 5 mg/L (Macan, 1974). The lowest values for hardness in this study were 21.1 mg CaCO,/L at station 42 and 19.5 mg CaCO,/L at station 26 (Figure 1). Melanoides arthu- rit, M. tuberculata, M. lutosa, and Physastra nasuta were present at one or both of these stations. The water temperature varied from 22 to 32°C. Al- though generally the water was warmer nearer the river mouth, inland species such as Fijidoma maculata and Flu- viopupa pupoidea were found in temperatures up to 29°C (station 11). It is possible that they were restricted to inland streams because they require a low temperature (22-23°C) for reproduction. Gastropods were absent from long stretches of the larg- er rivers (stations 6, 12, 28, 22) due to unstable bottom substrates and to the depth of the often turbid water. The freshwater clam Batissa violacea Lamarck was often pres- ent under such conditions. DISCUSSION All the species found have been reported from other Pa- cific islands (RIECH, 1937; STARMUHLNER, 1970, 1976). However, four species, Gyraulus montrouziert, Planorba- rius corneus, Clithon spinosus, and Neritina squamipicta, have not been reported previously from Fiji. In addition, Fijidoma maculata has been previously recorded only from swift flowing parts of the Rewa and Lami river systems, Viti Levu (MorRISON, 1954). In this survey it was found in the headwaters of the Rewa, at stations 7 and 8 in the Wainimala River, and at station 11 in the Wainibuka River where it reached a density of 2250/m’. Decreases in total ion concentration and temperature in the mountain streams of Madagascar, Sri Lanka, and New Caledonia similar to those observed in the Waiku- bakuba-Tavua river system were reported by STAR- MUHLNER (1979). These trends were absent in some of the longer river systems of Fiji. When Starmihlner sampled the Vago Creek (station 25) in 1971 (STARMULNER, 1976) he reported a water Page 210 speed of 0.5-1 m/s, a temperature of 23.6°C, a total ion concentration of 45 ws, and the following species present: Neritina pulligera, Septaria porcellana depressa, Thiara amarula, Melanoides aspirans and M. tuberculata. This is not dissimilar to the findings of the present study when the water speed was 40-80 cm/s, the temperature 24°C, the total ion concentration 62.8 ws, and the same species were found as in 1971 plus Septaria suffreni, Thiara terp- sichore, Neritina macgillivrayi, and N. petiti. Starmtihlner also sampled at Nausori (station 3) where he found the temperature was 27.4°C compared with 27°C in this study. The species that he found were Neritina turtoni, Septaria lineata and Thiara bellicosa. In this study 7. bellicosa was absent but Neritina squamipicta and Melanoides aspirans were present. Starmtihlner sampled at 8 stations in the Suva area and J. A. McLean sampled at 4 stations in the interior near Nadarivatu (STARMUHLNER, 1976). Starmiihlner found 18 species, three of these were not discovered in this sur- vey. These were Clithon olivaceus (Récluz), Neritina ca- nalis (Sowerby), and Septaria macrocephala (Le Guillou). All three were found during 1983 in clear torrential streams on the relatively undeveloped Fiji islands of Ova- lau, Taveuni, Kadavu, and Gau. It is possible that they have become rare on Viti Levu due to the increase in road building and logging. Both activities disturb the soil which is then washed into the rivers and streams during heavy rains and increases the turbidity of the water. Mud is deposited on rocks and stones where it inhibits the growth of algae, the main food source of these gastropods. ACKNOWLEDGMENTS I wish to thank the University of the South Pacific for providing a research grant for this work, the Institute of Natural Resources, USP, for analyzing the water sam- The Veliger, Vol. 28, No. 2 ples, and Mrs. Jean Maybin for helping with the sam- pling. LITERATURE CITED BENTHEM-JUTTING, W. S. S. VAN. 1956. Systematic studies on the non-marine Mollusca of the Ind-Australian Archi- pelago. Treubia 23(2):259-427. ELLIOTT, J. M. 1977. Some methods for the statistical analysis of samples of benthic invertebrates. Freshwater Biological Association Scientific Publication No. 25. 156 pp. FRANC, A. 1956. Mollusques terrestres et fluviatilis de L’Ar- chipel Néo-Calédonien. Mém. Mus. Natl. Hist. Natur., Sér. A Zool. 3(1):200 pp. Haynes, A. 1982. Ecological and behavioural studies on the water snail Potamopyrgus jenkinsi (Smith) in the Upper Avon. Doctoral Thesis, The Open University. 336 pp. Macan, T. T. 1974. Freshwater ecology. 2nd ed. Longman Group Ltd.: London. 343 pp. MacKERETH, F. J. H., J. HERON & J. F. TALLING. 1978. Water analysis: some revised methods for limnologists. Freshwater Biological Association Scientific Publication No. 36. 20 pp. Morrison, J. P. E. 1954. The relationships of old and new world melanians. Proc. U.S. Natl. Mus. 103(3325):357- 394. Mousson, A. 1870. Faune malacologique terrestre et fluviatile des Iles Viti, publiée d’aprés les envois de M. le Dr. E. Graeffe. J. Conchol. 18(2):179-236. RiEcH, E. 1937. Systematiche, anatomische, okologische und tiergeographische Unterschungen tiber die Susswassermol- lusken Papuasiens und Melanesiens. Arch. Naturgesch. (N.F.) 6(36):40-101. STARMUHLNER, F. 1970. Etudes Hydrobiologiques en Nou- velle-Calédonie. O.R.S.T.O.M., Ser Hydrobiol. 4(3/4):3- 127. STARMUHLNER, F. 1976. Beitrage zur Kenntnis der Stisswas- ser-Gastropoden pazifischer Inseln. Ann. Naturhist. Mus. Wien 80:473-656. STARMUHLNER, F. 1979. Distribution of freshwater molluscs in mountain streams of tropical Indo-Pacific Islands (Mad- agascar, Ceylon, New Caledonia). Malacologia 18:245-256. The Veliger 28(2):211-215 (October 1, 1985) THE VELIGER © CMS, Inc., 1985 A Bibliography and List of Molluscan Names of Josiah Keep by EUGENE COAN Research Associate, Department of Invertebrate Zoology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118 Abstract. A list of the books and papers by the early west coast malacologist Josiah Keep and a list of the 12 molluscan names that he introduced are presented. Two neotypes are designated. JOSIAH KEEP was an early malacologist on the west coast of the United States whose particular contribution was as a popularizer of the study of shells. The several editions of his West Coast Shells were responsible for recruiting the interest of many a student and amateur. Josiah Keep was born in Paxton, Massachusetts, on May 11, 1849. He received a Bachelor’s degree from Am- herst College in 1874 and a Master’s from the same in- stitution in 1877. That year he also married and moved to California. There he taught at the Golden Gate Acad- emy for one year, then Alameda High School for seven years, where he was principal from 1881 to 1885. In 1885, he became Professor of Natural Sciences at Mills College in Oakland, California, with which he was associated for the rest of his life. He came to specialize in courses in geology and astronomy, but his real love was the Mollusca. Between 1881 and 1910 he published sev- eral editions of his handbook on the shells of the west coast (KEEP, 1881, 1887d, 1888c, 1892, 1893, 1904, 1910b; 1935, posthumous edition by KEEP & BaILy). The interest that they elicited was one of the cornerstones of malacol- ogy in the western states. He died in Pacific Grove, California, on July 27, 1911, where he is buried. (For more information on his life, see ANONYMOUS, 1911; DALL, 1911a, b.) Here I present what is intended to be a complete bib- liography of his papers on the Mollusca. In addition, care- ful examination of the 1887 edition of West Coast Shells reveals that he inadvertently introduced several Carpenter manuscript names, some of which have escaped previous detection. These probably got onto collection labels in Cal- ifornia through identified specimens returned by Philip Carpenter to Henry Hemphill, another early west coast collector. As evidence of this, there are three lots identified by Carpenter in the California Academy of Sciences from the Hemphill collection with three of these manuscript names on them. This material is cited below, but there is no evidence that Keep ever saw these particular lots. In each case, Keep probably had specimens in his own col- lection labeled with these names. In addition, Keep intro- duced three Hemphill manuscript names. After his death, Keep’s personal collection was sold by his family to the Institute of Geology & Paleontology of Tohoku University in Sendai, Japan, in 1915, where it is housed today.' The collection has some 3000 lots that be- longed to him. The larger portion of the separate Mills College Collection went to the Department of Paleontol- ogy at the University of California at Berkeley. A smaller part went to the Department of Invertebrate Zoology at the California Academy of Sciences. I have examined these last two collections, as well as spot-checked the collection of the United States National Museum of Natural His- tory, where Keep sent some specimens. Aside from the already isolated syntype of Alvania aequisculpta in the NMNH, I could not find Keep type material in any of them. Drs. Tamio Kotaka and Kenshiro Ogasawara of the Institute of Geology & Paleontology at Tohoku Uni- versity have thus far been unsuccessful in finding type specimens of these 12 taxa in the Keep collection there. Class Bivalvia marginata, Crassatella—KEEP 1887d:179, ex Carpenter MS. No locality given. Type material—USNM 15578, neotype (Coan, 1984: ' Some workers were evidently misled into believing that Keep’s collection went to Tokyo and was lost during World War II (for example, A. G. Smith, in ANONYMOUS, 1968). Page 212 233), the same specimen that is the lectotype of Psephis salmonea CARPENTER, 1864:539; 611; 641. San Diego, San Diego Co., Calif. Remarks—A synonym of Halodakra (Stohleria) sal- monea (Carpenter, 1864). There is a specimen identi- fied by Carpenter with this name on it from the Hemp- hill collection in the California Academy of Sciences (CASIZ 036681), but there is no evidence that Keep ever saw it. The Veliger, Vol. 28, No. 2 columbiana, Physella—KEEP, 1887d:120, ex Hemphill MS. Columbia River, Oregon/Washington. Type material—Not located. Remarks—Should evidently be Physella columbiana Keep, 1887, according to BURCH (1982:53; 159, fig. 639). It has often been dated from HEMPHILL (1890: 27), and it was misspelled as Physella “‘columbella” by KEEP (1904:152). graciliente, Evalea—KEEP, 1887d:52-53, ex Carpenter MS. No locality given; presumably California. Type material—Not located. Neotype (herein), USNM 842108, designated from USNM 46152. Bahia Class Gastropoda aequisculpta, Aluania—KEEP, 1887d:65, ex Carpenter MS. No locality given. Type material—USNM 219564, syntype. San Die- go, San Diego Co., Calif.; “on mossy rocks at low tide”’; H. Hemphill; sent to the USNM by Keep in 1910. According to a letter from Bartsch to Keep (16 Aug. 1910) in the Archives at Mills College, four additional specimens were returned to Keep. The USNM syntype was figured by BARTSCH (1911:358-359; 362; pl. 32, fig. 7). Remarks—Manzonia (Alvinia) aequisculpta (Keep, 1887, ex Carpenter MS), according to PONDER (1985: 48; 150, figs. 101G-I). castanea, Chemnitzia —KEEP, 1887d:52; fig. 33, ex Car- penter MS. No locality given. Type material—Not located. DALL & BARTSCH (1909:101) say that they borrowed the “types” from Keep, but BARTSCH (1912:322) later claimed to have examined only a single specimen. In any event, the type lot came from San Diego, San Diego Co., Calif. Remarks—Turbonilla (Pyrgiscus) castanea (Keep, 1887, ex Carpenter MS), according to PALMER (1958: 252). Figured by DALL & BARTSCH (1909:pl. 9, figs. 1, 1a), a specimen from San Pedro, Los Angeles Co., Cal- if. If workers become worried by the brevity of Keep’s description, a neotype could be designated. A lot of four specimens from the Hemphill collection is in the California Academy of Sciences with this name on it identified by Carpenter (CASIZ 049331), but there is no evidence that Keep ever saw it. Coincidentally, DALL & BARTSCH (1907:509-510; 534; pl. 47, fig. 7) named a different and new species Turbonilla (Pyrgiscus) castanea, and DALL (1908:131) renamed it 7. (P.) castanella because of its homonymy with Keep’s taxon. columbiana, Fluminicola nuttalliana var.—KEEP 1887d:63, ex Hemphill MS. Rivers of Oregon and Washing- ton. Type material—Not located. Remarks—Should apparently be Lithoglyphus colum- bianus (Keep, 1887), according to TAYLOR (1975:60), or Fluminicola columbiana Keep, 1887, according to BuRCH (1982:22; 93, fig. 145). It has often been dated from PILsBRyY, 1899:121; 123; 125. Todos Santos, Baja California Norte. Figured by DALL & BARTSCH (1909:pl. 18, figs. 7, 7a). Remarks—Should apparently be Odostomia (Chrys- allida) gracilientis (Keep, 1887, ex Carpenter MS). (Since Odostomia is treated as a feminine noun, an -7s ending would be appropriate.) It is not a secondary homonym of O. interstincta gracilenta MONTEROSATO, 1878:93; Keep’s species has an “‘-ze” in the stem, whereas Mon- terosato’s has only an “-e,” and the two adjectives are placed into different termination groups (ICZN Code Art. 57e, f). However, DALL & BARTSCH (1909:160- 161; 243; pl. 18, figs. 7, 7a) named O. (C.) virginalis as a replacement name for Keep’s taxon. They thought the two names were homonyms, misspelling both as graci- lienta. (They incorrectly dated Monterosato’s taxon as 1884.) They also inappropriately selected a type for their taxon which, as a replacement name, should retain the same type specimen as the replaced homonym. Although Keep’s taxon is virtually a nomen dubium because of its scanty description, Dall & Bartsch have essentially given it status. I think that the most nomen- claturally stable solution is to make their “type” of O. virginalis the neotype of Keep’s Evalea graciliente; this then simultaneously makes it the neotype of Dall & Bartsch’s unnecessary replacement name. insculpta, Oscilla—KEEP, 1887d:52, ex Carpenter MS. No locality given, but presumably southern California. Type material—Not located. Neotype (herein), USNM 106501. Punta Abreojos, Baja California Norte. Figured by DALL & BARTSCH (1909:pl. 20, figs. 8, 8a). Remarks—A secondary homonym of Odostomia in- sculpta DE Kay, 1844:115-116; 263; pl. 31, fig. 297. DALL & BARTSCH (1909:183; 244; pl. 20, figs. 8, 8a) proposed O. (Jolaea) eucosmia expressly as a replace- ment name, but they inappropriately designated a “type” for their taxon. As with the preceding, the most nomenclaturally sta- ble solution is to make their “type” a neotype of Keep’s taxon, which in turn makes it a neotype of theirs. The correct name for the species is Odostomia (Jolaea) eucos- mia Dall & Bartsch, 1909. interclathrata, Clathurella—KEEP, 1887d:65, ex Carpenter MS. No locality given, but presumably California. E. Coan, 1985 Page 213 Type material—Not located. Remarks—Because this has not been cited since its first appearance and because of its ambiguous descrip- tion, it should probably be regarded as a nomen dubium. subquadrata, Amphisphyra—KEEP, 1887d:125, ex Carpen- ter MS. No locality given, but presumably Cali- fornia. Type material—Not located. There are three speci- mens in the California Academy of Sciences identified by Carpenter from the Hemphill collection (CASIZ 049330), but there is no evidence that Keep ever saw them. Remarks—Workers on opisthobranchs may want to consider whether this should be regarded as the earliest name for Diaphana californica DALL, 1919:299. tincta, Tegula gallina—KEEP, 1887d:84. No locality given, but presumably southern California. Type material—Not located. Remarks—A synonym of Tegula gallina (FORBES, 1852:271). This varietal name has sometimes been dat- ed from PILsBRY, 1889:169-170, ex Hemphill MS. Keep probably also got the name from Hemphill, but he didn’t credit it to him. Class Polyplacophora decoratus, Callistochiton —KEEP, 1887d:112, ex Carpenter MS. No locality given, but presumably southern California. Type material—Not located. Remarks—An overlooked introduction of this name, according to Ferreira (in litt., 26 March 1984), which has generally been dated from PILsBRy, 1893:269-270. See also FERREIRA (1979:448-449). fimbriatus, Callistochiton —KEEP, 1887d:112, ex Carpenter MS. No locality given, but presumably southern California. Type material—Not located. Remarks—An overlooked introduction of this name, making it a senior synonym of Callistochiton crassico- status PILSBRY, 1893:264-265, according to Ferreira, in litt., 26 March 1984. See also FERREIRA (1979:447- 448). This name is not preoccupied by Chiton fimbriatus SOWERBY, 1840:293-294, a Peruvian chiton. ACKNOWLEDGMENTS I appreciated the help and advice of Barry Roth and James T. Carlton in the initial phases of this project. I would also like to thank the following individuals who provided information or commented on drafts of the manuscript: Antonio J. Ferreira, Paul Finnegan, Terrence Gosliner, Jane Ing, Tamio Kotaka, David Lindberg, James H. McLean, Kenshiro Ogasawara, Winston Ponder, and the late Joseph Rosewater. BIBLIOGRAPHY anpb LITERATURE CITED All works cited in the text, relevant works about Keep, and papers by Keep that pertain to biology are listed here. Volume, bulletin, and monograph numbers are in bold face; series num- bers, in parentheses, precede volume numbers; issue numbers, in parentheses, follow volume numbers; supplemental informa- tion, such as second methods of listing volumes, part numbers, and parenthetical statements are given in brackets. Plates and portraits are listed, but not text figures, maps, charts, and tables. Exact publication dates are given when possible. ANONYMOUS. 1904. West American shells [concerning Keep’s new book]. Nautilus 18(5):59-60 (6 Sept. 1904). ANONYMOUS. 1911. In memory of Professor Josiah Keep. Pamphlet from memorial service, Sept. 3, 1911. With Mills Bull. (1)4:35 pp.; 1 port. (Dec. 1911). ANONYMOUS. 1968. [About the acquisition of Mills College collection by Calif. Acad. Sci.]. Calif. Acad. Sci., Casual Crier 2(1):1-2 (1 July 1968). BARTSCH, PAUL. 1911. The Recent and fossil mollusks of the genus Alvania from the west coast of America. U.S. Natl. Mus., Proc. 41 (1863):333-362; pls. 29-32 (15 Nov. 1911). BARTSCH, PAUL. 1912. Additions to the west American pyram- idellid mollusk fauna, with descriptions of new species. U.S. Natl. Mus. Proc. 42(1903):261-289; pls. 35-38 (17 May 1912). BuRCH, JOHN BayARD. 1982. Freshwater snails (Mollusca: Gastropoda) of North America. U.S. Environmental Pro- tection Agency, Off. Resh. & Develop., Envtl. Monitoring & Support Lab., EPA-600/3-82-026:vi + 294 pp.; 775 figs. (April 1982). CARPENTER, PHILIP PEARSALL. 1864. Supplementary report on the present state of our knowledge with regard to the Mol- lusca of the west coast of North America. Brit. Assoc. Adv. Sci., Rept. 33 [for 1863]:517-686 (post-1 Aug. 1864). Coan, EUGENE V. 1984. The Bernardinidae of the eastern Pacific (Mollusca: Bivalvia). Veliger 27(2):227-237; 10 figs. (5 Oct. 1984). DALL, WILLIAM HEALEY. 1908. Note on TJurbonilla castanea and Odostomia montereyensis. Nautilus 21(11):131 (7 March 1908). DALL, WILLIAM HEALEY. 1911a. Professor Josiah Keep. Sci- ence 34(873):371 (22 Sept. 1911). DALL, WILLIAM HEALEY. 1911b. Professor Josiah Keep. Nau- tilus 25(6):61-62; frontis. (19 Oct. 1911) [a reprint of the preceding]. DALL, WILLIAM HEALEY. 1919. Descriptions of new species of Mollusca from the North Pacific Ocean in the collection of the Untied States National Museum. U.S. Natl. Mus., Proc. 56(2295):293-371 (30 Aug. 1919). DALL, WILLIAM HEALEY & PAUL BarRTSCH. 1907. The py- ramidellid mollusks of the Oregonian faunal area. U.S. Natl. Mus., Proc. 33(1574):491-534; pls. 44-48 (31 Dec. 1907). DALL, WILLIAM HEALEY & PAUL BARTSCH. 1909. A mono- graph of west American pyramidellid mollusks. U.S. Natl. Mus., Bull. 68:xii + 258 pp. (13 Dec. 1909). De Kay, JAMES ELLSWORTH. 1844. Natural history of New York. Zoology of New-York, or the New-York fauna; ... pt. V: Mollusca. Albany (State Geol. Surv.) viii + 271 pp.; 40 pls. FERREIRA, ANTONIO J. 1979. The genus Callistochiton Dall, 1879 (Mollusca: Polyplacophora) in the eastern Pacific, with the descripton of a new species. Veliger 21(4):444-466; 3 pls.; 9 figs. (1 April 1979). FoRBES, EDWARD. 1852. On the marine Mollusca discovered Page 214 during the voyage of the Herald and Pandora, by Capt. Kel- lett, R.N., and Lieut. Wood, R.N. Zool. Soc. London, Proc. for 1850 [pt. 18] (217):270-272 (24 Jan. 1852); (218):273- 274; pls. 9, 11 (post-24 Jan. 1852). HEMPHILL, HENRY. 1890. New forms of western limniades. Nautilus 4(3):25-27 (6 July 1890). KEEP, JOSIAH. 1881. Common sea-shells of California. Upton Bros.: San Francisco. 64 pp.; 16 pls. 1886a. Eminent naturalists. I. [Thomas Say]. West Amer. Sci. 2(18):85-86 (Sept. 1886). 1886b. Eminent naturalists.—II. Rafinesque. West Amer. Sci. 2(19):99-102 (Oct. 1886). . 1886c. Eminent naturalists.—III. Augustus A. Gould, M.D. West Amer. Sci. 3(20):6-8 (Dec. 1886). 1887a. Eminent naturalists.—IV. Isaac Lea, LL.D. West Amer. Sci. 3(21):25-28 (Jan. 1887). 1887b. Eminent naturalists. V. Hugh Miller. West Amer. Sci. 3(22):47-49 (Feb. 1887). 1887c. Eminent naturalists. VI. Linnaeus. West Amer. Sci. 3(25):118-119 (May 1887). 1887d. West coast shells. A familiar description of the marine, fresh water, and land mollusks of the United States, found west of the Rocky Mountains. Bancroft Bros.: San Francisco. 230 pp.; 182 figs.; frontis. (July 1887). . 1887e. Beauties of the sea. West Amer. Sci. 3(28): 153-155 (Aug. 1887). 1888a. Cabinet notes. Conchologists Exchange 2(8): 107-108 (Feb. 1888). 1888b. George W. Tryon, Jr. West Amer. Sci. 4(35): 37-38 (March 1888). 1888c. West coast shells. A familiar description of the marine, fresh water, and land mollusks of the United States, found west of the Rocky Mountains. Samuel Carson: San Francisco. 230 pp.; 182 figs.; frontis. . 1889. Summer studies in conchology. Nautilus 3(5): 54-56 (1? Oct. 1889). . 1890a. A word to young collectors. Nautilus 3(10): 115-117 (12 March 1890). 1890b. The Haliotis. Nautilus 4(2):13-15; 3 figs. (27 June 1890). 1890c. The Tryons’ Handbook for young concholo- gists. San Francisco. 8 pp.; 2 figs. 1891. Mollusks of the San Francisco markets. Nau- tilus 4(9):97-100 (11? Jan. 1891). 1892. West coast shells. A familiar description of the marine, fresh water, and land mollusks of the United States, found west of the Rocky Mountains. S. Carson: San Fran- cisco. 230 pp.; 182 figs.; frontis. 1893. West coast shells. A familiar description of the marine, fresh water, and land mollusks of the United States, found west of the Rocky Mountains. H. S. Crocker: San Francisco. 230 pp.; 182 figs.; frontis. 1895. A study of fossil shells. Nautilus 9(1):7-10 (2 May 1895). 1896. West Coast species of Haliotis. Nautilus 9(11): 129-132 (10 March 1896). 1897. A tray of shells from Denmark. Nautilus 10(11): 124-127 (7 March 1897). 1899a. Caring for shells. Nautilus 12(11):132 (5 March 1899). 1899b. Pomatia aspersa in California. Nautilus 13(5): 60 (31 Aug. 1899). 1900. To West Coast conchologists. Nautilus 14(1): 10 (2 May 1900). The Veliger, Vol. 28, No. 2 1901a. Conchology. Conchologist 1(1):1-2 (Jan. 1901) [concerning this journal, see ROTH & CARLTON (1970)]. 1901b. Exotic mollusks in California. Nautilus 14(10): 114-115 (1 Feb. 1901). 1901c. Shells and sea-life. Western Series Readers 8. Whitaker & Ray: San Francisco. 200 pp.; 87 figs.; frontis.; 12 photos.; 1 etching (post-6 Feb. 1901). . 1902. Helix aspersa increasing in California. Nautilus 15(10):119 (5 Feb. 1902). 1904. West American shells. A description in familiar terms of the principal marine, fresh water and land mollusks of the United States found west of the Rocky Mountains, including those of British Columbia and Alaska. Whitaker & Ray: San Francisco. 360 pp.; 303 figs.; frontis. (post-11 July 1904). 1905. Edible mollusks of the Pacific. Pacific Fisher- man 3(1):19-21; 6 figs. (Jan. 1905). 1907. [On the loss of copies of West American Shells in the San Francisco fire.] Nautilus 20(12):144 (12 April 1907). . 1910a. List of the most common mollusks found around Monterey Bay. Hancock Bros.: San Francisco. 20 pp. (July 1910). 1910b [1911]. West coast shells (revised edition). A description of the principal marine mollusks living on the west coast of the United States, and of the land shells of the adjacent region. Also a chapter on the fresh water mollusks of the Pacific slope by Harold Hannibal. Whitaker & Ray- Wiggin: San Francisco, Calif. 346 pp.; 3 pls.; frontis.; 300 figs. (Dec. 1910, according to TAYLor, 1975:298). 1935 [posthumous]. The story of the pecten as told by himself. Whimsical reprints number 4 from Shells and Sea Life, a book for children, written by Josiah Keep in 1901. Eucalyptus Press: Mills College, Calif. 13 pp. (Oct. 1935). 1946 [posthumous]. The story of the pecten as told by himself. A chapter from Shells and Sea Life, written for children by Josiah Keep in 1901. Eucalyptus Press: Mills College, Calif. 10 pp. (Dec. 1946). 1949 [posthumous]. The poetry of shells. Eucalyptus Press: Mills College, Calif. 25 pp. (Oct. 1949). KEEP, JOSIAH [POSTHUMOUS] & JOSHUA L. BalILy, JR. 1935. West Coast shells: a description in familiar terms of the principal marine, fresh-water, and land mollusks of the United States, British Columbia, and Alaska, found west of the Sierra. Stanford Univ. Press: Stanford, Calif. & Oxford Univ.: London. xii + 350 pp.; 334 figs. (post-1 Feb. 1935). MONTEROSATO, TOMMASO DI MARIA ALLERI [MARCHESE DI]. 1878. Eumerazione e sinonimia delle conchiglie Mediter- ranee. Palermo, Giorn. di Scienz. Natur. ed Econ. 13:61- AS: PALMER, KATHERINE EVANGELINE HILTON (VAN WINKLE). 1958. Type specimens of marine Mollusca described by P. P. Car- penter from the West Coast (San Diego to British Colum- bia). Geol. Soc. Amer., Mem. 76:viii + 376 pp.; 35 pls. (8 Dec. 1958). Pitspry, HENRY AuGusTus. 1889. [Trochidae, pt. 2]. Manual Conchology (1)11(42):65-128; pls. 15-32 (5 July 1889). Pitspry, HENRY AuGuUsTUS. 1893. Polyplacophora. Lepido- pleuridae, Ischnochitonidae, Chitonidae, Mopaliidae. Man. Conch. (1)14(56, 56a):209-350 + i-xxxiv; pls. 41-68 (1 July 1893). Pitssry, HENRY AuGusTus. 1899. Catalogue of the Amnico- E. Coan, 1985 lidae of the western United States. Nautilus 12(11):121- 127 (5 March 1899). PitsprY, HENRY AuGuSTUS. 1904. West American Shells [a review]. Nautilus 18(8):95-96 (17 Dec. 1904). PONDER, WINSTON F. 1985. A review of the genera of the Rissoidae (Mollusca: Mesogastropoda: Rissoacea). Austra- lian Mus., Rec. Suppl. 4:221 pp; 153 figs. (12 Feb. 1985). RotuH, BARRY & JAMES T. CARLTON. 1970. A forgotten pe- riodical of West American conchology. Nautilus 84(1):31- 32 (16 July 1970). Page 215 SOWERBY, GEORGE BRETTINGHAM, II. 1840. Descriptions of some new chitons. Mag. Natur. Hist. (n.s.) 4(42):287-294; pl. 16 (June 1840). TayLor, DWIGHT WILLARD. 1975. Index and bibliography of late Cenozoic freshwater Mollusca of western North Amer- ica. Univ. Michigan, Mus. Paleo., Claude W. Hibbard Mem. Vol. 1 [Papers on Paleo. no. 5]:284 pp. The Veliger 28(2):216-220 (October 1, 1985) THE VELIGER © CMS, Inc., 1985 NOTES, INFORMATION & NEWS Concerning Carpenter’s “First Duplicate Series” of Mazatlan Shells by Eugene Coan Research Associate, Department of Invertebrate Zoology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118 and Joseph Rosewater Department of Invertebrate Zoology, U.S. National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 It has evidently not come to the attention of many workers on the marine molluscan fauna of the tropical eastern Pacific that an important collection of the mollusks studied by Philip P. Carpenter in the preparation of his Mazatlan catalogue (CARPENTER, 1857b), formerly at the New York State Museum, is now housed in the Division of Mollusks of the U.S. National Museum of Natural History (NMNH). Although this transfer, which occurred in Sep- tember 1955, was mentioned by BRANN (1966:12), many workers have overlooked the material because it is stored in special cases apart from both the main and the type collections. This collection, which Carpenter called the “first du- plicate series” of the Reigen collection, is of special sig- nificance in part because it contains syntypes of many of Carpenter’s new taxa from Mazatlan (or paralectotypes if lectotypes have already been selected). In general, work- ers have properly chosen to select lectotypes from among the primary collection of Mazatlan mollusks in the British Museum (Natural History) in London. However, there are some instances in which the BM(NH) specimens are lost, broken, or have distintegrated. When this has oc- curred, the material in the NMNH may have intact syn- types for lectotype designation. For some species, there are more specimens in the NMNH_ than there are in the British Museum (Natural History) holdings. The collection in the NMNH was listed by CARPENTER (1860), and later discussed and listed by PALMER (1951). After its transfer to the NMNH, it was arranged and catalogued by Florence Ruhoff just before she left that institution. It is now housed in two cabinets near the end of the mollusk collection. As Carpenter originally request- ed when he turned the material over to the New York State Museum, the collection is stored as an intact unit containing both the type material of his new species as well as specimens of other species. Most of the specimens originally mounted on Carpenter’s glass slides have been removed and stored in regulation containers like those used for the balance of the NMNH collection. The orig- inal species and tablet numbers were carefully recorded on new specimen labels, as these data are of great impor- tance in recognizing types. Bibliography and Literature Cited BRANN, D.C. 1966. Illustrations to “Catalogue of the collec- tion of Mazatlan shells” by Philip P. Carpenter. Paleo. Res. Inst.: Ithaca, New York. 111 pp.; 60 pls. (1 April 1966). CARPENTER, P. P. [For a complete bibliography of Carpenter’s papers on the Mollusca, see COAN (1969); however, the most relevant to understanding his work on Mazatlan are included here.] 1855. List of four hundred and forty species of shells from Mazatlan. British Assn. Adv. Sci., Rept. 24 [for 1854]:107-108. CARPENTER, P. P. 1857a. Report on the present state of our knowledge with regard to the Mollusca of the west coast of North America. British Assn. Adv. Sci., Rept. 26 [for 1856]: 159-368 + 4 pp.; pls. 6-9 (pre-22 April 1857). [Many of the names Carpenter validated in the following work are listed here first as nomina nuda; see especially pp. 241-281.] CaRPENTER, P. P. 1857b. Catalogue of the collection of Ma- zatlan shells, in the British Museum: collected by Frederick Reigen, .... London (British Museum) i-iv + ix-xvi + 552 pp. (1 Aug. 1857) [Warrington ed., i-viii + i-xii + 552 pp., published simultaneously] [repr.: Paleo. Res. Inst., 1967]. CaRPENTER, P. P. 1860. Catalogue of the Reigen collection of Mazatlan Mollusca, presented to the State Cabinet ..., being the first duplicate of the collection presented to the British Museum. Regents Univ. State of New York, 13th Ann. Rept. on the Condition State Cabinet Natur. Hist. [for 1859]:21-36 (post-10 April 1860). CARPENTER, P. P. 1864. Supplementary report on the present state of our knowledge with regard to the Mollusca of the west coast of North America. British Assn. Adv. Sci., Rept. 33 [for 1863]:517-686 (post-1 Aug. 1864) [see especially pp. 542-548]. Coan, E. V. 1969. A bibliography of the biological writings of Philip Pearsall Carpenter. Veliger 12(2):222-225 (1 Oct. 1969). KEEN, A. M. 1968. West American mollusk types at the Brit- ish Museum (Natural History) IV. Carpenter’s Mazatlan collection. Veliger 10(4):389-439; pls. 55-59 (1 April 1968). PALMER, K. E. H. (VAN WINKLE). 1951. Catalogue of the first duplicate series of the Reigen collection of Mazatlan shells in the State Museum at Albany, New York. New York State Mus., Bull. 342:79 pp.; 1 pl. (Jan. 1951). Notes, Information & News Two Little-Known Italian Papers on Galapagos Intertidal Zonation and Mollusks by Matthew J. James Department of Paleontology, University of California, Berkeley, California 94720 Two papers that resulted from a 1971-1972 Italian ex- pedition to the Galapagos Islands were kindly brought to my attention by Dr. E. V. Coan. These papers are not likely to be known to workers interested in the marine biota of the Galapagos because they were published in Italian and in a journal not often encountered in libraries. The first paper, published in 1974, is by Francesco Cinelli and Paolo Colantoni (CINELLI & COLANTONI, 1974) and deals with some observations on the marine benthic zonation of the rocky coast of the Galapagos Is- lands. The authors describe the specific organisms, both invertebrates and algae, that were found in the supralit- toral, midlittoral, and infralittoral zones at nine stations on six islands. This zonation information is compared to the littoral zonation pattern known to occur in the Med- iterranean Sea. In addition to documenting the spatial distribution, Cinelli and Colantoni comment on the bio- geographic affinities of the organisms (including several molluscan taxa) found within the three zones. The supra- littoral and midlittoral organisms are considered to have tropical affinities, while the infralittoral organisms are considered to have temperate or cold-temperate affinities. The second paper, published in 1979, is by Marco Ta- viani (TAVIANI, 1979), and concerns the chitons, gastro- pods, and bivalves collected by the Italian expedition. Taviani documents the occurrence of 3 chitons, 52 gastro- pods, and 9 bivalves from 14 stations on 10 islands. The text provides information and observations about each species, and the plates have good illustrations. No new species are described. In addition, Taviani discusses the origin and composition of the molluscan fauna from geo- logical, paleontological, and geographic perspectives. Both papers are recommended to anyone interested in the Galapagos marine invertebrate biota, particularly mollusks. The bibliographies of both papers contain many useful Galapagos references and also five references to additional papers resulting from the same Italian expe- dition, including one on chitons. An address for Marco Taviani, to whom reprint requests for both papers can be sent, is: Laboratorio di Geologia Marina del C.N.R., Via Zamboni 65, 40127 Bologna, Italia. Literature Cited CINELLI, F. & P. COLANTONI. 1974. Alcune observazioni sulla zonazione del benthos marino sulle coste rocciose delle Isole Galapagos (Oceano Pacifico). Museo Zoologico dell’Universita di Firenze [Florence]: Galapagos, Studi e Page 217 ricerche. Spedizone ‘L. Mares-G.R.S.T.S.. Gruppo Ri- cerche Scientifiche e Techniche Subacquee. 22 pp., 17 figs. TAvIANI, M. 1979. I molluschi marini raccolti dalla spedizione “L. Mares-G.R.S.T.S.” alle Isole Galapagos 1. Gastropoda e Bivalvia. Museo Zoologico dell’Universita di Firenze [Florence]: Galapagos, Studi e ricerche. Spedizone ‘L. Mares- G.R.S.T.S.’ Gruppo Ricerche Scientifiche e Techniche Su- bacquee. 61 pp., 90 figs. Some Additional Notes on the Distributions of Eastern Pacific Donacidae by Eugene Coan Research Associate, Department of Invertebrate Zoology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118 In redistributing the research materials from the office of the late Dr. Joseph P. E. Morrison, curators at the U.S. National Museum of Natural History came across a num- ber of lots of eastern Pacific Donax. Morrison had evi- dently planned to work on this group and had isolated some interesting specimens for examination, specimens that I did not have a chance to see during my study of that group (Coan, E. 1983. The eastern Pacific Donacidae. Veliger 25(4):273-297). Two of these lots provide new distributional records: Donax caelatus caelatus—Occurs as far south as Isla San José, Panama (8°15'N, 79°8’W) (USNM 598877a). I had previously seen specimens only from as far south- east as Golfito, Costa Rica. Donax dentifer—Occurs as far north as Tapachula, Chia- pas, Mexico (14°43'N, 92°26’W) (USNM 591610). This extends the known distribution from Guatemala north- ward into Mexico. An Extension of the Known Depth Range for Sepia elegans Blainville, 1827 (Cephalopoda: Sepioidea) by Angel Guerra Instituto de Investigaciones Pesqueras, Muelle de Bouzas s/n, Vigo, Spain The cuttlefish Sepia elegans Blainville, 1827, has a geo- graphical distribution extending in the eastern Atlantic from 15° to 55°N, and throughout the Mediterranean Sea. The total depth range of the species was listed as 60 to 450 m in the Atlantic Ocean, and from 20 to 250 m in the Mediterranean Sea (MANGOLD-WIRZ, 1963). ROPER et al. (1984) have pointed out that Sepza elegans is a small, demersal species with a depth range from 30 to 430 m. Page 218 The purpose of this note is to show that the shallow limit of the bathymetric distribution of Sepza elegans must be changed. We frequently have captured this species from 2 to 45 m depth in the Ria de Vigo (Vigo estuary) (42°15'N-8°48’W). The Ria de Vigo is a drowned tectonic valley. Its water circulation is estuarine. Interchanges with Atlantic water are very important. Sepia elegans occupies habitats in the central and outer parts of the estuary but is not found in the inner reaches of the estuary. Sepia officinalis is more euryhaline and lives and spawns throughout the estuary including its inner part, where there are large fluctuations in salinity (GUERRA, 1984; in prep.). Literature Cited GUERRA, A. 1984. Cefalopodos de la Ria de Vigo. Resultados preliminares. Cuadernos de Area de Ciencias Marinas. Seminario de Estudos Galegos 1:333-348. MANGOLD-WIRZ, K. 1963. Biologie des Céphalopodes ben- tiques et nectoniques de la Mer Catalane. Vie et Milieu (Suppl.) 13:1-285. Roper, C. F. E., M. J. SWEENEY & C. E. NAUEN. 1984. FAO species catalogue. Vol. 3. Cephalopods of the world. An annotated and illustrated catalogue of the species of interest to fisheries. FAO Fish. Synop., No. 125, Vol. 3, 277 pp. California Malacozoological Society California Malacozoological Society, Inc., is a non-profit educational corporation (Articles of Incorporation No. 463389 were filed January 6, 1964 in the office of the Secretary of State). The Society publishes a scientific quarterly, The Veliger. Donations to the Society are used to pay a part of the production costs and thus to keep the subscription rate at a minimum. Donors may designate the Fund to which their contribution is to be credited: Operating Fund (available for current production); Sav- ings Fund (available only for specified purposes, such as publication of especially long and significant papers); En- dowment Fund (the income from which is available. The principal is irrevocably dedicated to scientific and educa- tional purposes). Unassigned donations will be used ac- cording to greatest need. Contributions to the C.M.S., Inc., are deductible by donors as provided in section 170 of the Internal Revenue Code (for Federal income tax purposes). Bequests, lega- cies, gifts, and devises are deductible for Federal estate and gift tax purposes under section 2055, 2106, and 2522 of the Code. The Treasurer of the C.M.S., Inc., will issue suitable receipts which may be used by donors to substan- tiate their tax deductions. Page Charges Although we would like to publish papers without charge, high costs of publication require that we ask authors to The Veliger, Vol. 28, No. 2 defray a portion of the cost of publishing their papers in The Veliger. We wish, however, to avoid possible financial handicap to younger contributors, or others without fi- nancial means, and to have charges fall most heavily on those who can best afford them. Therefore, the following voluntary charges have been adopted by the Executive Board of the California Malacozoological Society: $30 per printed pages for authors with grant or institutional sup- port and $10 per page for authors who must pay from personal funds (2.5 manuscript pages produce about 1 printed page). In addition to page charges, authors of papers containing an extraordinary number of tables and figures should expect to be billed for these excess tables and figures at cost. It should be noted that even at the highest rate of $30 per page the Society is subsidizing well over half of the publication cost of a paper. However, authors for whom the regular page charges would present a financial handicap should so state in a letter accompa- nying the original manuscript. The letter will be consid- ered an application to the Society for a grant to cover necessary publication costs. We emphasize that these are voluntary page charges and that they are unrelated to acceptance or rejection of manu- scripts for The Veliger. Acceptance is entirely on the basis of merit of the manuscript, and charges are to be paid after publication of the manuscript, if at all. Because these contributions are voluntary, they may be considered by authors as tax deductible donations to the Society. Such contributions are necessary, however, for the continued good financial health of the Society, and thus the contin- ued publication of The Veliger. Reprints While it was hoped at the “birth” of The Veliger that a modest number of reprints could be supplied to authors free of charge, this has not yet become possible. Reprints are supplied to authors at cost, and requests for reprints should be addressed directly to the authors concerned. The Society does not maintain stocks of reprints and also can- not undertake to forward requests for reprints to the au- thor(s) concerned. Patronage Groups Since the inception of The Veliger in 1958, many generous people, organizations, and institutions have given our journal substantial support in the form of monetary do- nations, either to The Veliger Endowment Fund, The Ve- liger Operating Fund, or to be used at our discretion. This help has been instrumental in maintaining the high qual- ity of the journal, especially in view of the rapidly rising costs of production. At a recent Executive Board Meeting, we felt we should find a way to give much-deserved recognition to those past Notes, Information & News and future donors who so evidently have our best interests at heart. At the same time, we wish to broaden the basis of financial support for The Veliger, and thus to serve our purpose of fostering malacological research and publica- tion. Accordingly, it was decided to publicly honor our friends and donors. Henceforth, donors of $1000.00 or more will automatically become known as Patrons of The Veliger, donors of $500.00 or more will be known as Sponsors of The Veliger, and those giving $100.00 or more will become Benefactors of The Veliger. Lesser donations are also sincerely encouraged, and those donors will be known as Friends of The Veliger. As a partial expression of our gratitude, the names of donors in these different categories will be listed in a regular issue of the journal. Of course, we will honor the wishes of any donor who would like to remain anonymous. The Treasurer of the California Malacozoological Society will provide each member of the new patronage groups with a receipt that may be used for tax purposes. We thank all past and future donors for their truly helpful support and interest in the Society and The Veli- ger. Through that support, donors participate directly and importantly in producing a journal of high quality, one of which we all can be proud. Notes to Prospective Authors The increasing use of computers to prepare manuscript copy prompts the following notes. We request that the right margin of submitted papers be prepared “ragged,” that is, not justified. Although right-justified margins on printed copy sometimes look “neater,” the irregular spac- ing that results between words makes the reviewer’s, ed- itor’s, and printer’s tasks more difficult and subject to error. Similarly, the automatic hyphenation capability of many machines makes for additional editorial work and potential confusion; it is best not to hyphenate words at the end of a line. Above all, manuscripts should be printed with a printer that yields unambiguous, high-quality copy. With some printers, especially some of the dot-matrix kinds, copy is generally difficult to read and, specifically, the letters “a, p, g, and q” are difficult to distinguish, especially when underlined as for scientific names; again, errors may result. Other reminders are (1) that three copies of everything (figures, tables, and text) should be submitted to speed the review process, and (2) absolutely everything should be double-spaced, including tables, references, and figure leg- ends. Because The Veliger is an international journal, we oc- casionally receive inquiries as to whether papers in lan- guages other than English are acceptable. Our policy is that manuscripts must be in English. In addition, authors whose first language is other than English should seek the assistance of a colleague who is fluent in English before submitting a manuscript. Page 219 Subscription Rates and Membership Dues At its regular Annual Business Meeting on October 19, 1984, the Executive Board of the California Malacozoo- logical Society, Inc., set the subscription rates and mem- bership dues for Volume 28 of The Veliger. For affiliate members of the Society, the subscription rate for Volume 28 will remain unchanged at US$22.00; this now includes postage to domestic addresses. For libraries and nonmem- bers the subscription rate will increase very slightly to US$44.00, also now with postage to domestic addresses included. An additional US$3.00 is required for all sub- scriptions sent to foreign addresses, including Canada and Mexico. Affiliate membership in the California Malacozoologi- cal Society is open to persons (no institutional member- ships) interested in any aspect of malacology. There is a one-time membership fee of US$2.00, after payment of which, membership is maintained in good standing by the timely renewal of the subscription. Send all business correspondence, including subscription orders, membership applications, payments for them, and changes of address to C.M.S., Inc., P.O. Box 9977, Berke- ley, CA 94709. Moving? If your address is changed it will be important to notify us of the new address at least six weeks before the effective date and not less than s7x weeks before our regular mailing dates. Send notification to C.M.S., Inc., P.O. Box 9977, Berkeley, CA 94709. Because of a number of drastic changes in the regula- tions affecting second class mailing, there is now a sizable charge to us on the returned copies as well as for our remailing to the new address. We are forced to ask our members and subscribers for reimbursement of these charges; further, because of increased costs in connection with the new mailing plate, we also must ask for reim- bursement of that expense. The following charges must be made: change of address and re-mailing of a returned issue— $2.75 minimum, but not more than actual cost to us. We must emphasize that these charges cover only our actual expenses and do not include compensation for the extra work involved in re-packing and re-mailing re- turned copies. Erratum: Volume 27, Number 3 (January 2, 1985) Due to an oversight by the author, an error occurred in the article by Paul S. Mikkelsen, 1985, A comparison of two Florida populations of the coquina clam, Donax var- Page 220 iabilis Say, 1822 (Bivalvia: Donacidae). II. Growth rates. Veliger 27(3):308-311. Because the error is in a result, and thus may cause problems for future investigators, we make it known here. On page 308, eighth line of “Results,” change “7.3” to a3 Sli Opinions: International Commission on Zoological Nomenclature The following Opinions of potential interest to our read- ers have been published by the International Commission on Zoological Nomenclature in the Bulletin of Zoological Nomenclature, Volume 42, Part 2, on 27 June 1985: Opinion No. 1306 (p. 146). Ledella bushae Warén, 1978, is the type species of Ledella Verrill & Bush, 1897 (Mollusca, Bivalvia). Opinion No. 1315 (p. 165). Eolis alderi Cocks, 1852, is the type species of Aeolidiella Bergh, 1867 (Mollusca, Gastropoda). Sale of C. M. S. Publications All back volumes still in print, both paper-covered and cloth-bound, are available only through “The Shell Cab- inet,” 12991 Bristow Road, Nokesville, VA 22123. The same applies to the supplements still in print, with certain exceptions (see below). Prices of available items may be obtained by applying to Mr. Morgan Breeden at the above address. Volumes 1 through 13, 24, 26, and 27 are out of print. Supplements still available are: part 1 and part 2, sup- plement to Vol. 3, and supplements to Volumes 7, 11, 14, 15, and 16; these can be purchased from ‘““The Shell Cab- inet” only. Copies of the supplement to Volume 17 (“Growth rates, depth preference and ecological succes- sion of some sessile marine invertebrates in Monterey Harbor” by E. C. Haderlie) may be obtained by applying to Dr. E. C. Haderlie, U.S. Naval Post-Graduate School, Monterey, CA 93940; the supplement to Volume 18 (“Chitons”’) is available from ‘““The Secretary,” Hopkins Marine Station, Pacific Grove, CA 93950. Some out-of-print editions of the publications of C. M. S. are available as microfiche reproductions through Mr. Steven J. Long. The microfiches are available as negative films (printed matter appearing white on black back- The Veliger, Vol. 28, No. 2 ground), 105 mm x 148 mm, and can be supplied im- mediately. The following is a list of items now ready: Volumes 1-6: $9.95 each Volumes 7-12: $12.95 each Supplement to Volume 6: $3.95; to Volume 18, $6.95 Send orders to Mr. Steven J. Long, Shells and Sea Life, 1701 Hyland, Bayside, CA 95524. Donations We wish to acknowledge and thank some of the many friends of The Veliger who have contributed financial re- sources in years past. Through their support, these gen- erous donors have participated importantly in producing a journal of high quality. Thank you. Patrons—$1000 or more Mrs. Jean M. Cate Dr. Eveline DuBois-Reymond Marcus Dr. and Mrs. Rudolf Stohler Dr. Steven Wainwright Sponsors—$500 or more Dr. J. Gibson-Smith Benefactors—$100 or more Mrs. Twila L. Bratcher Dr. Eugene V. Coan Conchological Club of Southern California Mrs. Sandra Gardner Mrs. Richard H. Hagemeyer Dr. Cadet Hand Dr. Harold W. Harry Dr. David R. Lindberg Dr. Victor L. Loosanoff San Diego Shell Club Dr. Ralph I. Smith Dr. Joan E. Steinberg Friends Dr. Kikutaro Baba Dr. Robert D. Barnes Dr. H. A. Bern Dr. James T. Carlton Dr. John S. Hensill Dr. Carol N. Hopper Dr. Hiroshi Hoshikawa Mr. John P. McQueen Mr. Paul S. Mikkelsen Drs. Alan C. and Susanne E. Miller Mrs. Gordon Neiswanger Dr. Robert Robertson Dr. W. D. Russell-Hunter Dr. Louella R. Saul Dr. Earl Segal Dr. Edmund H. Smith Dr. Thomas R. Waller Dr. Druid Wilson Information for Contributors Manuscripts Manuscripts must be typed on white paper, 812” by 11”, and double-spaced throughout (including references, figure legends, footnotes, and tables). If computer generated copy is to be submitted, margins should be ragged right (7.e., not justified). To facilitate the review process, manuscripts, including figures, should be submitted in triplicate. The first mention in the text of the scientific name of a species should be accompanied by the taxonomic authority, including the year, if possible. Underline scientific names and other words to be printed in italics. Metric and Celsius units are to be used. The sequence of manuscript components should be as follows in most cases: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, lit- erature cited, figure legends, figures, footnotes, and tables. The title page should be on a separate sheet and should include the title, author’s name, and address. The abstract should describe in the briefest possible way (normally less than 200 words) the scope, main results, and conclusions of the paper. Literature cited References in the text should be given by the name of the author(s) followed by the date of publication: for one author (Smith, 1951), for two authors (Smith & Jones, 1952), and for more than two (Smith ef al., 1953). The “literature cited” section must include all (but not additional) references quoted in the text. References should be listed in alphabetical order and typed on sheets separate from the text. Each citation must be complete and in the following form: a) Periodicals Cate, J. M. 1962. On the identifications of five Pacific Mitra. Veliger 4:132-134. b) Books Yonge, C. M. & T. E. Thompson. 1976. Living marine molluscs. Collins: London. 288 pp. c) Composite works Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135. In: R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal invertebrates of California. Stanford Univ. Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Figures and plates Figures must be carefully prepared and should be submitted ready for publication. Each should have a short legend, listed on a sheet following the tables. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photographs for half-tone plates must be of good quality. They should be trimmed off squarely, arranged into plates, and mounted on suitable drawing board. Where necessary, a scale should be put on the actual figure. Preferably, photographs should be in the desired final size. It is the author’s responsibility that lettering is legible after final reduction (if any) and that lettering size is appropriate to the figure. Charges will be made for necessary alterations. Processing of manuscripts Upon receipt each manuscript is critically evaluated by at least two referees. Based on these evaluations the editor decides on acceptance or rejection. Acceptable manuscripts are returned to the author for consideration of comments and criticisms, and a finalized manuscript is sent to press. The author will receive from the printer two sets of proofs, which should be corrected carefully for printing errors. At this stage, stylistic changes are no longer appropriate, and changes other than the correction of printing errors will be charged to the author at cost. One set of corrected proofs should be returned to the editor. An order form for the purchase of reprints will accompany proofs. If reprints are desired, they are to be ordered directly from the printer. Send manuscripts, proofs, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. CONTENTS — Continued A bibliography and list of molluscan names of Josiah Keep. EUGENE| | GOAN (0 oie NUR NSUSNE Sa eee TET Ne Ce Ge ae 211 NOTES, INFORMATION & NEWS Concerning Carpenter’s “‘first duplicate series” of Mazatlan shells. EUGENE COAN AND’ JOSEPH) ROSEWAGER | 2.0) higy a iy ea ee 216 Two little-known Italian papers on Galapagos intertidal zonation and mol- lusks. MATTHEW J. JAMESG 560.) 5790 ie EEN al Sree ae A auc scien alte ae ZT Some additional notes on the distributions of eastern Pacific Donacidae. EUGENE GOAN) fi hah sos. cucs cote beaten bets oe St URE eae ON GO Tena cad PAT An extension of the known depth range for Sepia elegans Blainville, 1827 (Cephalopoda: Sepioidea). ANGEL GUERRA. 5 054 Mie ste an Oe ere ht Sn a 217 ISSN 0042-3211 VELIGER, A Quarterly published by \ CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. — Berkeley, California R. Stohler, Founding Editor Volume 28 January 2, 1986 Number 3 CONTENTS A revision of the genus Acanthopleura Guilding, 1829 (Mollusca: Polyplacoph- ora). PRNGRONTLO Me PER RE NRA ve scott aye. sud emul eco taleeal Rustliaee Jere ue i ech LUM dee a Leyte Ro 221 Cerithidea reidi, spec. nov., from Western Australia. IUGEIARDE Se EL OUBRIG Kear cin ce eP teenies 8S oie Wall MM emee ra elgates aly ty APY ao Hct 280 Two new bulimulid land snail species from Isla Santa Cruz, Galapagos Islands. Sir ESN Aly CEVAINMIBERS Ws over tooiey hts nh sel ikon Rie tier 3 core inet GUEST ea 287 Nassarius (Gastropoda: Neogastropoda) from the Galapagos Islands. ERIZABERH CROZIERY NESBIGT AND WILLIAM, PITT 03) 5 2 ess 9 San a 294 On Pleurobranchomorpha from Italian seas (Mollusca: Opisthobranchia). RICCARDO CAMMANEO=VMEGEL 2). 5.25 oi 6 os ise testeyeeeiss2! figs Sa) aap erie) AW ahtaeen 302 Swimming tracks of Aplysia brasiliana, with discussion of the roles of swimming in sea hares. EA SeEMANTINBON meric sie ieee AU a ht gees STlUNtai 1 8) Be eee 310 A short-term study of growth and death in a population of the gastropod Strom- bus gibberulus in Guam. CEERAT EI EOVERMEI AND) EDIE: ZIPSER) 2 ot ise al a tyne lay ace 314 Aspects of the reproduction of rocky intertidal mollusks from the Jordan Gulf of Aqaba (Red Sea). INSEE @MPEAUIIEING Siete ain: soaeernet earn Sc otlons Riel i ciaie tine me eNan.t s eoceeaitety Si awete a 318 CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly on the first day of July, October, January and April. Rates for Volume 28 are $22.00 for affiliate members (including domestic mailing charges) and $44.00 for libraries and nonmembers (including do- mestic mailing charges). An additional $3.00 is required for all subscriptions sent to foreign addresses, including Canada and Mexico. Further membership and subscrip- tion information appears on the inside cover. The Veliger is published by the Cali- fornia Malacozoological Society, Inc., % Department of Zoology, University of Cali- fornia, Berkeley, CA 94720. Second Class postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to C.M.S., Inc., P.O. Box 9977, Berkeley, CA 94709. THE VELIGER Scope of the journal The Veliger is open to original papers pertaining to any problem concerned with mol- lusks. This is meant to make facilities available for publication of original articles from a wide field of endeavor. Papers dealing with anatomical, cytological, distributional, eco- logical, histological, morphological, physiological, taxonomic, etc., aspects of marine, freshwater, or terrestrial mollusks from any region will be considered. Short articles containing descriptions of new species or lesser taxa will be given preferential treatment in the speed of publication provided that arrangements have been made by the author for depositing the holotype with a recognized public Museum. Museum numbers of the type specimen must be included in the manuscript. Type localities must be defined as accurately as possible, with geographical longitudes and latitudes added. Very short papers, generally not exceeding 500 words, will be published in a column entitled “NOTES, INFORMATION & NEWS”; in this column will also appear notices of meetings, as well as news items that are deemed of interest to our subscribers in general. Editor-in-Chief David W. Phillips, 2410 Oakenshield Road, Davis, CA 95616, USA Editorial Board Donald P. Abbott, Emeritus, Hopkins Marine Station of Stanford University Hans Bertsch, Universidad Autonoma de Baja California James T. Carlton, Williams College—Mystic Seaport Eugene V. Coan, Research Associate, California Academy of Sciences, San Francisco J. Wyatt Durham, University of California, Berkeley Cadet Hand, University of California, Berkeley Carole S. Hickman, University of California, Berkeley A. Myra Keen, Emerita, Stanford University David R. Lindberg, University of California, Berkeley James H. McLean, Los Angeles County Museum of Natural History Frank A. Pitelka, University of California, Berkeley Peter U. Rodda, California Academy of Sciences, San Francisco Clyde F. E. Roper, National Museum of Natural History, Washington Barry Roth, Santa Barbara Museum of Natural History Judith Terry Smith, Stanford University Ralph I. Smith, University of California, Berkeley Wayne P. Sousa, University of California, Berkeley T. E. Thompson, University of Bristol, England Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (no institutional memberships) interested in any aspect of malacology. As an afhliate member, a person may subscribe to The Veliger for US $22.00, which now includes mailing charges to domestic addresses. There is a one-time membership fee of US $2.00, after payment of which, membership is maintained in good standing by the timely renewal of the subscription; a reinstatement fee of US $1.00 will be required if membership renewals do not reach the Society on or before April 1 preceding the start of the new Volume. If a receipt is required, a self-addressed, stamped envelope (or in the case of foreign mem- bers, the envelope and two International Postal Reply coupons) should be included with the membership or subscription request. The annual subscription rate to The Veliger for libraries and nonmembers is US $44.00, which now includes mailing charges to domestic addresses. An additional US $3.00 is required for all subscriptions sent to foreign addresses, including Canada and Mexico. Memberships and subscriptions are by Volume only (July 1 to April 1) and are payable in advance to California Malacozoological Society, Inc. Single copies of an issue are US $30.00 plus postage. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: C.M.S., Inc., Post Office Box 9977, Berkeley, CA 94709. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. The Veliger 28(3):221-279 ( January 2, 1986) THE VELIGER © CMS, Inc., 1986 A Revision of the Genus Acanthopleura Guilding, 1829 (Mollusca: Polyplacophora) ANTONIO J. FERREIRA! Research Associate, Department of Invertebrate Zoology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118, U.S.A Abstract. Fifteen species of Acanthopleura Guilding, 1829 (=Corephium Gray, 1847a; Enoplochiton Gray, 1847a; Maugeria Gray, 1857; Sclerochiton Dall, 1881; Francisia Dall, 1882; Rhopalopleura Thiele, 1893; Mesotomura Pilsbry, 1893a; Amphitomura Pilsbry, 1893a; Liolophura Pilsbry, 1893a; Squamo- pleura Nierstrasz, 1905a; Clavarizona Hull, 1923; Acanthozostera Iredale & Hull, 1926; Planispina Taki, 1962) are here recognized, two of which are new to science. The fifteen species are A. granulata (Gmelin, 1791), A. spinosa (Bruguiére, 1792), A. echinata (Barnes, 1824), A. nigra (Barnes, 1824), A. gemmata (Blainville, 1825), A. hirtosa (Blainville, 1825), A. gaimardi (Blainville, 1825), A. loochooana (Broderip & Sowerby, 1829), A. brevispinosa (Sowerby, 1840a), A. japonica (Lischke, 1873), A. curtis- zana (Smith, 1884), A. miles (Carpenter im Pilsbry, 1893c), A. araucariana (Hedley, 1898), A. arenosa Ferreira, spec. nov., and A. rehderi Ferreira, spec. nov. Despite large specimen size, intertidal habitat, and world-wide, mostly tropical distribution, Acanthopleura is not represented in the fossil record. The genus-name Acanthopleura Guilding, 1829, began as a grouping of seven heterogeneous “‘sections,” each de- fined by the girdle characteristics of a given species. Al- though subsequent authors (SWAINSON, 1840; GRay, 1847a, 1857; SHUTTLEWORTH, 1853; Carpenter in DALL, 1882) altered GUILDING’s (1829) concept, it was left to PILsBRY (1893c:213-218) to restrict Acanthopleura to the “section” characterized by ‘“‘zona [=girdle] spinosa,” and typified by Chiton spinosus Bruguiére, 1792. Assigning species without teeth in the posterior valve to a new genus, Liolophura, PILsBRY (1893c) divided Acanthopleura into four subgenera based upon the characteristics of the in- sertion plate of the posterior valve: (1) Acanthopleura s.s. (type, C. spinosus Bruguiére), with ‘“‘a very long insertion plate cut into numerous teeth by short slits,” (2) Maugeria Gray, 1857 (restricted) (type, C. piceus Gmelin), with “the pectinated insertion plate cut into numerous teeth by slits similar to those of the head-valve,” (3) Amphitomura Pilsbry, 1893a (type, C. borbonica Deshayes), with “the insertion-plate very short, with blunt, crenulated edge, interrupted only by a single mopaloid slit on each side,” and (4) Mesotomura Pilsbry, 1893a (type, C. echinatum ' For reprints: 2060 Clarmar Way, San Jose, California 95128, U.S.A. Barnes), with “the long insertion plate deeply pectinated outside, its edge interrupted only by a single median-pos- terior slit.” Yet, Acanthopleura (sensu PILSBRY, 1893c), a wide- spread, tropical to subtropical group of large, accessible (intertidal to emergent), and abundant specimens, has re- mained problematical in regard to the biological species involved and their taxonomic arrangement. Complicated by the appearance of several ill-defined nominal genera, subgenera, species, and subspecies much before clear un- derstanding of the whole group had been attained, Acan- thopleura has been sorely in need of revision. But, likely, the difficulties in assembling sufficient material from ex- otic areas have been such as to stifle earlier attempts to study the group. This revision rests upon the examination of material in the museum collections of the California Academy of Sci- ences, San Francisco (CAS); Los Angeles County Mu- seum of Natural History (LACM); University of Colo- rado Museum, Boulder, Colorado (UCM); Academy of Natural Sciences of Philadelphia (ANSP); U.S. National Museum of Natural History, Washington, D.C. (USNM); British Museum (Natural History) (BMNH); Muséum National d’Histoire Naturelle, Paris (MNHN); Uppsala Universitets Zoologiska Museum, Sweden (UUZM); The Australian Museum, Sydney (AMS); Western Australian Page 222 _ The Veliger, Vol. 28, No. 3 Museum, Perth (WAM); National Museum of Victoria, Melbourne (NMV); Institut Royal des Sciences Natu- relles de Belgique, Bruxelles (IRScnB); Musée d’ Histoire Naturelle, Bucarest, Romania (MHNB); Istituto di Zoo- logia dell’Universita di Firenze, Italy (MF); and in the personal collections of Dr. John S. Pearse, University of California, Santa Cruz, California; Salle Crittenden, Oakland, California; Clay Carlson and Patty Jo Hoff, Merize, Guam; Ian Loch, Sydney, Australia; Richard A. Van Belle, Sint-Niklaas, Belgium; J. R. Penprase, West Hobart, Tasmania; and J. R. Penniket, Warkworth, New Zealand. Additional material and field observations were obtained by, and are presently in the collection of, A. J. Ferreira (AJF collection station numbers on file at CAS). An effort was made to study all available type speci- mens deemed necessary to resolve or clarify taxonomic problems; where type material is mentioned in the liter- ature but no attempt was made here to verify its existence or repository, reference is made only to information at hand regarding types “not examined”; where there is no information in the literature about type specimens and no attempt was made here to examine or locate type material, the types are said to remain “unascertained.” SYSTEMATIC TREATMENT Polyplacophora Gray, 1821 Neoloricata Bergenhayn, 1955 Ischnochitonina Bergenhayn, 1930a Chitonidae Rafinesque, 1815 Acanthopleura Guilding, 1829 Description: Mostly large, oval, depressed, round-backed chitons. Valves thick, heavy, beaked. Tegmental sculpture coarsely granular to wrinkled, often obliterated by ero- sion; tegmentum broadly inflexed at posterior margin of intermediate valves. Mucro central to posterior. Ocelli scattered throughout anterior valve, anterior ¥%3 to % of lateral areas of intermediate valves, and postmucro area of posterior valve. Gills mostly holobranchial (7.e., extend- ing along 90-100% of foot [SIMROTH, 1894:247]). Artic- ulamentum whitish to blue, brown, or black. Strong su- tural laminae. Insertion plates markedly pectinate on outside; on posterior valve insertion teeth may be absent in part or in total, often buttressed by transverse callus; slits 8-12 on anterior valve, 1 (2 in one species) on inter- mediate valves, 0-10 on posterior valve. Girdle thick, muscular, densely covered with calcareous elements, vari- able in length and in shape from spikes to spines, spine- lets, or scales. Radula major lateral teeth with discoid heads (with 4 cusps in one species). All species of Acanthopleura are confined to the inter- tidal zone. Specimens are often found on the surface of rocks, often exposed at low tide. Type-species: Chiton spinosus Bruguiére, 1792, by sub- sequent designation (GRAY, 1847b). Synonyms: Corephium Gray, 1847a (not Browne, 1789) Type: Chiton echinatus Barnes, 1824, by subsequent designation (GRAY, 1847b). Enoplochiton Gray, 1847a Type: Chiton niger Barnes, 1824, by monotypy. Maugeria Gray, 1857 Type: Chiton piceus Gmelin, 1791 (=Chiton granulatus Gmelin, 1791), by subsequent designation (PILSBRY, 1893c). Sclerochiton Dall, 1881 (not Kraatz, 1859) Type: Chiton miles Carpenter in Pilsbry, 1893c, by monotypy. Francisia Dall, 1882 Type: Chiton spinosus Bruguiére, 1792, by original des- ignation. Rhopalopleura Thiele, 1893 Type: Chiton aculeatus Linnaeus, 1758 (?=Chiton echi- natus Barnes, 1824), by monotypy. Mesotomura Pilsbry, 1893a Type: Chiton echinatus Barnes, 1824, by monotypy. Amphitomura Pilsbry, 1893a Type: Chiton borbonicus Deshayes, 1863 (=Chiton brevispinosus Sowerby, 1840a), by original desig- nation. Liolophura Pilsbry, 1893a Type: Chiton japonicus Lischke, 1873, by original des- ignation. Squamopleura Nierstrasz, 1905a Type: Chiton miles Carpenter in Pilsbry, 1893c, by sub- sequent designation (Pilsbry, 1893c). Clavarizona Hull, 1923 Type: Chiton hirtosus Blainville, 1825, by original des- ignation. Acanthozostera Iredale & Hull, 1926 Type: Chiton gemmatus Blainville, 1825, by original designation. Planispina Is. Taki, 1962 Type: Acanthopleura (Amphitomura) planispina Ber- genhayn, 1933 (=Acanthopleura gemmata [Blain- ville, 1825]), by original designation. Remarks: The arrangement of species here allocated to Acanthopleura differs significantly from that of PILSBRY (1893c). It is based on the appreciation of over-all simi- larities among species rather than on modifications of any single character. Most generic names here synonymized under Acanthopleura had been defined solely in terms of changes in the girdle elements or in the insertion plate of the posterior valve, ignoring similarities in body plan among species. As a result, species that in most respects appear extremely close, such as Chiton gemmatus Blain- ville, 1825, and C. gaimardi Blainville, 1825, had been placed in different genera (even in different subfamilies) on account of modifications in the posterior valve, not- withstanding extreme closeness in all other features. Such situations are here corrected, and 15 species of Acantho- pleura recognized: CMe EeGreira,) (986 Acanthopleura granulata (Gmelin, 1791) Acanthopleura spinosa (Bruguiére, 1792) Acanthopleura echinata (Barnes, 1824) Acanthopleura nigra (Barnes, 1824) Acanthopleura gemmata (Blainville, 1825) Acanthopleura hirtosa (Blainville, 1825) Acanthopleura gaimardi (Blainville, 1825) Acanthopleura loochooana (Broderip & Sowerby, 1829) Acanthopleura brevispinosa (Sowerby, 1840a) Acanthopleura japonica (Lischke, 1873) Acanthopleura curtisiana (Smith, 1884) Acanthopleura miles (Carpenter in Pilsbry, 1893c) Acanthopleura araucariana (Hedley, 1898) Acanthopleura arenosa Ferreira, spec. nov. Acanthopleura rehderi Ferreira, spec. nov. Acanthopleura has no fossil record. The paleontological literature contains only two possible references to the ge- nus as here understood: SMITH (1960:67) mentioned (but did not cite) Acanthopleura in “‘Pleist., S. Am. (Bol.)”; and CossMAN (1888:20) allocated one posterior valve from the Eocene of the Paris Basin to Enoplochiton, as E. rochebru- nei, which “is probably no chiton at all” (VAN BELLE, 1983:128). Acanthopleura spinosa (Bruguiére, 1792) Figures 1 to 6, and 112-S Chiton spinosus BRUGUIERE, 1792:25, pl. 2, fig. 1; LAMARCK, 1819:321; BuRRow, 1815:185, pl. 26, fig. 4; Mawe, 1823:1, 3-4, pl. 1, fig. 3; Woop, 1825:4, pl. 1, fig. 38; BLAINVILLE, 1825:550; SowERBY, 1840b:1, 10, sp. no. 2, fig. 151; REEVE, 1842:12, pl. 134, fig. 151; 1847, pl. 9, fig. 51; ADAMS & ADAMS, 1858:475; CHENU, 1859: 381, fig. 2868; PAETEL, 1869:66, 1873:80; FISCHER, 1885:881 (in subgen. Acanthopleura); ARNOLD, 1901: 322 (fig. only) (reprinted, 1968); Lamy, 1923:260. Maugeria spinosa: GRAY, 1857:184. Francisia spinosa: TRYON, 1883:343, pl. 85, fig. 81; HADDON, 1886:30. Acanthopleura spinosa: PILSBRY, 1893c:220, pl. 45, figs. 80- 87; THIELE, 1893:373, pl. 30, fig. 3; HmpaLco, 1905: 272; NIERSTRASZ, 1905a:101, 1905b:152; Horst & SCHEPMAN, 1908:526; HEDLEY, 1910:352; IREDALE, 1910b:158, 1914b:668; AsHBy, 1918:86, 1922a:31-32, 1926:384, 388-389; IREDALE & HULL, 1926:264-265, pl. 38, figs. 1-2 (reprinted, 1927:127-128, pl. 16, figs. 1-2); THIELE, 1929:22; BERGENHAYN, 1930a:32-33; LeELoupP, 1933a:24-25, 1933b:2-3; Kuropba, 1941:71; LELoup, 1952:59-61, text fig. 20, pl. 6, fig. 4; ALLAN, 1959:239, fig. 6b; Is. Taki, 1962:47; Iw. Tak1, 1964: 412; ANG, 1967:415-416, pl. 7, figs. 1-4; Wu, 1969: 103-104, figs. 1, 7-15; WELLS, 1981:253. Acanthopleura spinosa montebelloensis ASHBY, 1922a:32. Type material and type locality: Chiton spinosus Bruguiére, 1792: Type specimens (2) at MNHN (fide Lamy, 1923), not examined; locality, not originally stated but presumed to be “isle Maria, baye Page 223 de l’Est,” Australia, according to label accompanying types (LAmMy, 1923). Acanthopleura spinosa montebelloensis Ashby, 1922a: Types’ (Ashby coll. No. 5888) whereabouts unknown (not at WAM [WELLS, 1977]); locality ‘““Monte Bello Island,” Western Australia (20°25'S, 115°32’E). Material examined: AUSTRALIA: Yardie Creek, North West Cape, 2 specimens (WAM 56-74); Monte Bello Is., 1 specimen (WAM 69-74); Roseman Id., Dampier Arch., 2 specimens (WAM 65-74); Ken- drew Is., Dampier Arch., 7 specimens (WAM 1341-78); Port Hedland, 1 specimen (WAM 66-74); Flaccount Bay, 1 specimen (WAM 618-67); Walsh Point, Admiralty Gulf, 3 specimens (WAM 79-78; WAM 1107-78); Pt. Gantheaume, near Broome, 2 specimens (WAM 10884); Riddal Beach, Broome, 2 specimens (WAM 60-74); Broome, 1 specimen (NMV); Kuri Bay, 1 spec- imen (WAM 55-74); Yampi Sound, 4 specimens (WAM 64- 74); Augustus Id., Bonaparte Arch., 2 specimens (WAM 58- 74); “N.W. Australia,” 4 specimens (WAM 1106-78; NMV); Darwin, 1 specimen (NMV); Nightcliff, Darwin, 1 specimen (WAM 61-74); Fanny Bay, N.T., 5 specimens (NMV); Rain- bow Cliffs, Gove, N.T., 1 specimen (I. Loch coll.). NEW GUINEA: Sowek Soepiori, Schouten Is., 1 specimen, 75 mm long (ANSP 207130). SULAWESI (=Celebes): 32 km E of Manado, 3 specimens (CAS 009806; CAS 012252). TAIWAN: Wan-li-tong, Ping-tung Co., 1 specimen (UCM DZ SD2)' PHILIPPINES: Hundred Islands, Lingayen Gulf, Luzon, 16 specimens (AJF 464; AJF 465). Description: Excellent accounts of Acanthopleura spinosa were given by PILsBRY (1893c), IREDALE & HULL (1926), and Wu (1969). Among 60 specimens of Acanthopleura spinosa here ex- amined, largest 80 mm long (in alcohol) (WAM 1107- 78). Body width/length mean 0.66. Tegmentum uniform dark chestnut in color. Lateral areas hardly defined, not raised (Figures 1, 2). Tegmental sculpture of minute granules on anterior valve, postmucro area of posterior valve, and lateral areas of intermediate valves. Central areas almost smooth except for discrete transverse, con- centric growth rugae. Mucro obtuse, moderately poste- rior; postmucro convex with marked slope. Width of teg- mental surfaces of valves i/viii, 1:22. Ocelli round to oval, 40-50 um in diameter, scattered throughout anterior valve, anterior % to % of lateral areas of intermediate valves, and postmucro area. Gills with 50-70 plumes per side. Articulamentum reddish-brown in middle, lighter to al- most white at periphery. On valve viii, tegmentum length/ width mean 0.60; articulamentum wider than tegmentum (width of articulamentum/width of tegmentum on valve viii, mean 1.49). Sutural laminae wide; sinus well defined. Insertion plates relatively long, strongly pectinate on out- side; on valve i, length of insertion plate/length of teg- mentum, mean 0.36. On valve viii (Figures 3, 4), insertion teeth in part (%) fused to somewhat rounded, transverse callus; in specimen 50 mm long, insertion teeth (at midline Page 224 The Veliger, Vol. 28, No. 3 /\. Jjo WON, MISO of valve) 1.3 mm long on outside but extend only 0.4 mm beyond transverse callus on inside of valve. Slit formula 12/16-1-8/12; some intermediate valve(s) (particularly valve II) show 2 slits on one or both sides. Sinusal plate often pectinate on valve ii, but not so on other valves; on valve viii, relative width of sinus (width of sinus/width of sutural plates), mean 1.0. Girdle upper surface with abundant, dark red, slender spines, sharply pointed, up to 16 mm long, 0.7 mm thick, in large specimens (Figure 5); girdle covered otherwise with dark red elements, translucent to opaque, as small as 30-50 um long, variable in length and shape, repre- senting all stages of growth from scalelike to spine. Girdle bridges (see Ferreira, 1983a) empty, 7.e., without signifi- cant elements. Undersurface paved with juxtaposed rows of transparent, squarish scales, 50 x 50 um, roughly striated, inner edge somewhat convex as if articulating upon concave outer edge of adjacent scale. Radula averaging 55% of specimen length (range 50- 63%, SD = 7.9, n= 5) and 96 rows of teeth (range 80- 110, SD = 11.9, n=5). In specimen 75 mm long (AJF 464: Lingayen Gulf, Philippines), median tooth (Figure 6) 100 wm wide at anterior blade; first lateral teeth some- what rectangular, outer edge deeply concave, about 220 um at anterior blade; major lateral teeth with discoid head, 360 wm wide; outer marginal teeth 320 wm long, 350 wm wide (length/width, 0.9). Distribution: Acanthopleura spinosa has been reported from Poulo Dama, Gulf of Thailand (LELoupP, 1952), Taiwan (Wu, 1969), Philippines (HIDALGO, 1905; ANG, 1967), Java, Timor (Horst & SCHEPMAN, 1908), Pisang Islands (LELOup, 1933b), Amboine Id. [=Ambon] (Horst & SCHEPMAN, 1908), New Guinea (NIERSTRASZ, 1905b; Horst & SCHEPMAN, 1908; LELOouP, 1933b), and New Caledonia (IREDALE & HULL 1926). In Australia it has been reported along the northwestern coast from the Monte Bello Islands (IREDALE, 1914b) to Cape York (HADDON, 1886). From the data, A. spinosa appears to be confined to the central Indo-Pacific from 25°N to 22°S, and from 105°E to 142°E. The northernmost verified record is Wan- li-tong, Taiwan (25°11'N, 121°41’E) (UCM 27522); the southernmost and westernmost verified record, North West Page 225 Cape, Western Australia (21°45’S, 114°10’E) (WAM 1107-78); and the easternmost verified record, Gove (12°18’S, 136°55’E). The report of A. spinosa in New Cal- edonia (IREDALE & HULL, 1926) requires corroboration; collecting trips to New Caledonia (A. J. Ferreira, July 1980) and eastern New Guinea, Trobriand Is., and New Britain Id. (A. J. Ferreira, August 1981) failed to find the species. Acanthopleura spinosa is confined to the intertidal zone, with specimens often exposed at low tide. Remarks: Acanthopleura spinosa is very distinct from all other species of Acanthopleura, posing no problems in identification. The dark reddish color and the long, thin, sharp girdle spines are diagnostic. It is the only Acantho- pleura with two slits, though inconstant, on intermediate valves. Acanthopleura spinosa occurs sympatrically with A. gemmata and A. miles throughout its range. Field obser- vations show that where A. spinosa and A. gemmata coexist (AJF 464-465: Gulf of Lingayen, Luzon, Philippines), estimated population sizes are in the proportion of 1:100. Acanthopleura gemmata (Blainville, 1825) Figures 7 to 23, and 113-C Chiton gemmatus BLAINVILLE, 1825:544. Acanthopleura gemmata; IREDALE, 1914b:668; AsHBy, 1918: 86 (in subgen. Amphitomura), 1922b:29-31, 1923b:230, 1926:384, 388-389, 1928:171-172, pl. 12, figs. 6, 7; Way & PURCHON, 1981:313; WELLS, 1981:253; CHELAZZI et al., 1983:115-125; FERREIRA, 1983b:278- 282, fig. 30. Acanthopleura gemmata queenslandica ASHBY, 1922a:30, 1926: 384, 1928:171-172. Acanthopleura gemmata maudensis ASHBY, 1928:172, pl. 12, figs. 8, 9. Amphitomura gemmata: ASHBY, 1920:291; HULL, 1925:114- alls Acanthozostera gemmata: IREDALE & HULL, 1926:263, pl. 37, figs. 33, 34 (reprinted, 1927:126-127, pl. 15, figs. 33-34); Is. Taki, 1947:1269, fig. 3607, 1960:197, pl. 90, fig. 3, 1962:46-47; ANG, 1967:418-421, pl. 14, figs. 1-5; Wu, 1975:70-71, figs. 14-28. Chiton spiniger SOWERBY, 1840a:287-288, Suppl. pl. 16, fig. Explanation of Figures 1 to 4 and 7 to 10 Figure 1. Acanthopleura spinosa (Bruguiére, 1792). Gove, N.T., Australia (AJF coll.); specimen 16 mm long. Figure 2. Acanthopleura spinosa (Bruguiére, 1792). Same speci- men as in Figure 1. Close-up of lateral areas and girdle. Figure 3. Acanthopleura spinosa (Bruguiére, 1792). Same speci- men as in Figure 1. Posterior aspect of posterior valve. Figure 4. Acanthopleura spinosa (Bruguiére, 1792). Same speci- men as in Figure 1. Ventral aspect of posterior valve. Figure 7. Acanthopleura gemmata (Blainville, 1825). Bunaken Id., North Sulawesi, Indonesia (AJF 711); specimen 20 mm long. Figure 8. Acanthopleura gemmata (Blainville, 1825). Koror, Pa- lau, Western Caroline Islands (AJF 409); specimen 55 mm long. Dorsal aspect of posterior valve. Figure 9. Acanthopleura gemmata (Blainville, 1825). Same spec- imen as in Figure 8. Posterior aspect of posterior valve. Figure 10. Acanthopleura gemmata (Blainville, 1825). Same spec- imen as in Figure 8. Ventral aspect of posterior valve. Page 226 mm Figure 5 Acanthopleura spinosa (Bruguiére, 1792). Hundred Islands, Lin- gayen Gulf (AJF 464); specimen 35 mm long. Girdle spine. 2, 1840b:1, 10, sp. no. 3, fig. 68, 1841:61; REEVE, 1847, pl. 14, fig. 75; Jay, 1850:99; ADAMS & ADAMS, 1854: 475; CHENU, 1859:381-382, fig. 2871; TILLIER & Bavay, 1905:176, 179. Maugeria spinigera: GRAY, 1857:184. Chiton (Acanthopleura) spiniger: SMITH, 1884:81; 1891:420; FISCHER, 1885:881; MARTENS, 1887:199. Acanthopleura spiniger: DALL, 1879:298, fig. 28; TRYON, 1883:343, pl. 86, fig. 94; MosELEy, 1884:145; 1885:18, pl. 6, figs. 1-3, 6-9; HADDON, 1886:23-25; PILsBRY, 1893¢:221-226, pl. 48, figs. 22-32, pl. 49, figs. 33, 34; PELSENEER, 1899:8, 24-25; STURANY, 1904:267, 280; HIDALGO, 1905:272; IREDALE, 1910b:158; ODHNER, 1919:21, 42; AsHBy, 1922a:31; DAUTZENBERG, 1923: 58, 1929:552; BUCKNILL, 1930:529, fig. 10; LELoup, 1933a:19-23, 1933b:1-2, 1952:41, text fig. 15, pl. 2, 1981:9-10, fig. 3. Acanthopleura spinigera: Horst & SCHEPMAN, 1908:526- 527; THIELE, 1893:372, pl. 3, fig. 30; MELVILL & The Veliger, Vol. 28, No. 3 STANDEN, 1899:180; THIELE, 1909:8, 1911:398-399; NIERSTRASZ, 1905a:99-101, 1905b:151-152 (in part), 1906:511-513; Sykes, 1907:34; SmMitH, 1910:211; ASHBY, 1923a:226; PALLARY, 1926:28, pl. 4, figs. 4a, 4b; TOMLIN, 1927:291; BERGENHAYN, 1930a:32-33, pl. 8, fig. 75, pl. 9, fig. 84, 1930b:39-42, pl. 3, figs. 55- 61, 70-74; Lamy, 1938:88; SOLEM, 1953:215; GREEN- FIELD, 1972:37-47. “Chiton aculeatus Gmelin” Quoy & GAIMARD, 1835:373- 376; (Atlas, 1833), pl. 74, figs. 1-5. Chiton (Acanthopleura) macgillivray: ADAMS, 1855:120; PILSBRY, 1893c:224-225 (as syn. of A. spiniger). Acantopleura [sic] vaillantii ROCHEBRUNE, 1882:192. Acanthopleura rawakiana ROCHEBRUNE, 1882:195-196. Acanthopleura balansae ROCHEBRUNE, 1882:197; Lamy, 1923: 265. Acanthopleura glareosa SMITH, 1884:81, nomen nudum (as syn. of A. spiniger). Acanthopleura haddoni WINCKWORTH, 1927:206, pl. 28, figs. 1-4; LeLoup, 1937a:172-176, figs. 17-19; REES & STUCKEY, 1952:185; LELOupP, 1960:38-39; BoscH & BoscH, 1962:145; PEARSE, 1978:92-101, 1979:50, fig. 9; LELOuP, 1980b:6. Acanthopleura planispina BERGENHAYN, 1933:36-38, text fig. 12, pl. 1, fig. 11, pl. 3, figs. 51-52, 54-59 (in subgen. Amphitomura). Planispina planispina: Is. Taki, 1962:47. Acanthopleura bergenhayni LELOUP, 1937b:3, 1980c:1-5, figs. 1-3. Acanthozostera virens ANG, 1967:421-422, pl. 15, figs. 1-5. “Acanthopleura brevispinosa (Sowerby) ANG, 1967:416-417, pl. 13, figs. 1-5. Type material and type locality: Chiton gemmatus Blainville, 1825: Type material from “New Holland” presumed lost (ASHBY, 1928, doc. cit.); neo- type, designated by AsHBy, 1928:172, pl. 12, fig. 6 (=holotype of Acanthopleura gemmata queenslandica Ashby, 1922a), “in coll. Ashby” (fide IREDALE & HULL, 1926:264), not examined; locality of neotype, Dunk Is- land, Queensland, Australia (17°56'S, 146°06’E). Chiton spiniger Sowerby, 1840a: Lectotype (BMNH 1842.5.10.1654) and _ paralectotype (BMNH 1842.5.10.1652) designated herein; locality “Cagayan ... Misamis, Island Midinao” and “island Siquijor,” Philippines (SOWERBY, 1841:61), here restricted to Si- quijor Id. (9°10’N, 123°33’E). Chiton macgillivray: Adams, 1855: Types unascertained; lo- cality Fiji Islands. Acantopleura vaillantu Rochebrune, 1882: Lectotype and paralectotype (at MHNH) designated herein; locality “Canal de Suez,” Egypt. Acanthopleura rawakiana Rochebrune, 1882: Lectotype (at MHNH) designated herein; locality “Rawak. Terre des Papous” (? Rawak, Sumba Id., Lesser Sunda Is- lands, 9°55’S, 119°50’E). Acanthopleura balansae Rochebrune, 1882: Lectotype and 3 paralectotypes (at MHNH) designated herein; locality “Australie ... Nouvelle Caledonie,” here restricted to New Caledonia. Acanthopleura gemmata queenslandica Ashby, 1922a: Holo- type, ‘“‘in coll. Ashby” (fide IREDALE & HULL, 1926: 264), not examined; locality Dunk Island, Queensland, Australia (17°56'S, 146°00’E). Gn |perreina, O86 Page 227 ee 4 OlOrum Figure 6 Acanthopleura spinosa (Bruguiére, 1792). Hundred Islands, Lingayen Gulf, Philippines (AJF 464); specimen 75 mm long. Radula median and first lateral teeth, and head of major lateral tooth. Acanthopleura haddoni Winckworth, 1927: Types unascer- tained; locality Aden (12°45’N, 45°12’E). Acanthopleura gemmata maudensis Ashby, 1928: Holotype, “Ashby Coll.” (fide AsHBy, 1928), not examined; lo- cality Maud’s Landing, Western Australia (23°06’S, 113°48’E). Acanthopleura planispina Bergenhayn, 1933: Holotype (UUZM no. 114a) and 4 paratypes (UUZM no. 114b); locality Bonin Islands (27°00'N, 142°10’E). Acanthopleura bergenhayni Leloup, 1937b: Holotype (BMNH 19823); locality ““N. C. Australia.” Acanthozostera virens Ang, 1967: Holotype (University of the Philippines, U.P. Am.-112), not examined; locality, Philippines, but exact “locality, collector, date not re- corded” (ANG, 1967:422). Material examined: AUSTRALIA, W.A.: Dorre Id., Shark Bay, 8 specimens (WAM 469-74); Maud’s Landing, 3 specimens (AMS 112352; WAM 9326); Ningaloo, Point Cloates, 1 specimen (WAM 34-74); Yar- die Creek, North West Cape, 7 specimens (WAM 23-74; WAM 1743-78; WAM 1749-78); Exmouth Gulf, 1 specimen (WAM 26-74); Hermite Id., Monte Bello Is., 5 specimens (WAM 19- 74; WAM 27-74); Long Id., near Onslow, 1 specimen (AMS C69355); “West Australia,’ 1 specimen, 65 mm long (AMS C44883); Monte Bello Id., 5 specimens (WAM 5887); Kendrew Id., Dampier Arch., 1 specimen (WAM 74-7); Barrow Id., 2 specimens (WAM 24-74; WAM 605-67); Whitnell Bay, Dam- pier Penins., 1 specimen (WAM 1111-78); Wood Id., 2 speci- mens (WAM 36-74); Cockatoo Is., 3 specimens (WAM 18-74); Broome, 8 specimens, largest 50 mm long (AMS C69100; WAM 8987; NMV); Buccaneer Arch., 36 specimens (AMS C42222); Lacepede Is., 1 specimen (WAM 30-74). AUSTRALIA, N.T.: 1 specimen (AMS C77617); Anson Bay, 3 specimens, largest 54 mm long (AMS C31532); Fannie Bay, Darwin, 1 specimen (WAM 29-74); “N.C.,” holotype of A. bergenhayni (BMNH 19823); Darwin, 14 specimens (WAM 38- 74; NMV; AMS C112353; AMS C10722); Cape Wessel, 1 specimen (AMS C77821); Port Essington, 3 specimens, largest 70 mm long (AMS C85096; AMS C90471); Cape Arnhem, 1 specimen, 50 mm long (AMS C135475). AUSTRALIA, Qld.: Swears Id., South Wellesby Is., Gulf of Carpentaria, 3 specimens (AMS C15820); Albatross Bay, Wei- pa, 1 specimen (AMS C109287); Darnley Id., 1 specimen (AMS C517900; Bamfield Point, Prince of Wales Id., Torres Strait, 1 specimen, 36 mm long (AMS C110626); Murray Id., Torres Strait, 9 specimens (AMS C29619; AMS C112345); Cape York, 1 specimen (Penprase coll.); Campwin Beach, 2 specimens, larg- est 45 mm long (AMS C135481); Lizard Id., 1 specimen (AMS C135478); Darnby Is., 1 specimen (AMS C51790); Cooktown, 1 specimen (C109286); Two Isles, 2 specimens (AMS C109286); Hope Id., 1 specimen (WAM 27990); Port Douglas, 3 specimens (Van Belle coll.; AMS C76027); Low Isles, 3 specimens (AJF 603); Michaelmas Cay, off Cairns, 1 specimen (AMS C53575); Alma Bay, Magnetic Id., 2 specimens (NMV); Armit Is., 5 specimens (NMV); Bowen, 1 specimen (AMS C109288); Palm Id., 4 specimens (AMS C9303); Brampton Id., Whitsunday Pas- sage, 4 specimens (AMS C109290); Linderman Id., Whitsunday Passage, 13 specimens (AMS C109293); Heron Id., Capricorn Group, 3 specimens (AMS C109189; AMS C109292; CAS 012261); Wilson Id., Capricorn Group, 2 specimens, largest 110 mm long (AMS C135519); Hillsborough Channel, 1 specimen (AMS C125472); Gatecombe Head, 3 specimens, largest 75 mm long (AMS C18778); Keppel Bay, 2 specimens (AMS C109191; AMS (C109295); Fairfax Id., Bunker Group, 1 specimen (AMS C69053); Port Curtis, 1 specimen (AMS C109294). PAPUA NEW GUINEA: Manubada Id., Port Moresby, 6 specimens (A JF 608); Madang, 1 specimen (AJF 623); Kaibola, Kiriwina Id., Trobriand Is., 2 specimens (AJF 611); Rabaul, New Britain Id., 2 specimens (AJF 615); Blanche Bay, New Britain Id., 1 specimen, 35 mm long (AMS C3160); Gigira Id., Louisiade Arch., 6 specimens (AMS C82857); Misima Id., Louisiade Arch., 9 specimens (AMS 112348); Schouten Is., 16 specimens, largest 40 mm long (ANSP 207593; ANSP 207128); Yule Id., 3 specimens, largest 70 mm long (ANSP 84007). INDONESIA NEW GUINEA: Padaido Is., 2 specimens, larg- est 30 mm long (ANSP 206192; ANSP 205047); Japen Id., 7 specimens, largest 53 mm long (ANSP 205227; ANSP 208964); Biak Id., 7 specimens, largest 30 mm long (ANSP 206295). SULAWESI (=Celebes): Utara, 3 specimens, 48 mm long (CAS 009875; CAS 012252); Bunaken Id., 58 specimens (AJF 706; AJF 711); Manado Tua Id., 23 specimens (AJF 708); Manado Bay, 10 specimens, largest 48 mm long (AJF coll., leg. S. Mot- ley). THAILAND: Lower Siam, Butang Arch., 1 specimen (NMV); Hey Id., S of Phuket, 3 specimens (AJF 865). MALAYSIA: Telor Id., 22 specimens (AJF 719); Kedah, 1 specimen (Penprase coll.). SUMATRA: Padang, 1 specimen, ca. 50 mm long (ANSP 84316); Pagai, Mentavi Is., 1 specimen (NMV 4055; NMV 4056). Page 228 BALI: 5 specimens (AMS C60813). BORNEO: Marudu Bay, 2 specimens (ANSP 255743; ANSP 255742); Sapi Id., 1 specimen, 62 mm long (ANSP 275040). SOLOMON IS.: 1 specimen (Van Belle coll.); Tulagi, 9 spec- imens (AMS C30640); Florida Id., 2 specimens (AMS C11109); Skutland Id., 1 specimen, ca. 60 mm long (ANSP 310058). NEW CALEDONIA: 23 specimens (AMS C112346; AMS C112347; NMV); thio, 1 specimen (NMV); “Hargraves Coll.,” 1 specimen (AMS 11314); Noumea, 4 specimens (AJF 530; AJF 534); Baie des Citrons, Noumea, 2 specimens (AMS C72643); Roche al la Voile, Noumea, 1 specimen (AMS C72650); Bou- rail, 4 specimens (AJF 539); Poindimie, 5 specimens (AJF 537); Faden Reef, Heinghene, 1 specimen (AMS C112344); Ile des Pins, 22 specimens (AJF 532; AMS C4343). PHILIPPINES: Rita Id., Ulugan Bay, Palawan (South China Sea), 3 specimens (AJF 822); “Auson” Id., off Port Barton, Palawan (South China Sea), 8 specimens, largest 56 mm long (AJF 820); Binumsalian Bay, Palawan, 4 specimens (AJF 814); Tawitani, Sulu Arch., 6 specimens (CAS 002383; CAS 009881); Laminusa Id., Siasi Group, Sulu, 3 specimens (CAS 009877); Luuk, Sulu, 2 specimens (CAS 0122200; Juruck Bay, Cagayan Sulu Id., Sulu, 7 specimens (CAS 012245); Cebu, 1 specimen (Van Belle coll.); Nonoc, 25 specimens (CAS 012795); Siquijor Id., 1 specimen disarticulated (AMS C135473), Maloh, Negros Id., 36 specimens (AJF 451); Sumillon Id., 10 specimens (AJF 453); Liloan Point, Cebu Id., 7 specimens (AJF 454); Apo Id., 2 specimens (AJF 455); San Jose, Negros Id., 6 specimens (A JF 456); Punta Cruz, Bohol Id., 14 specimens (AJF 459); Loon Id., off Bohol Id., 20 specimens (AJF 460); Inamora Id., off Bohol Id., 1 specimen (AJF 461); Mactan Id., off Cebu Id., 21 specimens (AJF 462); Sulpha Id., off Cebu Id., 10 specimens (AJF 463); Ambulong Id., Mindoro, 15 specimens (AJF 791; AJF 793); Ilin Point, lin Id., Mindoro, 4 specimens (AJF 797); Hundred Islands, Lingayen Gulf, Luzon, 34 specimens (AJF 464; AJF 465). TAIWAN: Orchid Id., near Hungtou Rock Formation, Taitung Co., 2 specimens, largest 41 mm long (UCM 28842); Keelung, 17 specimens (CAS 009880; CAS 016698). OKINAWA, Japan: Cape Ata, 32 specimens (CAS 002383; CAS 012253); Buckner Bay, 1 specimen, 25 mm long (CAS 002163); Okuma, 3 specimens (CAS 012221; S. Crittenden coll.); Meijo, 1 specimen (CAS 012242); White Beach, 2 specimens, largest 62 mm long (AMS C135474). BONIN IS., Japan: Holotype and 4 paratypes of Acanthopleura planispina Bergenhayn) (UUZM Nos. 104, 104a); 2 specimens (CAS-SU 2871). PALAU: Koro, 55 specimens (AJF 409); Ngesil, 5 specimens (Carlson & Hoff coll.). YAP: 4 specimens (NMV; Crittenden coll.). GUAM: Pago Bay, 2 specimens (Carlson & Hoff coll.); Bile Bay, 4 specimens (Carlson & Hoff coll.). FIJI: 1 specimen, 50 mm long (AMS C135477); Tai Id., Nadi Bay, Viti-levu, 2 specimens (AJF 284); Naindi Bay, near Sa- vusavu, Vanua-levu, 5 specimens (A JF 528); 1 specimen (NMV); Yasawa Is., 1 specimen (Van Belle coll.); Namuya Levu, Ya- sawa Group, | specimen (Penprase coll.); Wambia, Ono, 5 spec- imens (AMS C112349). TONGA: Tongatapu, N coast, 16 specimens, largest 55 mm long (CAS 046669; AJF coll., leg. M. Wolterding, Apr. 1984); Fanga Tavi Beach, Eva Id., 5 specimens (CAS 046670); Pan- gaimotu Id., 3 specimens, largest 48 mm long (AJF 766); Fafa Id., 5 specimens, largest 55 mm long (AJF coll., leg. M. Wol- terding, May 1984); Afa Id., 1 specimen, 58 mm long (AJF 768); Tongatapu Id. and Valitoa Id., 2 specimens (CAS 009883); Vava’u Id., 2 specimens (Penprase coll.). The Veliger, Vol. 28, No. 3 MARQUESAS: Eiao, 1 specimen (AJF coll., ex C. Richard, Ecole Pratique des Hautes Etudes, Paris). ISRAEL: Elat, Gulf of Aqaba, 2 specimens (Van Belle coll.); Na’ama Bay, Sinai Peninsula, 31 specimens (AJF 434). EGYPT: Wadi-el-Dom, Gulf of Suez, 6 specimens (AJF coll., leg. Dr. J. Pearse). OMAN: Masgat, 11 specimens, largest 45 mm long (K. Gud- nason coll.). AFARS ET ISSAS: Djibouti, 4 specimens (MHNB; CAS 009874). SOMALIA: Gesira, 5 specimens, largest 50 mm long (MF 4106); Sar Uanle, 4 specimens, largest 50 mm long (MF 4107). COMORO IS.: Grand Comore Id., 1 specimen (CAS 001263). KENYA: Mombasa Beach, 5 specimens (AJF 593); Malindi, 17 specimens (AJF 594); Wamatu, 10 specimens (AJF 595); Twiga, 15 specimens (ANSP 276916; ANSP 287347); Wassini Id., 7 specimens (AJF 597). TANZANIA: Mbudya Id., 6 specimens (MHNB); Chumbe Id., 4 specimens (ANSP 213823); Tumbatu Id., 1 specimen (ANSP 212991); Zanzibar Is., 19 specimens (ANSP 212326; ANSP 213024; ANSP 213319; ANSP 214527); Dar-es-Salaam, 19 specimens (Van Belle coll.; CAS 9882; ANSP 283100). MADAGASCAR: Pointe de Tafondro, 3 specimens (ANSP 258101); Nossi Be, 136 specimens (ANSP 257344; ANSP 257601; ANSP 258553; ANSP 258627; ANSP 258911; ANSP 25864; ANSP 258965); Nossi Iranja, 35 specimens (ANSP 257084; ANSP 257085). Description: BLAINVILLE’s (1825) original description of Chiton gemmatus would have been quite sufficient to iden- tify the species were it not for uncertainties developed on account of its considerable intraspecific variation, very wide geographic distribution, and the presence of several other closely related species. Even PILsBRy’s (1893c) and IRE- DALE & HULL’s (1926) accounts, good as they are, fail to characterize the species unequivocally. Among over 1,210 specimens of Acanthopleura gemmata here examined, largest 110 mm long (dry) (AMS €135519: Wilson Id., Capricorn Group, Australia) (largest report- ed, 120 mm long [live] [IREDALE & HULL, 1926]). Spec- imens (Figures 7-10) depressed, round-backed, large; live specimen 82 mm long observed to shrink to 70 mm after two week preservation in 10% formalin in seawater fol- lowed by three months in isopropyl alcohol. Body width/ length, mean 0.66. Intermediate valves beaked; posterior edge of valve ii forming 100-120° angle. Tegmentum grayish-green to grayish-brown, usually extensively erod- ed. Lateral areas poorly defined, hardly raised, sculptured with low-profile, irregular, round to elongate granules, sometimes coalesced into arched wrinkles; anterior valve and postmucro area of posterior valve similarly sculp- tured. Central areas almost featureless except for smaller to obsolete granules in pleural areas, and thin, ill-defined, transverse lamellae appressed across jugal areas. Mucro central (in small specimens) to somewhat posterior (in larger ones); postmucro strongly convex, at 45—90° slope. Ocelli round to oval, 50-70 wm in diameter, randomly distributed throughout anterior valve, postmucro area of posterior valve, and anterior 4 to % of lateral areas of intermediate valves. On valve i, tegmentum length/width, A. J. Ferreira, 1986 (ee imm Figure 11 Acanthopleura gemmata (Blainville, 1825). Yeppoon, Qld., Aus- tralia; specimen 40 mm long. Girdle spines. mean 0.5. On valve vill, articulamentum often wider than tegmentum; tegmentum length/width, mean 0.5. Widths of tegmental surfaces of valves i/vili, mean 1.1. Gills with 40-60 plumes per side. Articulamentum color fairly constant for specimens from a given locality, but varying with locality from bluish- white to brown. Sutural laminae well developed, relatively long, subtriangular on valve 11 to subrectangular on valve vill. Sinus well formed; sinusal plate mostly smooth; rel- ative width of sinus (width of sinus/width of sutural lam- ina) on valve viii, 0.9. Insertion plates strongly pectinate on outside. On valve i, insertion teeth irregularly spaced, sometimes fused together; in midline, length of insertion plate/length of tegmentum, mean 0.2. On valve viii, pec- tinations extremely variable, resulting in incomplete slit- ting and poor definition of teeth, particularly towards midline; teeth often recurved forward, anteriorly fused to but extending beyond buttressing, transverse, round cal- lus. Slit formula (not always clearly determinable), 8/11- 1-6/10. Eaves thick (0.5 mm on midline of valve viii of specimen 50 mm long), moderately spongy. Girdle thick, musculous, wide, often banded, shrinking appreciably with preservation; at level of valve iv, girdle may measure 75% of width of valve in live specimens, 50% of valve in alcohol preserved specimens, 30% in dry specimens. Upper surface crowded with white to dark gray, brown, or black spinelets (Figure 11), pointed to blunt, straight to curved, somewhat conical, about 3 x 0.6 mm in average-sized specimen (up to 7 x 1 mm in large specimens), with smaller to minute spinelets in between; Page 229 t s100um Figure 12 Acanthopleura gemmata (Blainville, 1825). Same specimen as in Figure 11. Scales of girdle undersurface. in some specimens, pointed, crystalline, needle-like ele- ments (up to 200 x 30 wm) may be seen, isolated or in clusters interspersed amidst spinelets. Girdle bridges, empty. Undersurface paved with imbricate, transparent, squarish scales (Figure 12), about 40 x 40 wm (becoming elongate towards outer margin), with 8-10 coarse stria- tions radiating from outer edge of scale. Radulae averaging 45% of specimen length (range 37- 57%, SD = 7.2%, n= 13) and 63 rows of mature teeth (range 45-85, SD = 11.5, n = 13). Radular features (Fig- ure 13) rather constant in 23 specimens examined: in specimen 52 mm long (AJF 594: Malindi, Kenya) median tooth 80 um wide at anterior blade; first lateral teeth about 450 um long, 230 wm wide at anterior blade; major lateral teeth with tubercle 170 wm long at anterior part of inner edge, and discoid head 350 um wide; outer marginal teeth 300 um long, 210 wm wide (length/width, 1.5). Distribution: Among intertidal chiton species, Acantho- pleura gemmata seems to have the widest range (Figure 113-C). It has been recorded, albeit by synonymous names, from about 32°E to 140°W, an east-west range of some 20,000 km. In the western Indian Ocean, it has been reported from the gulfs of Aqaba and Suez in the Red Sea (ROCHEBRUNE, 1882; NIERSTRASZ, 1905b; TILLIER & Bavay, 1905; SykEs, 1907; Horst & SCHEPMAN, 1908; LELouP, 1933a; PEARSE, 1978) down the east coast of Africa, from Djibouti (LELouP, 1933a) through Somalia, Kenya, to Dar-es-Salaam, Tanzania (FERREIRA, 1983b), the Comoros Is. (NIERSTRASZ, 1905a, 1906), and the west coast of Madagascar (ODHNER, 1919; DAUTZENBERG, 1923, 1929), as well as at Aden and Barim Id., Yemen (WINCKWORTH, 1927), and Oman (LELOuP, 1937a; BOSCH & Boscu, 1962). It has been reported at the Andaman Is. (LELOUP, 1952), west coast of Malaysia (Way & PURCHON, 1981), Sumatra (BERGENHAYN, 1930a; LELOUP, 1952), Java (Horst & SCHEPMAN, 1908; BERGENHAYN, Page 230 \ 1400um Figure 13 Acanthopleura gemmata (Blainville, 1825). Same specimen as in Figure 11. Radula median and first lateral teeth. 1930a; LELoupP, 1933a), Sunda Is. (BERGENHAYN, 1930a; LELoup, 1933a), Lombock, Flores, Borneo (NIERSTRASZ, 1905a), Timor (Lamy, 1923), Amboine, Morotai, Sangi Is. (Horst & SCHEPMAN, 1908), Sulawesi (NIERSTRASZ, 1905a; LELoup, 1933a, b), Halmahera Id., New Guinea (Quoy & GAIMARD, 1835; NIERSTRASZ, 1905a; HORST & SCHEPMAN, 1908; LELoupP, 1933a), New Ireland and Tonga (Quoy & GAIMARD, 1835); Philippines (SOWERBY, 1841; HIDALGO, 1905; LELouP, 1933b; ANG, 1967), Tai- wan (Is. TAKI, 1947, 1962; Wu, 1975), Yaeyama Is. (Is. TAKI, 1938), Bonin Is. (BERGENHAYN, 1933, as Acantho- pleura planispina), Solomon Is. (LELOuP, 1933a), Fiji (ADAMS, 1855; PILSBRY, 1893c; BERGENHAYN, 1930a; LELoup, 1933a, 1952), and New Caledonia (ROCHE- BRUNE, 1882; PILSBRY, 1893c; LAMy, 1923). In Australia, A. gemmata was reported from the “Torresian Region, . the whole coastline from Darwin east and south to Port Curtis, and west and south to Bunbury” (IREDALE & HULL, 1926:264); but AsHBy (1928:171) pointed out that, on the west coast, the species occurs southward not to Bunbury but to “a point between Carnarvon and Maud’s Landing, north of Shark Bay.” Reports of the species at the Society Is. (LELOuP, 1933a) and Hong Kong (BERGENHAYN, 1930a) have not been corroborated (VAN BELLE, 1980, 1982; A. J. Ferreira, field trips to Society Is. [Moorea, Sept. 1974, Tahiti, Aug. 1980 and Dec. 1983, Bora-Bora, Dec. 1983], and Hong Kong, Sept. 1982, Jan. 1983). The report of A. gemmata at Cape of Good Hope, South Africa (NIERSTRASZ, 1905b, 1906) is obviously in error. In the Indian Ocean, the northernmost verified record is Elat, Gulf of Aqaba, Red Sea (29°33'N, 34°57’'E); the southernmost verified record is Nossi Iranja, 50 km SW of Nossi Be, Madagascar (13°20'S, 48°15’E). Reported records from Mahakamby (=Mahajamba) and Tulear, Madagascar (23°21'S, 43°40’E) (ODHNER, 1919, as A. spi- niger) require confirmation (Kaas, 1979). On the main- land of Africa, the species has been found as far south as Dar-es-Salaam, Tanzania (6°48'S, 39°17'E) (FERREIRA, 1983b, and herein). The Veliger, Vol. 28, No.3 There seems to be a wide distributional gap between the Red Sea and western Indian Ocean population on one side, and the Indo-Pacific population on the other. Acan- thopleura gemmata has not been reported between Oman and the Andaman Is.; the species was not found on South Male Atoll, Maldives, or the southwest coast of Sri-Lanka (A. J. Ferreira, field trip, Feb. 1983). The recording gap, approximately between 58°E and 92°E, does not seem to be a collecting artifact due to inadequate sampling but a distributional disjunction, perhaps intermittent, for which no reasonable explanation is at hand. In the central and eastern Indo-Pacific, Acanthopleura gemmata ranges from the Andaman Islands (92°45’E) to Tonga (175°W). It seems to be absent (another distribu- tional gap?) in the Society Is. (A. J. Ferreira, above) and Samoa Is. (A. J. Ferreira, field trips to Tutuila Id. and Upolu Id., March 1976 and Dec. 1983), reappearing in the Marquesas Islands (9°S, 139.5°E), its easternmost rec- ord. The northernmost verified record is the Bonin Islands (27°00'N, 142°10’E). In Australia, the southernmost ver- ified record on the east coast is Port Curtis, Qld. (24°00'S, 151°30'E); on the west coast, Dorre Island, Shark Bay, W.A. (25°09'S, 113°07’E). Acanthopleura gemmata is confined to the intertidal and low subtidal zones, 0-2 m, with specimens often exposed at low tide. Reports of the species at subtidal depths (HaDpDoNn, 1886, at 11 m; NIERSTRASZ, 1905a, at 27 m) are obviously in error. Remarks: As should be expected from a species with such a wide geographical distribution, Acanthopleura gemmata shows considerable intraspecific variation, particularly in the color of the tegmentum (from lighter grays to darker browns), color of the articulamentum (from blue to brown), tegmental sculpture (from granules to wrinkles, variable in shape and disposition), and in the girdle spinelets (from white to gray, brown, or black; from short and stubby to long and pointed). In regard to the latter, there seems to be some positive correlation between the length of the spinelets of a given specimen and the degree of erosion of the valves; specimens with the tegmentum uneroded tend to have longer spinelets, a phenomenon which seems to vary with locality but not necessarily with habitat (z.e., exposure to surf). The type material of Chiton spiniger consists of 2 syn- type specimens, well preserved, dry, and flat. The speci- men here designated lectotype (Figures 14-16) is accom- panied by a pink museum label which says, in part, “Figured Syntype / Loc. — ? [Philippines] / Coll. — ? Purch. of H. Cuming.” The specimen, dirty brown, 58 mm long, 35 mm wide; tegmentum clean, not encrusted or significantly eroded, shows abundant coarse, irregular granules in both central and lateral areas as well as in the anterior valve and postmucro of the posterior valve where they are arranged in concentric rows. Ocelli round to oval, 40-60 wm in diameter, throughout anterior valve, post- mucro area of posterior valve, and anterior 4 to % of em Rerreinas, 1986 lateral areas of intermediate valves. Mucro central, prom- inent; postmucro convex, with sharp, near vertical slope. Girdle with abundant calcareous spinelets up to 2.4 mm long, 0.4 mm thick. Soft parts absent; articulamentum dark brown; the number ““42.5.10.1654” is written in black ink on underside of girdle. The other syntype specimen (BMNH 1842.5.10.1652), here designated paralectotype, 73 X 58 mm, has very similar features. Two other spec- imens, dry, flat, well preserved, part of the H. Cuming collection, were examined. The larger specimen (BMNH 19824/1), 80 x 47 mm, with yellow museum label, cor- responds to the figured specimen in REEVE (1847: pl. 14, sp. & fig. 75); the smaller specimen (BMNH 19825/1), 35 x 19 mm, carries yellow museum label stating in part “Loc. Island of Siquijor [Philippines] found under stones at low water.” Pilsbry (zn ASHBY, 1922a:29) recognized Acanthopleura gemmata, apparently as senior synonym of A. spiniger; but ASHBY (1922a) recommended retaining the name spiniger for the Sumatra specimens he had examined, reserving gemmata for the Australian population. Further, ASHBY (Joc. cit.) postulated that specimens from Dunk Id., on account of differences in the tail valve, belonged to “a distinct geographical race” for which he proposed the name A. gemmata queenslandica (not queenslandica Pilsbry, 1894b) (see Remarks on A. arenosa). As noted by HADDON (1886), the specimen(s) cited by Quoy & GAIMARD (1835) as “‘Chiton aculeatus” are of A. gemmata. The type material of Acantopleura vaillanti Roche- brune, 1882, consists of two specimens, dry, curled, soft parts removed, 45 and 40 mm in (estimated) length, show- ing remnants of glued paper. Accompanying label reads “rec. Vaillanti—Types [on red background] / Acantho- pleura vaillant: Rochbr. / = A. spinigera Sow. / Mer rouge / Bull. Soc. Philom. 1882:192.” The specimens agree with ROCHEBRUNE’s (1882) description of the species; the larg- er is here designated lectotype, the smaller (Figure 17) paralectotype. The type material of Acanthopleura rawakiana Roche- brune, 1882, consists of a single specimen (Figure 18). The label reads, in part, “Syntype [on red background] / Acanthopleura rawakana [sic] ROCH. 1882 / Terre des Papous / Bull. Soc. Philom. Paris 1882 p. 195.” The specimen, here designated lectotype, agrees with ROCHE- BRUNE’S (1882) description, but the alleged locality, Ra- wak, could not be found on maps of New Guinea, new or old (Judy Kelly, National Library Service, Boroko, Papua New Guinea, 2m litt. 4 Dec. 1981), and is presumed to refer to a village of that name at Sumba Id., Lesser Sunda Islands. The type material of Acanthopleura balansae Roche- brune, 1882, comprises 4 specimens, dry, somewhat curled, estimated lengths from 40 to 50 mm, showing bits of glued paper. Accompanying label reads, “rec. Balansa 1872. XIV-221 — Types [on red background] / Acanthopleura balansae Roch. / = A. Spinigera Sow. / rec.: Caledonie / Page 231 Bull. Soc. Philom. Paris 1882:197.” The specimens, with soft parts intact, agree with ROCHEBRUNE’s (1882) de- scription of the species, the published dimensions corre- sponding to those of the largest specimen in the lot, here designated as lectotype; the 3 other specimens in the lot, one illustrated (Figure 19), are here designated paralec- totypes. The holotype of Acanthopleura bergenhayni Leloup, 1937b (BMNH 19823), is poorly preserved in alcohol; somewhat curled, estimated 60 mm long, 45 mm wide (including girdle); valves ili, v, vi, and vii in place, others missing except for part of valve viii showing pectinate teeth; valves bluish-gray, extremely eroded; girdle almost denuded of spinelets; spinelets white or grayish-black. Ac- companying museum label reads, in part, “Holotype / Acanthopleura bergenhayni / 1 specs. Acc. no.: / Leloup, 1937 / Loc. N. C. Australia / Coll. Antartic Exped., the Admiralty.” The specimen (Figures 20, 21) agrees with LELoup’s (1937b:3) description, except in dimensions (Leloup [loc. cit.] described it as 53 x 40 mm), and cor- responds to the current concept of A. gemmata. The type material of Acanthopleura planispina Bergen- hayn, 1933 (at UUZM) consists of holotype and 4 para- types, well preserved in alcohol. With holotype, type-writ- ten label reads, “‘Uppsala Univ. Zool. Mus. / Typesamlingen / nr. 144a / Molluscaa”; three other hand- written labels add, “... Type—ex. J. R. M. Bergenhayn 1933... Prof. Sixten Bocks Japan—Ex. 1914 / Bonin Islands (Ogasawara) / Taki ura Ebbestrand 28.7 / for- mal konserv ....”” Holotype (Figure 22), well preserved, flat, 36 mm long, 22 mm wide (including girdle); pulling down girdle (which, obviously, had been done before) shows pectinate insertion plate and single, well defined slit in midline. Contrary to BERGENHAYN’s (1933) asser- tion, girdle spinelets do not appear distinct in any partic- ular way. Paratypes, the anterior valve of one here illus- trated (Figure 23), very similar to holotype; largest about 44 mm long, disarticulated, missing valve viii; second larg- est 39 mm long, with valves v, vi, and vii in place, but others missing; third largest curled, estimated 25 mm long, missing valve i; smallest 16 mm long, with all valves in place. The material corresponds to that described and il- lustrated by BERGENHAYN (1933) and to specimens of A. gemmata from other localities. From the description and illustrations, Acanthozostera virens Ang, 1967, agrees clearly with the present under- standing of Acanthopleura gemmata. A number of other species regarded by authors as syn- onyms of Acanthopleura gemmata are here cast aside as insufficiently characterized: (1) Chiton aculeatus LINNAEUS, 1758:667, no. 3, from “Asia,” is unrecognizable (HANLEY, 1855; DopGE, 1952). (2) Chiton magnificus SOWERBY, 1840b:2, sp. 11, fig. 52 (not Deshayes, 1827), is a nomen nudum. (3) Chiton granatus REEVE, 1847: pl. 5, sp. & fig. 24, of unknown locality, is inadequately described. Carpenter (in PILSBRY, 1893c:224-225, pl. 48, figs. 29, 30) stated of The Veliger, Vol. 28, No. 3 Page 232 em) eierreinas, 1986 the tail-valve of the type “should be examined in order to tell whether it is a specimen of spiniger or of borbonica .... It is a nomen dubium. (4) Chiton piceus REEVE, 1847: pl. 13, sp. & fig. 70, from “New Holland” (=Australia) (not Gmelin, 1791, from St. Thomas, West Indies) may apply to several sim- ilar species of Acanthopleura in Australia. Carpenter (in PILsBRY, 1893c:226, pl. 49, figs. 37, 38) examined ‘“‘four specimens from Australia” regarded as types, but left no diagnostic clues. It remains a nomen inquirendum. (5) Chiton obesus SHUTTLEWORTH, 1853:61, 69, is a nomen nudum. (6) Chiton cunninghami REEVE, 1847, is here regarded as a nomen inquirendum. The holotype (BMNH 1951.1.25.6) (Figures 24-27), accompanied by a pink museum label which reads, in part, “... Loc. Australia, on the rocks / leg. Cunninghami / Coll. Cuming Acc. 1829,” comprises 8 disarticulated valves which, reassem- bled, suggest a living specimen (with girdle) about 110- 120 mm long, 70-80 mm wide. Valves thick, beaked; teg- mentum dark brown with whitish midline band; anterior valve with irregular, round granules about 300-400 um in diameter, in concentric rows; lateral areas hardly de- fined, with similar granules; central areas with similar granules in pleural regions, somewhat fused into antero- posterior riblets; jugum smooth, almost shiny; valve viii moderately inflated, posterior edge slightly sinused in middle third; mucro not prominent, somewhat posterior; postmucro convex; ocelli round to oval, 60 wm in diameter; articulamentum bluish-gray, darker in middle; sutural laminae subtriangular (on ii) to subrectangular (on viii); sinus well defined, sinusal plate coarsely pectinate; inser- tion teeth on valve viii, teeth very underdeveloped, as if sunken in middle third; slit formula 10-1-6. Radula and girdle not available. Carpenter (in PILSBRyY, 1893c:225) described the specimen’s girdle (apparently still available at the time) as “‘dried in around the valves, and the hairs are worn off except in the sutures, where they are short, crowded and black.” The species was regarded by Car- penter (77 PILsBRY, 1893c) as a synonym of A. spiniger Page 233 (=A. gemmata), a possibility that may be seriously consid- ered if a pathologic posterior valve is assumed; IREDALE & HUuLu (1926) and Kaas & VAN BELLE (1980) consid- ered it a synonym of A. brevispinosa, a supposition that goes contrary to the objective evidence. In the western Indian Ocean, Acanthopleura gemmata is sympatric through much of its range with A. brevispi- nosa. In the central Indo-Pacific, A. gemmata is sympatric with A. spinosa, A. araucariana, A. miles, A. curtisiana, and A. loochooana; and is parapatric in the north with A. ja- ponica, in the south with A. gaimardi, A. arenosa, and A. hirtosa. In some localities, as personally observed in Fiji, Phil- ippines, Palau, and at the Trobriand Islands, Papua New Guinea, specimens of Acanthopleura gemmata are often the object of human predation, and are so actively searched by the natives as a delicacy that entire populations have been nearly wiped out. Acanthopleura brevispinosa (Sowerby, 1840) Figures 28 to 34, and 112-B Chiton brevispinosus SOWERBY, 1840a:287, Suppl. pl. 16, fig. 1, 1840b:1, 10, sp. no. 4, fig. 136; REEVE, 1847, pl. 9, sp. & fig. 52. Acanthopleura brevispinosa: PLATE, 1898:167, pl. 11, figs. 111-112; Pitspry, 1893c:231-232, pl. 47, figs. 18-21 (in subgen. Amphitomura); NIERSTRASZ, 1905a:102, 1906:511-515; AsHBy, 1931:49, pl. 7, fig. 82; (?) FI- SCHER, 1939:36; BARNARD, 1963:344; Kaas, 1979:868; LeLoup, 1980c:8-11, figs. 5, 7, map 1 (in part); CHELAZZI et al., 1983:115-125. [Non: ROCHEBRUNE, 1882:240; Lamy, 1936:267 (=Plaxiphora mercatoris Leloup, 1936); FISCHER, 1978:49 (in part); ANG, 1967:416-417, pl. 13, figs. 1-5 (=A. gemmata).] Chiton borbonicus DESHAYES, 1863:37, figs. 12-13; Mar- TENS, 1880:300 (in subgen. Acanthopleura); VIADER, 1937:58. Acanthopleura borbonica: PILSBRY, 1893a:105, 1893c:230-231, pl. 45, figs. 76-79 (in subgen. Amphitomura); THIELE, 1893:372, pl. 30, fig. 31; NrerRsTRASZ, 1905a:102-103; Explanation of Figures 14 to 22 Figure 14. Acanthopleura gemmata (Blainville, 1825): Chiton spi- niger Sowerby, 1840a; lectotype (BMNH 1842.5.10.1654). Dor- sal aspect of valves i and 11. Figure 15. Acanthopleura gemmata (Blainville, 1825): Chiton spi- niger Sowerby, 1840a; lectotype (BMNH 1842.5.10.1654). Dor- sal aspect of valves iii and iv. Figure 16. Acanthopleura gemmata (Blainville, 1825): Chiton spi- niger Sowerby, 1840a; lectotype (BMNH 1842.5.10.1654). Dor- sal aspect of valves vii and viii. Figure 17. Acanthopleura gemmata (Blainville, 1825): Acantho- pleura vaillantii Rochebrune, 1882; paralectotype (MNHN). Figure 18. Acanthopleura gemmata (Blainville, 1825): Acantho- pleura rawakiana Rochebrune, 1882; lectotype (MNHN). Figure 19. Acanthopleura gemmata (Blainville, 1825): Acantho- pleura balansae Rochebrune, 1882; paralectotype (MNHN). Figure 20. Acanthopleura gemmata (Blainville, 1825): Acantho- pleura bergenhayni Leloup, 1937b; holotype (BMNH 19823). Dorsal aspect of specimen with only valves iii and v in place. Figure 21. Acanthopleura gemmata (Blainville, 1825): Acantho- pleura bergenhayni Leloup, 1937b; holotype (BMNH 19823). Fragment of posterior aspect of posterior valve. Figure 22. Acanthopleura gemmata (Blainville, 1825): Acantho- pleura planispina Bergenhayn, 1933; holotype (UUZM). Page 234 The Veliger, Vol. 28, No. 3 ? ~ UW) ge im Loe wae \ . \ VO @ . AGS Herreira,, 1986 Melvill, 1909:119; LELoup, 1941:9 (in subgen. Amphi- tomura), 1980c:5-8, figs. 4-6, map 1 (in part). Acanthopleura afra ROCHEBRUNE, 1882:192. Type material and type locality: Chiton brevispinosus Sowerby, 1840: Types unascertained; locality “Ins. Johanna [=Anjouan Id., Comoro Is.], E. Africa” (12°13'S, 44°29’E). Chiton borbonicus Deshayes, 1863: Types unascertained; lo- cality Reunion Id. (21°06’S, 55°38’E). Acanthopleura afra Rochebrune, 1882: Lectotype and para- lectotype (MNHN) here designated; locality Madagas- car (Rochebrune [1882:192] gave “Cap de Bonne Es- pérance ...; Madagascar,” as localities, but the first is clearly in error). Material examined: SOMALIA: Mogadiscio, 5 specimens, largest 35 mm long (MF 4108); Gesira, 7 specimens, largest 60 mm long (MF 4109). KENYA: Mombasa Beach, 10 specimens, largest 67 mm long (AJF 593); Malindi, 6 specimens (AJF 594); Wamata, 13 spec- imens (AJF 595); Wassini Id., 5 specimens (AJF 597). TANZANIA: Fumba, Zanzibar, 3 specimens (ANSP 213319); Mbudy Id., 3 specimens (M. Bacescti coll.). SEYCHELLES: 2 specimens (ANSP 310686); Port Ternary, Mahé Id., 1 specimen, 53 mm long (ANSP 311232); Anse Etoile, Mahé Id., 1 specimen, 50 mm long (ANSP 310382); North West Bay, Mahé Id., 10 specimens (AJF 564). COMOROS: 1 specimen (ANSP 220840); Pamandzi Id., 2 specimens (ex Van Belle coll.); Grand Comore Id., 4 specimens (CAS 000489, CAS 001253, CAS 009876). MADAGASCAR: Syntypes (2) of Acanthopleura afra Roche- brune, 1882 (MNHN); 1 specimen disarticulated, estimated 50 mm long, valves and radula (IRScNB I.G. 9247); 1 specimen, ca. 70 mm long, cited by Leloup (1980c) (IRScN I.G. 9247). MAURITIUS: 3 specimens (ANSP 35915; ANSP 35952); Souillac, 3 specimens 26-32 mm long (ANSP 274063); Gris- Gris, 4 specimens, 18-33 mm long (ANSP 274088); Pte. Fay- ette, 8 specimens, 25-45 mm long (ANSP 273693); Vacoas Pt., 5 specimens (ANS 274168); Pointe-aux-Roches, 9 specimens (ex B. Smith, March 1979), 26 specimens (AJF 586). REUNION: St. Gilles-les-Bains, 3 specimens (ex Van Belle coll.). Description: SOWERBY (1840a) described specimens of Chiton brevispinosus as “rather flat, oval, narrowed in front; Page 235 the valves are rounded and smooth at the beaks, and gran- ulated at the sides, in undulating, concentric ridges; ... the numerous short black spines studding (the girdle) are tipped with light yellow points ... a pretty relief to the general black colour of the shell” (p. 287). PILsBRy (1893c) pointed out “the valves concentrically wrinkled-grained at the sides of the central areas, and the ill-defined lateral areas ... cut into granules by concentric and radiating grooves. End valves finely grooved radially, finely wrin- kled concentrically; mucro posterior, prominent and rath- er acute. [Articulamentum] blackish-brown or purple- brown except [for white] sutural and insertion plates .. . . Sinus broad, deep, rounded. [Slits 7/8-1-2]; anterior teeth moderately long, finely pectinated outside; posterior teeth very short, blunt, obsoletely pectinated .... [Girdle] nar- row, clothed with white-tipped black spinelets ...” (p. 2510) Among 138 specimens of Acanthopleura brevispinosa here examined, largest 67 mm long (in alcohol) (AJF 593: Mombasa Beach, Kenya). Body width/length, mean 0.61 (SD = 0.05; n = 25). Specimens (Figures 28-31) oval, de- pressed, round-backed. Intermediate valves beaked; pos- terior edge of valve 11 forming 120-160° angle. Tegmen- tum purple-brown to dark brown or black, often with two parajugal cream-white bands. Tegmental sculpture of coarsely round granules, larger (up to 250 wm in diame- ter) and better defined at periphery, arranged in radial rows. Anterior valve with 30-50 rows of granules. Central areas featureless except for dull shagreened surface and, occasionally, some fine transverse, concentric rugae. Lat- eral areas not elevated, poorly defined except for 6-10 rows of granules. Mucro central to slightly posterior; post- mucro convex, at 45-90° slope. Ocelli round to oval, 50 um in diameter, randomly distributed throughout anterior valve, postmucro area of posterior valve, and anterior 4 to % of lateral areas of intermediate valves. Ratio of teg- mental surfaces of valves i/viii, mean 1.06. On valve viii, length of tegmentum/width of tegmentum, mean 0.37 (SD = 0.04; n = 11). Articulamentum dark brown to black, or bluish-white with dark brown area in middle of valve. Sutural laminae Explanation of Figures 23 to 31 Figure 23. Acanthopleura gemmata (Blainville, 1825): Acantho- pleura planispina Bergenhayn, 1933; paratype (UUZM). Ante- rior aspect of anterior valve. Figure 24. Chiton cunningham: Reeve, 1847; holotype (BMNH 1951.1.25.6). Anterior aspect of anterior valve. Figure 25. Chiton cunningham Reeve, 1847; holotype (BMNH 1951.1.25.6). Dorsal aspect of intermediate valve. Figure 26. Chiton cunningham Reeve, 1847; holotype (BMNH 1951.1.25.6). Dorsal aspect of posterior valve. Figure 27. Chiton cunningham: Reeve, 1847; holotype (BMNH 1951.1.25.6). Ventral aspect of posterior valve. Figure 28. Acanthopleura brevispinosa (Sowerby, 1840a). Wa- matu, Kenya (AJF 595); specimen, 23 mm long. Figure 29. Acanthopleura brevispinosa (Sowerby, 1840a). Ma- lindi, Kenya (AJF 594); specimen, 43 mm long. Dorsal aspect of posterior valve. Figure 30. Acanthopleura brevispinosa (Sowerby, 1840a). Same specimen as in Figure 29. Posterior aspect of posterior valve. Figure 31. Acanthopleura brevispinosa (Sowerby, 1840a). Same specimen as in Figure 29. Ventral aspect of posterior valve. Page 236 1mm Figure 32 Acanthopleura brevispinosa (Sowerby, 1840a). Wamatu, Kenya (AJF 595); specimen 26 mm long. Girdle spines. well developed, relatively long, subtriangular on valve ii to subrectangular on valve vill. Insertion plates strongly pectinate on outer surface. On valve i, length of insertion plate/length of tegmentum, mean 0.17; insertion teeth ir- regularly spaced. On valve viii, relative width of sinus, 0.8; teeth poorly defined by incomplete and irregular slit- ting and underdevelopement of middle third of insertion plate; often only 2 slits, symmetrically placed; insertion teeth supported anteriorly by buttressing transverse, round, variably developed callus. Slit formula 7/11-1-2/6. Eaves moderately spongy. Gills with some 50 plumes per side. Girdle wide, muscular, mostly black, usually not band- ed. Upper surface profusely beset with calcareous, dark brown to black, blunt spinelets (Figure 32), often tipped with yellowish-white, up to 3 mm long, 0.3 mm thick, amidst much smaller to minute ones. Girdle bridges emp- ty. Undersurface covered with transparent, rectangular scales, about 50 um long, 35 um wide, with few, coarse longitudinal striations. Radulae averaging 62% of specimen length (range 53- 68%, SD = 6.6%, n= 6) and 93 rows of mature teeth (range 80-110, SD = 12.3, n= 6). In specimen 42 mm long (M. Bacescti coll.: Mbudy Id., Tanzania) median tooth 130 um wide at anterior blade; first lateral teeth 170 #m at anterior blade, outer edge deeply concave, outer posterior angle sharply pointed, inner posterior angle very elongate (Figure 33); major lateral teeth with discoid head 210 wm in diameter; outer marginal teeth 170 um long, 250 um wide (length/width, 0.7). Distribution: Acanthopleura brevispinosa is confined to the western Indian Ocean (Figure 112-B). Its range, as here verified, extends from the east coast of Africa to the Sey- chelles, Comoros, Madagascar, Reunion, and Mauritius; and from Gesira, Somalia (1°58'N) to Santa Carolina Id. (Kaas, 1979) and Bazaruto Id., Mozambique (21°40’S) The Veliger, Vol. 28, No. 3 (BARNARD, 1963). FISCHER’s (1939) report of the species at “mer d’Oman” (? Gulf of Oman) and Aden requires confirmation. Reports of the species at Cabo Verde Arch. (ROCHEBRUNE, 1882; PILsBRY, 1893c; LELOUP, 1980c), Cape of Good Hope (ROCHEBRUNE, 1882; PILSBRY, 1893c; NIERSTRASZ, 1905a, 1906; AsHBy, 1931), Philippines (ANG, 1967; FISCHER, 1978), Indochina and Hong Kong (FISCHER, 1978), as well as at Rio de Janeiro (Brazil), Ile du Prince (Gulf of Guinea), Poulo Condor (Con Son Is., Vietnam), Moluccas (Indonesia), and Fiji (LELOUvP, 1980c) are not credible and must be considered in error. Possibly in error, too, is the report of the species in the Red Sea (NIERSTRASZ, 1905a). Acanthopleura brevispinosa is confined to the intertidal zone, with specimens often exposed up to 2 m above low tide level. Remarks: The type material of Acanthopleura afra Roche- brune, 1882 (MNHN), is accompanied by a museum la- bel which reads, in part, “rec. Verreaux. XIV 230 Types [on red background] / Acanthopleura afra Rchb. 1882 / = A. borbonica / Cap de Bonne Espérance ....” It consists of 2 specimens, dry, curled, estimated length 60 and 50 mm, soft parts removed, showing indications of having been glued to cardboard. Girdle spinelets of smaller spec- imen mostly fallen off. The specimens agree with ROCHE- BRUNE’Ss (1882) description of the species. The larger spec- imen is here designated lectotype, the smaller, paralectotype (Figure 34). Of the two given localities, Cape of Good Hope is in error, and Madagascar, therefore, must be regarded as type locality. Chiton nebulosus Woop, 1828, figured (pl. 1, fig. 4), from “Isle of France,” but undescribed, is here considered a nomen dubium; on subjective grounds, Kaas & VAN BELLE (1980) suggested the figured specimen might be of Acanthopleura borbonica (=A. brevispinosa). Throughout its range, Acanthopleura brevispinosa is sympatric with A. gemmata from which it must be, there- fore, carefully differentiated. Distinctions based on size, color, shape, tegmental sculpture, or girdle elements are potentially deceiving, given the intraspecific variation of the two species. The study of the radula is indispensable, particularly in questionable specimens. The radula of A. brevispinosa differs from that of A. gemmata, as well as from all other species of Acanthopleura, in (1) relatively longer size, (2) much greater number of rows of mature teeth, (3) wide, parallel-sided median teeth, (4) narrow, posteriorly pointed first lateral teeth with sharper outer edge protuberance, (5) relatively smaller head of major lateral teeth, and (6) proportionally shorter outer margin- al teeth. Ratio between width (at anterior blade) of me- dian tooth and width of head of major lateral teeth in A. brevispinosa, mean = 0.83 (n = 8; SD = 0.11; range 0.73- 1.00); in specimens of 12 other species of Acanthopleura pooled together, mean = 0.32 (n = 33; SD = 0.08; range 0.21-0.43) (P < 0.001). ewe bk erneira, 1986 Acanthopleura araucariana (Hedley, 1898) Figures 35 to 40, and 115-K Ischnochiton araucarianus HEDLEY, 1898:100-101, figs. 3-6; NIERSTRASZ, 1905a:21, 1908:145 (in subgen. Hetero- zona); IREDALE & HULL, 1926:261-262 (as syn. of Squamopleura miles) (reprinted, 1927:123-124); LELOuP, 1939b:1-6 (as syn. of S. miles); Kaas & VAN BELLE, 1980:19 (as syn. of S. miles). Sclerochiton araucarianus: THIELE, 1910a:96. Type material and locality: Ischnochiton araucarianus Hedley, 1898: Holotype (AMS C.4344); locality “Isle of Pines, New Caledonia” (22°37'S, 1167°30'B). Material examined: NEW CALEDONIA: Pines Id., holotype of Ischnochiton arau- carianus Hedley, 1898 (AMS C.4344); Pines Id., Kuto Beach, 55 specimens, largest 52 mm long (AJF 532). LOYALTY IS.: Lifou Id., 1 specimen (NMV). TONGA: Ha/alafu Beach, Tongatapu, 1 specimen, 25 mm long (AJF 767, leg. A. J. Ferreira & M. Wolterding, 3 Dec. 1983); Eua Id., intertidal zone up to 0.8 m above mid-tide water level, 16 specimens, largest 27 mm long (AJF coll., leg. M. Wolter- ding, 9 May 1984). Description: HEDLEY (1898) gave an excellent descrip- tion of Ischnochiton araucarianus: “Shell oval, depressed, valves rounded posteriorly, but the anterior ones more pointed. Colour greenish-grey, each valve shading poste- riorly into cream, with a median wedge of black, which is sometimes split with a central white stripe. Interior dull purple, shading posteriorly into brown. Girdle ... che- quered black and cream. Lateral areas elevated with about three obscure, diverging lines of granules, more prominent on the anterior valves. Central areas finely and evenly corded transversely. Anterior valve radiately tuberculated. Posterior valve ... with subcentral mucro, anterior area concentrically striated, posterior concentrically tubercu- lated, the mucro is eroded in specimens studied. Anterior and posterior valves with eight slits, median with one; teeth finely pectinated and roughened with minute grains. Scales of girdle radiately furrowed, somewhat apart, large and small intermingled, with a series of very small next the valves and along the margin. Gills extending along five-sixths of the foot. Length 38, breadth 22 mm” (HEDLEY, 1898:100-101). The holotype (AMS C.4344), is accompanied by a pink museum label that reads, in part, “... Loc.: Isle of Pines, New Caledonia / Ref.: Fig’d. P. L. S., NSW, 1898, pt. 1, pg. 100 / Oct. 1897, Coll. C. Hedley.” The specimen (Figures 35-38)—preserved dry, soft parts in place, part- plus fragment of girdle—agrees with HEDLEY’s (1898) description and illustration of the species. Among 74 specimens of Acanthopleura araucariana here examined, largest 52 mm long. Body width/length, mean Page 237 ss, 20041M Figure 33 Acanthopleura brevispinosa (Sowerby, 1840a). Same specimen as in Figure 32. Radula median and first lateral teeth, and head of major lateral tooth. 0.60 (SD = 0.05; n= 10). Specimens depressed, round- backed, often markedly eroded. Shell in tones of light grays. Lateral areas of intermediate valves markedly raised, with few, robust, coarse granules, up to 350 wm in diameter. Anterior valve and postmucro area of posterior valve sim- ilarly sculptured. Central areas with appressed, transverse lamellae, featureless otherwise. Mucro somewhat poste- rior; postmucro convex. Ocelli round to oval, 50-60 um in diameter, on anterior 2 of lateral areas, anterior valve, and postmucro area of posterior valve. Gills holobranchi- al, 40-50 plumes per side. Articulamentum white with dark brown discolorations at apex of valves. Sutural laminae well developed; sinus well defined. Insertion plates pectinate on outside. On valve viii, sinus relative width, 0.7; pectinations appreci- ably reduced in middle third, and 2 symmetrical slits at outer thirds; well developed transverse callus. Slit formula 8/9-1-2. Girdle thick, musculous, light gray, often with few sym- metrical black bands. Upper surface covered with coarsely striated scales (Figure 39), quite variable in size, up to 1300 um long, lightly imbricated or, more often, separated from each other by “nude” girdle; occasional hyaline spic- ules, single or bunched, amidst scales. Girdle bridges emp- ty. Undersurface paved with transparent, squarish scales, 40 x 40 um, outer edge convex, inner edge concave, ar- ranged as if articulated in rows. In specimen 32 mm long (AJF 532: Pines Id., New Caledonia), radula 10 mm in length, comprises 50 rows of mature teeth. Median tooth (Figure 40) 60 wm wide at anterior blade; first lateral teeth, 120 um wide at anterior blade; head of second major lateral teeth, discoid, 200 wm wide; outer marginal teeth, 200 um long, 180 wm wide (length/width, 1.1). Distribution: Acanthopleura araucariana is known only from Pines Id., adjacent Lifou Id., New Caledonia, and Eua Id. and Tongatapu Id., Tonga (21°10’S, 175°10'W) (Figure 115-K). It is confined to the upper intertidal zone, having been The Veliger, Vol. 28, No. 3 Page 238 ae kenreinas, 1986 observed up to 2 m above water level, often exposed to the sun, immediately above the area occupied by the sym- patric Acanthopleura gemmata (A. J. Ferreira, field obser- vations at Kuto Beach, Pines Id., New Caledonia, 22-24 July 1980). Remarks: The irregularity in the size of the girdle scales in Acanthopleura araucariana was noted by HEDLEY (1898), NIERSTRASZ (1905a) (who proposed a subgenus, Hetero- zona, on that account), and LELOuP (1939b). IREDALE & HULL (1926), having examined the type of Acanthopleura miles and specimens of A. araucariana from New Caledonia, regarded them as conspecific. Yet, differ- ences between A. araucariana and A. miles are apparent in (1) the lateral areas (markedly elevated in A. arauca- riana; hardly raised in A. miles), (2) the granules in lateral areas and end valves (large, elevated, coarse in A. arau- cariana; small, subdued, well defined in A. miles), and (3) the girdle scales (irregular in size, up to 1300 um long, often separated by naked girdle in A. araucariana; rela- tively regular in size, up to 700 um long, often imbricated in A. miles). These differences seem sufficient to justify separation at the species level. Acanthopleura miles (Carpenter in Pilsbry, 1893) Figures 41 to 47, and 115-M Chiton (Sclerochiton) miles Carpenter in PILSBRY, 1893b: 189, pl. 46, figs. 1-5; SMiTH, 1903:619; HEDLEy, 1910: 52% Sclerochiton miles: ASHBY, 1923d:231, pl. 18, fig. 3. Squamopleura miles: LELOUP, 1939c:1-6, figs. 1-4, 1940:1- 7 (in part). Chiton (Squamopleura) miles: SMITH, 1960:66. Squamopleura imitator NIERSTRASZ, 1905a:102-103, pl. 8, figs. 212-218, 1905b:153-154; Horst & SCHEPMAN, 1908:527; LELoup, 1933a:19, pl. 1, fig. 4, 1939d:4-8, figs. 3, 4, 10, 11, 20-27. Sclerochiton imitator: THIELE, 1910:95, pl. 10, figs. 24-28. Sclerochiton thiele. ASHBY, 1923e:233. Squamopleura carter’. IREDALE & HULL, 1926:260, pl. 27, figs. 18, 20, 28 (reprinted, 1927:123, pl. 15, figs. 18, 20, 28). Squamopleura stratiotes LELOuP, 1939d:9-12, figs. 5-7, 12- 15, 28. Squamopleura salisbury: LELOuP, 1939d:9-12, figs. 8, 9, 16- 19, 29. Page 239 (eet ee Figure 39 Acanthopleura araucariana (Hedley, 1898). Kuto Beach, Pines Id., New Caledonia (AJF 532); specimen 35 mm long. Girdle scales, outer side and inner side. “Sclerochiton curtisianus (Smith)” AsHBy, 1922a:34 (fide ASHBY, 1923d:232); ANG, 1967:412-414, pl. 9, figs. 1-5. “Squamopleura curtisiana (Smith) LELouP, 1933a:18-19, pl. 1, figs. 1, 2 (in part: specimen from Mansfield Id., fide LELoupP, 1939d:1-2, footnote); Wu, 1975:69-70, figs. 1-13. Type material and type locality: Chiton (Sclerochiton) miles Carpenter in Pilsbry, 1893b: Holotype (BMNH 198017); locality “Torres Strait” (Australia). Squamopleura imitator Nierstrasz, 1905a: Types unascer- tained, in Zoologischen Museum zu Amsterdam (fide NIERSTRASZ, 1905a); locality “Insel Remarksaja [? Su- matra, Indonesia, 4°52'’N, 95°22’E] and Java.” Sclerochiton thiele. Ashby, 1923: Types unascertained; lo- cality Sumatra. Squamopleura carteri Iredale & Hull, 1926: Holotype (WAM 11662); locality Point Cloates, Western Australia (22°43'S, 113°40'E). Squamopleura stratiotes Leloup, 1939d: Types unascertained; locality Trincomalee, Ceylon (8°34'N, 81°41’E). Squamopleura salisbury: Leloup, 1939d: Types unascer- tained; locality Hambantota, Ceylon (6°07'N, 81°07’E). Explanation of Figures 34 to 38 Figure 34. Acanthopleura brevispinosa (Sowerby, 1840a): Acan- thopleura afra Rochebrune, 1822; paralectotype (MNHN). Figure 35. Acanthopleura araucariana (Hedley, 1898): Ischnochi- ton araucarianus Hedley, 1898; holotype (AMS C-4344). Side Figure 36. Acanthopleura araucariana (Hedley, 1898): Ischnochi- ton araucarianus Hedley, 1898; holotype (AMS C-4344). Dorsal Figure 37. Acanthopleura araucariana (Hedley, 1898): Ischnochi- ton araucarianus Hedley, 1898; holotype (AMS C-4344). Dorsal view of valves i, v, and viii. Figure 38. Acanthopleura araucariana (Hedley, 1898): Ischnochi- ton araucarianus Hedley, 1898; holotype (AMS C-4344). Ventral view of valves i, v, and viii. Page 240 The Veliger, Vol. 28, No. 3 200 am Figure 40 Acanthopleura araucariana (Hedley, 1898). Same specimen as in Figure 39. Radula median and first lateral teeth. Material examined: AUSTRALIA, W.A.: Maud’s Landing, N of Carnarvon (24°53'S, 113°40’E), 2 specimens, ca. 30 mm long each (NMV); Point Cloates, holotype of Squamopleura carteri Iredale & Hull, 1926, ca. 35 mm long (WAM 11662); Dampier Arch., Kendrew Is., 8 specimens, largest 31 mm long (WAM 75-78; WAM 77- 78; WAM 1340-78); Broome Bay, 1 specimen (NMV). NEW GUINEA: Japen Id., Cape Tekopi, 1 specimen, 20 mm long (ANSP 272165). BORNEO (Sabah, Malaysia): Layang-Layangan, Labuan Id., 19 specimens, largest 36 mm long (AJF 759); Tanjong-Kubong, Labuan Id., 48 specimens, largest 40 mm long (AJF 760). TAIWAN: Coral reef southeast of Kungting, Pingtung Co., 18 specimens, largest 32 mm long (UCM 28885); Orchid Is., Tai- tung Hsien, 3 specimens, largest 30 mm long (UCM 28845); Hsiao-liu-chiu Id., Penghu Village, 2 specimens, largest 40 mm long (UCM 28864); Pingtung Hsien, 4 specimens, largest 32 mm long (UCM 28885). PHILIPPINES: Hundred Islands, Lingayen Gulf, Luzon, 75 specimens, largest 35 mm long (AJF 464; AJF 465); “Auson” Id., off Port Barton, Palawan, 1 specimen, 26 mm long (AJF 820). JAVA, Indonesia: Udjong Kuton, 1 specimen, 28 mm long (WAM 478-74). SRI-LANKA (=Ceylon): Trincomalee, 4 specimens, largest 31 mm long (CAS 001203); Dondra Head, 22 specimens, largest 22 mm long (AJF 734); Dondra, 21 specimens, largest 22 mm long (AJF 735); Tangala, 58 specimens, largest 26 mm long (AJF 736; AJF 738); Galle, 22 specimens, largest 21 mm long (AJF 740). Description: The original description of Chiton (Sclero- chiton) miles is quite adequate to identify the species: ““Shell . rugose, oval, depressed ... dorsal ridge rounded ... mucro [posterior], nearly flat; apices of the valves prom- inent, obtuse .... Central areas transversely pretty reg- ularly rugulose, the wrinkles appressed; lateral areas hardly elevated, moderately well defined, conspicuously rugose, rugae subradiating, granose; the end valves simi- larly sculptured ... {slits, 11-1-9/11] .... Teeth of pos- terior valves directed forward, strongly callosed inside above the slits, sulcate outside; the rest of the valves hav- ing the teeth sulcate outside and pectinated at the margins .... Eaves moderate, solid .... Sinus deep, wide, wavy, smooth .... Girdle [with] large, solid, more or less sep- arated scales which are striated outside .. .”” (Carpenter in PILSBRY, 1893b:189). Holotype (BMNH 198017), dry, in excellent condition, flat, 30 mm long, 18 mm wide, 5 mm high, shows evidence of having been glued to cardboard; soft parts removed, posterior valve disarticulated; girdle scales variable in size, up to 600 um long, discretely striated. Accompanying la- bel reads, in part, “Chiton (Sclerochiton) miles (Carpenter MS) Pilsbry / HOLOTYPE / Torres Straits / H. Cum- ing coll. No. 42 / 1 specs Acc. no: 1829 ....” The spec- imen (Figures 41-44) agrees with Carpenter’s (77 PILSBRY, 1893b) description and illustrations. Among 176 specimens of Acanthopleura miles here ex- amined, largest 40 mm long (live) (AJF 760: Labuan Id., Borneo). Body width/length, mean 0.63 (SD = 0.05; n= 30). Specimens round-backed, often depressed; valves moderately beaked, posterior edge angled (130-150° on valve ii). Tegmentum brownish-gray usually with dark Jugal stripe. Lateral areas variably raised, from flat (in Sri-Lanka specimens) to elevated (in Taiwan specimens), with relatively small, round, well defined granules; diag- onal line of lateral areas often defined by row of granules; tegmental surface between granules coarsely microgran- ular. Central areas featureless except for some 30 trans- verse, appressed trabeculae, 60-100 wm thick, often inter- rupted on pleural areas by 4-6 poorly defined oblique riblets. Similar granules on end valves. Mucro central to slightly posterior; postmucro convex, at 30-45° slope. Widths of tegmental surfaces of valves i/vili, mean 1.2. On valve i, tegmentum length/width, mean ratio 2.1. On valve vili, tegmentum’s length/width, mean 1.9. Eaves thick (0.5 mm at midline of valve viii of specimen 30 mm long), relatively solid. Ocelli round to oval, 50-60 um in diameter, randomly distributed amidst granules of ante- rior valve, postmucro area of posterior valve, and anterior % to % of lateral areas of intermediate valves. Gills with 40-50 plumes per side. Articulamentum brown to white. Insertion plates strongly pectinate on outside. On valve i, length of inser- tion plate/length of tegmentum, mean 0.14; insertion teeth irregularly spaced. On valve viii, insertion plate with ir- regular pectinations becoming smaller to obsolete in mid- dle %, often with only 2 symmetrical slits (but in some specimens with as many as 5); well defined transverse round callus. Slit formula 5/10-1-1/5. Sutural laminae well developed, subtriangular on valve ii to subrectangular on valve viii. Sinus well defined; sinusal plate, smooth; relative width of sinus (width of sinus/width of sutural lamina) on valve viii, 0.5. Girdle thick, musculous, often all black. Upper surface covered with calcareous, opaque, oval, strongly convex scales (Figure 45), more or less separated (evident in live or wet preserved specimens) to somewhat imbricate, vari- able in size, often up to 600-700 um long (largest mea- sured, 1100 um long), with 6-15 discrete striations; amidst Aes Ferreira, 1986 Page 241 Explanation of Figures 41 to 44, and 47 Figure 41. Acanthopleura miles (Carpenter in Pilsbry, 1893c): Chiton (Sclerochiton) miles Carpenter in Pilsbry, 1893c; holotype (BMNH 1951.2.1.2). Dorsal aspect of valves i and ii. Figure 42. Acanthopleura miles (Carpenter in Pilsbry, 1893c): Chiton (Sclerochiton) miles Carpenter in Pilsbry, 1893c; holotype (BMNH 1951.2.1.2). Side view of valves iii, iv, and v. Figure 43. Acanthopleura miles (Carpenter in Pilsbry, 1893c): scales, occasional bunches of hyaline spicules, up to 100 x 30 um. Girdle bridges empty. Undersurface covered with transparent, rectangular scales, 50 x 30 um, with slightly convex outer edges, slightly concave inner edges; at outer Chiton (Sclerochiton) miles Carpenter in Pilsbry, 1893c; holotype (BMNH 1951.2.1.2). Dorsal aspect of posterior valve. Figure 44. Acanthopleura miles (Carpenter in Pilsbry, 1893c): Chiton (Sclerochiton) miles Carpenter in Pilsbry, 1893c; holotype (BMNH 1951.2.1.2). Ventral aspect of posterior valve. Figure 47. Acanthopleura miles (Carpenter in Pilsbry, 1893c): Squamopleura carteri Iredale & Hull, 1926; holotype (WAM 11662). margin, fringe of translucent spicules, up to 150 x 40 um, vaguely striated longitudinally. In a specimen 25 mm long (AJF 464: Lingayen Gulf, Philippines), radula measures 10 mm in length (40% of Page 242 L_ 4 1mm Figure 45 Acanthopleura miles (Carpenter in Pilsbry, 1893c). Hundred Is- lands, Lingayen Gulf, Philippines (AJF 464); specimen 25 mm long. Girdle scales, outer side and inner side. specimen length), comprising 40 rows of mature teeth; median tooth, 40 um wide at anterior blade, bulging pos- teriorly to 60 wm wide (Figure 46); lateral teeth 75 um wide at anterior blade, with outer edge deeply concave, outer-posterior corner protruding almost to a knob; major lateral teeth with discoid head, 160 wm in width; outer marginal teeth 140 wm long, 100 um wide (length/width, 1.4). Distribution: Widely distributed in the central Indo-Pa- cific (Figure 115-M), from 25°S to 27°N, and from 73°E to 142°E, Acanthopleura miles has been recorded, albeit by synonymous names, at Atu Atoll, Maldives Is. (SMITH, 1903), Ceylon (LELouP, 1939c, as Squamopleura stratiotes and S. salisbury:), Andaman Is. and Nicobar Is. (LELOUP, 1939c), Raja Id., Java, and Timor (NIERSTRASZ, 1905a, b; THIELE, 1909), Sumatra (THIELE, 1909; AsHBy, 1923d, as S. thielei Ashby, 1923d), Bali (LELouP, 1933a, as S. umitator), Philippines (ANG, 1967, as S. curtisiana), Lan- shu Id., Taiwan (Wu, 1975, as S. curtisiana), Mansfield Id. (LELOupP, 1933a, as S. curtistana), Torres Strait, Aus- tralia (Carpenter in PILSBRY, 1893b, type locality), Car- narvon, W. Australia (ASHBY, 1923d), Point Cloates, W. Australia (IREDALE & HULL, 1926, as S. carteri). The species is here recognized also at Taiwan, Philippines (Lingayen Gulf), Borneo, and New Guinea (AJF coll.). Acanthopleura miles is confined to the intertidal zone, 0-0.5 m, often on rocks exposed at low tide. Remarks: The identity of Acanthopleura miles has been much confused in the literature. IREDALE (1914b:125), noting that “Nothing like . . . miles has yet been seen from Torres Straits,” assumed the locality in error, a gratuitous assumption later echoed by AsHBy (1923) and IREDALE & HULL (1926). LELoup (1939c) introduced Squamopleu- ra siratiotes and S. salisburyi, from Ceylon, which he im- mediately (in the same paper!) synonymized as geographic forms of a single species, later (LELOUP, 1940) regarded simply as a variety (“forma”) of S. miles. As observed here, Acanthopleura miles differs from A. araucariana in (1) lateral areas (very elevated in A. arau- cariana; hardly raised in A. miles), (2) tegmental granules (coarse, large in A. araucariana; well defined, small in A. miles), and (3) girdle scales (larger and separated in A. araucariana; smaller and often imbricated in A. miles). The Veliger, Vol. 28, No. 3 L », 100m Figure 46 Acanthopleura miles (Carpenter in Pilsbry, 1893c); same speci- men as in Figure 45. Radula median and first lateral teeth. In the field, Acanthopleura miles was observed sharing the intertidal zone with A. gemmata, and A. spinosa (AJF 464, AJF 465, Lingayen Gulf, Philippines). Acanthopleura curtisiana (Smith, 1884) Figures 48 to 54, and 115-T Chiton (Ischnochiton) curtisianus SMITH, 1884:78, pl. 6, fig. D. Ischnochiton curtisianus: PILSBRY, 1892b:97, pl. 24, fig. 6. Liolophura curtisiana: PILSBRY, 1893c:242; HEDLEY & HULL, 1909:265. Sclerochiton curtisianus: THIELE, 1910:96, pl. 10, figs. 29- 35; IREDALE, 1910a:103-104, 1914b:125; AsHBy, 1922a: 34. Squamopleura curtisiana: HULL, 1923:160, 1925:114; IRE- DALE & HULL, 1926:259-260, pl. 37, figs. 5, 26-27 (reprinted, 1927:122-123, pl. 15, figs. 5, 26-27); Mackay, 1930:292-295; LELoup, 1933b:17-19, pl. 1, fig. 3 (in part: specimens from Aru Id.), 1939d:1-4, figs. 1-2. [Non: LELouP, 1933a:17-19, pl. 1, figs. 1-2; ANG, 1967: 412-414; Wu, 1975:69-70, figs. 1-13.] Enoplochiton torr: BAstow & GATLIFF, 1907:27-30, pls. 3- 4, figs. 1-12. Sclerochiton aruensis THIELE, 1910:96, pl. 10, figs. 36-41. Type material and locality: Chiton (Ischnochiton) curtisianus Smith, 1884: Holotype (BMNH 1881.11.10.32); locality Port Curtis, Queens- land, Australia (24°00’S, 151°30’E). Enoplochiton torr: Bastow & Gatliff, 1907: Types unascer- tained; locality Queensland, Australia. Sclerochiton aruensis Thiele, 1910: Types unascertained; lo- cality Aru Is., Indonesia (6°00’S, 134°30’E). Material examined: AUSTRALIA, Qld.: Utinga, Cape York, 15 specimens, largest 23 mm long (NMV); Port Curtis, holotype of Chiton (Ischno- chiton) curtisianus Smith, 1884 (BMNH 1881.11.10.32); Port Curtis, 9 specimens, largest 23 mm long (NMV); Cid Id., W of Aceje Kerreira, 1986 Whitsunday Id., 1 specimen, 18 mm long (CAS 044790); Shelly Beach, near Townsville, 10 specimens, largest 25 mm long (NMV); Gill’s Beach, Hinchinbrook Id., 5 specimens (ex J. R. Penprase coll.); Broad Sound, 17 specimens, largest 18 mm long (NMV); Moreton Bay, 8 specimens, largest 24 mm long (NMV). AUSTRALIA, N.T.: Port Darwin, 2 specimens (CASG-SU 3389). AUSTRALIA, W.A.: Point Cloates, 4 specimens, largest 40 mm long (cited by AsHBy, 1922a) (WAM 9336); Bay of Rest, Ex- mouth Gulf, 3 specimens, largest 25 mm long (WAM 1746-78); Dampier Arch., 2 specimens, largest 16 mm long (WAM 82- 78; WAM 1079-75; Barrow Id., 1 specimen, 19 mm long (WAM 605-67); Cape Preston, 1 specimen, 17 mm long (WAM 1113- 78); Broome, 1 specimen, 19 mm long (NMV). Description: SMITH (1884) described Chiton (Ischnochi- ton) curtisianus, based on a specimen 16 X 9 mm (exclud- ing girdle), as “Shell ... dark greyish ... [with] black broadish line from end to end down the middle of the back . with strong, concentric lines of growth .... [Girdle] alternately light and dark .. . . Valves arched, not carinate . very indistinct lateral areas ... mucro probably near the centre .... [Articulamentum] greenish blue, stained dark brown in the middle . . . [slits 10-1-0] ... . [Anterior valve insertion teeth] different-sized ... striated on both sides but more strongly externally, their edges being sharp, but not smooth .... [Posterior valve] much thickened within along the posterior edge, which is roughened by fine cross striae, there being no prominent teeth, and of course no notches .... [Girdle] covered with small sub- imbricating oval [scales] .... The granules of the surface have an irregular disposition, following to some extent the lines of growth” (SMITH, 1884:78-79). The holotype (BMNH 1881.11.10.32) consists of 8 dis- articulated valves; no girdle, no radula. Reassembled, the valves add up to 18 mm, indicating a live specimen (with girdle) about 20 mm long (larger than, but compatible with length given by SMITH [1884]). Museum pink label reads, in part, “... Holotype / Chiton (Ischnochiton) cur- tisanus Smith, 1884 ... / Loc. Port Curtis ... / leg. Coppinger / Pres. The Admiralty.” The specimen (Fig- ures 48, 49) agrees with SMITH’s (1884: pl. fig. D) de- scription and illustration: light brown, with darker jugal band flanked by lighter bands. Valves considerably erod- ed, round-backed; beak distinct on valve ii, indistinct on others; posterior edge angulate at 130° on valve 11, less so on valve iii, almost straight on valves iv—vil. Tegmentum on valve i sculptured with concentric rows of round to oval granules, 100-150 um in diameter, often coalesced into minute ridges. Lateral areas of intermediate valves slightly raised, poorly defined otherwise; sculpture with similar granules. Central areas with similar granules, more noticeable in pleural areas (particularly on valve ii) where they align into longitudinal rows; jugal areas with similar granules, though less well defined, and few coarse, trans- verse growth rugae. Posterior valve somewhat flattened, much eroded; mucro central (?). Width of valve i/width of valve viii, 1.15. Ocelli round to oval, 70-90 um in Page 243 diameter, throughout anterior valve, lateral areas of in- termediate valves, and postmucro area of posterior valve. Articulamentum white with intense, large, brown dis- coloration in the middle. Sutural laminae, subtriangular on all intermediate valves, more so on ii; sinus very wide in all valves; on valve viii, width of sinus/width of sutural laminae, 0.9. Insertion plates pectinate on outside. On valve i, 10 slits, teeth 0.4 mm long. Intermediate valves uni-slit. On valve viii, insertion teeth almost obsolete, re- duced to small pectinations on posterior aspect of round, transverse callus; only 2 (? 3) vestigial slits. Eaves rela- tively compact, 0.3 mm wide. Among 78 specimens examined, largest 40 mm long (dry) (WAM 9336: Pt. Cloates, Australia). Body width/ length, mean 0.66 (n = 20). Tegmentum grayish-green to grayish-brown, often with wide dark brown band at jug- um. Lateral areas poorly defined, hardly elevated, with round to elongate granules, 100-200 um in diameter, often in concentric rows. Central areas with smaller granules, 80-150 um in diameter in pleural areas, often aligned in 10-15 longitudinal rows, much less apparent at jugum. Intermediate valves clearly beaked on valve ii, less so pos- teriorly; posterior edge of valve ii forming 130-140° angle. Mucro central; postmucro moderately convex (Figure 50). Ocelli round to oval, 60-80 wm in diameter, randomly distributed throughout anterior valve, postmucro area of posterior valve, and anterior ¥% to 2 of lateral areas of intermediate valves. Widths of tegmental surfaces of valves i/viiil, mean 1.08. On valve i, tegmentum length/width, mean 0.48. On valve viii, tegmentum length/width, mean 0.48. Eaves thick (0.4 mm in midline of valve viii), rela- tively solid. Gills 25-30 plumes per side. Articulamentum white with diffuse light brown discol- orations in middle. Sutural laminae well developed, sub- triangular on valve ii to subrectangular on valve viii. Sinus wide. Insertion plates pectinate on outside. On valve i, length of insertion plate/length of tegmentum, mean 0.13. On valve vill (Figures 51, 52), insertion plate markedly underdeveloped to obsolete, often reduced to weak pecti- nations on posterior aspect of thick, smooth, transverse callus; 2 slits symmetrically placed at outer fourth of plate, weakly defined to obsolete. Slit formula 8/10-1-0/2. Girdle thick, musculous, wide, often banded. Upper surface with oval, vaguely striated, opaque, calcareous scales (Figure 53), up to 300-400 um in length in middle of girdle, progressively smaller to minute at margins; in living and alcohol preserved specimens, girdle scales are clearly separated from each other by about half of their width; occasional bunches of small, translucent spicules, up to 70 x 15 um, amidst the scales. Girdle bridges emp- ty. Undersurface covered with transparent, rectangular scales, 40 x 20 um, with coarse striations. In a specimen 17 mm long (NMV: Utinga, Cape York, Australia), radula 6 mm in length (35% of specimen length), comprising 35 rows of mature teeth. Median tooth (Figure 54), 35 wm wide at anterior blade, parallel-sided, The Veliger; Vol) 28) Nos Page 244 Explanation of Figures 48 to 52 Figure 48. Acanthopleura curtisiana (Smith, 1884): Chiton curtis- zanus Smith, 1884; holotype (BMNH 1881.11.10.32). Dorsal aspect of valves i, ii, v (?), and viii. Figure 49. Acanthopleura curtisiana (Smith, 1884): Chiton curtis- zanus Smith, 1884; holotype (BMNH 1881.11.10.32). Ventral side of posterior valve. Figure 50. Acanthopleura curtisiana (Smith, 1884). Cape York, globose posteriorly; first lateral teeth 65 wm wide at well developed anterior blade; second lateral teeth with discoid head 130 um in diameter; outer marginal teeth, 130 um long, 80 um wide (length/width, 1.6). Distribution: Acanthopleura curtisiana is confined to trop- ical northern Australian waters (Figure 115-T). It has been recorded in Queensland at Port Curtis (SMITH, 1884; IREDALE, 1910a; HULL, 1923), Gladstone (HEDLEY & HULL, 1909), Keppel Bay, Broad Sound, Magnetic Id. N.T., Australia (NMV); specimen 17 mm long. Dorsal aspect of posterior valve. Figure 51. Acanthopleura curtisiana (Smith, 1884). Same speci- men as in Figure 50. Ventral aspect of posterior valve. Figure 52. Acanthopleura curtisiana (Smith, 1884). Same speci- men as in Figure 50. Posterior aspect of posterior valve. (Mackay, 1930), Cape York (LELOupP, 1939c), Thursday Id. (IREDALE, 1910a), and (herein) at Moreton Bay (27°20’S), its southernmost record; in the Northern Ter- ritory, at Port Darwin (AsHBy, 1922a; and herein); in Western Australia at Point Torment, and Point Cloates (22°43'S), its southernmost record on the western coast. Reports of the species at Aru Is., Indonesia (6°S) (THIELE, 1909; LeLoup, 1933a), may constitute northernmost rec- ords. LELOuP’s (1933a) report of the species at Mansfield Id. was later retracted (LELOUP, 1939d). Reports of A. ACM eberrema, O86 n , 2504m Figure 53 Acanthopleura curtisiana (Smith, 1884). Exmouth Gulf, W.A., Australia (WAM 1746-78); specimen 23 mm long. Girdle scales, outer and inner sides. curtisiana in the Philippines (ANG, 1967) and Taiwan (Wu, 1975) are presumed in error for A. miles. Bathymetric range 0-1 m, on rocks often exposed at low tide. Remarks: Acanthopleura curtisiana has often been con- fused with A. miles. The two species, hitherto assigned to Squamopleura, are identical in size, shape, color, and hab- itat, but they differ in (1) size of girdle scales (up to 400 um long in A. curtisiana; up to 800 um long in A. miles), (2) central areas (with longitudinal rows of granules on pleural areas in A. curtisiana; without granules, feature- less except for fine, appressed, transversal lamellae at ju- gal areas in A. miles), (3) radula (median tooth parallel- sided in A. curtisiana; bulging posteriorly in A. miles). THIELE’s (1910) illustration of the median tooth of the radula of Acanthopleura curtisiana shows a two-pointed posterior end, instead of a globose end as here illustrated (Figure 54); the discrepancy may reflect intraspecific vari- ation or the difficulties in properly visualizing radular teeth under ordinary microscopy. Acanthopleura loochooana (Broderip & Sowerby, 1829) Figures 55 to 61, and 114-L Chiton loochooanus BRODERIP & SOWERBY, 1829:368. Liolophura loochooana: PILsBRY, 1893c:244; (?) Is. Taki, 1938: 411, 1962:46 (with Liolophura gaimardi platispinosa Leloup, 1939a, as syn.); KAAS & VAN BELLE, 1980:76. Type material and locality: Chiton loochooanus Broderip & Sowerby, 1829: Neotype (CAS 044306) here designated; locality “shore of Loo Choo Is.” (=Ryukyu Islands, Japan), here restricted to Buck- ner Bay, Okinawa, Japan (26°17'51.5”N, 127°54'26"E), intertidal zone. Other material examined: OKINAWA, Japan: Buckner Bay, at rocky intertidal zone, neo- type lot, 14 specimens, largest 26 mm, smallest 8 mm long, Jeg. E. V. Iverson, 23 Apr. 1975 (CAS 001627); Nagahama, 2 spec- imens, largest 25 mm long (CAS 019919; CAS 040204); Okuma, 1 specimen, ca. 50 mm long (S. Crittenden coll.). TAIWAN: Penghu Is., south of Chienshan Village, Penghu Co., Page 245 Figure 54 Acanthopleura curtisiana (Smith, 1884). Same specimen as in Fig- ure 53. Radula median and first lateral teeth. 4 specimens, largest 28 mm long (UCM 28893); Keelung, 18 specimens, largest 26 mm long, /eg. F. B. Steiner, Oct. 1963 & Sept. 1968 (CAS 009880; CAS 016698). HONG KONG: Repulse Bay, Hong Kong Id., 2 specimens, largest 18 mm long (AJF 686); Cheung-Sha, Lantau Id., 7 spec- imens, largest 34 mm long (AJF 685). Description: Chiton loochooanus was described as “Ch. valuis subscabrosis, areis marginalibus radiatim granosis, margin coriaceo superne granoso, granielevatis; long. 15/20, lat. 3/10 poll. Hab. in mari Sinensi, ad littora Insulae Loo- Choo. A very pretty little Chiton, whose margin is covered with small grains resembling very short, blunt spines” (BRODERIP & SOWERBY, 1829:368). Among 48 specimens here referred to Acanthopleura loochooana, largest 50 mm long (in alcohol) (S. Crittenden coll.: Okuma, Okinawa). Body width/length, mean 0.65. Round-backed; intermediate valves beaked; posterior edge of valve ii forming 140-160° angle. Tegmentum grayish- white to brown, often with dark gray jugal stripe. Lateral areas elevated, well defined, crowded with low, round granules which tend to align radially and coalesce; ante- rior valve and postmucro area of posterior valve similarly sculptured. Central areas with ill-defined sculpture of ap- pressed transverse lamellae, often broken into irregular granules or rugosities which, on pleural areas, tend to form forward-converging riblets. Mucro central (in small specimens) to posterior (in larger ones); postmucro slight- ly to markedly convex, slope varying with size of specimen (Figure 55). Ocelli round to oval, 60-70 um in diameter, throughout anterior valve, postmucro area of posterior valve, and anterior half of lateral areas. On valve i, teg- mentum length/width, mean 0.48. On valve viii, tegmen- tum length/width, mean 0.42. Widths of tegmental sur- faces of valves i/viii, mean 1.2. Gills with 35-40 plumes per side. Articulamentum brown with white discoloration at su- tural laminae and insertion plates. Sutural laminae well developed, relatively long, subtriangular on valve ii to sub- rectangular on valve viii. Sinus well formed; sinusal plate Page 246 ihe Veliger; VolyZ85 Nowe A. J. Ferreira, 1986 Page 247 1mm Figure 58 Acanthopleura loochooana (Broderip & Sowerby, 1829). Keelung, Taiwan (CAS 009880); specimen 29 mm long. Girdle scales, outer and inner sides. smooth; relative width of sinus on valve viii, 0.8. Insertion plates pectinate on outside. On valve i, insertion teeth irregularly spaced; in midline, length of insertion teeth/ length of tegmentum, mean 0.13. On valve viii, pectina- tions and teeth underdeveloped to obsolete in middle third, hardly extending beyond transverse round callus in outer third; slits of posterior valve not always clearly determin- able (Figures 56, 57). Slit formula 8/10-1-2/7. Eaves relatively thick (0.4 mm on midline of valve viii in spec- imen 20 mm long), moderately spongy. Girdle banded black and white, thick, musculous. Up- per surface girdle elements (Figure 58) irregular in size and shape; most elements spinelet-like, up to 2000 wm long, 600 wm wide, relatively flattened in cross section; other elements scale-like, 7.e., much smaller in all dimen- sions, particularly in height, vaguely striated on outer face; in living or wet preserved specimens, girdle elements often separated by “nude” girdle, with occasional hyaline, needle-like elements, single or bunched, up to 120 x 20 um. Girdle bridges empty. Undersurface paved with transparent, subrectangular scales, about 40 x 30 um, be- coming elongate to 70 x 30 um at periphery, with some ce 1 100 4m Figure 59 Acanthopleura loochooana (Broderip & Sowerby, 1829). Same specimen as in Figure 60. Radula median and first lateral teeth. coarse striations, convex outer edge, concave inner edge. Marginal fringe of 1 or 2 rows of translucent spicules, up to 300 x 50 um, finely striate. In specimen 29 mm long (CAS 009880: Taiwan), rad- ula measures 12.8 mmm in length (44% of specimen length), comprising 55 rows of mature teeth. Median tooth (Fig- ure 59) 80 wm wide at anterior blade; first lateral teeth 110 um wide at anterior blade; head of second lateral teeth discoid, 210 um wide; outer marginal teeth 200 um long, 150 um wide (length/width, 1.4). Distribution: The geographic range of Acanthopleura loo- chooana extends from Okinawa (26°31'N, 127°59’E) to Taiwan and Hong Kong (22°15'N, 114°10’E) (Figure 114- L). Listings of the species at Shizuoka, Japan (Is. TAKI, 1938, 1962) cannot be taken at face value. Acanthopleura loochooana seems to be confined to the intertidal zone, on top of rocks in moderate surf areas. Remarks: Chiton loochooanus, left unfigured, was consid- ered by PILSBRY (1893c:244) as “absolutely unrecogniz- able ... but ... perhaps a member of the genus Liolo- phura.” The original type material could not be found Explanation of Figures 55 to 57, and 60 to 65 Figure 55. Acanthopleura loochooana (Broderip & Sowerby, 1829). Specimen from the neotype-lot (CAS 001627). Dorsal aspect of posterior valve. Figure 56. Acanthopleura loochooana (Broderip & Sowerby, 1829). Same specimen as in Figure 55. Posterior aspect of posterior valve. Figure 57. Acanthopleura loochooana (Broderip & Sowerby, 1829). Same specimen as in Figure 55. Ventral aspect of posterior valve. Figure 60. Acanthopleura loochooana (Broderip & Sowerby, 1829); neotype (CAS 044306). Figure 61. Acanthopleura loochooana (Broderip & Sowerby, 1829); neotype (CAS 044306). Close-up of girdle. Figure 62. Acanthopleura gaimardi (Blainville, 1825). Magnetic Id., Qld., Australia (AJF 602); specimen 15 mm long. Figure 63. Acanthopleura gaimardi (Blainville, 1825). Magnetic Id., Qld., Australia (AJF 602); specimen 30 mm long. Dorsal aspect of posterior valve. Figure 64. Acanthopleura gaimardi (Blainville, 1825). Same spec- imen as in Figure 63. Posterior aspect of posterior valve. Figure 65. Acanthopleura gaimardi (Blainville, 1825). Same spec- imen as in Figure 63. Ventral aspect of posterior valve. Page 248 (Solene Morris, BMNH, zn Jitt., 7 July 1983) and is presumed lost. The finding of specimens from the type locality (Okinawa) compatible with BRODERIP & SOWERBY’s (1829) meager description of the species sug- gests the identification, albeit on subjective grounds. To obviate future uncertainties, a neotype (CAS 044306), 25 x 15 x 7 mm (including girdle) (Figures 60, 61), is here designated from neotype lot of 15 specimens pre- served in alcohol (CAS 001627). In the underdeveloped insertion teeth and pectinations of the posterior valve, Acanthopleura loochooana is similar to A. brevispinosa, A. miles, A. araucariana, A. curtisiana, and A. arenosa. The differences lie almost exclusively in the girdle elements: in A. miles, A. araucariana, and A. curtisiana, the girdle elements are clearly scales; in A. brev- ispinosa, they are thin and cylindrical spicules; and in A. arenosa, they are pointed spicules, almost conical, circular in cross section, and larger at base. In its relatively limited range, Acanthopleura loochooana is sympatric with A. japonica, A. miles, and A. gemmata. Specimens of A. loochooana have been found exposed at low tide, up to 30 cm above water level, sharing habitat with specimens of A. japonica (A. J. Ferreira, field obser- vations at Lantau Id. and Repulse Bay, Hong Kong, Sept. 1982). Acanthopleura loochooana is present in a relatively nar- row area in the zone of contact between A. japonica and A. gemmata. In this respect, A. loochooana seems to be related to A. japonica and A. gemmata in the vicinity of the Tropic of Cancer, as A. arenosa is to A. gemmata and A. gaimardi in the vicinity of the Tropic of Capricorn. Acanthopleura gaimardi (Blainville, 1825) Figures 62 to 67, and 114-D Chiton gaimardi BLAINVILLE, 1825:546. Liolophura gaimardi: PILSBRY, 1893c:240-241, pl. 53, figs. 30-35; NIERSTRASZ, 1905b:155 (in part: Sydney spec- imens only); Horst & SCHEPMAN, 1908:528 (in part: Sydney specimens only); ASHBy, 1918c:87, 1922¢:581; HULL, 1923b:198, pl. 28, figs. 1-4; AsHBy, 1926b:384; IREDALE & HULL, 1926:262, pl. 37, figs. 13-16, 19, 31 (reprinted, 1927:125, pl. 15, figs. 13-16, 19, 31); BERGENHAYN, 1930a:32, pl. 8, figs. 76-77; ALLAN, 1959: 238, fig. 6a; LELOUP, 1961b:42-44, text figs. 3, 5b, pl. 3, fig. 2; Wu, 1969:109, figs. 5a, 5b, 47-58. [Non: NIERSTRASZ, 1905a:108, 1905b:154-155, figs. 20- 21; Horst & SCHEPMAN, 1908:528 (in part); LELOuP, 1939a:3-7, fig. 5A.] Liolophura gaimardi queenslandica PILSBRY, 1894f:87-88; ASHBY, 1918c:87; Davis et al., 1979:2, 18. [Non: LELoup, 1961b:67, fig. 5-B3)]. Liolophura queenslandica: HULL, 1923b:199, pl. 28, figs. 5- 8, 1925:115; IREDALE & HULL, 1926:263, pl. 37, figs. 23-25, 30, 32 (reprinted, 1927:126, pl. 15, figs. 23-25, 30, 32); ALLAN, 1959:239. Chiton incanus GOULD, 1846:145 (reprinted, 1862:6), 1861 (Atlas):315, pl. 28, figs. 432, 432a. Maugeria incana: GOULD, 1862:248. Chiton (Acanthopleura) incanus: SMITH, 1884:81-82. The Veliger, Vol. 28, No. 3 Acanthopleura incana: HADDON, 1886:25-30. Liolophura incana: PILsBRY, 1893a:105. “Chiton piceus Gmelin” ANGaS, 1867:223. Type material and type locality: Chiton gaimardi Blainville, 1825: Possible types at MNHN (fide ASHBY, 1922c:581), not examined; locality Port Jackson, New South Wales, Australia (53°50’S 151°16’E). Liolophura gaimardi queenslandica Pilsbry, 1894f: Lectotype (ANSP 64853) and paralectotype (ANSP 355874) herein designated; locality Bundaberg, Queensland, Australia (24°52’S, 152°21’E). Chiton incanus Gould, 1846: Holotype (USNM 5823) and paratypes (MCZ 169189); locality “New South Wales,” Australia. 5) Material examined: AUSTRALIA, Qld. Townsville, 11 specimens, largest 42 mm long (AJF 600); Radical Bay, Magnetic Id., 14 specimens, larg- est 30 mm long (AJF 602); Yeppoon, 33 specimens, largest 59 mm long (AJF 356; AMS C135480); Stradbroke Id., 9 speci- mens, largest 36 mm long (AMS C135478; AMS C13008); Port Curtis, Bagara Beach, 1 specimen (AMS C109294); Bundaberg, 13 specimens, largest 45 mm long (AJF coll., leg. Gail Chap- man); Pt. Cartwright, 1 specimen, 28 mm long (NMV, ex Basset Hull coll.); Caloundra, 8 specimens, largest 61 mm long (NMV); Peel Is., Moreton Bay, 1 specimen, 38 mm long (AMS C109296). AUSTRALIA, N.S.W.: Holotype of Chiton incanus Gould, 1846 (USNM 5823); Flat Rock, north of Richmond River, 2 speci- mens, largest 50 mm long (AMS C50821); Byron Bay, 2 spec- imens, largest 50 mm long (CAS 012387); Sydney, 3 specimens, largest 32 mm long (WAM 1592-78; WAM 1590-78; CASG- SU 2868); Port Jackson, 11 specimens (WAM 1591-78; AMS C135482); Gunnamatta Bay, Port Hacking, 2 specimens, largest 35 mm long (AMS C135479). Description: BLAINVILLE’s (1825) description of Chiton gaimardi based on 3 specimens, “un pouce a quinze lignes” (27-34 mm) in length, collected by Quoy and Gaimard at Port Jackson, Australia, stresses the fact that insertion teeth are absent on the posterior valve but present and pectinate on the anterior valve. The species was further described and illustrated by PiLsBry (1893c), HULL (1923b), and IREDALE & HULL (1926), always with sig- nificance given to the callused, slitless and toothless pos- terior valve. Among 112 specimens of Acanthopleura gaimardi here examined, largest 61 mm long (in alcohol) (NMV: Ca- loundra, Qld., Australia). Specimens (Figures 62-65) de- pressed, round-backed. Body width/length, mean 0.60. Intermediate valves beaked; posterior edge of valve 11 forming 110-140° angle. Tegmentum grayish-green to grayish-brown often with wide black band at jugum. Lat- eral areas poorly defined, hardly raised, sculptured with inconspicuous, low-profile granules in vaguely defined ra- dial rows, or coalesced into obsolete concentric ridges; an- terior valve similarly sculptured. Central areas almost fea- tureless except for transversely appressed lamellae, about 50 um thick. Posterior valve rather flat, almost triangular on |peierreinas, 1986 ee ean eg Figure 66 Acanthopleura gaimardi (Blainville, 1825). Yeppoon, Qld., Aus- tralia (AJF 356); specimen 45 mm long. Girdle spines. in some specimens; mucro inconspicuous, decidedly pos- terior to terminal. Ocelli round to oval, 60-70 um in di- ameter, randomly distributed throughout anterior valve, postmucro area of posterior valve, anterior 2 to % of lat- eral areas of intermediate valves, and (in about 50% of specimens examined) pleural areas. On valve i, tegmen- tum length/width, mean 0.5. On valve viii, tegmentum length/width, mean 0.4. Widths of tegmental surface of valves i/vili, mean 1.1. Gills with 35-50 plumes per side. Articulamentum dark brown, often lighter on sutural laminae. Sutural laminae well developed, subtriangular on valve ii to subrectangular on valve viii. Sinus well formed; sinusal plate mostly smooth. Insertion plates of anterior and intermediate valves, strongly pectinate on outside. On valve i, insertion teeth irregularly spaced, sometimes fused together; in midline, length of insertion plate/length of tegmentum, mean 0.14. Valve viii without insertion teeth; posterior aspect of transverse callus flat, crescentic, smooth, without slits or pectinations. Slit for- mula 8/12-1-0. Eaves thick, projecting conspicuously be- yond articulamentum of posterior valve. Girdle, thick, musculous, wide, often banded black- brown and white. Upper surface crowded with white to dark brown spinelets (Figure 66), often tipped with white, pointed to blunt, close together, straight to curved, some- what conical, variable in size, from short and scale-like to 1.5 mm long, with vague to obsolete longitudinal stria- tions; in fresh or wet-preserved material, spinelets often seen standing apart from each other, separated by “‘nude” girdle; amidst spinelets, in the “nude” girdle, needle-like elements may be found, single or clumped, pointed, hya- line, up to 120 x 25 um. Girdle bridges empty. Under- surface paved with transparent, squarish to rectangular scales, about 60 x 40 um, vaguely striate. Radulae averaging 45% of specimen length, with 50 rows of mature teeth. In specimen 40 mm long (AJF 356: Yeppoon, Australia), median tooth (Figure 67) 50 wm wide at anterior blade; first lateral teeth 80 um at anterior blade; head of second lateral teeth discoid, 230 um wide; Page 249 100 um Figure 67 Acanthopleura gaimardi (Blainville, 1825). Same specimen as in Figure 66. Radula median and first lateral teeth. outer marginal teeth, 220 wm long, 110 um wide (length/ width, 2.0). Distribution: Acanthopleura gaimardi is confined to the eastern coast of Australia between Magnetic Id. (19°08'S) (AJF 602), the northernmost verified record, and Sydney (33°52'S) (IREDALE & HULL, 1926; Ferreira, herein), the southernmost verified record (Figure 114-D). The pres- ence of the species southward to Port Hacking (34°05'S) (HULL, 1923b) is credible but requires corroboration. Re- ports of the species in the Moluccas Is. (NIERSTRASZ, 1905b; Horst & SCHEPMAN, 1908), Japan (LELOUP, 1939a), and Taiwan (WU, 1969) are in error. Bathymetric range, confined to the intertidal zone, often exposed at low tide. Remarks: Acanthopleura quatrefagest ROCHEBRUNE, 1881a: 42 (misspelled as quatrefage: in ROCHEBRUNE, 1881b:117), from Table Bay, South Africa, was regarded by ASHBY (1931a) as a synonym of A. brevispinosa and by THIELE (1910), LELoup (1961b), and Kaas & VAN BELLE (1980) as a synonym of A. gaimardi. LELOUP (1939a, fig. 5-B) figured a valve of a supposed ‘“‘co-type.”” But type material has not been found at MNHN (A. Tillier, zn (itt. 15 April 1980) and may be presumed lost. Because ROCHEBRUNE’S (1881a, b) accounts are totally inadequate, the species is here regarded as a nomen dubium. Chiton incanus Gould, 1846, has been regarded as a synonym of C. gaimardi since PILSBRY (1893a). The ho- lotype (USNM 5823) is accompanied by museum labels that read, in part, “. . . Chiton incanus Gould / New South Wales ...,” and“... type. ..”; a handwritten note states “2/8/18 ... compared to undoubted Liolophura gaimardi Blainville 1825 which it certainly is / E. Ashby.” The single specimen, disarticulated with fragments of dry gir- dle, agrees with GOULD’s (1846, 1852) description and illustration of the species. The report of the species at Page 250 Stewart Id., New Zealand (SMITH, 1884) has not been corroborated. PitsBry (1894b:87) established a subspecies, Liolo- phura gaimardi queenslandica, based solely on “the uni- form black color of the girdle” and the “somewhat more slender [spinelets] than in Gazmardi.” The type material consists of 2 specimens, dry, soft parts removed. The larg- er specimen, 42 mm long, 29 mm wide (including girdle), is a 7-valve specimen; the number “64853” is written on the articulamental surface of valve v (?); anterior valve missing; posterior valve, disarticulated showing white ar- ticulamentum and broad, flat, toothless callus; girdle spinelets all black and relatively slender. The smaller specimen, curled, estimated length 20 mm, soft parts re- moved, articulamentum dark brown, is here designated lectotype (ANSP 64853); the larger, partly disarticulated, 7-valve specimen, paralectotype (ANSP 355874). PILSBRY’s (1894b) designation of a subspecies, queenslandica, cannot be maintained, despite HULL (1923b) having raised it to species rank upon hardly spelled out differences in the tegmental sculpture of specimens collected between Port Hacking (34°05’S) and Broken Bay (33°34'S). Acanthopleura gaimardi is remarkably similar to A. ja- ponica. Yet, neither Pitspry (1893c) nor subsequent workers devoted much space to contrasting them. Pilsbry stated only that Liolophura gaimardi differs from L. ja- ponica “in the differently colored interior and sutural plates, in the details of girdle-structure, etc.” (PILSBRY, 1893c:241), and that L. japonica differs “from L. incana [=A. gaimardi] by the uniform black color of the inner layer or articulamentum” (PILSBRY, 1893c:243). Diag- nostic criteria (besides locality) have been increasingly ob- scured by subsequent authors. NIERSTRASZ (1905b:154- 156, pl. 1, figs. 20-25) reported specimens of both L. gaimardi and L. japonica (as “var. tesselata’”’) at the Mo- luccas Is., none of which agrees with the concept of Lio- lophura Pilsbry, 1893a. LELoup (1939) identified Jap- anese specimens as L. gaimardi, and specimens from Indochina as L. japonica; and later LELOuP (1961:38) tab- ulated a single distinction between the two species, the girdle spinelets, said to be “‘of equal length” in japonica, “of different lengths” in gaimardi. Wu (1969) identified specimens from Taiwan as L. gaimardi, with no reference to L. japonica. In this study no reliable differences were found between putative specimens of Acanthopleura gaimardi from Aus- tralia and A. japonica from Japan in habitat, general shape, size, tegmental sculpture, articulamental features, slit for- mula, gills, girdle undersurface scales, or radula. How- ever, the specimens were seen to differ in (1) the shape of the posterior valve (flat, somewhat triangular, mucro al- most terminal in A. gaimardi; modestly elevated, some- what oval, mucro posterior but not terminal in A. japon- ica), (2) the color of articulamentum (dark brown in A. gaimardi; almost black in A. japonica), (3) the callus of the posterior valve (with no vestiges of teeth or pectina- tions in A. gaimardi; often with some symmetrical pecti- The Veliger, Vol. 28, No. 3 nations or vestigial teeth, particularly on outermost areas of the callus, in A. japonica), and (4) the spinelets of the girdle upper surface (pointed, irregular in size, rather conical in A. gavmardi; erect, blunt-ended, regular in size, often white-tipped, cylindrical or wide, and imbricating in A. japonica). Given the considerable variation observed in these characters and their relative unreliability, it re- mains difficult to decide whether there are two species, A. japonica and A. gaimardi, or disjunct populations of a sin- gle biological species of Acanthopleura. However, until and unless further studies (cytological, molecular, etc.) should produce evidence to the contrary, it is here recommended that the traditional view of two species, A. japonica and A. gaimardi, be maintained. Morphologically, Acanthopleura gaimardi is also very close to A. gemmata, from which it differs only in (1) the shape of the posterior valve (oval, elevated, mucro almost central in A. gemmata; triangular, depressed, mucro al- most terminal, in A. gazmardz), (2) insertion plate of pos- terior valve (with well developed teeth in A. gemmata; without teeth in A. garmardz), (3) the tegmental sculpture of the central areas (an unreliable distinction, particularly considering the often extremely eroded condition of spec- imens of either species), (4) the presence of ocelli in the pleural areas in A. gaimardi (an inconstant feature), not in A. gemmata, (5) the smaller average size of A. gaimardi, and (6) the geographic distribution. The differentiation between Acanthopleura gaimardi and A. gemmata is further complicated by the presence of a “transition” population between the two at their zone of contact, a population here regarded as of a different species, A. arenosa, but which further study may prove to be a gaimardi-gemmata hybrid (see Remarks on A. arenosa). Liolophura gaimardi platispinosa LELOUP, 1939a:3-7, figs. 2, 5, reported from Japan and Gulf of Tonkin, is here regarded as a nomen dubium, because description and illustrations do not permit certain assignment to a known taxon; type material not examined. Acanthopleura japonica (Lischke, 1873) Figures 68 to 72, and 115-J Chiton japonicus LISCHKE, 1873:22-23, 1874:71-72, pl. 5, figs. 8-11. Chaetopleura japonica: DUNKER, 1882:158. Acanthopleura japonica: THIELE, 1893:373, pl. 30, fig. 34. Acanthopleura (Liolophura) japonica: THIELE, 1910a:115. Liolophura japonica: PILSBRY, 1893c:242-244, pl. 53, figs. 41-44; NIERSTRASZ, 1905b:155, pl. 10, fig. 22; Horst & SCHEPMAN, 1908:528; THIELE, 1929:21; BERGEN- HAYN, 1933:39-40, pl. 1, fig. 12, pl. 3, figs. 60-67, text figs. 13a—c; Is. TakI, 1938a:398-404, pl. 15, fig. 3, pl. 32, figs. 15-16, pl. 33, figs. 1-8, pl. 34, figs. 1-4, 1947: 1269, fig. 3606, 1949:287, fig. 904; OKADA et al., 1954: 214, fig. 392; Is. Taki, 1960:197, pl. 90, fig. 2; LELoup, 1961:39-42, text figs. 1, 2, 5a, pl. 3, fig. 1; Is. Tax1, 1962:46; Iw. TakI, 1964b:412; VAN BELLE, 1982:473- 474; INABA, 1982:32. [Non: NIERSTRASZ, 1905b:155-156, pl. 10, figs. 23-25]. AS Je Ferreira, 1986 Liolophura japonica tessellata PILSBRY, 1893¢:243-244, pl. 53, figs. 45-46. [Non: Horst & SCHEPMAN, 1908:528]. Liolophura japonica tenuispinosa LELOUP, 1939a:1-3, figs. 1, 3, 4; 1952:59; VAN BELLE, 1980:33-35. Liolophura japonica unispinosa Is. TAKI, 1962:46 (nomen nu- dum). Liolophura japonica planispinosa Is. TAKI, 1962:46 (nomen nudum). “Liolophura gaimardi (Blainville)” Wu, 1969:109, figs. 5a, 5b, 47-58. Chiton defilippi TAPPARONE-CANEFRI, 1874:77; PILSBRY, 1893c:243-244 (as syn. of Liolophura japonica). Nuttallina allantophora DALL, 1919:502; SMITH, 1961:82, NOT253: [Non: “Nuttallina sp. cf. allantophora Dall, 1919,” STEINBECK & RICKETTS, 1941:555, pl. 26, fig. 6 (=Nuttallina crossota Berry, 1956).] Type material and type locality: Chiton japonicus Lischke, 1873: Types unascertained; local- ity Nagasaki, Japan (32°48’N, 129°55’E). Liolophura japonica tessellata Pilsbry, 1893c: Lectotype (ANSP 35969) and 2 paralectotypes (ANSP 355873) herein designated; locality Enoshima, Japan (35°18'N, 13922901) Liolophura japonica tenuispinosa Leloup, 1939a: Types un- ascertained; locality here restricted to Poulo Dama Is., Gulf of Thailand (9°40'N, 104°30’E). Chiton defilippi Tapparone-Canefri, 1874: Types unascer- tained; locality Japan. Nuttallina allantophora Dall, 1919: Holotype (USNM 110360a); locality ? Japan (given in error as “Los An- imas Bay,” Baja California, Mexico). Liolophura gavmardi platispinosa Leloup, 1939a: Types un- ascertained; locality Shikok Kamigari, Toso Pref., Ja- pan. Material examined: HONSHU: Japan, Shiriya, Aomori Pref., 2 specimens, largest 48 mm long (LACM 17-82); Takojima, Ishikawa Pref., 1 spec- imen, 25 mm long (LACM 11-82); Sagami Bay, 21 specimens, largest 37 mm long (CAS 009871; CAS 015103; CAS 12390; CAS 12394; CAS 012395; CAS 012400; CAS 031641; NMV 1026); Awaji, 1 specimen (NMV); Toshijima, Mie Pref., 1 spec- imen, 48 mm long (UCM 28917); Cape Bansho-zaki, near Seto Marine Biological Station, Wakayama Pref., 20 specimens, larg- est 47 mm long (LACM 19-82); Wakayama Pref., 2 specimens (ex Iw. Taki coll.); Ise Wan, 4 specimens, largest 52 mm long (CAS, acc. no. 1658); Shiju-shima Id., Hiroshima Pref., 2 spec- imens, largest 62 mm long (ex K. Y. Arakawa coll.). KYUSHU: North Kyushu, 8 specimens, largest 38 mm long (CAS 034180); Taujushima Id., near Amakusa Marine Biolog- ical Station, Kumamoto Pref., 4 specimens, largest 60 mm long (LACM 25-82); Amakusa Marine Biological Station, Tamioka Peninsula, 4 specimens, largest 51 mm long (LACM 26-82); Hana, 1 specimen, 18 mm long (CAS 030933); Moji, S of Ka- bura Shima, 28 specimens, largest 50 mm long (CAS 012370; CAS 012399); Nagasaki, 9 specimens, largest 30 mm long (CAS 012364; CAS 012362; CAS 12362). SOUTH KOREA: Pusan, 48 specimens, largest 32 mm long (CAS 012396; CAS 034552; CAS 053347); Dadas Po Beach, 15 specimens, largest 35 mm long (CAS 053345); Tanang Mal, 4 specimens, largest 30 mm long (CAS 053346); Seogwipo, 18 Page 251 specimens, largest 30 mm long (CAS 012391); Chesudo, 1 spec- imen, 20 mm long (CAS 000973). HONG KONG: Cheung-Sha, Lantau Id., 1 specimen, 33 mm long (AJF 685); Repulse Bay, Hong Kong Id., 3 specimens, largest 37 mm long (AJF 686). TAIWAN: Yehliu, Chilung Co., 1 specimen, 57 mm (UCM 27560); Chien-shaw, Pen-hu Co., 3 specimens, largest 23 mm long (UCM 28893). THAILAND): Sattahip, Gulf of Thailand, 4 specimens, largest 80 mm long (CAS 012397; CAS 030931); Ko-I-lao, 3 specimens, 70 mm long (CAS 012389; CAS 012398); Ko-Phai, 3 specimens, largest 50 mm long (CAS 012358); Ko-Sichang, 14 specimens, largest 55 mm long (CAS 012393; CAS 012403; CAS 012404). Description: LISCHKE (1873) described Chiton japonicus as “Testa ovata, parum convexa, atro-fusca, griseo strigata, minutissime rugulosa, granulis parvis concentrice, ad latera interdum radiatim ordinatis sculpta; areae laterales indis- tinctae; valva postica perbrevis, planata, acutimarginata mar- go incertus; valvae anticae, incisurisque profundoribus 8 ad 10 wrregulariter dwisus, valvae posticae integer, valuarum re- liquarum minute crenulatus et incisura unica bipartitus; l- gamentum spinis calcareis erectis, obtusis, griseis, fuscis et fulvis densissime obtectum; pagina valvarum fusca, paene ni- gra.—Long. 35, lat. 27 mill.” (pp. 22-23), distinguishing it carefully from the similar Chiton spiniger Sowerby (=A. gemmata). Among 175 specimens of Acanthopleura japonica here examined, largest 80 mm long (in alcohol) (CAS 030931: Sattahip, Gulf of Thailand). Body width/length, mean 0.64 (SD = 0.03; n = 10). Tegmentum dark brown. Lat- eral areas poorly defined, not elevated, sculptured with ill-defined, small granules. Central areas sculptureless ex- cept for subdued transverse growth lines. Posterior valve rather flat; mucro poorly defined, posterior to near ter- minal. Ocelli round to oval, 40-60 um in diameter, scat- tered through anterior valve, anterior 3 of lateral areas, and postmucro area (Figure 68). Gills with 40-50 plumes per side. Articulamentum dark brown. Sutural laminae well de- veloped, triangular on valve 11, becoming subrectangular on valve vill. Sinus well defined. Insertion plates pectinate on outside. Posterior valve with no teeth but wide, trans- verse, flat-surfaced callus (Figure 69); occasionally (par- ticularly in specimens from the Gulf of Thailand), a few, coarse pectinations or teeth seen at outer part of callus (one specimen shows large single tooth in middle of cal- lus). Slit formula 8/11-1-0. Girdle upper surface often banded brown-white, cov- ered with erect spinelets, all about same size and appear- ance, usually cylindrical, blunt-ended, white-tipped, up to 700 um long (Figure 70); in some populations, specimens have variable girdle elements, from “‘typically” slim and cylindrical spinelets, to flattened, scale-like, imbricate ele- ments (Figure 71). Girdle bridges empty. Undersurface paved with transparent, rectangular to squarish scales, 50 x 40 um, arranged in rows, with convex outer edge articulating with concave inner edge of adjacent scale. Radula averaging 35% of specimen length and 55 rows Page 252 The Veliger, Vol. 28, No. 3 Explanation of Figures 68, 69, and 73 to 76 Figure 68. Acanthopleura japonica (Lischke, 1873). Takojima, Ishikawa Pref., Japan Sea, Japan (LACM 82-11); specimen 18 mm long. Figure 69. Acanthopleura japonica (Lischke, 1873). Shiriya. Aomori Pref., Japan (LACM 82-17); specimen 25 mm long. Ventral aspect of posterior valve. Figure 73. Acanthopleura hirtosa (Blainville, 1825). Bremer Bay, W.A., Australia (WAM 47-74); specimen 25 mm long. Close- up of lateral areas of valves ili-iv and girdle. of mature teeth. In a specimen 38 mm long (CAS 012399: Unose Hana, Kyushu, Japan), radula 22 mm in length, comprising 60 rows of mature teeth. Median tooth (Fig- ure 72) 110 wm at anterior blade; first lateral teeth 170 wm wide at anterior blade; head of second lateral teeth Figure 74. Acanthopleura hirtosa (Blainville, 1825). Cockburn, W.A., Australia (WAM 201-74); specimen 40 mm long. Dorsal aspect of posterior valve. Figure 75. Acanthopleura hirtosa (Blainville, 1825). Same speci- men as in Figure 74. Posterior aspect of posterior valve. Figure 76. Acanthopleura hirtosa (Blainville, 1825). Same speci- men as in Figure 74. Ventral aspect of posterior valve. discoid, 300 um in diameter; outer marginal teeth 250 um long, 230 um wide (length/width, 1.1). Distribution: Acanthopleura japonica has been recorded at Hakodate, Hokkaido, Japan (41°45’N, 140°43’E) (Is. em eeerreinas, 1986 (eee ey mm Figure 70 Acanthopleura japonica (Lischke, 1873). Thailand (CAS 018358); specimen 41 mm long. Girdle spines. TAKI, 1938a; INABA, 1982), its northernmost record, along the coasts of Japan on the Sea of Japan, Inland Sea and Pacific Ocean, to Kyushu, the southern coast of Korea and Cheju-do (Is. TAk1, 1938a, 1962), Taiwan (Wu, 1969, as “Liolophura gaimardi”’), and Hong Kong (22°15’N, 114°10’E) (VAN BELLE, 1980, 1982; Ferreira, herein). A seemingly disjunct, perhaps relict population is present in the Gulf of Thailand (12°N, 102°E), with records at Poulo Dama Is., Poulo Condor (=Con Son Is.), Cap Saint- Jacques (=Vung-tau), Vietnam (LELOuP, 1939a, as Lio- lophura japonica platispinosa), and Sattahip, Thailand (Ferreira, herein) (Figure 115-J). The report of the species in the Moluccas (NIERSTRASZ, 1905b:155-156, pl. 10, figs. 23-25) is, from the illustrations, in error. Bathymetric range confined to the intertidal zone, on top of rocks often exposed at low tide. Remarks: PILsBry (1893c), segregated Liolophura japon- ica tessellata from the “typical japonica” on account of its “much narrower [girdle] ... conspicuously varied with alternate patches of white and scorched-brown or blackish . spinelets [which] are larger ... [and] vary much in size, being small toward the outer edge of the girdle, large and flattened toward the inner edge ...” (p. 243). Type material of L. 7. tessellata (ANSP 35060) consists of 3 dry specimens, curled, soft parts removed, estimated lengths 35-38 mm, eroded; articulamentum black; insertion plate of posterior valves inaccessible for inspection; girdle spine- Figure 71 Acanthopleura japonica (Lischke, 1873). Same specimen as in Figure 69. Girdle scale-like spines. Page 253 n 4 400 um Figure 72 Acanthopleura japonica (Lischke, 1873). Same specimen as in Figure 69. Radula median and first lateral teeth. lets as described and illustrated by PILsBRY (1893c). An old museum label on cardboard marked with a red dot to which the specimen had been glued, reads “Type of var. 35969 / L. Japonica Lischke. / var. tessellata Pils. / Fr. Stearns. Enoshima, Japan.” The least eroded specimen is here designated lectotype (ANSP 35969); the other two as paralectotypes (ANSP 355873). As observed by TakI (1938a:402), Liolophura japonica exhibits such wide variations in size, color, and width of girdle spinelets as to render meaningless alleged subspe- cies based on such variations (yet, TAKI [1962] later in- troduced two such subspecies, unispinosa and planispinosa, left undescribed). Still, attention must be called to two other phenotypes. Specimens from the Gulf of Thailand differ from those of Japan not only by attaining conspic- uously larger sizes, but in the girdle spinelets, which tend to be thinner, clearly cylindrical, closer together, equal- sized, and often white-tipped. Some specimens from Ja- pan (LACM 7-82: Akasaki, Noto Peninsula [37°21.5’N, 137°15'E]; LACM 25-82: Tsujushima Id., Kumamoto [32°33'N, 130°07.7’E]) and from Korea have girdle ele- ments that, being wide, flat, and clearly imbricate, are more properly called scales than spinelets (Figure 71), suggesting a distinct species (perhaps deserving a name such as “‘planispinosa,” left nudum by TAKI [1938a]) were it not for the presence of spinelets with an intermediate form, and the fact that such specimens do not appear to differ from those of “typical” Acanthopleura japonica in any other respect. Further study of these populations is indicated. Ornithochiton [sic] caliginosus Carpenter (7m PILSBRY, 1893c:243-244, pl. 54, figs. 41-45), based upon specimens from the China Sea and Hong Kong, has been regarded by Kaas & VAN BELLE (1980) as a synonym of Acantho- pleura japonica. However, as TAKI (1938a:402) pointed out, the specimen illustrated by PILSBRY (1893c) differs from A. japonica in the insertion plate of the posterior valve (specimen of A. loochooana?). The species-name, un- questionably referring to an Acanthopleura, is here sup- pressed as a nomen dubium. The placement of Chiton defilippi in the synonymy of Page 254 Liolophura japonica proposed by PILsBRY (1893c) on sub- jective grounds is here accepted. Nuttallina allantophora Dall, 1919, was shown to belong in the synonymy of Liolophura japonica, the type locality, “Los Animas Bay,” Baja California, Mexico, being in error (SMITH, 1977). Examination of color slides of the holotype (CASIZ Color Slide Series Nos. 1992, 1993) substantiates Smith’s conclusion. LELOup’s (1939a) assignment of specimens from Japan and the Gulf of Tonkin to Liolophura gaimardi platispinosa and specimens from Indochina and the Gulf of Thailand to L. japonica tenuispinosa was left unjustified. The true nature of the former (see Remarks on Acanthopleura gai- mardi) remains a matter of speculation. In Hong Kong, Acanthopleura japonica is found sharing the intertidal habitat with A. loochooana on rocks exposed to mild surf (personal observations, Sept. 1982). In Tai- wan, A. japonica is sympatric (or at least parapatric) with three other species of Acanthopleura, A. spinosa, A. loo- chooana, and A. miles. Acanthopleura hirtosa (Blainville, 1825) Figures 73 to 78, and 114-H Chiton hirtosus BLAINVILLE, 1825:546; PILSBRY, 1894a:106; Lamy, 1923:263. Acanthopleura (Liolophura) hirtosa: DuPuls, 1917:533-534. Liolophura hirtosa: Dupuis, 1918:531; AsHBy, 1926:384; LELouP, 1961:44-49, text figs. 4, 5c. Liolophura (Chiton) hirtosus: ASHBY, 1922b:579-580. Clavarizona hirtosa: HULL, 1923:199, pl. 28, figs. 9-12; IRE- DALE & HULL, 1926:261-262, pl. 37, figs. 9-12, 17, 21 (reprinted, 1927:124-125, pl. 15, figs. 9-12, 17, 21). Liolophura (Clavarizona) hirtosa: THIELE, 1929:21; VAN BELLE, 1983:129-130. Chiton georgianus Quoy & GAIMARD, 1835:379, pl. 75, figs. 25-30; IREDALE, 1910b:154. Liolophura georgiana: PILSBRY, 1893¢:241-242, pl. 53, figs. 36-40; Torr, 1911:100-101; AsHBy, 1921:45, 1922a: 32: Acanthopleura (Liolophura) georgiana: THIELE, 1911a:399- 400, fig. 3; Duputs, 1918:533-534. Plaxiphora pustulosa TORR, 1911:107, pl. 25, fig. 7. Type material and type locality: Chiton hirtosus Blainville, 1825: Type at MHNH (fide Dupuis, 1917; AsHBy, 1922c; Lamy, 1923); locality “mers de Pile King” (in error) corrected to King George Sound, Western Australia (35°03’S, 117°57’E) (AsHBy, 1922c; IREDALE & HULL, 1926). Chiton georgianus Quoy & Gaimard, 1835: Syntypes (4) at MNHN (fide AsuBy, 1922c); locality “port du Roi- Georges,” Western Australia (35°03'S, 117°57’E). Plaxiphora pustulosa Torr, 1911: Type “‘in coll. Torr” (fide IREDALE & HULL, 1926); locality Albany, Western Australia (35°02’S, 117°53’E). Material examined: AUSTRALIA, W.A.: Cape Cuvier, 1 specimen, 20 mm long (WAM 48-74); Point Quobba, 2 specimens, largest 35 mm long (WAM 43-74); Point Gregory, Peron Peninsula, Shark Bay, 3 The Veliger, Vol. 28, No. 3 specimens, largest 50 mm long (WAM N4726); Shark Bay, 1 specimen, 30 mm long (WAM 714-79); Dick Hartog Id., 1 specimen disarticulated (NMV, ex Ashby coll.); Kalbarri, 1 specimen, 35 mm long (WAM 1864-67, WAM-USNM Barrow Is. Exped., 18 August 1966); Abrolhos Is., 1 specimen, 40 mm long (AMS C31530); Geraldton, 1 specimen 30 mm long (WAM 7125); Dongara, 10 specimens, largest 30 mm long (WAM 6021); “W. Austr.,” 2 specimens (NMV); Carnac Is., 3 specimens, largest 55 mm long (WAM 50-74; WAM 704-79); Leighton, 2 specimens, largest 40 mm long (WAM 13446/7); Cockburn Sound, 6 specimens, largest 60 mm long (WAM 200-74; WAM 201-74; WAM 476-74); Port Gregory, 1 specimen, 40 mm long (WAM 46-74); Fremantle District, 1 specimen, 20 mm long (AMS C31531); Fitzgerald Inlet, 2 specimens (WAM 42-74); Point Peron, 4 specimens, largest 60 mm long (WAM 45-74; WAM 278/9-1938); Bunbury, 1 specimen, 20 mm long (WAM 49-74); Lucky Bay, 1 specimen, 50 mm long (WAM 710-79); Geographe Bay, 3 specimens, largest 30 mm long (AMS C18012); Cottesloe, 1 specimen, 24 mm long (NMV); Bunker Bay, 1 specimen, 40 mm long (WAM 713-79); Rottnest Id., 1 specimen 35 mm long (AMS C32119); Foul Bay, 1 specimen (AMS C121194); Garden Id., 2 specimens, largest 45 mm long (WAM 706-79, with label “ident. as Liolophura gaimardi by E. Ashby’’); Coweramup Bay, 1 specimen, 40 mm long (WAM 705-79); Augusta, 1 specimen, 50 mm long (WAM 708-79); Cape Leeu- win, 1 specimen disarticulated (NMV, ex Gatliff coll.); Nor- nalup, 2 specimens, largest 50 mm long (WAM 15568/69, leg. E. Ashby, August 1929); King George Sound, 8 specimens, larg- est 56 mm long (NMV, ex Basset Hull coll.); King George Sound, 5 specimens (NMV); Middleton Beach, King George Sound, 1 specimen, 42 mm long (AMS C69338); Albany, 4 specimens, largest 40 mm long (WAM N3272); Frenchman’s Bay, 1 specimen, 25 mm long (WAM 707-79); Bremer Bay, 4 specimens, largest 40 mm long (WAM 47-74); Hopetown, 2 specimens, largest 17 mm long (WAM 495-74); Esperance, 1 specimen, 37 mm long (WAM 51-74); Seven Mile Beach, Es- perance, 6 specimens, largest 40 mm long (NMV); Dempster Head, Esperance (NMV, ex Basset Hull coll.); Duke of Orleans Bay, 1 specimen, 40 mm long (WAM 709-79); Wilson Id., 1 specimen, 40 mm long (WAM); Mondrain Id., 5 specimens, largest 70 mm long (WAM 44-74; WAM 712-79). Description: Acanthopleura hirtosa is adequately charac- terized in the descriptions of BLAINVILLE (1825), PILSBRY (1893c), and IREDALE & HULL (1926). Among 96 specimens of Acanthopleura hirtosa here ex- amined, largest 70 mm long (dry) (WAM 44-74: Mon- drain Id.). Body width/length, mean 0.62. Tegmentum olive-green to brown, often with black patches and whitish band in jugal area. Anterior valve covered with round granules which tend to align in radial rows. Lateral areas of intermediate valves moderately elevated, with some- what elongate granules in about 10-15 ill-defined radial rows (Figure 73). Central areas with transverse, juxta- posed lamellae on jugal area superimposed by round gran- ules in parallel longitudinal rows, particularly well de- fined on pleural areas. Intermediate valves beaked; posterior edge of valve ii forming 100-110° angle. Mucro posterior to terminal; postmucro convex, sharply sloped. Ocelli round to oval, 20-50 um in diameter, throughout anterior valve, postmucro area of posterior valve, and an- terior % of lateral areas of intermediate valves. Widths of tegmental surfaces of valves i/vill, mean 1.17. On valve i, ene ierneinal9 86 ee imm Figure 77 Acanthopleura hirtosa (Blainville, 1825). Bayonet Head, Albany, W.A., Australia (WAM N3272); specimen 32 mm long. Girdle scales. tegmentum length/width, mean 0.51. Posterior valve somewhat triangular in outline (Figure 74); tegmentum length/width, mean ratio 0.38; articulamentum not wider than tegmentum. Eaves thick (0.5 mm in midline of valve viii of specimen 32 mm long), somewhat spongy. Gills with 25-35 plumes per side. Articulamentum dark brown and white. Sutural lami- nae well developed, subtriangular on valve ii to subrec- tangular on valve viii. Insertion plates pectinate on out- side. On valve i, insertion plate hardly extends beyond tegmental surface (measured in midline, length of inser- tion plate/length of tegmental surface, mean ratio, 0.10). On valve viii, insertion plate absent but with conspicuous, toothless, flat, wide callus (Figures 75, 76). Slit formula 8/11-1-0. Girdle banded black-white in most specimens; at valve iv, girdle width up to 36% of width of valve in alcohol preserved specimens. Upper surface covered with calcar- eous, somewhat conical scales (Figure 77), irregular in size, averaging 400 wm in length, 500 um in height, 200 um in thickness, with 8-12 fine, converging striae; clusters of 3-8 relatively opaque spicules 100-200 um long, 20- 30 um thick, may be seen amidst scales. Girdle bridges empty. Undersurface covered with transparent, subquad- rangular scales, 60 x 50 wm, vaguely striate. In specimen 40 mm long (WAM 201-74: Cockburn, SW Australia), radula measures 22 mm in length (55% of specimen length), comprising 68 rows of mature teeth. Median tooth (Figure 78) 100 um wide at anterior blade; first lateral teeth 150 um wide at anterior blade; head of second lateral teeth discoid, 260 wm in diameter; outer marginal teeth 270 um long, 190 um wide (length/width, 1.4). Distribution: Acanthopleura hirtosa is confined to south- western Australia (Figure 114-H) from Cape Cuvier (WAM 48-74) (24°14'S, 113°22’E), the northernmost ver- ified record, to Mondrain Id., Recherche Arch. (WAM 44-74, WAM 712-19) (34°08’S, 122°15’E), the eastern- most verified record on the south coast of Australia. HULL (1923) recorded the species from Point Cloates to Eyre. Bathymetric range 0-2 m. Remarks: Acanthopleura hirtosa is the only species of the genus in southwestern Australia. It is remarkably similar Page 255 L 3 200um Figure 78 Acanthopleura hirtosa (Blainville, 1825). Same specimen as in Figure 77. Radula median and first lateral teeth. to A. gaimardi from which it mainly differs in the girdle elements—conical scales in A. hirtosa, spinelets in A. gai- mardi. HULL (1923) separated the two species at the ge- neric level, erecting Clavarizona for hirtosa, but his action was refuted by AsHBy (1926) and LELOuP (1961). SMITH (1960:67) placed Clavarizona in the synonymy of Liolo- phura. Acanthopleura arenosa Ferreira, spec. nov. Figures 79 to 85, and 114-A Diagnosis: Specimens quite similar to those of Acantho- pleura gaimardi except for a moderately inflated (z.e., not as flat) posterior valve, with a rounder posterior edge, and insertion teeth definitely present although underdeveloped to obsolete. Type material: Holotype (CAS 044305) and 11 paratypes (CAS 044304; BMNH 1985065; USNM 848001; ANSP A10648; LACM 2107; AJF coll.). Type locality: Pebbly Beach (about 25 km south of Port Douglas), Queensland, Australia (16°35’S, 145°22’E), 0.5 m be- low to 1 m above low tide water (AJF 604, leg. A. J. Ferreira & Sandy Motley, 14 Aug. 1981). Other material: AUSTRALIA, Qld.: Buchan’s Point (30 km N of Cairns), 0 to 1 m above low tide water, 4 specimens, largest 25 mm long (AJF 605); Trinity Beach (some 15 km N of Cairns), 0 to 1 m above low tide water, 4 specimens, largest 28 mm long (AJF 606). Description: Holotype (Figures 79, 80) preserved in al- cohol, intact (except for disarticulated posterior valve), somewhat curled, 22 m long (if flattened), 14 mm wide (including girdle); dark brown on side, and well defined jugal band, whitish otherwise; valves somewhat beaked, round-backed. Tegmentum eroded at apex of valves; an- The Veliger, Vole 28 3Noms Page 256 ae henreinas 1986 Figure 84 Acanthopleura arenosa Ferreira, spec. nov. Paratype (AJF coll.), 24 mm long. Girdle spinelets, outer and inner sides. terior valve with roundish, poorly defined granules; lateral areas slightly elevated, with similar granules; central areas with weak, transversely appressed lamellae; posterior valve with tegmental surface semicircular, 7.5 mm wide, 3.6 mm long, with mucro (eroded) slightly post-central; gills holobranchial. Articulamentum of disarticulated posterior valve dark brown in middle, white at sutural laminae; sutural laminae subrectangular, about 2.8 mm wide, 1.5 mm long; sinus well defined, smooth, about 1.5 mm wide at anterior edge of tegmental surface; insertion plate re- duced to transverse, roundish callus, pectinations clearly cut in outer % but progressively ill-defined to absent in middle 43, showing symmetrical slit on each side; eaves solid. Girdle banded black and white, crowded with point- ed but short spinelets. Paratypes (Figures 81-83) dark brown, most with brown jugal band flanked by white areas; largest 30 mm long (live). Body width/length, 0.65. Valves beaked, round- backed. Anterior valve with concentric rows of small, ir- regular, round to ovate granules 100-200 um in diameter. Lateral areas poorly defined, hardly raised, similarly sculptured. Central areas (markedly eroded) featureless except for ill-defined, appressed, transverse lamellae. Pos- terior valve somewhat inflated (7.e., not flat), with semi- circular posterior edge; mucro not prominent, central to slightly posterior; postmucro convex, sloping. On valve i, tegmentum length/width, 0.5. On valve viii, tegmentum length/width, 0.4. Widths of tegmental surfaces of valves Page 257 Figure 85 Acanthopleura arenosa Ferreira, spec. nov. Same paratype as in Figure 84. Radula median and first lateral teeth. i/vili, 1.1. Ocelli round to oval, 50-60 wm in diameter. Gills with 30-35 plumes per side. Articulamentum brown to bluish-white. Sutural lami- nae well developed, subtriangular to subrectangular; sinus well formed; sinusal plate smooth; relative width of sinus, 0.6. Insertion plates strongly pectinate on outside. On valve i, insertion teeth irregularly spaced; in midline, length of insertion teeth/length of tegmentum, 0.18. On valve viii, pectinations and insertion teeth extremely subdued, ves- tigial to absent in middle %; single, well cut, symmetrical slits on each outer %. Slit formula 9/10-1-2. Eaves thick (0.5 mm on midline of valve viii), somewhat spongy. Girdle thick, musculous, banded blackish-brown and white. Upper surface with calcareous spinelets (Figure 84), somewhat pointed, straight to slightly curved, conical in outline, up to 1000 um high, 300 um thick at base. Undersurface with transparent scales, vaguely striate, squarish (40 x 35 wm) to elongate (60 x 30 wm) at pe- riphery. In paratype (AJF coll.) 24 mm long, radula 9 mm long (45% of specimen length), comprising 45 rows of mature teeth. Median tooth (Figure 85) 70 um wide at anterior blade; first lateral teeth 120 wm wide at anterior blade; Explanation of Figures 79 to 83 and 86 to 88 Figure 79. Acanthopleura arenosa Ferreira, spec. nov. Holotype (CAS 044305). Figure 80. Acanthopleura arenosa Ferreira, spec. nov. Holotype (CAS 044305). Close-up of lateral areas of valves iii-v and gir- dle. Figure 81. Acanthopleura arenosa Ferreira, spec. nov. Paratype (AJF coll.), 20 mm long. Dorsal aspect of posterior valve. Figure 82. Acanthopleura arenosa Ferreira, spec. nov. Same paratype as in Figure 81. Posterior aspect of posterior valve. Figure 83. Acanthopleura arenosa Ferreira, spec. nov. Same paratype as in Figure 81. Ventral aspect of posterior valve. Figure 86. Acanthopleura rehderi Ferreira, spec. nov. Holotype (USNM 842113). Figure 87. Acanthopleura rehderi Ferreira, spec. nov. Paratype (USNM 842114). Dorsal aspect of posterior valve. Figure 88. Acanthopleura rehderi Ferreira, spec. nov. Same paratype as in Figure 87. Ventral aspect of posterior valve. Page 258 major lateral teeth with discoid head, 180 wm wide; outer marginal teeth 150 um long, 140 um wide (length/width, ey)? Other material essentially as type material. Distribution: Acanthopleura arenosa is known only from the 80 km of coast between Peebly Beach and Trinity Beach, Queensland, Australia (Figure 114-A). Specimens are confined to the intertidal zone, exposed on rocks 0-1 m above low-tide water level. Remarks: Acanthopleura arenosa is extremely similar to A. gemmata and A. gaimardi; in fact, only the posterior valve shows reliable distinctions in configuration and par- ticularly in the characteristics of its insertion plate. The posterior valve of A. arenosa differs from that of A. gai- mardi in being rounder (1.e., not subtriangular), somewhat inflated (z.e., not as flat), with central to moderately pos- terior (7.e., not as terminal) mucro, and in the presence of insertion teeth and pectinations (completely absent in A. gaimardi) at the outer % of the insertion plate; and it differs from that of A. gemmata in the underdevelopment (to obsolescence) of the insertion teeth. Otherwise, speci- mens of A. arenosa conform well with what gemmata- gaimardi hybrids might be expected to be. Since, in ad- dition, specimens of arenosa have been found exclusively in the zone of contact and overlap between A. gemmata and A. gaimardi, the possibility that they might be part of a hybrid population must be considered. Hybridization in mollusks has been known in a few instances—in the prosobranch gastropods C'ypraea (SCHIL- DER, 1962) and Haliotis (OWEN, 1961; OWEN et al., 1971), the pulmonate gastropod Cerion (MAYR & ROSEN, 1956), and the pelecypods Ostrea (Crassostrea) (DAvis, 1950; IMaI & SAKAI, 1961), Pinctada (MatTsul, 1958), Mercenaria (MENZEL, 1962; MENZEL & MENZEL, 1965), and Tellina (Boss, 1964)—but not in chitons. Thus, unless further investigation should show otherwise, it seems appropriate to treat this population as a new species, which, judging from the abundance and relative uniformity of its speci- mens at the localities studied, is viable and self-reproduc- tive. The species is here named arenosa for the sandy beach- es where it was found, and after Cecily “Sandy” Motley, Davis, California, who assisted in the collecting. Acanthopleura rehderi Ferreira, spec. nov. Figures 86 to 92, and 113-R Diagnosis: Specimens small (to 2 cm) for the genus. An- terior valves with radial rows of round tubercles alternat- ing with rows of ocelli. Lateral areas similarly sculptured. Central areas with well defined, parallel, longitudinal rib- lets. Posterior valve flat, subtriangular, mucro posterior to terminal. Slit formula 8/9-1-0. Girdle with spinelets. Radula with 4-cuspid major lateral teeth. The Veliger, Vol. 28, No. 3 Type material: Holotype (USNM 842113) and paratypes (USNM 842114; CAS 060405). Type locality: Palmerston Id., Cook Islands (18°04’S, 163°10'W). Other material: NIUE: Avatolo, 1 specimen, ca. 22 mm long, leg. R. Sixberry (USNM 685399); Alofi, 2 specimens, largest ca. 18 mm long, leg. R. Sixberry (USNM 685343). Description: Holotype (Fig. 86), well preserved, dry, moderately curled, estimated 18 mm long (if flattened), 10 mm wide (including girdle), 4 mm high. Valves sub- carinate, posterior edge beaked and angled at about 150°. Tegmentum (somewhat eroded) light tan with grayish- green suffusions towards margin and brown stripe along jugal area. Anterior valve with about 24 radial rows of ground granules. Lateral areas of intermediate valves raised, with 3 or 4 similar rows of round granules better defined at sutural edge, rendering it crenulate. Central areas with 18-20 longitudinal, parallel riblets, close to- gether, as wide as interstices in between. Posterior valve rather flat, subtriangular; mucro posterior, almost termi- nal, somewhat pointed. Ocelli mostly oval, 50-60 um maximum diameter, aligned in radial rows on anterior valve and most of lateral areas. Gills with 35-40 plumes per side. Girdle upper surface covered with white, calcareous spinelets, close together, rather uniform in size, up to 700 um long, 130 wm thick. Paratypes quite similar to holotype. Disarticulated paratype (Figures 87, 88) ca. 20 mm long: Articulamen- tum white with brown discolorations at apex of valves; tegmentum reflected forward along posterior edge of valves; width of valve i/width of valve viii, 6.6 mm/6.2 mm = 1.1; sinus well defined, wide, deep, with pectinate sinusal plate; sutural laminae subrectangular; insertion plates pectinate on outside; slit formula 8-1-0; posterior valve with no teeth but well developed callus. Girdle upper surface with regular, straight, gently tapered, calcareous spinelets (Figure 89), mostly white, up to 900 um long, 140 um thick, with transverse “growth” striations and relatively abundant, glassy, sharply pointed spicules, up to 200 xX 15 um, scattered in between. Undersurface cov- ered with transparent, rectangular, coarsely striate scales (Figure 90), about 40 x 30 um, each with convex outer side that articulates into concave inner side of adjacent scale. Radula 5 mm long (25% of specimen length), com- prising 45 rows of mature teeth; median teeth (Figure 91) elongate, 40 um wide at anterior blade; first lateral teeth with elongate, bladed, anterolateral corner; second lateral teeth with basically discoid head, 140 um wide, but with four short, rounded cusps (unique feature among Acan- thopleura); spatulate teeth (Figure 92) with subrectangu- lar spatula; outer marginal teeth elongate, 125 x 80 um. A. J. Ferreira, 1986 ee SOOM Figure 89 Acanthopleura rehderi Ferreira, spec. nov. Same paratype as in Figure 87. Girdle spinelets. Specimens from Niue (USNM 685399; USNM 685343) agree in every respect with the types except for the slit formula, 9-1-0, of an 18 mm long specimen. Distribution: The species is known only from Palmer- ston, Cook Islands (“outside reef nr. village,” leg. R. Six- berry), type locality, and Niue (19°02’S, 169°52'W) (Fig- ure 113-R), presumably in the intertidal zone. Remarks: The specimens here referred to Acanthopleura rehderi are clearly distinct from those of A. nigropunctata Carpenter, 1865, described from the nearby Society Is- lands. Examination of the lectotype (USNM 19297) of the latter, here designated—an incomplete specimen miss- ing valves vii and viii but conforming well to CARPENTER’s (1865) account of nigropunctata, which indicated the pres- ence of slits in the posterior valve—corroborated its cur- rent generic assignment (since Carpenter im PILSBRY, 1893b:207) to Tonicza. Acanthopleura rehderi, like A. japonica, A. hirtosa, A. gaimardi, and A. nigra, has no insertion teeth in the pos- terior valve, and, as such, it would fit in Liolophura Pils- bry, 1893a. However, despite articulamental similarities, A. rehderi clearly differs from the other four ‘“Liolo- phura” in tegmental sculpture, girdle elements, and, above all, in its unique radula. The radula of Acanthopleura rehderi with 4-cuspid major lateral teeth constitutes a departure from all con- generics (and, to my knowledge, from all other chitonids), although the cusps seem to be based upon (or cut into) the discoid head characteristic of Acanthopleura. The pos- sible evolutionary significance of such a radular modifi- cation is unknown, but it is tempting to speculate that it may represent an adaptive response to a new set of dietary conditions, perhaps presaging a new line of speciation. The question deserves further investigation. Page 259 = 11504m Figure 90 Acanthopleura rehderi Ferreira, spec. nov. Same paratype as in Figure 87. Girdle undersurface scales. The species is here called rehderi after Dr. Harald A. Rehder, Professor Emeritus, National Museum of Nat- ural History (Smithsonian Institution), Washington, D.C., who generously provided the specimens for study. Acanthopleura granulata (Gmelin, 1791) Figures 93 to 97, and 113-G Chiton granulatus GMELIN, 1791:3205; Woop, 1815:9; BLAINVILLE, 1825:545; ORBIGNY, 1853:200. Acanthopleura granulata: HADDON, 1886:24-28; PILsBRY, 1893c:227-230, pl. 50, figs. 39-49 (in subgen. Mau- geria; DAUTZENBERG, 1900:220-221; DALL & Simpson, 1901:454; HAMILTON, 1903:138; NIERSTRASZ, 1905a: 102, 1905b:152 (in part); Horst & SCHEPMAN, 1908: 527 (in part); THIELE, 1909:3, 1910b:112; REMINGTON, 1922:121; BeRRy, 1925:173-175, pl. 12, figs. 1, 2; (?) PEILE, 1926:74; NIERSTRASZ, 1927:163; THIELE, 1929: 21; HUMMELINCK, 1933:303, 306; JOHNSON, 1934:14; LeELoup, 1937a:146-150, figs. 13-15a (in part), 1941: 44-45, pl. 1, fig. 1; SALISBURY, 1953:42; HIDALGO, 1956: 4-8, pls. 3, 4; OLSSON & McGinty, 1958:23; LEwIs, 1960:398, 410, fig. 8; WARMKE & ABBOTT, 1961:220, fig. 33f; CONDE, 1966:287; ALTENA, 1969:37; GLYNN, 1970:1-21; Kaas, 1972:117-122, text figs. 239-244, pl. 9, figs. 1-3; GOTTING, 1973:251, text fig. 2, pl. 11, fig. 14; ABBOTT, 1974:406, fig. 4755; BABOOLAL et al., 1981: 43, fig. 5; Mook, 1983:101-105. [Non: SUTER, 1905:70, 1913:44-45, 1915, pl. 2, fig. 21, pl. 5, fig. 2.] Chiton piceus GMELIN, 1791:3205; BLAINVILLE, 1825:545; SOwERBY, 1840b:1, 10, sp. no. 10, fig. 147; SAUSSAYE, 1853:416; SHUTTLEWORTH, 1853:78-79 (in subgen. Acanthopleura); SCHIFF, 1858:12-47, pls. 1-2. [Non: REEVE, 1847, pl. 13, sp. & fig. 70; ANGAS, 1867: 223.] Acanthopleura picea: MOSELEY, 1885:18, pl. 6, figs. 8, 9; DALL, 1889:174; THIELE, 1893:373, pl. 30, fig. 32. Chiton salamander SPENGLER, 1797:80-81. Acanthopleura salamander: THIELE, 1893:373, pl. 30, fig. 35. Page 260 The Veliger, Vol. 28, No. 3 Figure 91 Acanthopleura rehderi Ferreira, spec. nov. Same paratype as in Figure 87. Radula median tooth, first lateral teeth, and head of major lateral tooth. Chiton convexus BLAINVILLE, 1825:544. Chiton occidentalis REEVE, 1847, pl. 14, sp. & figs. 76; SAUSSAYE, 1853:416. Chiton (Acanthopleura) mucronulatus SHUTTLEWORTH, 1853: 79. Acanthopleura granulata mucronulata: DALL & SIMPSON, 1901: 454. Chiton (Acanthopleura) blauneri SHUTTLEWORTH, 1856:170- Ale Type material and type locality: Chiton granulatus Gmelin, 1791: Based upon CHEMNITZ, (1785:fig. 806); locality ‘““Oceano americano”’ (St. Thomas, West Indies, Caribbean Sea). Chiton piceus Gmelin, 1791: Based upon CHEMNITZ (1785: figs. 807, 810); locality “mari americano & rubro”’ (St. Thomas, West Indies, Caribbean Sea). Chiton salamander Spengler, 1797: Based upon CHEMNITZ (1785:fig. 806); locality St. Thomas, West Indies. Chiton convexus Blainville, 1825: Types unascertained; lo- cality “mers de l’archipel american.” lle 4 100,am Figure 92 Acanthopleura rehderi Ferreira, spec. nov. Same paratype as in Figure 87. Radula spatulate tooth. Chiton occidentalis Reeve, 1847: Types unascertained; local- ity ‘“Savannah-le-mer, West Indies.” Chiton (Acanthopleura) mucronulatus Shuttleworth, 1853: Types unascertained; locality Puerto Rico, West Indies. Chiton (Acanthopleura) blauneri Shuttleworth, 1856: Types unascertained; locality Puerto Rico, West Indies. Material examined: BAHAMAS: Grand Bahamas Id., 15 specimens (AJF coll., leg. A. J. Ferreira et al., May 1971); Bimini Id., 15 specimens, larg- est 86 mm long (AJF 290; AJF 291); Long Island, 10 specimens (AJF 248); San Salvador Id., 15 specimens (AJF 439); New Providence Id., 24 specimens (CAS 010094); Nassau, 4 speci- mens, largest 40 mm long (CAS 034777); Chub Cay, 2 speci- mens, largest 8 mm long (IRCZM 61:066); Gun Cay, 3 speci- mens, largest 65 mm long (IRCZM 61:016). FLORIDA KEYS: Bonefish Key, 18 specimens (AJF 426); Crawl Key, 2 specimens, largest 60 mm long (CAS 012259); between Windley and Plantation Keys, 4 specimens, largest 80 mm long (IRCZM 61:010). BONAIRE: 50 specimens (AJF 210; AJF 208; AJF 264). CURACAO: 20 specimens (AJF 260; AJF 263). PANAMA: Galeta, 2 specimens (AJF coll., /eg. H. Bertsch, Sept. 1974; IRCZM 61:013); Bocas del Toro, 10 specimens (AJF 216); Caledonia Bay, 5 specimens (LACM-AHF A 1-39). HONDURAS: Roatan Id., 24 specimens, largest 74 mm long (AJF 309; CAS 012137; CAS 012138; CAS 012148; CAS 012149; CAS 021294). NICARAGUA: Corn Id., 3 specimens (AJF coll., leg. B. Kea- gan, Sept. 1975). JAMAICA: Montego Bay, 20 specimens (AJF 253; AJF 254); Negril, 10 specimens (AJF 256). DOMINICAN REPUBLIC: Caracoles, 8 specimens (AJF coll., leg. B. Keagan, Jan.—Oct. 1976); Playas Bayabibe, 1 specimen (AJF coll., leg. B. Keagan, Sept. 1977). Aeala Herreinay 1986 BRITISH VIRGIN ISLANDS: Virgin Gorda Id., 1 specimen (AJF 297); Cooper Id., 4 specimens (AJF coll., leg. S. Motley, Feb. 1983); Peter Id., 3 specimens (AJF coll., leg. S. Motley, Feb. 1983). ANTIGUA: 14 specimens (CAS 012135; CAS 012136; CAS 012150). ST. LUCIA: Pigeon Id., 2 specimens (AJF coll., Jeg. B. Keagan, May 1977). DOMINICA: Anse de Mai, 3 specimens (AJF coll., leg. B. Keagan, May 1977). VENEZUELA: Puerto Mara, 2 specimens (AJF 347); Tortuga Id., 5 specimens, largest 102 mm long (LACM-AHF A20-39). CAYMAN ISLANDS: Grand Cayman Id., 34 specimens, larg- est 65 mm long (AJF 420; AJF 421; IRCZM 61:030; IRCZM 61:031); Cayman Brac, 5 specimens (AJF 424). TURK & CAICOS: Grand Turk Id., 20 specimens (AJF 443; AJF 444). MEXICO: Cozumel Id., 35 specimens (AJF 511; AJF 512; AJF 514). TOBAGO: Courland Point, 6 specimens (AJF 670); Bateau Bay, 2 specimens (AJF 672); Mt. Irvine Bay, 8 specimens (AJF 674); Store Bay, 12 specimens, largest 52 mm long (AJF 678). BARBADOS: Paradise Beach, 6 specimens (AJF 679); River Bay, 5 specimens (AJF 680); St. Lawrence, 1 specimen, 50 mm long (AJF 684); Bathsheba, 4 specimens, largest 39 mm long (CAS 012260). TRINIDAD: Maracas Beach, 3 specimens, largest 39 mm long (AJF 668). Description: Because there is only one Acanthopleura in the Caribbean, GMELIN’s (1791) description of Chiton granulatus in “Oceano Americano,” based upon a figure in CHEMNITZ (1758:fig. 806), has proved adequate to identify the species: “Ch. piceus supra planus, punctis ele- vatis numerosis in series digestis, limbo lato coriaceo spinoso; areis nigris albisque alternis.” The species has been re- peatedly described by authors working with the Carib- bean chiton fauna, and little remains to be added to such accounts (e.g., KAAS, 1972). Among 472 specimens of Acanthopleura granulata ex- amined, largest 102 mm long (in alcohol) (LACM-AHF A 20-39: Tortuga Id.,Venezuela). Body width/length, mean 0.64. Specimens (Figure 93) depressed, round- backed, large, beaked; posterior edge of valve i1 forming 100-120° angle. Tegmentum grayish-green to grayish- brown, often with dark longitudinal stripe in midline. Lateral areas poorly defined, hardly raised, sculptured with low, mostly round, coarse granules; anterior valve and postmucro area of posterior valve similarly sculp- tured. Central areas almost featureless except for smaller to obsolete granules in pleural areas, and thin, ill-defined, transverse lamellae appressed across jugal areas. Mucro central (in young specimens) to somewhat posterior (in older, larger ones); postmucro strongly convex, at 30—90° slope. Ocelli round to oval, 50-70 wm in diameter, throughout anterior valve, postmucro area of posterior valve, and anterior 4 to % of lateral areas of intermediate valves. On valve i, length of tegmentum/width of tegmen- tum, mean 0.6. On valve viii, articulamentum not wider than tegmentum; length of tegmentum/width of tegmen- Page 261 tum, mean 0.6. Widths of tegmental surfaces of valves i/viil, mean 1.2. Gills with 40-80 plumes per side. Articulamentum blue to blue-green, often with pur- plish-brown spot at apex of valves. Sutural laminae well developed, relatively long, subtriangular on valve ii to sub- rectangular on valve viii. Sinus well formed; sinusal plate mostly smooth; relative width of sinus on valve viii, 0.7. Insertion teeth irregularly spaced, sometimes fused to- gether; in midline, length of insertion plate/length of teg- mentum, mean 0.2. On valve viii (Figures 94, 95), pecti- nations of insertion plate variable, often with incomplete slitting and poor definition of teeth (usually better defined than in Acanthopleura gemmata); teeth recurving forward, extending considerably beyond (more so than in A. gem- mata) transverse callus. Slit formula 6/17-1-7/16. Eaves thick (0.5 mm wide on midline of valve viii of specimen 45 mm long), moderately spongy. Girdle, often banded, thick, musculous, wide, shrinking appreciably with preservation; at level of valve iv, girdle may measure 60% of valve in live specimens, 40% in al- cohol preserved specimens, 10% or less in dry specimens. Upper surface crowded with white or blackish, pointed to blunt, straight to curved, calcareous spinelets (Figure 96), up to 1.5 mm long in average specimens (up to 2.2 mm long in larger ones); occasional needle-like elements, pointed, crystalline, 200 x 30 wm, interspersed amidst spinelets. Girdle bridges empty. Undersurface paved with imbricated, transparent, squarish scales, about 40 x 40 um, becoming elongate towards outer margin, showing some 8-10 coarse striations and riblets that seem to ra- diate from outer edge of scale. Radulae averaging 43% of specimen length (range 38- 55%, SD = 8.4%, n=7) and 58 rows of mature teeth (range 40-70, SD = 9.3, n = 7). In specimen 43 mm long (AJF 248: Long Island, Bahamas), median tooth (Figure 97) 80 wm wide at anterior blade; first lateral teeth 220 um wide at anterior blade; head of major lateral teeth discoid, 280 wm wide; outer marginal teeth 250 um long, 180 um wide (length/width, 1.6). Distribution: Acanthopleura granulata is limited to the Caribbean Sea, having been recorded at practically every island or cay from the Florida Keys, Mexico, Central America coast to the Leeward Islands, from the Bahamas to the northern coast of South America (Figure 113-G). The northernmost verified record is Grand Bahama Id., Bahamas (26°40'N) (AJF coll., leg. A. J. Ferreira, May 1971); the report of A. granulata in Bermuda (PEILE, 1926) has not been corroborated in field work (A. J. Ferreira & W. E. Daily collecting trip to Bermuda, May 1977; Dr. John S. Pearse, personal communication upon field trip to Bermuda, July 1980) or museum material (Bermuda Aquarium, Natural History Museum, and Zoo, David D. Lonsdale, Curator: chiton collection on loan, Sept. 1979). The southernmost record is Trinidad (10°39'N) (BABOOLAL et al., 1981; Ferreira, herein); the westernmost record, Cozumel Id., Mexico (86°55’W) (HIDALGO, 1956; Page 262 ihe VeligersVoleZS Nome eae wermeinayy 19.86 ee a iimm Figure 96 Acanthopleura granulata (Gmelin, 1791). Barbados (AJF 609); specimen 35 mm long. Girdle spinelets. Ferreira, herein); the easternmost record, Barbados Id. (59°32'W) (THIELE, 1910b; Lewis, 1960; CONDE, 1966; Kaas, 1972; Ferreira, herein). Reports of A. granulata in the Magellan Strait (NIERSTRASZ, 1905a, b) and at the Cape of Good Hope (NIERSTRASZ, 1905b) are obviously in error. Acanthopleura granulata is confined to the intertidal zone, 0-1 m, often exposed, in crevices on coral limestone up to 1 m above low tide level. Remarks: Acanthopleura granulata is the only species of the genus in the Atlantic Ocean. Likely, A. granulata and A. gemmata stem from the same ancestral species separat- ed by the emergence of the Panama Isthmus. Still, A. granulata, a geographical isolate (MAyR, 1969), has not achieved sufficient phenotypic distance from A. gemmata to dispell the question of conspecificity. Although the question has not been previously addressed in the litera- ture—and PILsBry (1893c) went as far as allocating A. granulata and A. gemmata to different subgenera—the fact is that character-by-character comparison of Indo-Pacific specimens of A. gemmata with Caribbean specimens of A. granulata has failed to differentiate them in size, color, shape, tegmental sculpture, articulamental features, girdle elements, radula, and habitat. Page 263 U 1400 um Figure 97 Acanthopleura granulata (Gmelin, 1791). Long Island, Bahamas (AJF 248); specimen 45 mm long. Radula median and first lateral teeth. Admittedly, when compared with the Indo-Pacific Acanthopleura gemmata, a few, subtle, and variable char- acters do seem to earmark the Caribbean A. granulata: (1) the more regular and less often coalesced tegmental gran- ules, (2) the somewhat recticular appearance of the pleu- ral areas, (3) the less conspicuous lamellar sculpture of the central areas, (4) the brighter blue or bluish-green articulamentum, often with an apical dark brown spot, but never wholly brown, (5) the better defined teeth of the posterior valve, less recurved forward, extending farther beyond the callus, (6) the less well defined transverse cal- lus of the posterior valve, and (7) the girdle spinelets not quite as long as sometimes seen in A. gemmata. But, con- sidering the large intraspecific variation and wide geo- graphic range of the respective populations, these equiv- ocal distinctions can hardly be accepted as specific. Still, whether they are sibling species (sensu Mayr, 1969:183) or a single species cannot be decided on morphology alone and must await further and more sophisticated (electro- phoretic, immunologic, genetic, etc.) studies. Thus, given their total geographic separation, it is here recommended that the traditional view be accepted, and the Caribbean and Indo-Pacific populations continue to be addressed as different species. Chiton magellanicus GMELIN, 1791:3204, was based upon Explanation of Figures 93 to 95 and 98 to 102 Figure 93. Acanthopleura granulata (Gmelin, 1791). Caracoles, Dominican Republic (AJF coll.); specimen 15 mm long. Figure 94. Acanthopleura granulata (Gmelin, 1791). Villa del Mar Beach, Dominican Republic (AJF coll.); specimen 41 mm long. Posterior aspect of posterior valve. Figure 95. Acanthopleura granulata (Gmelin, 1791). Villa del Mar Beach, Dominican Republic (AJF coll.); specimen 41 mm long. Ventral aspect of posterior valve. Figure 98. Acanthopleura echinata (Barnes, 1824). Los Colora- dos, Chile (LACM 75-19); specimen 23 mm long. Figure 99. Acanthopleura echinata (Barnes, 1824). Same speci- men as in Figure 98. Close-up of anterior valves. Figure 100. Acanthopleura echinata (Barnes, 1824). Tumbes, Chile (AJF coll.); specimen ca. 70 mm long. Dorsal aspect of posterior valve. Figure 101. Acanthopleura echinata (Barnes, 1824). Same spec- imen as in Figure 100. Posterior aspect of posterior valve. Figure 102. Acanthopleura echinata (Barnes, 1824). Same spec- imen as in Figure 100. Ventral aspect of posterior valve. Page 264 an illustration in CHEMNITZ (1785, 8:279, pl. 95, figs. 797, 798) with Magellan Strait as locality. PILSBRY (1983c), KAAs (1972), and Kaas & VAN BELLE (1980) assigned the figured specimen to the West Indies (7.e., to Acanthopleura granulata); ROCHEBRUNE (1889) assigned it to Australia; NIERSTRASZ (1905b, 1906) to the Cape of Good Hope, South Africa. The name is here suppressed as a nomen dubium. Chiton unguiculatus BLAINVILLE, 1825:544, is regarded here also as a nomen dubium, having no locality, no illus- tration, and lacking descriptive elements to differentiate it from other species of Acanthopleura. Acanthopleura echinata (Barnes, 1824) Figures 98 to 105, and 113-E Chiton echinatus BARNES, 1824:71-72, pl. 3, figs. 4a, 4b; SOWERBY, 1840b:1, 9, sp. 1, fig. 47. Corephium echinatus: GRAY, 1847a:68, 1847b:169, 1857:184; DALL, 1879:280, fig. 30 (radula). Acanthopleura echinata: PILsBRY, 1893a:105, 1893c:217-219, pl. 47, figs. 6-17 (in subgen. Mesotomura); PLATE, 1898: 5-167, pls. 1-10, figs. 1-110; NrersTrasz, 1905a:102; SCHWEIKART, 1905:384-386, figs. 20, 26-28; Horst & SCHEPMAN, 1908:526; DALL, 1909:180, 248, pl. 23, fig. 6; AYRES, 1916:335; BERGENHAYN, 1930a:8, 31, 33; pl. 8, fig. 74; GiGoux, 1934:281; BouDET, 1945:130; LeELoup, 1956:55-58, figs. 28, 29; STUARDO, 1959:145- 146, 1964:82; MARINCOVICH, 1973:43, fig. 100; LELOuP, 1980a:1. Rhopalopleura echinata: THIELE, 1893:374 (as syn. of ““Rho- palopleura aculeata Linnaeus’’), 1909:6. Mesotomura echinata: THIELE, 1929:22. Chiton tuberculiferus SOWERBY, 1825:29 (nomen nudum). Chiton spiniferus FREMBLY, 1827:196-197, 1832 (plates):pl. 16, fig. 1; STEARNS, 1892:334 (in subgen. Corephium). Acanthochiton spinifera: STEARNS, 1894b:449. “Chiton aculeatus Linnaeus” REEVE, 1847:pl. 9, fig. 49 (with C. spiniferus and C. tuberculiferus as syn.). [Non: Linnaeus, 1758 (fide DoDGE, 1952:20-21).] “Rhopalopleura aculeata Linnaeus” THIELE, 1893:373-374, pl. 30, fig. 37, 1909:6. “Corephium aculeatum Linnaeus” MOSELEY, 1885:18-19, pl. 5, fig. 8, pl. 6, figs. 10-12 (aesthetes). Type material and type locality: Chiton echinatus Barnes, 1824: Types unascertained; locality “Coast of Peru,” here restricted to Callao, Peru (12°02’S, 77°05'W). Material examined: PERU: Talara, 4 specimens (CAS 010147); Paita, 8 spec- imens, largest 100 mm long (CAS 010145; LACM 72- 86); Lobos de Tierra Id., 1 specimen (LACM 74-10); Lobos de Afuera Id., 10 specimens, largest 75 mm long (LACM 74-5; LACM 74-6); Guanape Id., 2 specimens (LACM 74-2); Callao, 1 specimen (CAS 012793). CHILE: Arica, 2 specimens (CASG-SU 33761); Iquique, 1 specimen, 110 mm long (CAS 010146); Cumbres Bo- rascosas, T'arapaca Prov., 1 specimen, 45 mm long (LACM 75-14); Los Colorados, Antofagasta Bay, 4 specimens, The Veliger, Vol. 28, No. 3 largest 42 mm long (LACM 75-19); Antofagasta, 7 spec- imens, largest 100 mm long (LACM 75-15); Los Molles, Aconcagua Prov., 1 specimen, 44 mm long (LACM 75- 28); Islota Concon, Valparaiso Prov., 1 specimen, 132 mm long (LACM 75-31); Valparaiso, 9 specimens, largest 112 mm long (CAS 030914; CAS 030942); Punta Tumbes, Bahia de Concepcion, 7 specimens, largest 105 mm long (AJF coll., eg. E. Bay-Schmith). Description: Owing to the accompanying illustration, BARNES’ (1824) brief description of Chiton echinatus, based upon two specimens from Peru, is quite sufficient to iden- tify the species. Among 59 specimens of Acanthopleura echinata here ex- amined, largest 132 mm long (in alcohol) (LACM 75-31: Islota Concon, Valparaiso Prov., Chile) (largest specimen reported, 200 mm long [PLATE, 1898: Pajaros Id., off Co- quimbo, Peru]). Specimens (Figures 98-102) depressed, subcarinate. Body width/length, mean 0.57 (SD = 0.06; n = 17), specimens becoming relatively wider with growth (width/length ratio vs. specimen length, 7 = 0.58, P < 0.02, n = 17) (Figure 103). Tegmentum smooth to shiny (but often eroded), dark reddish-brown, with occasional small blue spots. Lateral areas hardly raised, smooth ex- cept for two radial rows, one of 5-9 round granules along diagonal line, another of 5-9 elongate granules indenting sutural edge. Anterior valve with some 10 radial rows of round granules; space between rows smooth. Central areas with raised, well defined, smooth jugal band bordered by shallow, longitudinal grooves with short, wavy, longitu- dinally oriented riblets on pleural areas. Mucro elevated, prominent, central to posterior; postmucro sharply sloped. Ocelli round to oval, 40-50 um in diameter, throughout anterior valve, postmucro area of posterior valve, and lat- eral areas of intermediate valves. Eaves somewhat spongy. Gills with 50-70 plumes per side. Articulamentum white, often with red discolorations at apex of valves. Central part of all valves show conspicu- ous, transverse, strongly engraved lines. Sutural laminae subrectangular; sinus well defined, pectinate; on valve viii, relative width of sinus, 0.44. Insertion plates strongly pec- tinate on outside; sinus minutely pectinate. Posterior valve with pectinate insertion plate with 1-3 slits (none in small specimens), irregularly and variably arranged, often with one particularly better defined on or near midline. Slit formula 8-1-0/3. Width of valves i/viii, 0.87. Valve viii tegmental surface length/width, 0.61. Girdle upper surface with erect, strong, spikelike spines (Figure 104), round in cross section, up to 8 mm long in large specimens (longer if not broken), often encrusted; in addition, abundant spinelet-like elements averaging 300 x 80 um, separated by about 100-200 um or more of “nude” girdle. Girdle bridges empty. Undersurface with trans- parent scales, about 35 x 35 um, with convex outer edge, concave inner edge, vaguely striate. In specimen 70 mm long (AJF coll.: Tumbes, Chile), radula 31 mm long (44% of specimen length), comprising 65 rows of mature teeth. Median tooth (Figure 105) 130 en peerreira, 1986 o Width/Length 0.6L. 0.5, Page 265 ee ee ee ee eee eee ee 10 20 30 40 Specimen Length [mm] Figure 103 Acanthopleura echinata (Barnes, 1824). ““Wideness” of specimen (body width/length ratio) as a function of specimen length (mm): Large specimens are relatively wider than smaller ones (r = 0.58; n = 17; P < 0.02). “um wide at anterior blade; first lateral teeth 200 um wide at anterior blade; head of second lateral teeth discoid, 550 uum wide; outer marginal teeth 350 wm long, 350 um wide (length/width, 1.0). Distribution: Acanthopleura echinata is confined to the western coast of South America (Figure 113-G), from Talara, Peru (4°35’S), the northernmost verified record (CAS 010147), to Punta Tumbes, Bahia de Concepcion, Chile (36°37'S), the southernmost verified record (AJF coll., leg. E. Bay-Schmith, Feb.-Sept. 1977). Reports of the species at the Galapagos Islands, Ecuador (PILSBRY, 1893c; STEARNS, 1894; NIERSTRASZ, 1905a; DALL, 1909), have not been confirmed (see SMITH & FERREIRA, 1977). Acanthopleura echinata is limited to the intertidal zone and shallow subtidal, 0-4 m, on rocks often exposed to heavy surf. Remarks: Among early authors, Chiton echinatus Barnes, 1824, occasioned some taxonomic confusions. SOWERBY (1825:29) named the species Chiton tuberculiferus as a re- placement for ‘“‘Aculeatus, Barnes” (!), apparently confus- ing the name echinatus with aculeatus, for nowhere did BARNES (1824) use the latter name. FREMBLY (1827) de- scribed and illustrated “tuberculiferus,” which SOWERBY (1825) had left as a nomen nudum, and renamed it spi- niferus “because the name aculeatus given to it by Barnes [!] was long since previously occupied; that of tuberculifer- us [Sowerby, 1825] was given from an old specimen, in which the spines were reduced in length by being broken, so that it is not applicable [!]; we have therefore now called it spiniferus” (p. 197). REEVE (1847b) compounded the confusion by stating that “the C. spiniferus of Frembly . . . is the old Linnaean C. aculeatus in fine condition . . . [and figured] in Chemnitz, Conch. Cab. v. 10. pl. 173. f. 1692.” It must only be added that the cited figure 1692 in CHEMNITZ (1788) does not conform at all, in morphology or locality (Nicobar Is.), to echinatus Barnes, 1824! Chiton echinatus was first allocated to Acanthopleura by PiLsBRY (1893a) in the monotypic subgenus Mesotomura Pilsbry, 1893a (=Corephium Gray, 1847a [not Browne, 1827]). The material here examined shows that PILSBRY’s (1893c) “single median-posterior slit,” characterizing Me- sotomura, is not a constant feature of A. echinata, which may have 1 to 3 slits (none, in small specimens) on the posterior valve, and not necessarily in the midline. In the uniqueness of its tegmental sculpture, articula- mentum, and girdle A. echinata differs sharply from all other species of Acanthopleura. It is a curious fact, then, that workers since PILSBRY (1893c) have easily accepted echinata as a member of the genus Acanthopleura while relegating species much closer to its type (A. spinosa), such as A. gaimardi, A. japonica, A. miles, or A. curtisiana, to other genera. Throughout its range, Acanthopleura echinata is sym- Page 266 j1mm Figure 104 Acanthopleura echinata (Barnes, 1824). Same specimen as in Fig- ure 100. Girdle spikelike spine. patric with A. nigra from which it clearly differs in teg- mental sculpture and girdle elements. Acanthopleura nigra (Barnes, 1824) Figures 106 to 111, and 113-N Chiton niger BARNES, 1824:71, pl. 3, fig. 3; GRay, 1828:6 (with C. coquimbensis Frembly as syn.). Enoplochiton niger: GRAY, 1847a:69, 1847b:169, 1857:181; MOSELEY, 1885:19, pl. 4, figs. 6-9; THIELE, 1893:375, pl. 30, fig. 40; Pitssry, 1893c:252-253, pl. 52, figs. 22- 29; PLATE, 1898:208-215, pl. 9, figs. 86-88, pl. 12, figs. 135-140; NIERSTRASZ, 1905a:106; Horst & SCHEP- MAN, 1908:528; DALL, 1909:181, 248, pl. 23, fig. 8; THIELE, 1929:21; BERGENHAYN, 1930a:32-34, pl. 8, figs. 78-79, pl. 9, figs. 81-82; GiGoux, 1934:281; LELOUP, The Veliger, Vol. 28, No. 3 Figure 105 Acanthopleura echinata (Barnes, 1824). Same specimen as in Fig- ure 100. Radula median and first lateral teeth. 1939b:6-9, figs. 5, 6, 1956:54-55; STUARDO, 1959:144, 146, 1964:82; MARINCOVICH, 1973:43, fig. 99. Chiton coquimbensis FREMBLY, 1827:197-198, 1832 (plates): pl. 16, fig. 2; REEVE, 1847, pl. 4, sp. & fig. 22. Type material and locality: Chiton niger Barnes, 1824: Types unascertained; locality “Coast of Peru,” here restricted to Iquique, Chile (20°13'S, 70°10'W). Chiton coquimbensis Frembly, 1827: Types unascertained; locality Coquimbo Bay, Peru (29°58'S, 71°21'W). Material examined: PERU: Talara, 2 specimens, largest 59 mm long (CAS 010148); Pisco, 3 specimens (CASG-SU 35328); Callao, 17 specimens (CAS 012791). CHILE: Independencia Bay, 2 specimens (LACM-AHF 380- 35); Arica, 3 specimens, largest 90 mm long (CASG-SU 33760); Iquique, 5 specimens, largest 131 mm long (LACM 64-16; LACM 75-12; CAS 010144); Cumbres Barascosas, Antofagasta Prov., 1 specimen, 93 mm long (LACM 75-14); Antofagasta, 7 specimens, largest 61 mm long (LACM 75-15); Coquimbo Bay, 1 specimen (CASG-SU 32955). Description: BARNES’ (1824) brief description but good illustration of Chiton niger is sufficient to identify the species. Among 41 specimens of Acanthopleura nigra here ex- amined, largest 131 mm long (in alcohol) (LACM 64-16: Iquique, Chile). Body width/length, mean 0.48 (n = 9). Specimens (Figures 106-109) round-backed, depressed. Tegmentum dark chocolate-brown, shiny, but easily erod- ed. Valves beaked; posterior edge of valve ii forming 110- 120° angle. Anterior valve with 4-6 concentric, zig-zagged furrows. Lateral areas elevated, well defined by strong round rib at diagonal line, with zig-zagged furrows as on anterior valve. Central areas well defined, smooth jugum ene kerreina,, O86 Page 267 Explanation of Figures 106 to 109 Figure 106. Acanthopleura nigra (Barnes, 1824). Iquique, Chile (LACM 75-12); specimen 32 mm long. Figure 107. Acanthopleura nigra (Barnes, 1824). Same specimen as in Figure 106. Dorsal aspect of posterior valve. bordered by narrow, depressed area with irregular, short, oblique furrows; para-jugal area smooth; pleural area with longitudinal, parallel furrows, not usually reaching an- terior border of valve. Mucro posterior, almost terminal. Ocelli round to oval, 20-30 um in diameter, throughout anterior valve and anterior half of lateral areas of inter- mediate valves. Gills with 70-80 plumes per side. Articulamentum dark chocolate-brown, with trans- verse, strongly engraved lines at middle of valves (also seen in A. echinata but not in any other Acanthopleura species). Sutural laminae somewhat elongate; sinus well defined, sinusal laminae pectinate. Insertion teeth strongly pecti- nate on outside. Posterior valve without insertion teeth but with well developed transverse callus. Slit formula 8/9-1-0. Girdle thick, musculous. Upper surface dark brown, conspicuously dotted with light brown scales (Figure 110); scales irregular in size (larger in middle % of girdle), up Figure 108. Acanthopleura nigra (Barnes, 1824). Same specimen as in Figure 106. Posterior aspect of posterior valve. Figure 109. Acanthopleura nigra (Barnes, 1824). Same specimen as in Figure 106. Ventral aspect of posterior valve. to 1.5-2 mm long in specimens 50 mm long (larger in larger specimens), vaguely striate, usually eroded at upper edge, clearly separated from each other by area as wide as scale (in alcohol preserved specimens); on outer % of girdle, scales much smaller, shorter, dark brown, erect, spine-like; girdle surface completely covered otherwise with minute, dark brown, lanceolate spicules, up to 100 um long, 25 wm thick. Girdle bridges, empty in middle third, but crowded with small, dark brown spiculoid elements (akin to those on girdle proper) in outer thirds. Under- surface covered with transparent squarish scales, about 40 x 40 um, in columnar arrangement, with coarse, ir- regular striations. In specimen 49 mm long, radula measures 15 mm in length (30% of specimen length) and comprises 60 rows of mature teeth. Median tooth (Figure 111) 100 um wide at anterior blade; first lateral teeth about 500 um long, 150 wm wide at anterior blade; head of major lateral teeth Page 268 Se We 2] ‘ yc 1mm Figure 110 Acanthopleura nigra (Barnes, 1824). Same specimen as in Figure 106. Girdle scales, outer and inner sides. discoid, 300 wm in width; outer marginal teeth 300 wm long, 250 um wide (length/width, 1.2). Distribution: Acanthopleura nigra is confined to the west- ern temperate coast of South America, the Peru-Chilean zoogeographic province (BRIGGS, 1974) (Figure 113-N), from Talara, Peru (4°34'S) (CAS 010148), the northern- most verified record, to Coquimbo Bay, Chile (29°58’S) (CASG-SU 32955), the southernmost verified record. Bathymetric range limited to the intertidal zone. Remarks: Chiton niger has been generically segregated from Acanthopleura by authors since GRay (1847a) and PiLsBry (1893c) on account of three taxonomic characters: (1) the girdle elements, (2) the articulamentum of the posterior valve, and (3) the ocelli. The girdle elements of Chiton niger have made for easy identification of the species; the light-colored, “rude” scales, clearly separated from each other by the fleshy girdle, do confer on C. niger a unique appearance. However, close observation of the scales shows that, although in shape, size, and placement (7.e., apart from each other) they have no similarity, they show no major departure from the girdle elements in other species of Acanthopleura. In ad- dition, their implantation in the girdle through a rough, irregular, and variously shaped facet (Figure 110) con- forms well to that of scales and spinelets in other species of Acanthopleura. The conspicuous separation of the scales by the girdle is a feature also seen, though less conspicu- ously, in other species of Acanthopleura (e.g., A. miles, A. curtisiana, A. araucariana). The absence of teeth in the posterior valve of Chiton niger led PILSBRY (1983c) to group the species with Lio- lophura and Onithochiton Gray, 1847a. Yet, toothless pos- terior valves are seen not only in other members of Acan- thopleura—A. gaimardi, A. japonica (the ‘“‘Liolophura”’), A. hirtosa, and A. rehderi)—and in Onithochiton, but also in other genera and other families, such as in the acantho- chitonid Cryptoplax Blainville, 1818, the schizochitonids Aulacochiton Shuttleworth, 1853 (=Lorica Adams & Adams, 1852, preoccupied by Lorica Bronn, 1848, a crus- tacean) and Componochiton Milne, 1963, and the mopaliid Plaxiphora Gray, 1847a, indicating that similar modifi- cations of the posterior valve have occurred more than once in the evolution of chitons. The ocelli in Chiton niger were said to be, in contrast The Veliger, Vol) 283INoms (eee 1400um Figure 111 Acanthopleura nigra (Barnes, 1824). Same specimen as in Figure 106. Radula median and first lateral teeth. to those of Acanthopleura species, “excessively minute” (MOSELEY, 1885:19) and ‘“‘extremely minute and oval in- stead of round” (PILSBRY, 1893c:252). These statements are here in part refuted. Careful evaluation of ocelli in Acanthopleura shows that in A. nigra the ocelli are smaller (average diameter: 25 wm in A. nigra, 45 um in other species of Acanthopleura) but not more “oval instead of round” than in other species of Acanthopleura. Thus, except for the fact that the girdle scales of Acan- thopleura nigra are distinct enough to immediately diag- nose the species, there seems to be no compelling reason to segregate the species in the monotypic Enoplochiton Pilsbry, 1893c, and so obscure its relationship with other members of Acanthopleura. Acanthopleura nigra is sympatric with A. echinata along the Peru—Chile coast. 30° 90° 150° Figure 112 Geographic distribution of: S = Acanthopleura spinosa (Bru- guiére, 1792); B = Acanthopleura brevispinosa (Sowerby, 1840). Aen|e kerreias 1986 Page 269 pS es ee 60° 120° 180° 120° 60° Figure 113 Geographic distribution of: C = Acanthopleura gemmata (Blainville, 1825); R = Acanthopleura rehderi Ferreira, spec. nov.; E = Acanthopleura echinata (Barnes, 1824); N = Acanthopleura nigra (Barnes, 1824); G = Acanthopleura granulata (Gmelin, 1791). DISCUSSION As here understood, the genus Acanthopleura corresponds to the subfamily Acanthopleurinae (sensu VAN BELLE, 1983:126-130), with the genera Liolophura, Enoplochiton, and Squamopleura suppressed as synonyms. Although the particular reasons for such an action were given previ- ously in the account of the respective species, they bear restating here since they may not be immediately apparent to chiton taxonomists. Liolophura Pilsbry, 1893a, erected to accommodate Chi- ton incanus (=C. gaimardi) and C. japonicus, was distin- guished from Acanthopleura by the presence of a “‘smooth crescentic callus in place of the insertion-teeth” (PILSBRY, 1893a:105) in the posterior valve. Despite obvious affinity with Acanthopleura, PILSBRY (1893a) placed the taxon “‘in the immediate vicinity of Onithochiton,” and grouped it, instead, with Enoplochiton and Onithochiton in the subfamily Liolophurinae Pilsbry, 1893c. Although Lio- lophurinae was rejected by some chiton workers (NI- ERSTRASZ, 1905a; THIELE, 1909, 1929; BERGENHAYN, 1930a, 1933), Liolophura, as a generic taxon, has re- mained in general usage. On introducing Liolophura, PILSBRY (1893a, c) assumed that in species of this genus the insertion teeth of the posterior valve had been replaced by a callus. But the interpretation was faulty. Close examination of specimens of Liolophura species suggests that the callus in the pos- terior valve appears not “in place of insertion-teeth” as PIusBry (1893a:105) asserted, but as a result of their dis- appearance. In fact, a transverse callus is present in the posterior valve of most other species of Acanthopleura, only less developed and “hidden” by the pectinations and teeth on its posterior aspect. In this respect, it may be noted that in specimens of A. japonica (previously segregated in Liolophura) vestigial insertion teeth are often seen as coarse pectinations or rugosities on the sides (occasionally, even in the middle) of an otherwise smooth, flat-surfaced, cres- centic callus, indicating by the irregularity of their pres- ence, position, shape, and size, their vanishing, relict na- ture. The taxonomic significance of the absence of teeth in “Liolophura” is further diminished by the observation that they are absent also in other genera and families (see Remarks on A. nigra), and that in several other species of Acanthopleura (A. brevispinosa, A. loochooana, etc.) the pos- terior valves show “in between” forms of insertion plates where insertion teeth are present but underdeveloped. Be- cause members of Liolophura and Acanthopleura (sensu Pilsbry) do not seem to differ in any other major char- acter, the two genera are here regarded as synonymous. Enoplochiton Gray, 1847a, erected to accommodate Chi- ton niger, has been accepted by chiton workers without dissent. Apparently, the unique appearance of C. niger, with its conspicuously large, light-colored (7.e., eroded) scales distinctly separated by the “velvety” girdle, has been regarded as demonstration of sufficient evolutionary dis- tance to justify assignment to a distinct genus. However, in the course of this study it became apparent that over- all similarities between C. niger and members of Acantho- pleura are greater than generally assumed. The girdle scales of C. niger, peculiar looking as they are, cannot be re- Page 270 30° 15° 0° 30° 750 120° 135° 150° Figure 114 Geographic distribution of: H = Acanthopleura hirtosa (Blain- ville, 1825); L = Acanthopleura loochooana (Broderip & Sowerby, 1829); A= Acanthopleura arenosa Ferreira spec. nov.; D= Acanthopleura gaimardi (Blainville, 1825). garded as a “major modification of the girdle armature” (AsHBY, 1929:160) of Acanthopleura; in fact, they do not seem to be more evolutionarily “distant” from the spines of A. spinosa (type species of Acanthopleura) than, say, the spinelets of A. gemmata, or the spikes of A. echinata. And since the modification in girdle elements is not accom- panied by changes in “some more stable character” (ASHBY, loc. cit.)—in their large size, oval body shape, heavy and beaked valves, shape and distribution of ocelli (though smaller in diameter), pectinate insertion teeth, radula with discoid major lateral teeth, and intertidal habitat, speci- mens of C. niger do conform well with other Acanthopleura species—it seems that segregation of C. niger in the mono- typic Enoplochiton is unjustified. Acanthozostera Iredale & Hull, 1926, and Planispina Taki, 1962, have been long regarded as synonyms of Acanthopleura (SMITH, 1960; VAN BELLE, 1983). Clavarizona Hull, 1923, was erected to accommodate a single species, Chiton hirtosus, distinguished from Liolo- phura for “the girdle covering which consists of . . . scales” (HULL, 1923:199). But, as already pointed out by LELOuP (1961:47), the girdle elements in C. hirtosus do not differ from those of other Liolophura “except in dimensions .. . [they appear] not as scales... but as spines .. . [and] their arrangement on the girdle is not regular like that of the The Veliger, Vol. 28, No. 3 scales of the Chitoninae and the Ischnochitoninae; their implantation is that of the spines... .”” Thus Clavarizona has been synonymized to Liolophura (ASHBY, 1926; SMITH, 1960; LELOoupP, 1961), although accepted by THIELE (1929) and VAN BELLE (1983) as a subgenus of Liolophura. Squamopleura Nierstrasz, 1905a (=Sclerochiton Car- penter zm Pilsbry, 1893b [not Kraatz, 1859]), was intro- duced also on account of the girdle scales: spinelets (‘‘kal- kigen Stacheln und Dornen”) in Acanthopleura, scales (‘“Kalkschuppen”’) in Squamopleura. THIELE (1929), and presently VAN BELLE (1983), regarded the finding of needle-like spicules amidst the girdle scales of Squamo- pleura species as of generic significance. Again, close study of the girdle elements of species hitherto allocated to Squa- mopleura (Chiton miles and C. curtisianus) revealed that they are essentially the same as those of C. hirtosus (1.e., variable in shape and size, and implanted like a spine, in a manner already observed by LELOup [1961b] for C. hirtosus), and, further, that the needle-like spicules amidst larger girdle elements are no different from those to be found in virtually all species of Acanthopleura. SMITH (1960), probably overlooking the presence of ocelli in Squamopleura species, regarded it as a subgenus of Chiton Linnaeus, 1758. Acanthopleura attains greatest species diversity in the Central Indo-Pacific, in the “fertile triangle” (BRIGGs, 1974:14) of the Indo-Malayan region (EKMAN, 1953), with a “center of origin” at Taiwan where five species (A. spinosa, A. gemmata, A. japonica, A. miles, and A. loo- chooana) have been recognized. In the central Indo-Pacific, the north-south distribution of Acanthopleura maps out in a curiously symmetrical manner: Acanthopleura gemmata, the most common species, is present virtually everywhere in the tropics. To the north of the Tropic of Cancer (approximately), in Korea, Tai- wan, and Japan, A. gemmata is replaced in the intertidal zone by A. japonica; and to the south of the Tropic of Capricorn (approximately), A. gemmata is similarly re- placed by A. gaimardi on the east and A. hirtosa on the west coasts of Australia. What is intriguing is the realiza- tion that the three species “replacing” A. gemmata in tem- perate waters, 7.e., A. japonica, A. gaimardi, and A. hirtosa, all differ from A. gemmata in a single major feature, the absence of insertion teeth in the posterior valve. It is cu- rious, too, that, in the northern hemisphere, A. loochooana (with posterior valve insertion teeth present but consid- erably underdeveloped) is confined to the zone of contact between A. gemmata and A. japonica; and, similarly, in the southern hemisphere, A. arenosa (with underdevel- oped insertion teeth in the posterior valve) is limited to the zone of contact between A. gemmata and A. gaimardi. This correlation between the reduction or loss of insertion teeth and high latitude (both north and south) is impres- sive. It is tempting to speculate that the implied difference in water temperature may be a factor influencing the ar- ticulamental changes, although the possible adaptive value of such changes is obscure. Aca |eHerreina® 1986 Page 271 90° 120° 150° 180° Figure 115 Geographic distribution of: T = Acanthopleura curtisiana (Smith, 1884); J = Acanthopleura japonica (Lischke, 1873); M = Acanthopleura miles (Carpenter in Pilsbry, 1893); K = Acanthopleura araucariana (Hedley, 1898). The differential diagnosis of Acanthopleura species may be quite difficult at times, particularly when one is faced with dry, and eroded specimens. The fifteen species of Acanthopleura here recognized are remarkably similar to each other, bespeaking their congeneracy, and explaining the difficulties encountered in unraveling the group. The following summary of comparative points and comments regarding the morphology of Acanthopleura species may be of value: (1) The relatively large size of the specimens, drab and eroded tegmentum, and intertidal habitat (on top of rocks, beaten by surf, emergent at low tide, often exposed to sun) are common to all Acanthopleura species. (2) The valves, heavy and beaked, are usually (easily ?) eroded; but even among pristine, young specimens, inter- specific distinctions in the tegmental sculpture are minor and hard to draw out (except in A. spinosa, A. echinata, and A. nigra), making for difficulties in the differential diagnosis of the species. In all species, the tegmentum recurves forward at the ventroposterior portion of the valve, forming an underfold or hypotyche (see HOARE et al., 1983). (3) The ocelli are essentially identical in distribution, shape, and size in all species, although somewhat smaller in A. nigra. They are assumed to be light-sensitive organs (BoyLeE, 1969, 1972). In this respect, it is curious to ob- serve that in Tonzcia, Onithochiton, or Schizochiton, whose specimens are hardly ever eroded, the ocelli are likely to remain functional throughout life; but in Acanthopleura their usefulness appears short-lived, because they are soon destroyed by erosion (though a few are usually seen at the periphery of the growing shell). The possible sensory function of the hypotyche (HOARE e¢ al., 1983:996), well developed in Acanthopleura, may conceivably compensate for the loss of sensory input resulting from erosion of the ocelli. No ocelli have been found in the hypotyche of Acan- thopleura species. (4) The significant differential features in the articu- lamentum of Acanthopleura species are limited to the pos- terior valve, where the insertion teeth may be well devel- oped (A. spinosa, A. gemmata, and A. granulata), absent (A. gaimardi, A. japonica, A. hirtosa, A. rehderi, and A. nigra), or poorly and irregularly developed (in the other species). In A. echinata and A. nigra, sympatric species Page 272 along the Peru-Chile coast, the articulamental surface shows an area of strongly engraved transverse lines in the middle of the valves; curiously, this feature is not present in any other species of Acanthopleura, but is seen in species of the genus Mopalia Gray, 1847a. (5) The girdle upper surface main elements may be spikelike (A. echinata), spines (A. spinosa), spinelets (A. gemmata, A. granulata, A. brevispinosa, A. arenosa, A. loo- chooana, A. gaimardi, A. japonica, and A. rehderi), or scales (A. miles, A. araucariana, A. curtisiana, A. hirtosa, and A. nigra). The undersurface shows no significant dis- tinctions among species. (6) With the exception of A. brevispinosa and A. reh- deri, the radula is remarkably constant in Acanthopleura, revealing no species-specific features. Acanthopleura has no fossil record, a rather intriguing fact for a group so widely distributed, and containing spec- imens that, in most of the fifteen species known, are rel- atively large and abundant in the intertidal zone. How- ever, the presence of Acanthopleura in the Caribbean, as A. granulata, indicates that the genus antedates the closure of the Pacific-Atlantic seaway which, as it may be inferred from GEISTER’s (1977) work on corals, did not take place until the late Pleistocene. A diagnostic key to the species of Acanthopleura is here suggested. In the case of very similar species where the question of conspecificity is still at issue (A. garmardi vs. A. japonica, and A. granulata vs. A. gemmata) it was thought preferable to use couplets based upon broad geographic localities rather than give artificial importance to some minor and/or inconstant character. In this manner it is hoped the diagnostic key will better fulfill its main pur- pose: to provide a practical tool for the nonspecialist. Diagnostic Key of Acanthopleura Species 1. Posterior valve with no insertion teeth ........ 2 Posterior valve with insertion teeth ........... 6 2eaGindleswithescalesirmras er ereerec ine ernest re 3 Girdle withyspinelets’ 42s aceeeeies aac: 4 3. Girdle scales large (1-2 mm long in large spec- imens), separated from each other by girdle (Reru=Chilecoast)) a. eee A. nigra Girdle with moderate size scales (less than 0.5 mm long even in large specimens) relatively close together (southwest Australia) ... A. hirtosa 4. Radula major lateral teeth with tetracuspid head Rae pre Wey creche chinese errr oatatelia ie riein nd A. rehderi Radula major lateral teeth with discoid head .. 5 5. Specimens from north of Tropic of Cancer (or Gulfrof (hailand) eases eee nore A. japonica Specimens from south of Tropic of Capricorn mids ataianeiilec day ste eUopeRee cue re hae eee ctoeeae A. gaimardi 6. Posterior valve with well developed insertion teeth eal er oahs Bette ey aoe Sycucuct Sea cae ae Cee eee ea ES 7 Posterior valve with poorly developed insertion [colt (Nie cater cu tena RR RR ert icon ik AMR EG BLS Eco 8 The Veliger, Vol. 28, No. 3 7. Specimens from the Caribbean Sea ... A. granulata Specimens from the Indo-Pacific ...... A. gemmata 8. Girdle with spikelike elements (Peru—Chile) coast Bi dhatks pvssev alae ce ootce SHEE eal eg Rea A. echinata Girdle with scales, spines or spinelets but not Spikelike’ elements) 32.) ooo ae eee eeeee 9 9. Girdle with long, blackish, thin spines; tegmen- tum reddish; intermediate valves often two- slitted said 22 4a aay eee ee A. spinosa Girdle with scales or spinelets; tegmentum gray to dark brown; intermediate valves uni-slitted er ere NINE ORES op'6'0:0-070-00.0.0 0 c 10 10. Tegmentum purple-brown to black with coarsely round granules throughout; girdle with equal- sized cylindrical, black (often white-tipped) spinelets; specimens length often attaining 50 mm (Indian Ocean) ............ A. brevispinosa Tegmentum gray to light brown; girdle elements not usually cylindrical; specimens length rare- ly attaining 50 mm (Indo-Pacific) .......... 11 11. Girdle with regular, equal-sized elements ..... 12 Girdle with irregular, unequal-sized elements .. 14 125 G = 100 S A cE 100p Figure 1 Teeth of Mediterranean Pleurobranchomorpha. A, Pleurobranchus membranaceus (inner lateral teeth); B, Plewro- branchus membranaceus (outer lateral teeth); C, Pleurobranchus testudinarius; D, Berthella ocellata; E, Berthella sp. 1; F, Berthella sp. 2; G, Berthellina citrina; H, Berthella aurantiaca. Page 304 served specimens are 30 mm (Genoa) and 35 mm (Poz- zuoli). In the 30 mm long specimen, the dimensions of the shell are 25 x 17 mm. The radular formula is 36-45 x 60(100).0.(100)60. The teeth nearest the rachis (Figure 1A) do not bear a subterminal spine, contrary to assertions by VAYSSIERE (1885), BERGH (1899), and THOMPSON (1976). The lateral teeth are longer and thinner (Figure 1B). The mandibular platelets (80-130 wm long) bear 3 large denticles (Figure 2A). The gill rachis is tuberculate and bears from 24 to 26 pinnae. A pedal gland is present. There is a flap of flesh around the genital apertures, which are contiguous. Diet: Nothing was found in the alimentary canals of the specimens studied. Discussion: This species has been recently recorded from France (HAEFELFINGER, 1960; VICENTE, 1967) and Spain (Ros, 1975). It seems to be uncommon in Italian Seas; however, PERRONE (1983) collected several specimens in the Gulf of Taranto. Pleurobranchus (Susania) testudinarius (Cantraine, 1840) Material collected: Tuscany Archipelago: Isle of Elba (October 1982, one specimen from 45 m); Montecristo Isle (June 1983, two specimens from 40 m). Description: This species (Figure 4) has a red ground color. The mantle is formed by large tubercles, which around their polygonal bases have a thin purple line. On top of the mantle there are two rows of larger tubercles. The Elba specimen was 15 cm in length, those from Mon- tecristo smaller (12 cm). In these three specimens a shell was not found. The radular formula is 220-250 x 220(250).0.(250)220. The hooked teeth are simple and uniform (Figure 1C). The jaws of the largest specimen are 15 mm long, and their platelets, 370 wm in length, have 2 or 3 lateral denticles (Figure 2B). The large tu- berculate gill (45 mm long in the Elba specimen) bears 18-20 pinnae. A pedal gland is present. Diet: Colonial ascidians (Didemnidae) were found in the stomachs, together with bivalves (Musculus marmoratus) and plant debris. Discussion: This species lives on sandy to muddy bottoms, but recently Ros & GIL (1984) found it in an underwater cave at Majorca (Balearic Isles). Recent records come also from France (HAEFELFINGER, 1960), Italy (SCHMEKEL, 1968), Israel (BARASH & DANIN, 1971), and Spain (Ros, 1975). Dhe VeligersVoleZ255Nows Berthella plumula (Montagu, 1803) Material collected: Ligurian Sea: Isle of Gallinara (Au- gust 1976, one specimen under a stone at 0.5 m depth). Description: This specimen measured 5 mm in length and was light yellow in color. The mantle has reticulate markings as reported by THOMPSON (1976). The oval, white internal shell covers all the contracted body. The radular formula is 40 x 55.0.55. The teeth are simple hooks, while the mandibular platelets exhibit 2 or 3 lat- eral denticles on either side of the cusp (Figure 2E). The gill has 16 pinnae on either side of the smooth rachis. Diet: No sponge spicules were found in the alimentary canal. This agrees with a diet of slime-sponges (e.g., Os- carella lobularis) as reported by DELALOI & TaRDy (1976) or colonial tunicates. Discussion: This species has been recorded in recent times from France (HAEFELFINGER, 1960), Italy (SCHMEKEL, 1968; PERRONE, 1983), and Israel (BARASH & DANIN, 1971). Berthella aurantiaca (Risso, 1818) Material collected: Tyrrhenian Sea: Gulf of Procchio, Isle of Elba (June 1978, one specimen in a meadow of Posidonia oceanica at 14 m depth). Description: The specimen was 6 mm long and yellow in color. Through the skin of the smooth mantle calcar- eous deposits could be discerned. A large oral veil is pres- ent, and the white internal shell covers the entire con- tracted body. The radular formula is 50 x 60.0.60. In the preserved state, this species can be distinguished from B. plumula by its larger mandibular elements (Figure 2D), but the radular teeth do not differ significantly (Figure 1H). The gill has 18 pinnae on either side of the smooth rachis. Diet: No sponge spicules were found in the alimentary canal. Discussion: Recent records have come from France (HAE- FELFINGER, 1960), Italy (SCHMEKEL, 1968; SORDI, 1969; PERRONE, 1983), Yugoslavia (STARMUHLNER, 1969), Is- rael (BARASH & DANIN, 1971), and Spain (Ros, 1975; ALTIMIRA et al., 1981; Ros & GILI, 1984). Berthella ocellata (Delle Chiaje, 1828) Material collected: Tyrrhenian Sea: Gulf of Naples (May 1979, in the Mitigliano Cave on the Sorrentine Peninsula at 12 m depth). Figure 2 Mandibular platelets of Mediterranean Pleurobranchomorpha (bar = 50 um). A, Pleurobranchus membranaceus; B, Pleurobranchus testudinarius; C, Berthella sp. 2; D, Berthella aurantiaca; E, Berthella plumula; F, Berthella sp. 1; G, Berthella ocellata; H, Berthellina citrina. Page 305 R. Cattaneo-Vietti, 1986 Page 306 Description: The ground color of this species is tan, and the mantle has many whitish spots (ocell:) surrounded by opaque white oval rings (MAZZARELLI, 1891). The spec- imen measured 18 mm in length and had a white oval internal shell, 4.5 mm long. The radular formula is 60 x 65.0.65. The teeth are squat and hook-shaped (Figure 1D). The mandibular elements are smooth (Figure 2G), which is unusual in this genus but has been reported also for B. monterosati (VAYSSIERE, 1885), a synonym. The gill has 15 pinnae on either side of the smooth rachis. Diet: In the Mitigliano Cave, Berthella ocellata feeds upon sponges, e.g., Plakina trilopha and Plakinastrella copiosa. Discussion: Recent records have come from Marseille (VICENTE, 1967), Tuscany (SORDI, 1969), and Costa Bra- va, Spain (Ros, 1975). Berthella sp. 1 Material collected: Ligurian Sea: Paraggi, Portofino Promontory (June 1978, one specimen from 13 m depth, hard bottom). Description: This specimen resembles B. ocellata, but the coloration is different: the brown mantle has only sparse white patches, which are not organized into ocelli. The body length is 20 mm, and the white, subtriangular shell is 7 mm long. The radular formula is 85 x 80.0.80; the teeth are simple hooks, like those of other species of Ber- thella (Figure 1E). The mandibular elements are smooth (Figure 2F). The smooth rachis of the gill bears 18 pin- nae. Diet: Spicules of the sponge Corticiwm candelabrum were found in the alimentary canal. Berthella sp. 2 Material collected: Ligurian Sea: off Alassio (May 1975, one specimen from 185 m, coll. G. Albertelli). Description: The shape of this specimen, which was 9 mm long and had a smooth, yellow mantle, resembles that of Berthellina citrina. A large oval veil is present, and the internal shell is approximately %4 the body length. The radular formula is 40 x 35.0.35. Teeth from the central part of the radula bear an additional subterminal spine, as in Berthella tupala Marcus, 1957, and many species of Pleurobranchus (Figure 1F). The mandibular platelets generally have one or two very distinct denticles on either side of the cusp and are more like the elements of Pleu- robranchus membranaceus than those of the other Berthella species (Figure 2C). The gill, lacking tubercles, has 20 pinnae on either side. An external penial sheath is present. The salivary glands are long and ribbon-shaped. Diet: There were no sponge spicules in the alimentary canal of the specimen. The Veliger, Vol. 28, No. 3 Berthellina citrina (Ruppell & Leuckart, 1828) Material collected: Ligurian Sea: Paraggi, Portofino Promontory (July 1979, one specimen from 25 m depth, under a stone). Tyrrhenian Sea: Gulf of Naples (June 1979, June 1980, June 1981, many specimens in sub- marine caves along the sorrentine Peninsula). Description: The color of this species is yellow-orange and it may be confused with Berthella aurantiaca on ex- ternal features alone. However, the tooth-shape is unmis- takable (Figure 1G). The oval shell is small and white; one specimen from Mitigliano Cave had no shell. The radular formula is 45-80 x 160(200).0.(200)160. The elongate teeth bear serrations on their posterior edge. (Figure 1G). The platelets are smooth (Figure 2H). The number of gill pinnae varies from 16 to 20 in specimens 6-11 mm in body length. A pedal gland is lacking. Diet: Observations on the diet of Berthellina citrina have been published by CATTANEO (1982). In the present col- lection, the specimen from Paraggi contained spicules of the sponges Hemimycale sp. and Batzella sp. Discussion: WILLAN (1983) discussed whether or not Ber- thellina citrina and Berthellina engeli Gardiner, 1936, are the same species. Here, they are considered synonyms, as did THOMPSON (1976). According to BURN (1962) these two species differ mainly in the shape of the jaw platelets, which are smooth in B. engeli and bear 1-3 indistinct lateral denticles in B. cztrina. In the same paper, however, Burn reports the presence of smooth platelets in B. cztrina from Australian waters: this is in agreement with THOMP- SON’s (1976) description of B. citrina from European waters. Recent records of Berthellina citrina include localities in France (HAEFELFINGER, 1960), Italy (SCHMEKEL, 1968), and Israel (EALES, 1970; BARASH & DANIN, 1977). Umbraculum mediterraneum (Lamarck, 1812) Material collected: Tyrrhenian Sea: Isle of Montecristo (July 1980; July 1982 in a submarine cave at 30 m depth). Sicily Channel: Isle of Linosa (August 1978 in a small cave at 4 m). Description: This species, unmistakable with a flat ex- ternal shell, is well described by MOQUIN-TANDON (1870) and PrRuvoT-FOoL (1954).The radular formula is up to 150 x 1000.0.1000. The teeth, numerous and all alike, are hook-shaped. In smaller specimens (length 30 mm) from the Isle of Montecristo, the gills are not present around the front of the body as in the larger individuals (length 85 mm). Diet: The diet consists of several sponges: Tethya citrina, Diplastrella unistellata, Jaspis johnston, Alectona millaris, Agelas sp., Aaptos aaptos, and Spirastrella cunctatrix. Discussion: This species is quite common in caves and in shaded habitats. Recent records of Umbraculum have come R. Cattaneo-Vietti, 1986 Page 307 Explanation of Figures 3 and 4 Figure 3. Pleurobranchaea meckeli, 10 cm length. Ligurian Sea, Genoa, on muddy bottom at 50 m depth. Figure 4. Pleurobranchus testudinarius, 15 cm length. Tyrrhenian from France (HAEFELFINGER, 1960; VICENTE, 1967), Turkey (SWENNEN, 1961), Italy (SCHMEKEL, 1968), Israel (BARASH & DANIN, 1971), and Spain (Ros, 1975). CONCLUSIONS If we exclude several species imperfectly described by ForBEs (1844), PHILIPPI (1844), and VERANY (1846) and all considered incertae sedis by PRUVOT-FOL (1954), the Sea, Isle of Elba, on sandy bottom at 45 m depth. The rhino- phores are on the left side. Photo by R. Pronzato. Pleurobranchomorpha present in the Mediterranean Sea are as follows: Superfamily Pleurobranchacea Family Pleurobranchaeidae Pleurobranchaea meckeli Leue, 1813 Pleurobranchaea inconspicua Bergh, 1897 Pleurobranchaea vayssierei Marcus & Gosliner, 1984 Pleurobranchaea notmec Marcus & Gosliner, 1984 Page 308 Family Pleurobranchidae Pleurobranchus (Oscanius) membranaceus (Monta- gu, 1803) Pleurobranchus (Susania) testudinarius Cantraine, 1840 Pleurobranchus forskali (Ruppell & Leuckart, 1828) Berthella plumula (Montagu, 1803) Berthella aurantiaca (Risso, 1818) Berthella ocellata (Delle Chiaje, 1828) Berthella stellata (Risso, 1826) Berthella perforata (Philippi, 1844) Berthella elongata (Cantraine, 1835) Berthellina citrina (Ruppell & Leuckart, 1828) Superfamily Umbraculacea Family Umbraculidae Umbraculum mediterraneum (Lamarck, 1812) Family Tylodinidae Tylodina perversa (Gmelin, in L., 1791) Tylodinella trinchesei Mazzarelli, 1897 The list is probably incomplete because of the past no- menclatural confusion. Some of the taxonomic problems are, at present, insoluble, but progress continues to be made. Berthella stellata, for example, was recently vali- dated by detailed re-description of Adriatic specimens by THOMPSON (1981), while Marcus & GOSLINER (1984) provided an exhaustive review of the Pleurobranchaeidae. Unfortunately, many other problems persist. Berthella elongata was recorded by VICENTE (1967) from the Gulf of Marseille and by PERRONE (1983) from the Salentin coast of the Gulf of Taranto, but the specific validity seems to be uncertain because of the poorness of the descriptions. The status of Berthella perforata remains enigmatic also, and there is confusion between past records of Berthella plumula, B. aurantiaca, and Berthellina citrina. It is nec- essary to inspect the radula and jaws in order to differ- entiate between species of Berthella and Berthellina, and distinguishing between B. aurantiaca and B. plumula re- mains difficult, especially when the specimens have been preserved (TERRENI & CAMPANI, 1980). When alive, B. plumula can be seen to have conspicuous reticulate mark- ings on the dorsal mantle. The description of Gymnotoplax barashi Marcus, 1977, was undoubtedly based in error upon a distorted specimen of Pleurobranchus membrana- ceus, aS pointed out by WILLAN (1978). Although Tyylo- dinella trinchesei is listed here, PRUVOT-FOL & FISCH- ER-PIETTE (1934) and BERTSCH (1980) have raised justifiable doubts about its validity. Two further notes on the records of pleurobrancho- morphs warrant mention. First, 7ylodina perversa is a rare species, recently recorded from the Gulf of Naples (SCHMEKEL, 1968) and the Gulf of Taranto (PERRONE, 1983). In the Ligurian Sea it was photographed at 15 m depth off Portofino in October 1962 by Dr. G. Pulitzer- Finali. Second, Pleurobranchus forskali has been recorded in the eastern Mediterranean Sea (BARASH & DANIN, 1977) and it is an example of Lessepsian migration. The Veliger, Vol. 28, No. 3 The observations reported here confirm that the Pleu- robranchomorpha are carnivorous. Furthermore, those species that live on hard substrata show a more highly specific diet (usually species of sponge) than the more catholic species that live on soft bottoms. ACKNOWLEDGMENTS I thank Dr. T. H. Thompson of the University of Bristol for help with the preparation of this paper and my col- league Dr. M. Pansini for identifying the sponge species. LITERATURE CITED ALTIMIRA, C., M. F. HUELIN & J. Ros. 1981. Molluscs ben- tonics de les Iles Medes (Girona). Bull. Inst. Cat. Hist. Natur. 47(sec. Zool., 4):69-75. BarasH, A. & Z. DANIN. 1971. Opisthobranchia (Mollusca) from the Mediterranean waters of Israel. Isr. J. Zool. 20: 151-200. BarasH, A. & Z. DANIN. 1977. Additions to the knowledge of Indopacific Mollusca in the Mediterranean. Conchiglie (Milano) 13(5-6):85-116. BerGH, L. S. R. 1899. Nudibranches et Marsenia provenant des campagnes de la Princesse Alice (1891-1897). Res. Camp. Sci. Monaco 14:1-45. BERTSCH, H. 1980. A new species of Tylodinidae (Mollusca: Opisthobranchia) from the northeastern Pacific. Sarsia 65: 233-237. Burn, R. 1962. On the new pleurobranch subfamily Berthel- linae (Moll.: Gastropoda); a revision and new classification of the species of the New South Wales and Victoria. Mem. Nat. Mus. Victoria 25:129-148. CaTTANEO, R. 1982. Opisthobranch molluscs of the Sorrentine Peninsula caves. Boll. Mus. Ist. Biol. Univ. Genova, 50(suppl.):376-377. DELALOI, B. & J. TARDY. 1976. Régime alimentaire et éthologie predatrice de Berthella plumula (Montagu, 1803), Moll- usque Opisthobranche. Haliotis 6:273-280. EALes, N. B. 1970. On the migration of tectibranch molluscs from the Red Sea to the eastern Mediterranean. Proc. Mal- acol. Soc. Lond. 39:217-220. Forbes, E. 1844. Report on the Mollusca and Radiata of the Aegean Sea and on their distribution, considered as bearing on geology. Rep. British Assoc. Adv. Sci. 13:130-193. HAEFELFINGER, H. R. 1960. Catalogue des Opisthobranches de la Rade de Villefranche-sur-mer et ses environs (Alpes Maritimes). Rev. Suisse Zool. 67(27):323-351. Marcus, Ev. & T. GOSLINER. 1984. Review of the family Pleurobranchaeidae (Mollusca, Opisthobranchia). Ann. So. Afr. Mus. 93(1):1-52. MAZZARELLI, G. 1891. Intorno alle specie di Plewrobranchus del Golfo di Napoli. Boll. Soc. Natur. Napoli 5(1):69-76. MoaguIin-TANDON, G. 1870. Recherches anatomiques sur l?Ombrelle de la Méditerranée. Theses pr. Fac. Sc. Paris, n° 326:1-303. PERRONE, A. 1983. Opistobranchi (Aplysiomorpha, Pleuro- branchomorpha, Sacoglossa, Nudibranchia) del litorale sa- lentino (Mar Jonio) (Elenco-contributo primo). Thalassia Salentina Taranto 13:118-144. Puiuiprl, R. A. 1844. Enumeratio molluscorum Siciliae cum viventium tum in tellure tertiaria fossilium quae in itinere suo observavit. II. Halis Saxonum 1-4:1-303. R. Cattaneo-Vietti, 1986 PruvoT-Fot, A. 1954. Mollusques Opisthobranches. Faune de France 58:1-460. PRuvoT-FoL, A. & E. FISCHER-PIETTE. 1934. Sur la Tylodina citrina et sur la famille de Tylodinidae. Bull. Soc. Zool. France 59:144-151. Ros, J. 1975. Opisthobranquios (Gastropoda: Euthyneura) del litoral iberico. Inv. Pesq. 39(2):269-372. Ros, J. & J. M. Giti. 1984. Opisthobranches des grottes sous- marines de l’Ile de Majorca (Baleares). C.I.E.S.M., XXIX* Congrés-Assemblée pléniére, Lucerne, 11-19 Octobre 1984. Session commune Benthos-Pénétration de 1’Homme sous la mer: Le peuplement des grottes:i-4. SCHMEKEL, L. 1968. Ascoglossa, Notaspidea und Nudibran- chia im litoral des Golfes von Neapel. Rev. Suisse Zool. 75(6):103-155. SCHMEKEL, L. & A. PORTMANN. 1982. Opisthobranchia des Mittelmeeres: Nudibranchia und Saccoglossa. Springer- Verlag. 410 pp. Sorpi, M. 1969. Biologia delle secche della Meloria. II. Gas- teropodi Opistobranchi. Boll. Pesca Pisc. Idrobiol. 24(2): 105-114. STARMUHLNER, F. 1969. Zur Molluskenfauna des felslitorals bei Rovinj (Istrien). Malacologia 9(1):217-242. SWENNEN, C. 1961. On a collection of Opisthobranchia from Turkey. Zool. Meded. Leiden 38(3):41-75. Page 309 TERRENI, G. & E. CAMPANI. 1980. Sul ritrovamento di due Opistobranchi della subfamilia: Pleurobranchinae e la loro problematica identificazione. Quad. Mus. St. Nat. Livorno 1:33-40. TuHompson, T. E. 1976. Biology of opisthobranch molluscs. Ray Society, no. 151:1-207. THompson, T. E. 1981. Taxonomy of three misunderstood opisthobranchs from the northern Adriatic Sea. J. Moll. Stud. 47(1):73-79. VAYSSIERE, A. 1885. Recherches zoologiques et anatomiques sur les mollusques opisthobranches du Golfe de Marseille. I. Tectibranches. Ann. Mus. Hist. Natur. Marseille 2(3): 1-181. VERANY, G. B. 1846. Descrizione di Genova e del Genovesato. Regno animale: Molluschi. Comune di Genova 1(2):90- 110. VICENTE, N. 1967. Contribution a l’étude des gastéropodes opisthobranches du Golfe de Marseille. Rec. Trav. St. Mar. Endoume 42(58):133-179. WILLAN, R.C. 1978. An evaluation of the Notaspidean genera Pleurobranchopsis Verrill and Gymnotoplax Pilsbry (Opis- thobranchia: Pleurobranchinae). J. Conchol. 29:337-344. WILLAN, R.C. 1983. New Zealand side-gilled sea slugs (Opis- thobranchia, Notaspidea, Pleurobranchidae). Malacologia 23(2):221-270. The Veliger 28(3):310-313 ( January 2, 1986) THE VELIGER © CMS, Inc., 1986 Swimming Tracks of Aplysia brasiliana, with Discussion of the Roles of Swimming in Sea Hares P. V. HAMILTON Department of Biology, University of West Florida, Pensacola, Florida 32514, U.S.A. Abstract. Surface-swimming Aplysia brasiliana were tracked at two locations in southwest Florida in order to document the magnitude of movement that can be achieved by this method of locomotion. Twenty sea hares released near a shoreline swam for a median duration of 9.9 min and traveled a median distance of 52 m. One animal swam continuously for 114 min in a weak to slack current, and traveled 953 m. Sea hares released in a lagoon influenced by tidal currents showed a tendency to swim for shorter periods in stronger currents. The swimming tracks of some sea hares were influenced by physical features in their environment. INTRODUCTION ALTHOUGH MOST GASTROPOD mollusks are exclusively benthic and locomote only by crawling, some normally benthic species occasionally swim (FARMER, 1970). Twelve of the Aplysia species recognized by EALES (1960) have been observed swimming. Sea hares swim using two large flaps (the parapodia) which project dorsolaterally from the foot. The biomechanics of swimming and its neuro- motor control are described by VON DER PORTEN et al. (1982) and previous workers. Other research has examined the orientation of swim- ming and the modulation of water speed in Aplysia bra- siliana (HAMILTON & AMBROSE, 1975; HAMILTON & RUSSELL, 1982a, b; HAMILTON, 1984). Because these studies involved analyses of only short periods (60 or 90 sec) of swimming, and because the popular neurophysio- logical model, A. californica, apparently does not swim at all, some biologists may assume that aplysiid swimming is an ecologically insignificant behavior. The literature contains few details on the duration of uninterrupted swims, and no information on tracks of uninterrupted swims or on the effect of current conditions on swimming. In order to document the role of swimming as a means of achieving significant horizontal movement in sea hares, I present here some simple descriptive data for the uninter- rupted swimming tracks of 40 animals released at two locations exposed to different current conditions. Tracks were recorded for released animals, instead of for animals found swimming naturally, so that the time and location where swimming began could be known accurately. MATERIALS anpD METHODS Sea hares, Aplysia brasiliana Rang, were studied in Char- lotte County, Florida (for map see HAMILTON & RUSSELL, 1982a:fig. 1). Adult animals (200-970 g) were collected from shallow-water grassbeds or, within 3 h after sunrise, from beaches where they had become stranded by over- night high tides. They were maintained in floating cages up to 4 d before release, and were fed Hypnea and other red algae, their natural food (KRAKAUER, 1971). The first set of releases was conducted at Mote Beach, on the northeast shore of Placida Harbor. Currents at Mote Beach are primarily influenced by wind-driven waves rather than tidal changes (HAMILTON & RUSSELL, 1982a). Tall trees grow supratidally, and a band (10-25 m wide) of sand bottom slopes gradually (<3°) from the high tide line to lower intertidal and subtidal grassbeds. During the releases water depth varied from 20 to 60 cm at the single release point within the band of sand bottom, and current speeds were slack to weak. Each animal was held on the bottom, facing offshore, for 10 sec before re- lease. The second set of releases was conducted in B3 Lagoon, which comprises one of several connections between Gas- parilla Pass (which opens to the Gulf of Mexico) and Gasparilla Sound. The Lagoon is well protected from wind and waves, but strong tidal currents occur there. Each sea hare was removed from a floating cage anchored near the shore and transported quickly by boat to a release site. Animals were released at several sites near the center of the rectangular-shaped lagoon, depending on current con- P. V. Hamilton, 1986 Table 1 Characteristics of uninterrupted swims by 20 sea hares released at Mote Beach during slack to weak current con- ditions. Mini- Maxi- Variable mum mum Median Swim duration (min) 4.5 114.0 9.9 Swim distance (m) 24 953 52 Ground speed (m/min) 225 9.2 5.3 ditions, but all sites were in water deeper than 5 m. Each sea hare was hand-placed into the water and gently agi- tated until it began to flap the parapodia and swim freely. Current speed beneath a bridge at one end of the Lagoon was Classified as strong, moderate, weak, or slack for each release. At both Mote Beach and B3 Lagoon, swimming sea hares were tracked by rowing 2-4 m behind them in a small boat. An animal had to swim for at least 3 min after release to be tracked, for reasons described in the Results. Directions were measured to the nearest 1° with an aim- able prismatic compass. Directions to two or (usually) three landmarks were recorded from the point of release, along the swimming track at 3-min intervals thereafter, and from the point where an animal stopped swimming or was lost from sight. A map of each release area was made using a USGS map for Placida, Florida (N2645- W8215/7.5) and on-site measurements of distances and directions between landmarks. Data on directions to land- marks were used to plot positions along swimming tracks. Track lengths were measured using a curvimeter. For the Mote Beach tracks, a directness (or straightness) value was computed for each track as described in HAMILTON (1977). Median values and ranges are used to summarize durations and distances of swims because the frequency distributions of both variables were positively skewed. RESULTS About 60 sea hares were released at Mote Beach during slack to weak current conditions. A frequency distribution of the swim durations for these 60 animals would show two distinct groups or modes. About 40 sea hares swam for less than 1.5 min. Grassbeds located 5 to 6 m offshore from the release point were reached by all sea hares in about 1 min, and about 40 animals immediately dove to the bottom upon reaching these grassbeds. Twenty sea hares did not dive to the bottom upon reaching the grassbeds, and swam longer than 3 min; their mode was in the 9-10 min range. The swimming tracks for the latter group of 20 sea hares are summarized in Table 1. These sea hares trav- Page 311 Figure 1 Sea hare swimming tracks may be influenced by physical fea- tures in their environment. A. Swimming tracks of the six sea hares that were released at Mote Beach and began swimming in the onshore direction. All six eventually reversed their head- ing, thus avoiding becoming stranded. Dashed lines for animals #7 and #16 indicate swimming tracks subsequent to brief stops on the bottom. B. Swimming track of one animal (#30), released in B3 Lagoon, that reversed its heading after passing beneath the bridge and swam back into the Lagoon against a weak out- going tide. This type of response has been observed in other sea hares. Scale bars are 50 m. Page 312 Table 2 Characteristics of uninterrupted swims by 20 sea hares released at B3 Lagoon during two current conditions. Mini- Maxi- Variable mum mum Median Current moderate to strong (n = 9) Swim duration (min) 3.0 33.5 7.8 Swim distance (m) 88 673 237 Ground speed (m/min) 15.6 43.5 30.3 Current slack to weak (n = 11) Swim duration (min) 6.0 87.0 21.0 Swim distance (m) 24 469 120 Ground speed (m/min) 2.9 10.1 6.7 eled at a median ground speed of 5.3 m/min for a median duration of 9.9 min, and covered a median distance of 52 m. One animal swam for 114 min and traveled 953 m from the release point. The track directness (straightness) values for 14 of the 20 sea hares were greater than 0.9, thus revealing considerable ability to maintain a relatively straight track over time. Five of the six sea hares with values less than 0.9 started swimming in the onshore di- rection, and subsequently reversed their headings (see be- low and Figure 1A). About 80 sea hares were released in B3 Lagoon. As for Mote Beach, a frequency distribution of the swim dura- tions for these 80 animals would show two distinct modes. About 60 sea hares dove to the bottom and were lost from sight within about 10 sec after release. Twenty animals did not dive immediately, and swam longer than 3 min; their mode was in the 7-8 min range. The range of current conditions occurring in B3 La- goon permitted examination of the effect of current on swimming tracks at a single location. The 20 tracked sea hares were divided into two groups according to the cur- rent conditions during their release. Their swimming tracks are summarized in Table 2. Because swimming Aplysia cannot achieve water speeds greater than about 14 m/min (HAMILTON, 1984), the median ground speed of 30.1 m/min for sea hares released in moderate to strong cur- rents clearly reflects the contribution of tidal current to horizontal displacement. Although it is not surprising that sea hares traveled at lower ground speeds and moved shorter distances when released in slack or weak currents, it is surprising that such sea hares swam for longer periods than those released in stronger currents. The swim du- rations for the moderate-strong group (median = 7.8 min) and slack-weak group (median = 21.0) are significantly different (Mann-Whitney U = 77.5; P < 0.05). The tracks of two groups of sea hares seemed to be influenced by physical features in their environment. At Mote Beach, six of the 20 tracked sea hares began swim- ming in the onshore direction (Figure 1A), but all even- tually reversed their heading and began swimming off- The Veliger, Vol. 28, No. 3 shore. Animals #6, #13, #16, and #20 all entered water shallower than about 12 cm, but did not touch bottom, before reversing their headings. Animal #6 reversed its heading at about T = 2 min, and so it had already moved back offshore when its T = 3 min position was recorded. Animal #16 stopped swimming soon after heading off- shore. At B3 Lagoon, undisturbed sea hares swimming on an outgoing tide have often been observed passing beneath the bridge at one end of the Lagoon. Although such ani- mals do not appear to have similar headings when they are still some distance from the bridge, many swing around and adopt an up-current heading as they get about 10-15 m from the bridge. Despite their swimming efforts, a strong tidal current carries them backward, beneath the bridge and out into Gasparilla Pass. Five of the 20 sea hares tracked in B3 Lagoon passed beneath bridges, and adop- tion of an up-current heading was observed in two of the five. The heading change was quite striking for animal #30, which was released on a weak outgoing tide, and which eventually swam back into the Lagoon from be- neath the bridge (Figure 1B). Thirty-eight of the 40 sea hares tracked at Mote Beach and B3 Lagoon dove to the bottom or were lost from sight in water 1-5 m deep. Most of these animals began making short “excursion” dives down to as much as 50-100 cm beneath the surface, a few minutes before their final dive or loss from sight. Although the rate of parapodial flap- ping was not recorded for any animal, sea hares seemed to swim normally during both the descending and ascend- ing phases of excursion dives. DISCUSSION Sea hares released at both study sites formed two distinct and natural groups according to swimming time: those which swam only until they reached a nearby grassbed (less than 90 sec) or which immediately dove to the bot- tom, and those which swam longer than 3 min. The pur- pose of this study was to document how far and how long Aplysia brasiliana are capable of swimming. Consequently, this study focused on the behavior of the second group. The data presented here should be considered minimum estimates of the swimming capabilities of the tracked sea hares. Wave height, sky conditions, water turbidity, and swimming depth all influenced how long an animal could be kept in view. All distances traveled by the tracked sea hares are underestimated because positions were recorded only once every 3 min, and the actual paths during these 3-min intervals were never perfectly straight. The influence of current speed in B3 Lagoon on ground speed and swimming distance was expected, and is prob- ably due to passive displacement effects of current on swimming animals. However, the significant influence of current speed on swimming duration must involve an ac- tive response by animals, and this suggests an ability to detect current speed. It would be interesting to learn how current speed is detected. Sea hares were released from a P. V. Hamilton, 1986 drifting boat, so they were essentially up-to-speed with the water mass from the time they commenced swimming. The Thrust Modulation Response (TMR) of Aplysia brasiliana involves detection of current direction, and it is influenced by current speed (HAMILTON, 1984). The heading reversals of those sea hares that almost stranded on the shore at Mote Beach (Figure 1A), and the up-current headings adopted by some sea hares as they pass beneath bridges in B3 Lagoon (Figure 1B), may both depend on visual detection of objects above the water’s surface (e.g., bridge, treeline). HAMILTON & RUSSELL (1982b) demonstrated that, under field experimental con- ditions, an unblocked view of the sky is required for sea hares to maintain a consistent swimming direction. Lit- torina irrorata, a gastropod possessing eyes of a similar design yet half the size of those possessed by Aplysia, can detect bar-shaped targets filling as little as 1° of visual arc (HAMILTON & WINTER, 1982; HAMILTON et al., 1983). Regardless of the mechanisms involved in the heading reversals of sea hares at Mote Beach, this response sug- gests that those few animals that oriented onshore during previous studies (HAMILTON & AMBROSE, 1975; HAM- ILTON & RUSSELL, 1982a) eventually would have turned around had they been allowed to swim for longer than 60 or 90 sec. The adaptive function of swimming in Aplysia has not been studied systematically, but several hypotheses exist. In at least some other opisthobranchs (e.g., EDMUNDS, 1968), swimming seems to serve as an escape response from benthic predators. However, the sea hare’s swim- ming capabilities seem far too sophisticated for just this function, and no evidence supports this hypothesis exclu- sively. It is clear that sea hares that are stranded on grad- ually sloping beaches or sand bars can use swimming to move into deeper water, if they are resubmerged by a subsequent high tide before succumbing to desiccation and insolation stresses. Swimming probably enables sea hares to move within or between grassbeds on a daily basis. The adaptive value for such movements could involve searches for prospective mates, concentrations of algal food, or more suitable physicochemical conditions. Finally, swimming may facilitate seasonal migration in Aplysia brasiliana. An incursion of sea hares into shallow water occurs in south Florida during the early spring (HAMILTON et al., 1982), the period of peak algal abun- dance. By late summer, sea hares are not found in shallow water, where water temperatures approach or exceed the lethal limit (27-31°C). A similar seasonal pattern of A. brasiliana abundance in shallow water was reported for the southern hemisphere by SAwAyA & LeEAnHy (1971), who also suggested a migration hypothesis. Seasonal movements between deep and shallow water could be as- Page 313 sisted by directionally advantageous tidal currents. Direct evidence of migration, in the form of movement records of tagged individuals, is lacking for all opisthobranchs and most other mollusks suggested to migrate (HAMILTON, 1985). ACKNOWLEDGMENTS I thank Dr. C. N. D’Asaro, R. Haywood, and T. Pickard for useful discussions. Support was provided by NSF Grants BNS79-16358 and BNS83-08186. LITERATURE CITED EA.Es, N. B. 1960. Revision of the world species of Aplysia (Gastropoda: Opisthobranchia). Bull. Brit. Mus. Natur. Hist. Zool. 5:268-404. EpMuNDs, M. 1968. On the swimming and defensive response of Hexabranchus marginatus (Mollusca, Nudibranchia). J. Linn. Soc. Lond. Zool. 47:425-429. FARMER, W. M. 1970. Swimming gastropods (Opisthobran- chia and Prosobranchia). Veliger 13(1):73-89. HaMILTON, P. V. 1977. The use of mucous trails in gastropod orientation studies. Malacol. Rev. 10:73-76. HAMILTON, P. V. 1984. Factors influencing the water speed of swimming sea hares, Aplysia brasiliana. Anim. Behav. 32: 367-373. HAMILTON, P. V. 1985. Migratory molluscs, with emphasis on swimming and orientation in the sea hare, Aplysia. Pp. 212-216. In: M. A. Rankin (ed.), Migration: mechanisms and adaptive significance, Suppl. to Contrib. Mar. Sci., Vol. 27. HaMILTON, P. V. & H. W. AMBRosE. 1975. Swimming and orientation in Aplysia brasiliana (Mollusca: Gastropoda). Mar. Behav. Physiol. 3:131-144. HaMILTON, P. V., S. C. ARDIZZONI & J. S. PENN. 1983. Eye structure and optics in the intertidal snail, Littorina irrorata. J. Comp. Physiol. 152A:435-445. HamMILTOoN, P. V. & B. J. RUSSELL. 1982a. Field experiments on the sense organs and directional cues involved in off- shore-oriented swimming by Aplysia brasiliana Rang (Mol- lusca: Gastropoda). J. Exp. Mar. Biol. Ecol. 56:123-143. HamMILTon, P. V. & B. J. RUSSELL. 1982b. Celestial orienta- tion by surface-swimming Aplysia brasiliana Rang (Mollus- ca: Gastropoda). J. Exp. Mar. Biol. Ecol. 56:145-152. HaMILTon, P. V., B. J. RUSSELL & H. W. AMBROSE. 1982. Some characteristics of a spring incursion of Aplysia brasi- liana into shallow water. Malacol. Rev. 15:15-19. HAMILTON, P. V. & M. A. WINTER. 1982. Behavioural re- sponses to visual stimuli by the snail Littorina irrorata. Anim. Behav. 30:752-760. KRAKAUER, J. M. 1971. The feeding habits of aplysiid opis- thobranchs in Florida. Nautilus 85:37-38. Sawaya, P. & W. M. Leany. 1971. Fisioecologia e etologia de Aplysia L. (Mollusca—Opisthobranchia). Boletim de Zool. e Biol. Marinha, N.S. 28:1-17. VON DER PorTEN, K., D. W. Parsons, B. S. ROTHMAN & H. PINSKER. 1982. Swimming in Aplysia brasiliana: analysis of behavior and neuronal pathways. Behav. Neural Biol. 36:1-23. The Veliger 28(3):314-317 ( January 2, 1986) THE VELIGER © CMS, Inc., 1986 A Short-term Study of Growth and Death in a Population of the Gastropod Strombus gibberulus in Guam by GEERAT J. VERMEIJ anp EDITH ZIPSER Department of Zoology, University of Maryland, College Park, Maryland 20742, U.S.A. Abstract. A month-long study of tagged Strombus gibberulus gibbosus Roding, 1798, in Pago Bay, Guam, during the spring of 1981 revealed very high mortality (at least 11%), most of which (93%) was due to shell breakage. No differences in shell size, lip thickness, or age were detectable among snails that died, sustained sublethal injury, and survived unscathed. It is concluded that the thickened adult lip was ineffective in conferring resistance to shell-breaking agents (chiefly xanthid crabs) in this population of S. gibberulus. Juvenile growth rates were very high (mean 0.16 mm/day). Snails cease to grow in length once the lip assumes the flared adult form. INTRODUCTION SMALL STROMBID GASTROPODS pose a curious enigma in that they sustain heavy mortality due to breakage despite the strongly thickened and flared outer lip that character- izes the adult shell. Inspection of “dead” shells (including fragments) suggests that breakage accounts for 50 to 100% of the mortality in most populations of Strombus gibberulus Linnaeus, 1758, in the tropical western Pacific (VERMEIJ, 1979, 1982). Although the thick lip proved to be an ob- stacle to the predaceous crab Calappa hepatica (L.) in the laboratory, it appeared to be ineffective against other crabs such as Carpilius maculatus (Linnaeus) and Daldorfia hor- rida (Linnaeus) (ZIPSER & VERMEIJ, 1978; VERMEIJ, 1982). If such powerful agents of breakage are commonly encountered by S. gibberulus, how is the thick lip main- tained by selection, and how can populations persist in the face of heavy mortality? These are the questions that prompted the present short-term study. MATERIALS anpD METHODS We studied a sample of 230 individuals of Strombus (Gib- berulus) gibberulus gibbosus Roding, 1798, from the in- shore part of the reef flat in a cove near the center of Pago Bay, Guam, Mariana Islands. The study area is covered by at least 3 cm of water at all times and is floored by a thin layer of white sand. There are several intertidal raised limestone ledges with undercut margins. Strombus gibber- ulus is found in and on the sand. The dorsum of the adult shell is always exposed above the sand, and is frequently fouled and pitted by small algae. All individuals observed during the week of 21-28 May 1981 were collected and marked, and were subsequently censused periodically until final recovery from 21-25 June. We measured shell length (distance from apex to tip of siphonal canal) and lip thickness (at a point midway be- tween the eye-notch and the posterior end of the lip). Each shell was inspected for the presence of repaired injuries (scars) on the body whorl. These scars appear as irregular traces that depart from the normal course of the growth lines. They result from damage to the outer lip, which is repaired by the mantle edge before growth recommences. In order to mark the snails, two methods were used. First, the anterior dorsal shell surface was filed clean, and a numbered beetag (Fabrik fur Bienenzuchtgerate, 7056 Weinstadt-Endersbach, West Germany) was affixed with underwater epoxy. The dorsal and ventral surfaces of the spire were also filed and labeled with a permanent mark- er. Secondly, a spot of nail polish was painted on the ventral surface of the penultimate whorl, so that individ- uals that had lost their tags could still be recognized as belonging to the marked sample. If their measurements matched those of a labeled individual that had not hitherto been recovered, the tagless snails were retagged with that number. After the epoxy had dried sufficiently in air, the animals were returned to the field site for release. The 3-h period of exposure to air that was required for label- ing apparently had no adverse effect, for snails began to kick actively as soon as they were placed in buckets of seawater for transport back to the field. The incidence of scars was assessed in two ways. The proportion of repaired individuals was calculated as the G. J. Vermeij & E. Zipser, 1986 Table 1 Characteristics of the Strombus gibberulus population at the time of marking. Characteristic Juveniles Adults Number tagged 21 209 Proportion of repaired snails 38% 17% Frequency of repair 0.52 0.18 Lip thickness 0.2-0.8 mm 1.3-2.5 mm number of snails with scars on the body whorl divided by the total number of snails in the sample. The frequency of repair is the number of scars divided by the number of individuals in the sample. The latter figure is by definition higher than is the former if some members of the scarred population have multiple scars. If large size and a thick lip conferred resistance to shell- breaking agents in the field, individuals surviving un- scathed or in damaged condition would be expected to be larger and to have thicker lips than would individuals that had died as a result of shell breakage. This hypothesis was tested against the null hypothesis of no difference by comparing survivors with individuals recovered as broken shells with respect to the shell length and lip thickness of these individuals at the time of labeling. Because lip thick- ness in adults was independent of shell length, t-tests were used throughout. RESULTS Characteristics of the Population Most of the population of Strombus gibberulus is com- posed of thick-lipped adults (Table 1). Shell repair on the body whorl was much more frequent in juveniles than in adults (Table 1). Two factors may contribute to this dif- ference. In the first place, adult lips are less susceptible to sublethal damage than are thin delicate lips of juveniles. Secondly, the scars on the body whorl of a juvenile come to lie on the penultimate whorl or even in the spire whorls of adults, and would thus not be counted in the adult shell. The overall frequency of repair in the population (0.21) is similar to that in several other Guamanian populations of S. gibberulus, but it is higher than are the frequencies of repair in populations of this species from other parts of the tropical western Pacific (VERMEIJ, 1982). Given the high incidence of scars in the Pago Bay population, the potential for selection in favor of breakage-resistant traits is high. Mortality and Its Causes We recovered 89 tagged living snails one month after the 230 original animals had been released. Another 7 snails with apical marks but without their numbered tags were also recovered, so that a total of 96 individuals (42% Page 315 Table 2 Comparison of dead and surviving members of the Strom- bus gibberulus population one month after initial marking. Means are given with standard deviations. Characteristic Dead individuals Survivors Shell length at time of marking (mm) 36.21 + 3.41 36.89 + 3.32 Lip thickness at time of marking (mm) 1.71 + 0.38 1.85 + 0.33 Percentage of juveniles 6.3% 7.3% Frequency of repair 0.24 0.22 of the original sample) was found again after one month. The percentage of recapture of juveniles (33%) was not significantly lower than that of adults (43%). Two sur- vivors (one juvenile and one adult) sustained sublethal shell injury during the period of observation. Of the 134 individuals not recovered alive after one month, 26 were found dead during our study. The mini- mum mortality for the period of observation was, there- fore, 11%. Most (93%) of the dead, recovered individuals had been crushed. The 17 broken shells that we were able to identify by tag did not differ from individuals that were recovered alive after one month in shell length, lip thick- ness, proportion of juveniles, or frequency of repair (Ta- ble 2). The ratio of juveniles to adults in the sample of broken shells was 1/16 = 0.063, as compared to 7/96 = 0.073 among survivors. The xanthid crabs Carpilius maculatus and Eriphia se- bana (Shaw & Nodder) live under ledges and large boul- ders in the study area, and appear to have been largely responsible for the breakage-related deaths of Strombus gibberulus. Nearly all broken shells had the ventral or dorsal portion of the body whorl broken away, or had severed spires, and the lip was frequently either missing or cut in half. These types of damage are typical of the prey of large xanthids with molarlike dentition in the crusher claw (ZIPSER & VERMEIJ, 1978). The spiral peel- ing characteristic of the prey of Calappa hepatica, which also occurs in the study area, was observed in only one dead individual. The porcupinefish Diodon hystrix (Lin- naeus) hunts over the Pago Bay reef flat at night, and we have recovered fragments of S$. gibberulus from its gut con- tents. The broken shells that we recovered in the field, however, were too large to have been passed through the digestive system and defecated by the porcupinefish, so that the contribution of that predator to the mortality of S. gibberulus could not be established. Another agent of mortality whose impact we were un- able to assess is man. QuoYy & GAIMARD (1834:67) al- ready observed that Strombus gibberulus was a favorite item of food for the people of Guam. Toward the end of our study, we encountered a girl, about ten years old, clutching six labeled snails and many unlabeled individ- Page 316 Table 3 Growth rates of Strombus gibberulus juveniles and adults over a period of 30 days. Means are given with standard deviations; numbers of individuals are given in parenthe- ses. Characteristic Juveniles Adults Growth in shell length (mm) 3:33 22114) (9) —0.035 + 0.3 (81) Growth in lip thickness (mm) 0.81 + 0.53 (9) 0.063 + 0.21 (81) Daily growth in length (mm) 0.16 + 0.072 (7) uals. We persuaded her to release the marked individuals, but the possibility that many of our labeled snails died in the soup-pot or on the grill cannot be eliminated. Growth Measurements of labeled individuals revealed that ju- venile growth rates in Strombus gibberulus were extremely high, whereas growth ceases once the adult lip has as- sumed the adult configuration (Table 3). In two juveniles whose lip edge was marked, growth rate in the spiral direction was calculated to be 0.6 and 0.7 mm/day. At these rates, hatchlings could be expected to reach a length of 30 mm (minimum adult length in the Pago Bay pop- ulation) in about six months. Because small juveniles probably grow even faster than did the rather large ju- veniles we monitored, all of which exceeded 27 mm in initial length (Table 1), the time required to reach mini- mum adult size may well be substantially shorter. DISCUSSION Although the high incidence of scars in the Pago Bay population of Strombus gibberulus suggests a high potential for selection in favor of breakage-resistant traits such as a thick lip, we were unable to detect differences in shell length or lip thickness between snails that survived after one month and those that we recovered as dead broken shells. If selection in favor of resistance to breakage took place, it involved characters we did not examine. The thick lip may function as a deterrent to predation in other populations of Strombus gibberulus. The species is common in grassbeds and on sand flats where boulders under which large shell-crushing crabs find shelter are absent. We suspect that the lip is chiefly effective against calappid crabs, which live in sand, and that the boulder- strewn sand patches of the Pago Bay reef flat are very different in terms of the types of predators encountered by S. gibberulus from inshore environments where rocky bottoms are absent. The Pago Bay population may well persist not by virtue of morphological adaptation, but by The Veliger, Vol. 28, No. 3 exceptionally high juvenile growth rates and presumably by high rates of larval settlement, although the latter point requires confirmation. It is also possible that the thick lip is not adaptive in resisting predators in any population of Strombus gibber- ulus, and that it is instead retained as a legacy of a well- established pattern of determinate growth that was fixed very early in the history of the Strombidae. The combi- nation of determinate growth, modified adult outer lip, and short post-larval life-span is found not only in S. gibberulus, but in many other small Indo-west-Pacific strombids and tropical cerithiids as well (HOUBRICK, 1974). Like the larger and longer-lived members of these and other gastropod families, these small species have high growth rates in the juvenile phase. Both adult size and juvenile growth rate may vary by a factor of two to three among individuals from the same site (RANDALL, 1964; FRANK, 1969; SPIGHT et al., 1974; YAMAGUCHI, 1977; VERMEIJ, 1980). If, as seems likely from studies of other gastropods, growth and size are partially determined by genetic factors, these tropical species should display sub- stantial genetic variation on which selection could act. We realize that the present study concerns only one population during one month in a possibly atypical hab- itat. It will be interesting to repeat our study in environ- ments in which calappid crabs are the chief predators, and to monitor populations over the course of an entire year or more. ACKNOWLEDGMENTS We thank A. R. Palmer and two anonymous reviewers for constructive criticism of this manuscript, and the staff of the Guam Marine Laboratory for the use of their facilities. A grant to Vermeij from the Program of Bio- logical Oceanography, National Science Foundation, sup- ported the present research. LITERATURE CITED FRANK, P. W. 1969. Growth rates and longevity of some gas- tropod mollusks on the coral reef at Heron Island. Oecologia 2:232-250. Houprick, R.S. 1974. Growth studies on the genus Cerzthium (Gastropoda: Prosobranchia) with notes on ecology and mi- crohabitat. Nautilus 88:14-27. Quoy, J. R. C. & J. P. GaiMarD. 1834. Voyage de décou- vertes de |’Astrolabe. Zoologie, 3¢me tome. J. Tastu: Paris. RANDALL, J. E. 1964. Contribution to the biology of the queen conch, Strombus gigas. Bull. Mar. Sci. Gulf Caribbean 14: 246-295. SpiGHT, T. M., C. BIRKELAND & A. Lyons. 1974. Life his- tories of large and small murexes (Prosobranchia: Murici- dae). Mar. Biol. 24:229-242. VERMEIJ, G. J. 1979. Shell architecture and causes of death in Micronesian reef snails. Evolution 33:686-696. VERMEIJ, G. J. 1980. Gastropod growth rate, allometry, and adult size: environmental implications. Pp. 379-394. Jn: D. C. Rhoads & R. A. Lutz (eds.), Skeletal growth of aquatic G. J. Vermeij & E. Zipser, 1986 organisms: biological records of environmental change. Ple- num: New York. VERMEYJ, G. J. 1982. Gastropod shell form, repair, and break- age in relation to predation by the crab Calappa. Malaco- logia 23:1-12. YAMAGUCHI, M. 1977. Shell growth and mortality rates in the Page 317 coral reef gastropod Cerithium nodulosum in Pago Bay, Guam, Mariana Islands. Mar. Biol. 44:249-263. ZipsER, E. & G. J. VERMEIL. 1978. Crushing behavior of tropical and temperate crabs. J. Exp. Mar. Biol. Ecol. 31: 155-172. The Veliger 28(3):318-327 ( January 2, 1986) THE VELIGER © CMS, Inc., 1986 Aspects of the Reproduction of Rocky Intertidal Mollusks from the Jordan Gulf of Aqaba (Red Sea) by NEIL C. HULINGS Marine Science Station, P.O. Box 195, Aqaba, Jordan Abstract. Reproductive and spawning periodicity, type of spawn, and spawning behavior of an intertidal chiton, 11 prosobranch gastropods, and one pulmonate gastropod, and two pelecypods from the Jordan Gulf of Aqaba are given. Reproduction was continuous in seven species, restricted to the warmer period and lowered sea level in seven, and in one species occurred during the colder and higher sea level period. No direct relationship between temporal reproduction and tide level or vertical position in the intertidal zone was seen. INTRODUCTION THE LITTLE INFORMATION available on the fauna of the rocky intertidal of the Red Sea, including the Gulf of Aqaba, is concerned primarily with zonation (SAFRIEL & LIPKIN, 1964; FISHELSON, 1971; AYAL & SAFRIEL, 1980; SAFRIEL et al., 1980). SAFRIEL (1969) reported on various aspects of the ecology of Nerita spp. and JORNE & SAFRIEL (1979) on the behavior of Nerita polita Linnaeus. Data on the reproduction of Red Sea intertidal mollusks are lack- ing except for very limited data given by SAFRIEL (1969) on Nerita polita and FAO (1972) on Ostrea forskali Chem- nitz. Various aspects of the reproduction of 15 intertidal mollusks from the Jordan coast of the Gulf of Aqaba have been investigated. The aspects include reproductive and spawning periodicity, type of spawn, and spawning be- havior. The relationships of these to external environ- mental factors including position in the intertidal zone, temperature, changes in sea level, and primary production are discussed. MATERIALS anD METHODS A minimum of 10 specimens of each species was collected around the middle of each month for at least 12 months. The species were usually collected from the same locality each month, in some cases from two different localities. Collections were spatially random but biased toward larg- er sizes to ensure obtaining sexually mature individuals. An investigation of the minimum size of sexual maturity of most species was conducted. In the laboratory, specimen length was measured to the nearest 0.1 mm using vernier calipers; microscopic mea- surements to the nearest 0.01 mm were made with an ocular micrometer. The shells of the gastropods were cracked using a ham- mer, those of the bivalves opened, the foot of the chiton removed, and the whole animal of limpets removed to examine the gonads under a dissecting microscope. Either teasing or microdissection of the gonads and/or associated structures was employed to determine the presence or ab- sence of gametes. When necessary, fresh preparations were examined under a compound microscope. Additional investigations including ones on the depo- sition of eggs and hatching were conducted in the labo- ratory for some of the species. Specimens were kept in individual aerated seawater aquaria. The terminology used for the zones in the rocky inter- tidal is that of STEPHENSON & STEPHENSON (1949) and SAFRIEL & LIPKIN (1964). The littorinid zone of Safriel & Lipkin is within the supralittoral fringe of Stephenson & Stephenson; the chthamalid and Tetraclita zones of Sa- friel & Lipkin are within the midlittoral zone of the Ste- phensons. FISHELSON (1971) included the entire rocky in- tertidal of the Red Sea in the infralittoral and referred to it as the Tectarius armatus-Tetraclita squamosa rufotincta community. Adult specimens of all 15 species studied as well as egg capsules, masses, and ribbons are deposited in the refer- ence collection of the Marine Science Station, Aqaba, Jor- dan, and are available for examination upon request. In addition, voucher specimens of all species have been de- posited in the Division of Mollusks, U.S. National Mu- seum of Natural History, Washington, D.C. N. C. Hulings, 1986 Page 319 DESCRIPTION oF THE ROCKY INTERTIDAL ZONE The region of the Jordan Gulf of Aqaba is within the very warm portion of the Saharan bioclimatic zone. The terrestrial component exerts greater influences on the in- tertidal and shallow marine zones than vice versa. The greater influences are due to the limited areal extent of the Gulf with respect to the surrounding terrestrial area, a narrow and clear-cut interface between the terrestrial and marine environments, land-to-Gulf hot and dry winds reducing transport of moisture in the opposite direction, and low rainfall. The prevailing winds are N-NNE, with Beauforts 2 through 4 occurring 84% of the time (HULINGs, 1979). Mean air temperatures range from about 16°C in January to 32°C in August (Jordan Meterological Department) and surface water temperatures from 20°C in February to 27°C in August-September (Morcos, 1970). In the absence of river runoff and in combination with low rain- fall (35 mm/yr, Jordan Meterological Department) and dry and hot winds, the evaporation rate is high (up to 4 m/yr, ANATI, 1976) resulting in high, constant salinity of 40.5 to 41.0 ppt (PALDoR & ANATI, 1979). There is a major period of primary productivity during December- January and a minor period during May-June (HULINGS & ABU HILAL, 1983). LEVANON-SPANIER et al. (1979) consider the northern Gulf of Aqaba to be oligotrophic from April through November. According to Morcos (1970) the tides of the Gulf of Aqaba are influenced by those of the Red Sea proper and the direct effects of the moon and sun are comparatively small. In addition, the tides are usually out of phase with the moon (Hulings, unpublished data). The tide in the Jordan Gulf is mixed, with diurnal inequality of the highs averaging 4.2 cm and that of the lows 4.7 cm (Hulings, unpublished data). The spring tide range averages about 1.0 m, the range of the neaps about 50 cm (FISHELSON, 1973; Hulings, unpublished data). Fluctuation of sea level, up to 1 m, occurs annually, being higher during the period December through May and lower from July through October (FISHELSON, 1973; Hulings, unpublished data). Although the vertical ranges and variations in tides are small, the net effect of the level and duration of submer- gence and emergence on the generally low profile beaches is considerable. The substratum of the rocky intertidal includes boul- ders (ranging from granitic to huge masses of conglom- erate or fossil reef), mostly medium to large-sized granitic and dike multi-colored pebbles underlain by sand, and platform or slab. The latter is extremely variable and includes sandstone, conglomerate (calcium carbonate ce- mented sand, gravel, and pebble mixtures) and beach rock (usually gravel sized and continuously forming). Eroded fossil coral reefs, composed mostly of a variety of coral heads surrounded by solidified calcium carbonate detritus (FRIEDMAN, 1965), are also included as part of the slab substrata or platform beaches. It is common to find mix- tures of two or more of the slab substrata at any particular locality. The horizontal profile of the slab beaches is gen- erally very gentle, whereas that of the pebble beaches is somewhat steeper. In addition, wave action is minimal on the usually protected slab beaches and moderate on the more exposed pebble beaches. ASPECTS oF REPRODUCTION Class Polyplacophora Family CHITONIDAE Acanthopleura haddoni Winkworth, 1927 Habitat: Most common in the Tetraclita (midlittoral) zone. Specimens examined: 375 averaging 50 mm long from June 1982 through May 1984 for state of reproduction; 39 specimens 16-42 mm long for minimum length at which gonads appear. Sex ratio: 1.0 male:0.8 female. Reproduction: The testes are dark red when immature, pinkish when mature; immature ovaries are tan, mature ovaries dark to black brown. Gametes were present in both sexes from June through December. Among the fe- males, all were ovigerous from September through No- vember; during other months, ovigerous and nonovigerous individuals occurred. Ova with a chorion (KUME & KaTsuMA, 1957) were found from August through No- vember. The majority of the males had enlarged testes with sperm from June through October. The major period of reproduction (the time that most specimens of both sexes had gametes) is from June through October. The smallest female with ovaries was 24 mm long, the smallest male with testes 27 mm long. Class Gastropoda, subclass Prosobranchia Family PATELLIDAE Cellana radiata (Born, 1778) Nomenclature: According to Dr. J. Rosewater (personal communication), Cellana radiata is synonymous with and has priority over C. rota (Gmelin, 1791) reported by Sa- FRIEL & LIPKIN (1964), FISHELSON (1971), MERGNER & SCHUHMACHER (1974), and MASTALLER (1979). Habitat: Characteristic of the Tetraclita (midlittoral) zone. Specimens examined: 597 averaging 33.0 mm long (range, 16.7-48.0) from March 1982 through May 1984 for state of reproduction; 129 specimens 3.6-23.5 mm long for minimum length with gonads. Sex ratio: 1.0 male:1.0 female. Page 320 Reproduction: Gametes were present in most individuals of both sexes each month. The smallest male was 9.5 mm long, the smallest female 12.0 mm; the smallest male with sperm was 10.2 mm long, the smallest female with ova 12.0 mm long. Comments: Rao (1973) reported a sex ratio of 1.0 male: 0.8 female in Cellana radiata from tropical southeast India. Developing or spawning gonads were present each month during a 12-month period with spawning occurring from June to February or March. Sexual maturity occurred at a length of 9 mm, and sex distinction could be made in individuals 10 mm long. RAo (1976) later reported con- tinuous breeding in C. radiata and sexual maturity at a shell length of 10-15 mm. Family TROCHIDAE Monodonta dama (Philippi, 1848) Habitat: A mobile midlittoral species ranging from above the chthamalid zone to the Jetraclita (midlittoral) zone. Specimens examined: 296 from April 1982 through May 1983 for state of reproduction; 62 specimens 5.7-12.3 mm high for minimum size with gonads. Sex ratio: 1.0 male:1.1 female. Reproduction: Dark green ovaries with ova and cream- colored testes with sperm occurred each month. The smallest male with testes was 8.1 mm long, the smallest female with ovaries 9.0 mm long and was ovigerous. Spawn: Spawning was observed on several occasions in a seawater table with continuously circulating seawater. Most spawning individuals were paired or in clusters, and masses of gametes were released in spurts by both sexes. The ova, averaging 0.13 mm in diameter, were surround- ed by a membrane averaging 0.15 mm in diameter. Spawning occurred irrespective of lunar or tidal cycles. Early development: Veligers averaging 0.19 mm long and lacking eyes and an operculum appeared within 24 h after spawning. Family NERITIDAE Nerita forskalu Recluz, 1844 Nomenclature: This species has been reported from the Jordan Gulf of Aqaba as Nerita sanguinolenta Menke, 1820, by MERGNER & SCHUHMACHER (1974) and N. al- bicilla Linnaeus, 1758, by MASTALLER (1979). SAFRIEL (1969) noted that the species is N. forskali based on shell morphology. Habitat: Has a wide vertical distribution on a variety of solid substrata and in varied microhabitats, including tide pools, on pebble beaches, and slab, within the Tetraclita (midlittoral) zone. Occurs lower than the other of the sym- patric pair, Nerita polita Linnaeus. The Veliger, Vol. 28, No. 3 Specimens examined: 350 averaging 18.4 mm _ long (range, 7.7-24.8) from April 1982 through May 1983 for reproductive state; 41 specimens 5.9-11.6 mm long for presence of gonads. Sex ratio: 1.0 male:1.1 female. Reproduction: Ova and spermatophores, the latter av- eraging 6.3/female (range, 0-16), and sperm were present in the seminal vesicle (BERRY et al., 1973) each month. The smallest male with testes was 7.7 mm long, the small- est female with ovaries 8.7 mm long. Copulation was observed in the field every month ex- cept August and September and occurred only during sub- mergence. The copulating behavior and the action of the cephalic penis were essentially the same as that described by IRIKI et al. (1963). Spermatophores were occasionally found in aquaria indicating unsuccessful copulation. Spawn: The egg capsules have the basic structure de- scribed by ANDREWS (1935). They are reddish brown in color and the presence of the irregular shaped, dark red spherulites give the capsules a faceted appearance. The capsules are more elliptical than round, averaging 2.2 mm wide X 2.8 mm long. Egg capsule deposition was noted during July, October, and January through May; there were no observations during the other months. The capsules were found on a wide variety of exposed surfaces including glass, metal, shells of other gastropods (living and dead), and the girdle of chitons, as well as on rocky surfaces from small pebbles to slab. The deposition of capsules on a wide variety of surfaces coincides with its wide-ranging distribution and the fact that it usually does not go into hiding during emergence. Early development: Various stages of embryonic devel- opment from uncleaved ova, averaging 0.15 mm in di- ameter, to veligers with eyes, averaging 0.20 mm long, were found enclosed within the membrane that lined the capsule. The average number of ova-veligers per capsule was 117 (range, 60-211). Hatched veligers with eyes and operculum averaged 0.21 mm long. Exit of the veligers from the capsule ap- peared to be through the top. Nerita polita Linnaeus, 1758 Habitat: Occurs above Nerita forskalu in the midlittoral, approximately equivalent to the chthamalid zone and is restricted to areas of pebbles underlain by sand. Specimens examined: 347 averaging 17.7 mm long (range, 9.0-22.8) from April 1982 through May 1983 for state of reproduction; 24 specimens 5.7-11.0 mm long for minimum size with gonads. Sex ratio: 1.0 male:1.4 female. Reproduction: Ova and spermatophores (the latter av- eraged 5.8/female, range, 0-20) and sperm in the seminal N. C. Hulings, 1986 vesicle (BERRY et al., 1973) were present each month. The smallest male with testes was 8.0 mm long, the smallest female with ovaries 9.6 mm long. Copulation was observed in the field from May through July and September through March and was essentially like that described by IRIKI et al. (1963) in terms of the behavior and action of the cephalic penis. Copulation oc- curred only among emergent pairs and during day and night. SAFRIEL (1969) and HUGHEs (1971) reported cop- ulation in Nerita polita from May through the beginning of August during investigations from April through the beginning of August. Spawn: The egg capsules, having the basic structure de- scribed by ANDREWS (1935), are white due primarily to the irregular and small spherulites, and have a finely granulated appearance. They are circular, averaging 1.7 mm in diameter. The average number of ova—veligers per capsule was 35 (range, 22-55). Capsule deposition was noted from May through July and during December; deposition is, however, probably more frequent. The capsules are deposited in groups on the sides of large pebbles that are well-anchored in the sand. Deposition is always beneath the surface of the sand and on relatively smooth surfaces. The site of deposition is in keeping with the snail’s behavior of burrowing into sand and attaching, by means of the foot, to the sides of pebbles during maximum ebb and maximum flood tides. Emergence occurs during flooding and ebbing tides during day and night. Variable patterns of behavior in Nerita polita, ranging from primarily nocturnal (SAFRIEL, 1969; HUGHES, 1971) to no nocturnal activity (ZANN, 1973), have been reported. Early development: Uncleaved ova, averaging 0.19 mm in diameter, through veligers, averaging 0.25 mm long, with eyes and an operculum were observed enclosed in the membranous lining of the capsule. Veligers apparently leave the capsule near the base. Hatched veligers averaged 0.25 mm long and were kept alive for 14 d, during which no significant morphological changes occurred. Family LITTORINIDAE Littorina scabra scabra (Linnaeus, 1758) Nomenclature: According to Dr. J. Rosewater (personal communication) the species and subspecies names are pro- visional, pending the publication of a revision by Dr. D. Reid. Habitat: Found at only one locality, on boulders in the main port of Aqaba. It occurred in the transitional litto- rinid—chthamalid (supralittoral fringe—-midlittoral) zones from May 1982 through January 1983. Specimens examined: 61 averaging 12.6 mm long (range, 7.2-18.1). Sex ratio: 1.0 male:1.3 female. Page 321 Reproduction: The ovoviviparous females were ovigerous and the males contained sperm in the vas deferens each month. Early development: The ctenidial brood pouch (ROSE- WATER, 1970) contained various stages of development during each month except September and January. The stages ranged from early cleavage to free, eyeless veligers averaging 0.11 mm long. Nodilittorina millegrana (Philippi, 1848) Nomenclature: ROSEWATER (1970) states that Littorina novaezelandiae Reeve, 1857, reported by SAFRIEL & Lip- KIN (1964) and Littorina urieli described by B1GGs (1966) from Eilat, Israel, are synonyms of Nodilittorina mille- grana. Habitat: Characteristic of the littorinid (supralittoral fringe) zone and most common on near vertical surfaces of boulders. It is the second highest mollusk in vertical distribution in the rocky intertidal zone. Specimens examined: 371 averaging 9.0 mm long (range, 3.2-14.3) from April 1982 through May 1983 for repro- ductive state; 270 specimens 1.0-6.5 mm long for mini- mum size of sexual maturity. Sex ratio: 1.0 male:1.7 female. Reproduction: Males with sperm in the vas deferens and females with ova in the oviduct were found each month. The smallest male possessing a rudimentary penis was 2.5 mm long and the smallest with a fully developed penis and sperm in the vas deferens was 3.3 mm. The smallest ovigerous female was 4.1 mm long. Spawn: Nodilittorina millegrana produces pelagic egg cap- sules. Each capsule contains a single egg, is transparent and dome-shaped, being flattened on one side and elevated on the opposite in three tiers (two rings) (TOKIOKA & HaseE, 1953; ROSEWATER, 1970). The capsules averaged 0.19 mm in diameter and 0.08 mm in height. The area containing the egg averaged 0.08 mm in diameter. Early development: Based on observations of capsules isolated in culture dishes, the development from zygote (average, 0.07 mm in diameter) into the veliger stage oc- curred within about 24 h. During the next 24 h, veligers having an operculum but lacking eyes left the capsule via a torn area on the flattened side. Veligers remained alive for 5 d following hatching and averaged 0.13 mm long. Neither eyes nor any other major morphological feature developed during this period. Nodilittorina subnodosa (Philippi, 1847) Nomenclature: What SAFRIEL & LIPKIN (1964) reported as Tectarius armatus Issel from the littorinid zone at Eilat, Israel, and FISHELSON (1971) reported is Nodilittorina subnodosa. According to Dr. J. Rosewater (personal com- Page 322 munication), 7. armatus appears to be a fossil and is prob- ably a trochid. Habitat: The highest occurring mollusk in the rocky in- tertidal zone and characteristic of the upper littorinid (supralittoral fringe) zone. It is most common on nearly horizontal slab substrata. Specimens examined: 339 averaging 9.0 mm long (range, 4.3-13.2) from May 1982 through May 1983 for state of reproduction; 112 specimens 2.2-5.0 mm long for mini- mum size of sexual maturity. Sex ratio: 1.0 male:2.1 female. Reproduction: All females contained ova in the oviduct and all males had sperm in the vas deferens from June through September. Females with and without ova and males with and without sperm were found in May and October. The major period of reproduction is considered to be June through September. From October through April there was no noticeable degeneration in the size of the penis, as PALANT & FISHELSON (1968) saw during the non-reproductive period of Littorina neritoides (Linnaeus, 1758). The smallest male having a recognizable, rudimentary penis was 2.4 mm long and the smallest male with a fully developed penis and sperm in the vas deferens was 3.4 mm. The smallest ovigerous female was 4.1 mm long. Spawn: Each pelagic egg capsule contains a single egg. The capsules are similar in shape and transparency to those of Nodilittorina millegrana. They differ, however, in having four tiers (three rings) and being larger, averaging 0.25 mm in diameter and 0.15 mm in height. The area containing the zygote averaged 0.11 mm in diameter. Early development: In culture dishes, veligers hatched within about 24 h after isolation of uncleaved ova (aver- aging 0.07 mm in diameter). The newly hatched veligers averaged 0.11 mm long and had an operculum but lacked eyes. The veligers remained alive for 5 d, at which time they still lacked eyes, but the shell length had increased to 0.13 mm. Family PLANAXIDAE Planaxis sulcatus (Born, 1780) Habitat: A wide-ranging component of the chthamalid- Tetraclita (midlittoral) zones on pebble and slab substrata. Specimens examined: 409 averaging 15.8 mm _ long (range, 8.7-21.6) from May 1982 through November 1983 for state of reproduction; 139 specimens 7.5-13.9 mm long for minimum size of sexual maturity. Sex ratio: 1.0 male: 1.2 female. Reproduction and early development: This species is ovoviviparous with a brood chamber at the termination of The Veliger, Vol. 28, No. 3 the pallial oviduct in the dorsal head-foot (RISBEC, 1935; THORSON, 1940). During December and through March, neither the males nor females had gametes. The gonads were reduced in size but retained the coloration of dark- orange to rust-colored testes and cream ovaries character- istic of the reproductive period. Beginning in April, males with sperm in the vas def- erens and some females with ova (0.12 mm in diameter) in the oviduct appeared, and gametes were present in some individuals of both sexes through November. From May through October, the brood chamber was filled with stages of development from uncleaved zygotes (average diameter, 0.12 mm) within a membrane to free (7.e., not enclosed in a membrane) veligers lacking eyes but having an oper- culum and averaging 0.13 mm long. In most cases, the contents of the brood chamber were uniform in that only one developmental stage, or closely sequential stages, were present. In a few cases, the brood chamber contained un- cleaved ova and a few free veligers, indicating that the brood chamber is filled following the release of veligers. When the brood chamber was filled, the oviduct usually contained ova. The major period of reproduction, based on the occurrence of a majority of mature individuals and brooding females, is from June through September. The smallest male, based on gonad coloration, was 8.7 mm long, and sperm were present in the vas deferens. The smallest female with a cream-colored ovary was 9.6 mm long, the smallest with ova in the oviduct 10.0 mm long, and the smallest with the brood chamber occupied was 10.3 mm long. Comments: RIsBEC (1935) described the reproduction of Planaxis sulcatus from New Caledonia and THORSON (1940) did so from Bushire, Iran, Persian Gulf. The en- tire process described by Risbec and Thorson (through the free veliger stage, including the lack of eyes in hatched veligers) and the process found in specimens from the Jordan Gulf of Aqaba are essentially the same. However, the suppression of the pelagic larval stage and the devel- opment of the veligers into small snails in the brood cham- ber found by Thorson in the Persian Gulf was not found in the Gulf of Aqaba. Both RisBEc (1935) and THORSON (1940) reported a low number of males compared to females. Based on this finding, and coupled with not having seen Planaxis sul- catus copulating, Thorson proposed parthenogenetic de- velopment. Planaxis sulcatus lacks a penis; actual sperm transfer was not observed in this study, nor is the mech- anism known. Clustering, however, is a common pattern of behavior during submergence and emergence. The clus- ter may include a large number of individuals; in one case a total of 867 specimens were counted in one cluster. In another case a 1:1 male to female ratio was found within a cluster (compare with the above 1.0:1.2 ratio). MAGNUs & HAACKER (1968) attributed the clustering behavior of P. sulcatus to physical factors, including prevention of N. C. Hulings, 1986 drying during exposure and protection against wave ac- tion. It is proposed herein that another aspect of clustering is the transfer of sperm, probably during submergence. Based on the above, parthenogenetic development, as proposed by THORSON (1940), is considered unlikely. The high female to male sex ratio found by RISBEc (1935) and THORSON is considered a sampling problem. My experi- ence is that a skewed sample with respect to sex in this and other species with an approximate 1:1 ratio is not uncommon. This is especially the case in small-sized sam- ples. Family CERITHIIDAE Cerithium caeruleum Sowerby, 1855 Habitat: This is the largest of the cerithiids occurring in the intertidal zone. It usually occurs between the two cer- ithiids described below, from above to below the Tetraclita (midlittoral) zone on smooth fossil reef bottoms and other slab having a sand cover. Specimens examined: 416 averaging 27.3 mm _ long (range, 14.0-34.2) from August 1982 through May 1984 for reproductive state; 101 specimens 10.0 to 23.0 mm long for minimum size of sexual maturity. Sex ratio: Distinction between male and female can only be made when gametes are present. For 161 with gametes, the sex ratio was 1.0 male:0.6 female. Reproduction: Males with sperm in the vas deferens were found from January through September but most com- monly from February through August. Females with ova in the oviduct were present from April through September but most commonly from April through August. The lat- ter period is considered to be the major period of repro- duction. The smallest male, based on the presence of sperm, was 17.3 mm long, and the smallest female having ova was 20.2 mm long. The actual process of sperm transfer was not observed. Pairing was observed frequently in the field and labora- tory and was similar to that of Cerzthiwm muscarum Say, 1832, described by HousRIcK (1973). Spawn: Specimens kept in aquaria from mid-April through early June deposited egg masses on the aquarium sides as well as on pebbles and clumps of algae during the day and night on 27 of 54 days. The deposition was not related to a particular lunar or tidal cycle. The masses were typically pale yellow in color and arranged in a continuous linear series of tight folds or loops up to 4 mm high and 50 mm long. The arm of the folds was 0.4 to 0.5 mm thick. The individual capsules suspended in a gelatinous matrix averaged 0.15 mm in diameter, and the contained zygote was 0.09 mm in diameter. Early development: Within 5 to 6 d following deposition of egg masses, veligers, averaging 0.14 mm long, hatched. Page 323 The veligers were kept alive for 4 d after hatching and did not develop eyes. Clypeomorus bifasciata (Sowerby, 1855) Nomenclature: According to Dr. R. S. Houbrick (per- sonal communication), this species is probably the same as Clypeomorus morus (Bruguiére, 1792) reported by MASTALLER (1979). Habitat: Characteristic of the chthamalid (midlittoral) zone and is the highest occurring of the three intertidal cerithiids. Specimens examined: 332 averaging 15.9 mm long (range, 9.8-20.5) from June 1982 through May 1984, except August 1982, for state of reproduction; 51 speci- mens 9.0-13.0 mm long for minimum size of sexual ma- turity. Sex ratio: The sexes could be determined only when ga- metes were present; for 179 specimens, the ratio was 1.0 male: 1.4 female. Reproduction: Some males with sperm in the vas defer- ens were found each month except December 1982 and 1983 and February 1983. Ovigerous females were absent or least abundant from September through February— March and most common from April through August. The latter period is considered as the major period of reproduction. Both the smallest male with sperm in the vas deferens and the smallest female with ova in the ovi- duct were 9.2 mm long. Spawn: In an aquarium, Clypeomorus bifasciata deposited white egg masses on the underside of clumps of the alga Enteromorpha. The masses were usually 2-3 mm wide, variable in length, and the mass had an irregular shape. The individual capsules within the gelatinous mass were aligned in rows perpendicular to the width. The capsules averaged 0.13 mm in diameter, and the contained zygote was 0.08 mm in diameter. Early development: Hatching of eyed veligers occurred 3-4 d after egg mass deposition and were 0.13-0.14 mm long. They were kept alive for 10 d following hatching and no significant morphological changes occurred. Clypeomorus petrosa gennesi (Fisher & Vignal, 1901) Nomenclature: According to Dr. R. S. Houbrick (per- sonal communication), this taxon is synonymous with Clypeomorus tuberculatus (Linnaeus, 1758) reported by MASTALLER (1979). Habitat: Occurs below the Tetraclita (midlittoral) zone, most commonly in depressions with sand occurring in the fossil reef substratum. Page 324 Specimens examined: 287 averaging 20.2 mm _ long (range, 14.2-24.5) from June 1982 through March 1984 for state of reproduction; 105 specimens 7.8-19.1 mm long for minimum size with gametes. Sex ratio: 1.0 male:0.5 female based on 201 specimens with gametes. Reproduction: Males with sperm in the vas deferens were found during each month except December 1982 and Jan- uary 1983. Females with ova in the oviduct were most common from May through September; some ovigerous females with few ova were found in April and from Oc- tober through January. The major period of reproduction is considered to be from May through September. The smallest male with sperm in the vas deferens was 14.4 mm long, the smallest ovigerous female 15.8 mm long. Pairing was observed on numerous occasions in the field and laboratory but actual sperm transfer was not; pairing was essentially the same as noted above for Cerithium caeruleum. Spawn: In an aquarium, deposition of yellowish egg masses occurred on the underside of Enteromorpha clumps or, in the absence of algae, on the sides of the aquarium. The sand and detritus covered masses were a continuous string 2-3 mm wide and with individual capsules sus- pended in a gelatinous matrix. The capsules averaged 0.13 mm in diameter, the zygotes 0.09 mm in diameter. Early development: Veligers with eyes hatched 5 d after deposition and averaged 0.13 mm long. The veligers re- mained alive for 1 wk and no significant morphological changes occurred. Class Gastropoda, subclass Pulmonata Family SIPHONARIDAE Siphonaria laciniosa Linnaeus, 1758 Nomenclature: According to BARASH & DANIN (1972), Siphonaria laciniosa and S. kurracheensis Reeve, 1856, are synonymous. Habitat: Most common in the chthamalid zone and be- tween the Tetraclita and chthamalid (midlittoral) zones. In the absence of a recognizable chthamalid zone, it occurs most commonly above the Tetraclita zone. Specimens examined: 245 from March 1983 through May 1984 averaging 14.0 mm long (range, 9.1-22.2). Reproduction: The presence of gametes was determined by teasing apart the gonad (ovotestis) and the hermaph- roditic duct. Sperm were present every month except July, August, and September, whereas the presence of ova was restricted to December through May, being most abun- dant from January through March. The latter period is considered the major period of reproduction. During the major period of reproduction, the gonad The Veliger, Voli 28) Nowe was bright yellow and reached a maximum size of 3.2 mm wide X 5.2 mm long. By contrast, when ova were absent, the gonad was burnt orange and greatly reduced in size, the minimum being 0.8 mm wide x 1.6 mm long. Actual copulation was not observed but behavior simi- lar to that reported for Siphonaria japonica (Donovan, 1824) by HIRANO & INABA (1980) was observed. Spawn: Siphonaria laciniosa deposits benthic egg ribbons. The ribbons were typically dome-shaped, yellowish, and coated externally with sand. The shape of the ribbon var- ied from a “C” to a loose coil and ranged in size from 1.2 mm wide x 10 mm long to 1.8 x 54 mm. Larger ribbons were seen in the field but not measured. Capsules con- taining the zygote were suspended within the ribbon, were ovate in shape, and averaged 0.17 mm wide X 0.23 mm long. The deposition of ribbons in the field and laboratory lasted from early January to late April. In the field, the ribbons were deposited on exposed surfaces, coated with sand, and exposed to the air during spring ebb tides. De- position was not related to a particular lunar or tidal cycle but occurred only at night. The latter is consistent with other behaviors of Siphonaria laciniosa, which is a homing species active only after sunset and when submerged (HULINGS, 1985). Early development: Hatching of operculate but eyeless veligers occurred about 10 d after deposition. Eyes ap- peared within 2 to 3 days. The hatched veligers averaged 0.18 mm long (range, 0.17-0.19) and remained alive for 2 wk, during which no obvious morphological changes occurred. Comments: THORSON (1940) reported two siphonarians from the Persian Gulf, Siphonaria kurracheensis (=S. la- ciniosa) and S. sipho Sowerby. The descriptions of the shape, size, and color of the egg ribbon, the size and shape of the egg capsule, and the embryological development and hatching reported for S. sxpho by Thorson are almost iden- tical to those of S. /aciniosa from the Jordan Gulf of Aqa- ba. In addition, Thorson reported direct development in S. kurracheensis (=S. laciniosa), 1.e., the veliger stage was completed in the capsule and at hatching, crawling juve- niles emerged. This is in contrast to the hatching of veli- gers in §. laciniosa in the northern Gulf of Aqaba. Class Bivalvia Family MYTILIDAE Brachidontes variabilis (Krauss, 1848) Habitat: Most characteristic between the chthamalid and Tetraclita (midlittoral) zones or in the middle midlittoral (SAFRIEL et al., 1980). This species typically occurs as dense beds in depressed areas of beach rock or fossil reef. Specimens examined: 247 from October 1982 through May 1984 averaging 19.5 mm long (range, 12.8-27.3). N. C. Hulings, 1986 Page 325 Sex ratio: 1.0 male:0.8 female. Reproduction: The gonads penetrate into the mantle. The coloration of the mantle lobe containing the ovaries is typically reddish, although variable in shade, and that containing the testes is cream colored. Gametes were pres- ent in at least some if not all of the specimens of both sexes each month. In addition, at least some individuals of both sexes had enlarged gonads each month. Comments: WILSON & HODGKIN (1967) found spawning of B. variabilis in Western Australia to be restricted to March-April although they noted that spawning proba- bly continued throughout the summer. They found in B. variabilis, compared to other mytilids, significant differ- ences in the time of year and the length of spawning, as well as the beginning of gametogenesis and the presence of a “reproductively neutral phase.” Family OSTREIDAE Ostrea forskali Chemnitz, 1785 Nomenclature: According to MASTALLER (1979), Ostrea cucullata Born, 1780, is synonymous with O. forskali. The former species has been variously placed in Saccostrea (BRALEY, 1982), Crassostrea (FAO, 1972), or Lopha (MASTALLER, 1978). Habitat: Characteristic of the 7etraclita (midlittoral) zone. Specimens examined: 144 averaging 39 mm long (range, 19-62) from April 1982 through June 1983. Sex ratio: 1.0 male:2.6 female based on 43 specimens with gametes. Reproduction: Ovigerous females occurred from June through November and in January. The greatest abun- dance occurred from July through October-November. Males with sperm were found from June through Octo- ber. The major period of reproduction is considered to be from July through October-November. The possibility of hermaphroditism in Ostrea forskali was not investigated. Comments: FAO (1972) found the greatest density of spat occurred during December—January, although Ostrea forskali spat were often absent or in the minority compared to those of other oysters. BRALEY (1982) reported low level and continuous reproduction in O. forskali with, however, peaks in November-December, March-April, and late June. He also found no correlation between reproduction and temperature, and a planktonic larval life of 3-4 wk. DISCUSSION Various aspects of the reproduction of 15 dominant species of mollusks from the rocky intertidal zone along the Jor- danian coast of the Gulf of Aqaba have been investigated. The species range in vertical distribution from the supra- littoral fringe to the lower midlittoral. The fauna can be divided into continuous reproducers (z.e., those that reproduce the year round) and restricted reproducers, with periods of reproduction being indicated by the majority of specimens having ova in the oviduct and sperm in the vas deferens or similar structures at the same time. The continuous reproducers include Cellana radiata, Monodonta dama, Nerita forskalu, Nerita polita, Lit- torina scabra scabra (probable), Nodilittorina millegrana, and Brachidontes variabilis. The restricted reproducers in- clude Acanthopleura haddoni, Nodilittorina subnodosa, Planaxis sulcatus, Cerithium caeruleum, Clypeomorus bifas- ciata, Clypeomorus petrosa gennesi, Siphonaria laciniosa, and Ostrea forskali.. Cycles within the continuous reproducers may exist; in addition, there may be longer or shorter periods of reproduction within the restricted reproducers. For example, among the latter group, there were often specimens with or without gametes in the ducts at the beginning and end of the major period of reproduction. There were also species in which sperm were present before and after the presence of ova (P. sulcatus, C. caeru- leum, C. bifasciata, C. petrosa gennesi, and S. laciniosa). An investigation of gametogenesis (in progress) may provide additional information on the above and other aspects of reproductive periodicity. The temporal patterns of reproduction noted above are not related to tide level or a species’ vertical position in the intertidal zone. For example, the highest species in the supralittoral fringe, Nodilittorina subnodosa, had a restrict- ed period of reproduction, while the next highest, N. mil- legrana, reproduced the year round. The supralittoral fringe—-upper midlittoral Littorina scabra scabra is assumed to reproduce continuously. Within the midlittoral, both continuous and restricted reproducers were found. Water temperature in the northern Gulf of Aqaba has a narrow annual range, 20 to 27°C, whereas average an- nual air temperature has a much wider range, 16 to 32°C. Among the species investigated there was no direct rela- tionship between temperature and reproduction or spawn- ing in the continuous reproducers. They reproduced and spawned throughout the annual range in air and water temperature. Among the restricted reproducers, 7 out of 8 of the species reproduced when annual water and air temperatures were warmer (generally May through Oc- tober). In two species (Cerithium caeruleum and Clypeo- morus bifasciata), however, reproduction in April coincided with a 4°C increase in annual air temperature but little increase in water temperature. The other restricted re- producer, Siphonaria laciniosa, reproduced during the coldest period of annual air and water temperatures. Thus, the pattern of the relationship between reproduction or spawning and temperature is highly variable. If there is a relationship, as seems to be the case for some species, air temperature, having a wide range, may be more sig- nificant than water temperature, having a narrow range. And as noted previously, the terrestrial environment ex- erts greater influence on the intertidal zone of Jordan than the marine environment. Page 326 Vertical migration of most of the mobile species occurs with the change in sea level, from high during December- May to low during July—October (Hulings, unpublished data). The continuous reproducers, including the sessile Brachidontes variabilis, reproduced irrespective of changes in sea level. Among the restricted reproducers, including the sessile Ostrea forskali, all reproduced during lowered sea level except Siphonaria laciniosa. The latter, a per- manent homer and non-migrant (HULINGS, 1985), repro- duced during the period of higher sea level, a period dur- ing which the egg ribbons were submerged more often than exposed. There was no consistent pattern between tide level or vertical position and type of spawning except in species of the supralittoral fringe. Both Nodilittorina millegrana and N. subnodosa deposited pelagic egg capsules. Littorina sca- bra scabra, transitional between the supralittoral fringe and the upper midlittoral, brooded and hatched veligers. Within the midlittoral, a wide variety of spawning pat- terns occurred. The absence of a relationship between reproduction (or spawning) and tidal levels and lunar cycles may result from the tidal levels and lunar cycles being out of phase. In addition, the annual changes in sea level modify the tide levels. HULINGS (1985) found that activity patterns in Cellana radiata and Siphonaria laciniosa were not related to tidal level (except that these animals are active only when submerged) or lunar cycles. The hatching of veligers occurred before, during, and after the relatively short periods of primary productivity, as well as during the extended period of oligotrophic con- ditions. Hatching occurred up to 7 d following deposition of the spawn, and the resulting veligers were small, less than 0.20 mm in length. It appears that the length of veliger life is short, based on their small size and the generally low primary productivity in the area. Indirect development is characteristic of all the species, based on direct observation or literature sources. Devel- opment through, and hatching of, veligers was observed in all species except Acanthopleura haddoni, Cellana radia- ta, Brachidontes variabilis, and Ostrea forskali. Among the veligers were those with, without, or developing eyes prior to or after hatching. Veligers with eyes prior to hatching included those of Nerita forskalu, N. polita, Clypeomorus bifasciata, and C. petrosa gennesi. Those lacking eyes prior to and up to 1 wk following hatching included Monodonta dama, Littorina scabra scabra, Nodilittorina millegrana, N. subnodosa, Planaxis sulcatus, and Cerithium caeruleum. The veligers of Siphonaria laciniosa developed eyes 2 to 3 d after hatching. The significance of the presence or absence of eyes in the veligers at hatching is not known. ACKNOWLEDGMENTS The author wishes to thank Dr. Richard Houbrick and the late Dr. Joseph Rosewater of the U.S. National Mu- seum of Natural History for species identifications and The Veliger, Vol. 28, No. 3 helpful comments on the manuscript. Thanks are also expressed to Dr. Elias Salameh, Department of Geology, University of Jordan, for rock-type determinations. Fi- nancial support provided by the Office of the Dean of Research, Yarmouk University, is gratefully acknowl- edged. LITERATURE CITED AnaTI, D. A. 1976. Balances and transports in the Red Sea and the Gulf of Elat (Aqaba). Israel J. Earth Sci. 25:104- 110. ANDREWS, E. A. 1935. The egg capsules of certain Neritidae. J. Morphol. 57:31-59. AYAL, Y. & U.N. SAFRIEL. 1980. Intertidal zonation and key- species associations of the flat rocky shores of Sinai, used for scaling environmental variables affecting cerithiid gas- tropods. Israel J. Zool. 29:110-124. BarasH, A. L. & Z. DANIN. 1972. The Indo-Pacific species of Mollusca in the Mediterranean and notes on a collection from the Suez Canal. Israel J. Zool. 21:301-376. Berry, A. J., R. Lim & A. S. Kumar. 1973. Reproductive systems and breeding conditions in Nerita birmanica (Ar- cheogastropoda: Neritacea) from Malayan mangrove swamps. J. Zool. 170:189-200. Biccs, H. E. J. 1966. A new species of Littorina from Eilat, Israel, and notes on its affinities with Littorina novaezelan- diae Reeve. J. Conchol. 26:137-139. BRALEY, R. D. 1982. Reproductive periodicity in the indige- nous oyster Saccostrea cucullata in Sasa Bay, Apra Harbor, Guam. Mar. Biol. 69:165-173. FAO. 1972. Report to the Government of Israel on the poten- tial for oyster culture at Elat on the Gulf of Aqaba. Based on the work of P. R. Walne, FAO/TA Consultant Rep. FAO/UNDP(TA), 3076. 13 pp. FISHELSON, L. 1971. Ecology and distribution of the benthic fauna in the shallow waters of the Red Sea. Mar. Biol. 10: 113-133. FISHELSON, L. 1973. Ecological and biological phenomena in- fluencing coral-species composition on the reef tables at Ei- lat (Gulf of Aqaba, Red Sea). Mar. Biol. 19:183-196. FRIEDMAN, G. M. 1965. A fossil shoreline reef in the Gulf of Elat (Aqaba). Israel J. Earth Sci. 14:86-90. Hirano, Y. & A. INABA. 1980. Siphonaria (pulmonate limpet) survey of Japan. I. Observations on the behavior of Sipho- naria japonica during breeding season. Publ. Seto Mar. Biol. Lab. 25:323-334. Houpsrick, R. S. 1973. Studies on the reproductive biology of the genus Cerithium (Gastropoda: Prosobranchia) in the western Atlantic. Bull. Mar. Sci. 23:875-904. Hucues, R.N. 1971. Notes on the Nerita (Archeogastropoda) population of Aldabra Atoll, Indian Ocean. Mar. Biol. 9: 290-299. Hutincs, N.C. 1979. Currents in the Jordan Gulf of Aqaba. Dirasat 6:21-33 Hutincs, N. C. 1985. Activity patterns and homing in two rocky intertidal limpets, Jordan Gulf of Aqaba (Red Sea). Nautilus 99:75-80. Hutincs, N. C. & A. ABU HILAL. 1983. The temporal dis- tribution of nutrients in the surface waters of the Jordan Gulf of Aqaba. Dirasat 10:91-105. IrIkI, S., S. NISHIWAKI & T. TocHimoTo. 1963. On the pe- culiar mode of spermatophore transfer in Nerita albicilla L. (Prosobranchia, Neritidae). Venus 22:290-292. N. C. Hulings, 1986 JorNeE, J. & U. N. Sarriet. 1979. Linear and non-linear diffusion models applied to the behavior of a population of an intertidal snail. J. Theor. Biol. 79:367-380. Kumeg, M. & D. KatsuMa. 1957. Invertebrate embryology. Bai Fukan Press: Tokyo. LEVANON-SPANIER, I., E. PADAN & Z. Retss. 1979. Primary production in a desert-enclosed sea—the Gulf of Elat (Aqa- ba), Red Sea. Deep-Sea Res. 26:673-685. Maanus, D. B. E. & U. HAACKER. 1968. Zum Phanomen der orstsunsteten Ruhrversammlungen der Strandschnecke Planaxis sulcatus (Born) (Mollusca, Prosobranchia). Sarsia 34:137-148. MaASTALLER, M. 1978. The marine molluscan assemblages of Port Sudan, Red Sea. Zool. Meded. 53:117-144. Mastat_er, M. 1979. Beitrage zur Faunistik und Okologie Mollusken und Echinodermen in den Korallenriffen bei Aqaba, Rotes Meer. Doctoral Dissertation, Ruhr-Univer- sitat Bochum, Fed. Republic Germany. 344 pp. MERGNER, H. & H. SCHUHMACHER. 1974. Morphologie, Okologie und Zonierung von Korallenriffen bei Aqaba, (Golf von Aqaba, Rotes Meer). Helgo. Wiss. Meeresunt. 26:238- 358. Morcos, S. A. 1970. Physical and chemical oceanography of the Red Sea. Oceanogr. Mar. Biol. Ann. Rev. 8:73-202. PALANT, B. & L. FISHELSON. 1968. Littorina punctata (Gme- lin) and Littorina neritoides (L.), (Mollusca, Gastropoda) from Israel: ecology and annual cycle of genital system. Is- rael J. Zool. 17:145-160. PALpor, N. & D. A. ANaTI. 1979. Seasonal variation of tem- perature and salinity in the Gulf of Elat (Aqaba). Deep- Sea Res. 26:661-672. Rao, M. B. 1973. Sex phenomenon and reproduction cycle in the limpet Cellana radiata (Born) (Gastropoda: Prosobran- chia). J. Exp. Mar. Biol. Ecol. 12:263-273. Rao, M. B. 1976. Studies on the growth of the limpet Cellana Page 327 radiata (Born) (Gastropoda: Prosobranchia). J. Moll. Stud. 42:136-144. RIsBEC, J. 1935. Biologie et ponte de mollusques gastéropodes Néo-Calédoniens. Bull. Soc. Zool. France 60:387-417. RosEWwaATER, J. 1970. The family Littorinidae in the Indo- Pacific. Part I. The subfamily Littorininae. Indo-Pacific Mollusca 2:417-506. SAFRIEL, U. 1969. Ecological segregation, polymorphism and natural selection in two intertidal gastropods of the genus Nerita at Elat (Red Sea, Israel). Israel J. Zool. 18:205-231. SAFRIEL, U. N., A. GILBOA & T. FELSENBERG. 1980. Distri- bution of rocky intertidal mussels in the Red Sea coasts of Sinai, the Suez Canal and the Mediterranean coast of Israel, with special reference to recent colonizers. J. Biogeog. 7:39- 62. SAFRIEL, U. N. & Y. LipKIN. 1964. On the intertidal zonation of the rocky shores at Eilat (Red Sea, Israel). Israel J. Zool. 13:187-190. STEPHENSON, T. A. & A. STEPHENSON. 1949. The universal features of zonation beween tidemarks on rocky coasts. J. Ecol. 37:289-305. THORSON, G. 1940. Studies on the egg masses and larval de- velopment of Gastropoda from the Iranian Gulf. Danish Sci. Invest. Iran 2:159-238. Tokioka, T. & T. HaBe. 1953. A new type of Littorina cap- sula. Publ. Seto Mar. Biol. Lab. 3:55-56. WILSON, B. R. & E. P. HODGKIN. 1967. A comparative ac- count of the reproductive cycles of five species of marine mussels (Bivalvia: Mytilidae) in the vicinity of Fremantle, Western Australia. Aust. J. Mar. Freshwater Res. 18:175- 203. ZANN, L. P. 1973. Relationship between intertidal zonation and circa-tidal rhythmicity in littoral gastropods. Mar. Biol. 18:243-250. NOTE ADDED IN PROOF: Too late for inclusion in the text, information has been obtained on the reproduction of another intertidal gastropod from the Jordan Gulf of Aqaba. Nerita undata Linnaeus, 1758 (Family NERITIDAE) Habitat: Only two specimens found, both on boulders above the Tetraclita (midlittoral) zone. MASTALLER (1979) reported finding only one specimen. Specimens examined: One female 19.8 mm long, October 1982; one female 28.0 mm, August 1985. Reproduction: No spermatophores like those in Nerita forskalu and N. polita nor any other type were found. Spawn: The female collected in August 1985 was kept in a seawater table with con- tinuously circulating water. In September 1985 the specimen deposited 11 egg capsules near to and just under the base of a permanently submerged pebble. ‘The capsules are white and composed of mostly round spherulites averaging 0.07 mm in diameter (range, 0.05-0.08). The shape of the capsules is more elliptical than round, averaging 2.7 mm wide X 3.5 mm long (range, 2.4—4.0). Early development: Development from uncleaved ova to veligers with eyes and oper- cula occurred in about 3 w. The capsules contained an average of 76 larvae (range, 70-85); the veligers averaged 0.34 mm long. Hatching occurred about 4 w after de- position of the capsules. The Veliger 28(3):328-339 ( January 2, 1986) THE VELIGER © CMS, Inc., 1986 NOTES, INFORMATION & NEWS Consumption of Pelagic Red Crabs by Black Abalone at San Nicolas and San Miguel Islands, California by Glenn R. VanBlaricom U.S. Fish and Wildlife Service, Institute of Marine Sciences, University of California, Santa Cruz, California 95064, U.S.A. and Brent S. Stewart Hubbs Marine Research Institute, 1700 South Shores Road, San Diego, California 92109, U.S.A. Black abalone, Haliotis cracherodit Leach, 1817 (Proso- branchia: Haliotidae), are common in rocky intertidal habitats from northern California to southern Baja Cali- fornia Sur, Mexico (MCLEAN, 1978). Black abalone feed primarily on drifting fragments of kelps and other algae (Cox, 1962; LEIGHTON & BOOLOOTIAN, 1963; ABBOTT & HADERLIE, 1980). Fragments of foraminiferans, bryozo- ans, hydroids, sponges, and sea urchins occasionally occur in the guts of black abalone, but ingestion of animal parts is thought to be incidental to consumption of algae (LEIGHTON & BOOLOOTIAN, 1963). Intentional capture and consumption of macroinvertebrates has not been re- ported, to our knowledge, for any species of abalone. During a morning low tide on 3 June 1984, Van- Blaricom searched for evidence of feeding among several hundred black abalone in a rocky cove (33°16.5'N, 119°33.5'W) at the west end of San Nicolas Island, Cal- ifornia. About 15% of the abalone observed were feeding on fragments of kelps. In addition, three abalone were consuming pelagic red crabs (Pleuroncodes planipes Stimp- son, 1860) (Anomura: Galatheidae). The abalone were all >100 mm in maximum shell diameter, and the crabs were 40-50 mm in total length. The abalone held the crabs against the substratum with the anterior portion of the foot, as they do when consuming algal fragments. When the abalone were removed from the rocks, it was noted that abdominal and posterior thoracic tissues (including the exoskeleton) of the crabs had been rasped away. Crabs held by abalone were dead, but were moist, flexible, and bright red in color. Crabs in the water near black abalone were alive, active, and bright red in color. Dead crabs were only seen high on a nearby beach and were dry, brittle, and bleached. Therefore, it seems likely that ab- alone captured the crabs alive, although post-mortem cap- ture cannot be ruled out. Stewart made similar observations on 10 February 1985 while examining a group of approximately 300 black ab- alone near Otter Harbor (34°3.5'N, 120°25’W) on the north shore of San Miguel Island, California. Many of the abalone were feeding on kelp fragments. Live and dead red crabs and body parts were scattered throughout the rocky intertidal zone. Five abalone held red crabs (dead in all cases) with the anterior part of the foot, and several others held only fragments of crabs. Therefore, it seems likely that some abalone were feeding on crabs captured post-mortem. During normal oceanographic conditions, pelagic red crabs are common in the coastal waters of Baja California south of 29°N latitude (BoyD, 1967). During El Nino- Southern Oscillation (ENSO) periods, anomalous north- ward currents carry populations of red crabs to Califor- nia. As a result, mass strandings of red crabs become common south of Pt. Conception (LONGHURST, 1966), and can occur farther north (GLYNN, 1961). The 1982-83 ENSO was perhaps the strongest of the century (CANE, 1983), producing striking warming of the coastal waters of California (FIEDLER, 1984). Stranded red crabs were observed frequently at San Nicolas Island from January 1983 through November 1984 (STEWART et al., 1984; VanBlaricom & Stewart, personal observations), and at San Miguel Island from January 1983 through February 1985 (STEWART et al., 1984; Stewart, personal observa- tions). Mainland strandings occurred as recently as March 1985 at sites as far north as Monterey Bay (Jameson, Baldridge & Deutsch, personal observations). Strandings and nearshore concentrations of red crabs in California have provided unusual feeding opportunities for gulls (STEWART et al., 1984), sea otters (Deutsch, personal ob- servations), and intertidal sea anemones (VanBlaricom, personal observations), in addition to black abalone. We thank the Command of the Pacific Missile Test Center, U.S. Navy, for allowing access to San Nicolas Island, and the Superintendent of Channel Islands Na- tional Park for allowing access to San Miguel Island. R. Dow, J. Vanderwier, and C. Harrold provided logistic support, and R. Saunders assisted in the field. We thank D. Lindberg for encouragement, J. Estes, R. Jameson, and two anonymous reviewers for comments on the manu- script, and P. Himlan for clerical expertise. GRVB was supported by the Denver Wildlife Research Center of the U.S. Fish and Wildlife Service and BSS was supported by contracts from the U.S. Air Force and the National Marine Fisheries Service. LITERATURE CITED AsBotTt, D. P. & E. C. HADERLIE. 1980. Prosobranchia: ma- rine snails. Pp. 230-307. Jn: R. H. Morris, D. P. Abbott Notes, Information & News & E. C. Haderlie (eds.), Intertidal invertebrates of Califor- nia. Stanford Univ. Press: Stanford, Calif. Boyp, C. M. 1967. The benthic and pelagic habitats of the red crab, Pleuroncodes planipes. Pacific Sci. 21:394-403. Cane, M. A. 1983. Oceanographic events during El Nino. Science 222:1189-1195. Cox, K. W. 1962. California abalones, family Haliotidae. Cal- if. Dept. Fish and Game, Fish Bull. 118:1-113. FIEDLER, P. C. 1984. Observations of the 1982-83 El Nino along the U.S. Pacific coast. Science 224:1251-1254. GLyNN, P. W. 1961. The first recorded mass stranding of pelagic red crabs, Plewroncodes planipes, at Monterey Bay, California, since 1859, with notes on their biology. Calif. Fish and Game 47:97-101. LEIGHTON, D. & R. A. BOOLOOTIAN. 1963. Diet and growth in the black abalone, Haluotis cracherodi. Ecology 44:227- 238. LonGuHursT, A. R. 1966. The pelagic phase of Pleuroncodes planipes Stimpson (Crustacea, Galatheidae) in the Califor- nia Current. California Cooperative Oceanic Fisheries In- vestigation Reports 11, 1 July 1963 to 30 July 1966:142- 154. McLean, J. H. 1978. Marine shells of southern California. Natur. Hist. Mus. Los Angeles Co., Sci. Ser. 24 (Revised edition):1—-104. STEWART, B. S., P. K. YOCHEM & R. W. SCHREIBER. 1984. Pelagic red crabs as food for gulls: a possible benefit of El Nino. Condor 86:341-342. Soviet Contributions to Malacology in 1980 by Kenneth J. Boss Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138, U.S.A. and M. G. Harasewych Section on Biochemical Genetics, Clinical Neurogenetics Branch, National Institute of Mental Health, Bethesda, Maryland 20205, U.S.A. INTRODUCTION As in past years, herein is provided a listing of malaco- logical papers by Soviet scientists included in, and fre- quently abstracted by, the 1980 issues of the Referativnyy Zhurnal (see Veliger 27[3]:339-346 for the last such list- ing and references to previous ones). We follow the categorical arrangements as utilized by the Referativnyy Zhurnal itself, although occasionally we may place selected titles in more approriate categories. Certain publications this year are major contributions to the field, the most important of these being Golikov’s monograph of the Buccininae of the world in which he treats 93 species and subspecies (several as new) in great detail; this extensive work is illustrated by plates showing the shells as well as enlargements for sculptural detail; also presented are figures of egg capsules, radulae, oper- Page 329 cula, and anatomy as well as maps and charts indicating geographical, bathymetric, and ecological ranges and pa- rameters. The bibliography includes nearly 1100 cita- tions. Although a number of new species were introduced, several papers also established new family-level taxa or revisionary arrangements of previously studied groups. Thus, Starobogatov and Izzatullaev divided the freshwa- ter prosobranch family Thiaridae into three independent familial units: Thiaridae s.s., Melanatriidae, and Mela- noididae, new family, on the configuration of the pallial gonoducts; further, they subdivided the widely distributed, often parthenogenetic Melanoides tuberculatus into four species, two of which are new. Among “hydrobioid” taxa Izzatullaev discussed the little known pomatiopsid taxa of Tadzhikistan, describing two new species, one in Kain- arella and another in Pseudocaspua. Special attention to mollusks of the Kuril Islands is reflected in Gul’bin’s paper on prosobranchs and Siren- ko’s on chitons, the latter work considering the chiton fauna off a single island, Simushir; the densities of these animals are high (e.g., 3100/m? for Juvenichiton albocin- namoneus). Further, an entire book by Volova, Golikov & Kusakin was devoted to the shelled gastropods of the geo- graphically adjacent Peter the Great Bay; 119 species in 43 families were noted and figures, descriptions, ranges, and ecological notes provided. Among cephalopods, considerable attention was given to the exploitation of the neritic niche with papers by Nigmatullin on the economically important ommastre- phids and by Nesis on sepiids and loliginids. Further, in a short review of the whiplash squids of the family Chi- roteuthidae by Nesis, the new genus Asperoteuthis was established. Popov & Skarlato reviewed the bivalve family Cardi- tidae in the North Pacific, describing a new species of Cyclocardia, while Kafanov reconsidered the living car- diids in the Black Sea, making several nomenclatorial al- terations. Of particular interest to those working on car- diids is a paper by Zaiko, Zaiko & Krasnov who assert that temperature effects the number of ribs on the shell, rendering narrowly circumscribed rib-counts rather sus- pect for taxonomic purposes. Izzatullaev examined the five species in the freshwater bivalve family Corbiculidae in Central Asia. Kuznetzov, Kozaka & Isibasi investigated the relation- ship of gill-size to palp-size in several bivalves, concluding that the deposit feeding Tellinacea have proportionately much larger palps than suspension feeding bivalves like mytilids or venerids, an adaptation documented earlier by other authors. For a continental Palearctic freshwater fauna, that of Siberia seems extremely rich: Dolgin & Johansen dis- cussed in some detail 31 species and recorded 65 species of freshwater mollusks from northwestern Siberia, and even in the more isolated Kureyka River, a tributary of the Yenisey above the Arctic Circle, 41 species of fresh- Page 330 water mollusks were listed by Gundrizer! For more south- erly climes, Zatravkin enumerated 55 species of fresh- water mollusks, 39 gastropods, and 16 bivalves, from the Il’mensk Preserve in the southern Urals. ABBREVIATIONS BMV—Biologiya Morya (Marine Biology, Vladivostok). BPGF—Biochim. i populyatzion. genet. rib. (Biochemical and population genetics of fish). ES—English summary. GZ—Gidrobiologicheskii Zhurnal (Hydrobiological Journal). NDVS—Nauch. Dokl. Vyssh. Shkol. Biol. Nauk. (Scientific Re- ports of the Higher Educational School for Biological Sci- ences). NPS—Nov. probl. zool. naukhi i ilk otrazhenie vyzovsk. pre- podavanii. Tez. Dokl. Nauch. Konf. Zoologov. Ped. in-tov. ch. 1 Stavropol (New problems in zoological science and their effect on university teaching. Thesis Reports on the teaching of science conference. Zoology. Pedagogical Institute, Stavro- pol). PEMZ—Vopr. Evoluts. Morfol. Zhivotnykh. Kazan (Problems of the Evolutionary Morphology of Animals. Kazan). PMIN—Paleobiogeokhimya mor. bespozvonochnykh (Paleobio- geochemistry of Marine Invertebrates, Novosibirsk). TIO—Trudy Instituta Okeanologii. Akademiya Nauk SSSR (Transactions of the Institute of Oceanology, Academy of Sci- ences, USSR). ZEBF—Zhurnal Evolyutsionnoi biokhimii i fiziologii (Journal of Evolutionary Biochemistry and Physiology). ZZ—Zoologicheskii Zhurnal (Zoological Journal). 14th PSC—14 Tikhookean. nauch. Kongr. (14th Pacific Science Congress). GENERAL ALYAKRINSKAYA, I. O. 1979. On the survival of mollusks under conditions of dehydration. Dikhatel’n Belki Nekotor. Grupp. Sovrem. Zhivotnikh (Respiratory proteins of several groups of Recent animals). Moscow, pp. 151-155. [Duration of survival of five aquatic species of snails and ten of clams in an air environment was studied. Though many species survived for several days to over a week, Planorbis corneus sur- vived for two months, even in temperatures above 15°C. ] ALYAKRINSKAYA, I. O. 1979. Dissolution of shell hypostracum in several mollusks. Dikhatel’n Belki Nekotor. Grupp. So- vrem. Zhivotnikh (Respiratory proteins of several groups of Recent animals). Moscow, pp. 155-159. [A significant increase in the concentration of calcium in the hemolymph is displayed in the Black Sea bivalve mollusk Venus gallinae under conditions of dehydration and in the terrestrial Caucasian mollusk Caucasotachea atrolabiata during the summer, the source of which appears to be the internal layers of the shell. Appended is a list of mollusks that utilize dissolution of shell hypostracum during interruptions of normal conditions of res- piration. ] ARTYUSHENKO, O. T. & I. V. MEL’NICHYK. 1979. Paleobo- tanical and malacofaunistic characteristics of the Quaternary deposits of the basal canyon near Mt. Snyatin (Pre-Carpathi- ans). Ukrainian Botanical Journal 36(6):528-532, 622. (In Ukrainian, with Russian and English Summaries). [Palynological and molluscan data showed appreciable differ- ences between forest and plain horizons in neanthropogenic de- posits in the Pre-Carpathians. | The Veliger, Vol. 28, No. 3 BEREZKINA, G. N. 1979. Some data on the biology of Limnaea atra in the Smolensk Region. NPS, pp. 49-51. BERGER, V. YA. 1979. Euryhaline marine mollusks: morpho- logical and functional aspects. 14th PSC, Sect. F, pp. 5-6. Do.ain, V. I. & B. G. JOHANSEN. 1979. Ecological and mor- phological characteristics of new and little known freshwater mollusks of northwestern Siberia. Nov. Dannie o Faune i Flo- re Sibiri (New Contributions on the Fauna and Flora of Si- beria). Tomsk, pp. 47-61. (The distribution, abundance, and morphological characteristics of 31 mollusks are discussed as is the role of these animals in the diets of fish and birds.] Do tain, V. I. & B. G. JOHANSEN. 1980. Ecological and geo- graphical characteristics of the mollusks of northwestern Si- beria. Nov. Dannye v Prirode Sibiri (New Contributions to the Natural History of Siberia). Tomsk, pp. 30-42. [Ecological and zoogeographic data on 65 species of freshwater mollusks are presented. ] FROLENKOVA, O. A. & N. D. KruGLov. 1979. On the mor- phology of the egg capsules in the molluscan families Acro- loxidae, Bulinidae, and Planorbidae. NPS, pp. 181-183. GRIDNEY, E. A. & E. A. KAZANNIKOV. 1979. On the mainte- nance of pond snails (lymnaeids) under laboratory conditions. NPS, p. 62. GuL’BIN, V. V. 1979. Sixth All-Union Meeting on the study of mollusks. Mollusks, principle results of their study. BMV, No. 6, p. 86. [96 papers were presented which dealt with various aspects of the ecology, physiology, and biology of marine bivalves, gastro- pods, and cephalopods as well as freshwater and terrestrial mol- lusks and their parasites. ] GUNDRIZER, V. A. 1979. Freshwater mollusks of the Kureyka River (Basin of the Lower Yenisey). Nov. Dannie o Faune i Flore Sibiri (New Contributions on the Fauna and Flora of Siberia). Tomsk, pp. 62-68. [The malacofauna of the Kureyka, a right bank tributary of the lower Yenisey, consists of 41 species, 8 recorded for the first time. Information is presented on their ecology, abundance, and role in the diets of fish.] Karanov, A. I. 1979. On conservatism and variability in growth temperatures in the shells of marine mollusks. BMV, No. 6, pp. 59-69 (ES). [Miocene specimens of Ciliatocardium ciliatum were shown, using an oxygen isotope method, to have higher average temperatures of growth than contemporary samples, indicating the trend of global cooling over Cenozoic time.] Karanov, A. I. 1979. On the ecological evolution of the mal- acofauna of the cool temperate shelf of the Northern Hemi- sphere and the paleoclimatological significance of marine bi- valves. Transactions of the Institute of Biology and Soil Science of the Far Eastern Scientific Center, Academy of Sciences of the USSR, 52/155, pp. 58-72. [An hypothesis is proposed to explain the evolution of this fauna since the Neogene, or late Miocene. Ecological parameters such as temperature are reconstructed on the basis of oxygen isotope analyses. | KAZANNIKOV, E. A. 1979. Freshwater mollusks of the Stav- ropol region. NPS, p. 81. KHOKHUTKIN, I. M. 1979. Sixth All-Union Meeting for the study of mollusks. 7-9 Feb. 1979. Ekologiya (Ecology), No. 6, p. 104. Notes, Information & News [Organized by the Zoological Institute, the conference heard 149 papers dealing with the ecology of mollusks, under the following subheadings: population ecology, species formation, intrapopu- lational variation in phenotype frequencies, population genetics methodology, evolutionary morphogenesis, co-evolution, radio- isotope tracers, growth, development, and other topics. } Koz.ova, L. E. & N. T. MANpDRIkovaA. 1980. Characteristics of composition of phragmacones of belemnites and shells of bivalve mollusks from Toarcian deposits in Yakut. PMIN, pp. 81-84. [Data on the mineralogical and chemical compositions of belem- nite phragmacones and bivalve shells collected together in the basin of the Vilyuy River were analyzed by infra-red spectros- copy, x-ray diffraction, and determination of specific heat; the mineralogical composition of carbonates was shown to be differ- ent.] KRIVOSHEINA, L. V. 1979. On the zoogeographic characteris- tics of the freshwater malacofauna of the Upper Irtish River basin. Priroda i Kh-vo Vost. Kazakhstana (Nature and Fish- eries of Eastern Kazakhstan), pp. 100-107. [93 species and subspecies are known from the basin of Upper Irtish, of these 15 have Palearctic distributions, 23 European- Siberian and 8 Siberian; one species is endemic. ] KruGLov, N. D. 1979. Reproductive biology and observations on protandry among lymaeids. NPS, pp. 91-92. Lur’e, A. A. & S. A. BEYER. 1980. A method for marking mollusks. MS Application, E. I. Marchinovski Inst. of Par- asitology and Tropical Medicine. [Radioactive cobalt and silver were applied, under a layer of water repellent lacquer, to the shells of Bithynia inflata, Biom- phalaria alexandrina, and Physa acuta, which were then found to be detectable in the field by scintillation radiometry. ] Lur’zE, A. A. & S. A. BEYER. 1980. On a new method of marking freshwater mollusks. ZZ 59(4):609-619 (ES). [Long-lived isotopes implanted in the shell and covered by a film of nail polish proved to be effective in following populations of Bithynia inflata for two years. | MoskvIcHEVA, I. M. 1979. Studies on the malacofauna of the Upper Zeya (River) Basin. [Amur Province, Western Siberia. ] NPS, p. 122. NATOCHIN, Yu. V., V. YA. BERGER, E. A. Lavrova, O. Yu. MIKHAILOVA & V. V. KHLEBOVICH. 1979. The roles of so- dium and potassium in the regulation of cell volume in littoral mollusks. 14th PSC, Sect. F, pp. 32-33. [Marine mollusks are capable of partial regulation of cell volume during changes in the salinity of the environment by regulation of intracellular levels of free amino acids and concentrations of electrolytes (Na, K, Cl).] NIKOLAEV, V. A. 1979. Clausiliid land snails from the central Russian hills. NPS, pp. 126-127. Potyakov, D. M. 1980. On the choice of a carrier for quan- titative spectral analysis of micro-elements in the shells of ma- rine mollusks. PMIN, pp. 139-143. [The carrier of choice for quantitative analyses of Fe, Mg, Mn, Sr, and Ba by emission spectroscopy was shown to be 10% PbCI.] REZNIK, Z. V. & N. TikHova. 1979. Terrestrial mollusks of highland pastures in the Urup regions [Western Caucasus] of the Stavropol Territory. NPS, pp. 144-145. SHAKHMAEV, N. K. 1979. Study of the mechanisms of the accumulation of manganese in freshwater mollusks. Khim. 1 Biokhim. Okislenne Sistem, Soderzhashch, d-elementy. Page 331 (Chemical and biochemical oxidative systems, maintaining d-elements). Chelyabinsk, pp. 38-39. [Mn accumulates by a variety of mechanisms in different organs, e.g., in gills, 28.8% by metabolism, 17.8% by adsorption, and 4.6% by diffusion; respective values are given for mantle, diges- tive and gonadal tissue.] ZATRAVKIN, M. N. 1980. Aquatic malacofauna of the I’mensk Preserve (Southern Urals). ZZ 59(3):452-455 (ES). [The 162 samples taken in June-August 1975 were found to contain 55 species (39 gastropods and 16 bivalves). In compar- ison, Tausson collected 62 species from 1937-1940. Their geo- graphic affinities were: 1 Irtish endemic, 17 European, 1 Sibe- rian, 2 southwestern European, 1 northwestern European, and 40 Euro-Siberian and Palearctic.] ZATRAVKIN, M. N., E. D. PAvLova & V. F. RopIonov. 1970. Gastropods of the Upper Volga. NPS, p. 74. POLY PLACOPHORA SIRENKO, B. I. 1979. Chitons (Polyplacophora) of the coastal waters of Simushir Island. Biology of the shelf of the Kuril Islands. Moscow, pp. 200-208. [16 species in 11 genera were found to occur in the near-shore waters off Simushir (to 70 m). Highest densities were 3100/m? by Juvenichiton albocinnamoneus and 800/m? by Spongioradsia subaleutica. Lepidozona thielei accounts for the greatest biomass (160 g/m’). The biogeographic composition of the fauna consists of high boreal species 65%, widely distributed boreal species 29%, and boreal-arctic species 6%. | GASTROPODA, GENERAL ANDRONNIKOV, V. B. 1980. Threshold temperatures of cellular thermonarcosis of littoral mollusks of coral islands and the temperature conditions of their environment. TIO 90:51-57 (ES). [Using pedal musculature of snails from supralittoral, littoral, and upper sublittoral zones from the Pacific Ocean, the author established the threshold temperature of thermonarcosis for dif- ferent species. ] GuL’BIN, V. V. 1979. Distribution of prosobranch gastropod mollusks on the Shelf of the Kuril Islands. Biology of the Shelf of the Kuril Islands, Moscow, pp. 209-221. [A study of the vertical distribution and relationship to substrate of prosobranch gastropod mollusks of the Shelf of the Kuril Islands showed: more warm-water species dwell in the upper zones of the sea, and their number declines with depth while the opposite is true of cold-water species. A vertical zonation of the shelf results that differs in different parts of the island chain. The most important factor influencing the distribution of mol- lusks appears to be the temperature of the water. The greatest number of species inhabit rocky substrates, the least gravel sub- strates; only 20% of gastropod mollusk species are restricted to one type of bottom, while the remainder can inhabit a variety of substrates. | KorRNYUSHIN, A. V. 1980. On the land mollusk fauna of the Black Sea Preserve. Vestnik Zoologii (Zoological Herald), No. 2, pp. 75-78. [25 species (2 prosobranchs) constitute the fauna, most of which are widely distributed Holarctic species; data on habitat, distri- bution, and predation by birds are included. ] SIRENKO, B.I. 1980. Gastropods of Scotta Reef. Biol. Korallov. Rifob. Morfol., Sistemat., Ekol. (Biology of Coral Reefs; Mor- phology, Systematics and Ecology). Moscow, pp. 87-112. Page 332 [192 species of gastropods are listed with notes on their ecological zonation. | VoLova, G. N., A. N. Goutrkov & O. G. KUSAKIN. 1979. Shelled gastropod mollusks of Peter the Great Bay. Vladivos- tok, Dal’nevost. Kn. Isd.-vo. (Far Eastern Book Press), 170 PP- GASTROPODA, PROSOBRANCHIA Barskov, I. S., M. A. GOLOVINOVA & V. N. GoRYACHEV. 1980. On the structure of the nacreous layers of deepwater Seguenzia (Mollusca: Gastropoda). Dokl. AN. SSSR (Reports of the Acad. Sci. USSR), 252(4):1015-1017 (ES). [The structure of the nacreous layer of this genus differ sharply from the columnar nacre of other gastropods, being reminiscent of some bivalves and imparting greater strength to the shell.] Go ikov, A. N. 1980. Fauna of the USSR. Mollusca. Vol. 5, part 2. The molluscan sub-family Buccininae in the World Ocean. Leningrad, Nauka (Science [Press]), 508 pp., 42 pls. [This monograph is a comprehensive study of the world fauna of Buccininae, including 3 genera, 3 subgenera, 93 species and subspecies, and numerous ecological forms and varieties. A full synonymy, description, figures, and an analysis of ecology and distribution are given for each species. A special section includes tables with species and subspecies diagnoses. Features of adap- tive and evolutionary morphogenesis are discussed in the main section. Utilizing data from historical geology and paleontology, the author presents a spatial-temporal scheme of the evolution of the Buccininae, which shows how it coincides with the evo- lution of ecosystems in the Northern Hemisphere during the Cenozoic. All vertical zones of epicontinental bodies of water of temperate and cold latitudes of the Northern Hemisphere are divided into biogeographic regions based on these data. In the ecology section, quantitative methods were used to examine the relationships between species and different types of substrates, vertical distribution, temperature, including optimal and surviv- able temperatures, and salinity. Special emphasis was placed on temperatures of growth, life-span, and productivity of the com- mon species. During studies of reproductive ecology, the egg capsules of many species were identified, and conditions optimal to reproduction and artificial propagation clarified. ] IZZATULLAEV, Z. 1979. On new species of gastropod mollusks of the family Pomatiopsidae (Mollusca: Discopoda) from Ta- dzhikistan. Reports Acad. Sci. Tadz, SSR 22(10):629-631 (Tadzhik Summary). [Previously, mollusks from underground springs of central Asia (genera Kainarella and Pseudocaspia) were included in the family Littorinidae Gray, 1857. Detailed studies of the holotypes al- lowed the author to clarify the systematic affinities of these species and to include them in the Pomatiopsidae Stimpson, 1865, a group whose representatives are widely distributed in the con- tinental waterways of eastern Asia. Described are Kaznarella lukharevi and Pseudocaspia rozae. The former is similar in shell form to K. minima from the southeastern region of Turkmenis- tan, but is distinguished on the basis of its irregular cylindrical shell shape, its round-oval aperture appressed to the wall of the penultimate whorl, the absence of surface sculpture, and by its larger size (holotype: shell 1.5 mm high, 0.65 mm wide; aperture 0.55 mm high, 0.5 mm wide). Number of whorls is 4. Pseudo- caspia rozae has an egg-shaped to conical, strong, light-brown shell. Dimensions of the holotype: shell 3 mm high, 1.2 mm wide; aperture 1.2 mm high, 1.1 mm wide. Number of whorls rc 5.] The Veliger, Vol. 28, No. 3 Kantor, Yu. I. 1980. Species composition and variability of the gastropod molluscan genus Buccinum in the White Sea. ZZ 59(4):518-528 (ES). [Seven species are recorded, two (B. maltzani, and B. finmar- chianum) for the first time. Shell shape and sculpture as well as penial configuration were utilized as distinguishing traits.] POBEREZHNI, E. S. & V. I. Maxksimov. 1979. On unusual forms of the operculum of the mollusk Benedictia limnaeoides from Lake Baikal. Hydrobiological and Icthyological Inves- tigations of Eastern Siberia, Irkutsk, pp. 186-188. [A triangular rather than an ovate operculum was found in a single female snail of a sample of 330 taken from 5 to 10 m; the authors claim this rare variant is connected to the polyploidy known to occur in Benedictia.| SEMENOV, O. Yu. 1979. Experimental studies of the biology of the mollusk Melanopsis praemorsa L. Vestn. LGU (Herald of Leningrad State University), No. 15, pp. 9-17 (ES). [Briefly investigated were the distribution and dispersal of this species, which is not affected by conditions of light, but does prefer warmer areas. |] STAROBOGATOV, YA. I. & Z. I. IZZATULLAEV. 1980. Mollusks of the family Melanoididae (Gastropoda: Pectinibranchia) of central Asia and adjacent territories. ZZ 59(1):23-31 (ES). [On the basis of the structure of the pallial portion of the female reproductive system, the authors propose to divide the family Thiaridae, as usually accepted, into three independent families: Melanatriidae Thiele, 1929, with a completely open pallial gonoduct, albumen gland in the renal portion, and presence of a bursa; Thiaridae Preston, 1915, with a massive pallial gono- duct reaching to the mid-length of the mantle cavity; and Mel- anoididae Starobogatov, fam. nov., with the pallial gonoduct represented by two parallel, non-glandular tubes reaching the mid-length of the mantle cavity. Analysis of Melanoides tuber- culatus from many stations in central Asia and Afghanistan sug- gests that it can be divided into five species on the basis of shell proportions and sculpture. Doubt is thus cast on the single species concept of M. tuberculatus throughout its extensive range of west- ern Africa to Polynesia. There are three species in the territory of the USSR: Melanoides pamiricus, M. kainarensis, and M. shah- daraensis; the last two are described in this paper. ] ZHIRMUNSKH, A. V., V. L. Kas’yANov & V. I. LUKIN. 1980. The Mollusk Haliotis or sea ear. Priroda (Nature), No. 8, pp. 44-406. {[Moneron Island, off the SW coast of Sakhalin in the Sea of Japan is the only place in the USSR where Haliotis discus is found; it also occurs in northern Honshu, Hokkaido, and the islands of Rebun and Rishiri off NW Hokkaido. In 1972 and 1976-77 biologists from the Far Eastern Center studied this population which was noted on cliffs in the Laminaria zone at depths of 0.5-2 m. Although population density was high in places, the species was, in general, sparse, and it is unclear whether the population is self-replenishing or whether an influx of larvae on a branch of the Chusumski Current from Rebun and Rishiri islands sustains it. Halzotis is fished commercially in many areas of the world, both for its tasty meat and for its beautiful shell. In Japan, the harvest is 5-7 thousand tons per year. It is proposed that the Moneron population be included in “The Red Book of the USSR.’’] GASTROPODA, PULMONATA, AQUATIC KrucLov, N. D. 1980. Reproductive biology of freshwater pulmonate mollusks. ZZ 59(7):986-995 (ES). Notes, Information & News [The reproductive biology of 27 species of freshwater pulmonates belonging to the families Lymnaeidae, Physidae, Bulinidae, and Planorbidae was found to pass through two stages of gonadal development: male gonadal maturity and hermaphroditic gonad- al maturity. The first copulation is always as a male. Products of sperm readsorption in the spermatheca are humorally trans- mitted and influence endocrine control over subsequent devel- opment of the female portion of the reproductive system. ] SHARKO, N. V. 1980. Adaptations to darkness in the eyes of the pond snail Lymnaea stagnalis. ZEBF 16(2):193-196. [Electrophysiological studies on the adaptation to darkness were conducted on isolated preparations of eyes of adult pond snails. Under stimulation by light flashes, an increase in the amplitude of the electroretinogram was observed with time. Initially the amplitude grows rapidly, later it stabilizes. The dynamics of the response to light stimulation of constant intensity depends on temperature, the optimum being 17-20°C.] SMIRENINA, L. K. 1979. On the problem of copulation among aquatic gastropods. Biol. Vnutr. Vod. (Biol. Internal Waters), Leningrad, No. 43, pp. 27-29. [In pairing experiments with Planorbarius corneus and Lymnaea stagnalis, the latter more easily found each other in aquaria, indicating better long-distance chemoreception in Lymnaea.| STAROBOGATOV, YA. I. & N. D. KRuGLov. 1979. On two species of pond snail, genus Limnaea, new to the fauna of the Soviet Union. NPS, pp. 162-163. GASTROPODA, PULMONATA, TERRESTRIAL AL’MUKHAMBETOVA, S. K. & K. K. UVALIEVA. 1979. Mollusks of the family Vertiginidae (Mollusca: Gastropoda) of south and southeastern Kazakhstan. Izv. AN KazSSR (Proceedings of the Kazakhstan Academy of Sciences). Biol. Ser., No. 4, pp. 35-40 (Kazakh Summary). [An ecological-faunistic survey of the mountain ranges of south and southeastern Kazakhstan revealed nine species (two de- scribed as new) of the family Vertiginidae. ] AL’MUKHAMBETOVA, S. K. & K. K. UVALIEVA. 1980. Mollusks of the family Pupillidae (Mollusca: Gastropoda) from south and southeastern Kazakhstan. Izv. AN KazSSR (Proceedings of the Kazakhstan Academy of Sciences). Biol. Ser., No. 2, pp. 27-32. [The ecology, biology, distribution, and variability were studied in pupillid species which are separated by features of the repro- ductive system. ] Davtoy, S. SH. 1979. Inducers of feeding behavior in Helix vulgaris (Stylommatophora: Helicidae). ZZ 58(10):1464-1469 (ES). [Starch and glycogen always elicit a feeding reaction; less fre- quently (10-40%) materials of animal origin cause it, suggesting potential carnivory. | Dmitrieva, E. F. & Ya. S. SHAPIRO. 1979. Studies of non- specific reactions of the reticulated slug to methaldehyde. Nauch. Tr. Leningr. S.-Kh. In-ta. (Scientific Transactions of the Leningrad S. Kh. Institute), No. 374, pp. 59-62. [The sensitivity of various organs of juvenile slugs to the toxin methaldehyde was investigated histochemically. ] IZZATULLAEV, Z. 1980. On the life cycle of the slug Lytopelte maculata (Boch. and Heynemann, 1874) (Mollusca: Gastro- poda) in Tadzhikistan. Izv. Acad. Nauk. Tadzh. SSR, Otd. Biol. (Proceedings Acad. Sci. Tadzhikistan SSR, Biological Sciences Section), No. 1, pp. 95-97. Page 333 [The structure, coloration, dimensions, and genitalia of Lytopelte maculata are described as are the details of reproduction, ecology, and distribution. ] IZZATULLAEV, Z. & A. A. SHILEYKO. 1980. A new species in the terrestrial molluscan genus Bradybaena from central Asia and observations on the genus Ponsadenia. Dokl. AN Tadzh. SSR (Reports of the Tadzhikistan Academy of Sciences) 23(4): 220-224 (Tadzhik Summary). [Bradybaena squamulosa (type-locality: Cholpon-Ata, near Lake Issyk-Kul in Kirgiz) is described as new based upon features of the reproductive system and upon small, round-triangular peri- ostracal scales that are similar to those on Ponsadenia hirsuta from the Terskey Mountains. Diagnoses are given for the sub- genera Tarbagataja and Ponsadenia. | RIMZHANOV, T. S. 1979. New contributions to the molluscan fauna of the family Bradybaenidae (Mollusca: Gastropoda) of the Zailiysky Mountains. Izv. AN KazSSR (Proceedings of the Kazakhstan Academy of Sciences). Biol. Ser., No. 6, pp. 51-57 (Kazakh Summary). [One species and one subspecies are described as new based on the morphology of the shell and the structure of the genital system. ] SAMIGIN, F. I. & L. D. KaRPENKo. 1980. Motor organization of defensive reflexes in mollusks. NDVS, No. 3, pp. 38-42. [Bodies of two motor neurons, responsible for contracting the respective right and left columellar muscles, were found in the right and left pedal ganglia of the grape snail.] SHAPIRO, YA. S. 1979. Terrestrial mollusks of the agrobioce- noses of Leningrad Province (Rept. 1). Nauch. Tr. Leningr. S-Kh. In-ta. (Scientific Transactions of the Leningrad Insti- tute), No. 374, pp. 62-65. [16 species of land mollusks representing 7 families were col- lected on agricultural lands of Leningrad Province.] SHIKOV, E. V. 1979. Effects of industrial activities of man on the distribution of terrestrial mollusks. The protection of na- ture in the Upper Volga, Kalinin, pp. 30-50. {1732 samples with over 60,000 individuals, dating from 1963 to 1979 and taken in the environs of Kalinin, Novgorod, Pskovsk, Leningrad, southern Mirmansk, and Moscow showed that the human factors most responsible for influencing terrestrial mol- lusks were fire, agriculture, alteration of waterways, and intro- duction of foreign species of snails. ] SHIkKov, E. V. 1979. Dependence of the distribution of slugs of the genus Deroceras Rafinesque, 1820 in the flood-plains of the large rivers of the Valdai Hills on the direction of pre- vailing winds. Ekologiya (Ecology), No. 5, pp. 97-99. [In the Staritsk region of Kalinin Province along the flood-plain of the Volga River there occur four species of Deroceras: agreste which is distributed in exact correspondence with the direction of the prevailing southwesterly winds, retzculatum occupies the most protected areas, and sturany: and laeve usually co-occur with both species. ] SHILEYKO, A. A. & Z. IZZATULLAEV. 1980. Taxonomic struc- ture of the terrestrial mollusks of the family Pupillidae in the fauna of the USSR and a description of a new species from central Asia. Dokl. AN Tadzh. SSR (Reports of the Tadzhik- istan Academy of Sciences) 23(5):282-285 (Tadzhik Sum- mary). [The diagnoses of three genera and two subgenera of the family are given, as is a description of Gibbulinopsis (Primipupilla) na- nosignata.| Page 334 TavasigEv, R. A. & T. A. TAVASIEVA. 1980. A new species of Caucasigena (Gastropoda: Hygromiidae) from the central Caucasus. ZZ 59(1):144-146 (ES). [Caucasigena schileykoi is described from limestone cliffs in beech forests at an altitude of 800 m above sea level in the North Ossetian Autonomous Republic; it is distinguished from C. ren- garteni by a sharp keel and by features of the reproductive sys- tem.] Uva.ieva, K. K. 1980. Ecological faunistic survey of the ter- restrial mollusks of the forest-steppe habitat. Zool. Inst. Acad. Sci. KazSSR, Alma-Ata, 22 pp., MS No. 2051-2080. [550 samples of land mollusks were collected on cattle pastures of collective farms in northern and central Kazakhstan and yield- ed 32 species, representing 17 genera and 13 families; 8 species are first reported from the area and 2 species are described as new.] ZEIFERT, D. V. & I. M. KHOKHUTKIN. 1979. Experimental studies on natural migrations in populations of autochthonous and introduced species of mollusks. Ecological studies of forest and meadow biocenoses in the Transural Plains. Info. mate- rials Talitsk. Hospital. Sverdlovsk, pp. 46-50. [Marked specimens of Bradybaena fruticum and Eobania vermic- ulata were used to show that spatial distribution is determined by the type of plant cover.] ZHULIDOV, A. V. 1980. On the concentration of gastropods (Mollusca: Pulmonata) on plots of stinging nettles containing increased levels of some chemical elements. Vestnik Zoologii (Zoological Herald), No. 2, pp. 78-79. [In the Voronezhky Preserve, heterogenous distribution was not- ed (mainly in Succinea putris and Eulota fruticum) in thickets of the stinging nettle Utrica pubescens; densities varied from zero to 178-211 snails/m? and it was shown that the snails preferred high levels of several trace elements. ] BIVALVIA ALYAKRINSKAYA, I. O. 1979. On the properties and sizes of shell crystals in bivalve mollusks. Dikhatel’n Belki Nekotor. Grupp. Sovrem. Zhivotnikh (Respiratory proteins of several groups of Recent animals). Moscow, pp. 142-150. [Dissolution rates of shell crystals were investigated at differing pH’s.} ANGELOV, A. 1976. Revision of the family Pisidiidae in Bul- garia. Annual Report, Faculty of Zoology, University of Sofia 69(1):109-119 (Bulgarian; German Summary). [242 samples from 145 collecting sites yielded three species of Sphaerium and ten of Pisidium; data include synonymies, descrip- tions, measurements, habitat characteristics, and distribution. ] DororeeEva, L. A. & A. V. KHABAKOV. 1980. Determination of environmental temperatures for Recent and late Quater- nary oysters, using the Ca/Mg method. Byul. Mosk. O-ba. Ispyt. Prirody. Otd. Geol. (Bulletin of the Moscow Naturalists Soc., Geol. Soc.) 55(4):106-113. [Accumulation of Mg in calcitic shells of Recent oysters is gov- erned by the temperature regime and is independent of salinity. Average temperatures of surface waters inhabited by late Qua- ternary oysters from the Karangatsky Horizon of the Kerchensk Peninsula were 22-23°C during the warm period of the year while the average annual temperatures were 15-16°C.] GeERASIMOVA, T. N. 1980. Seasonal changes in the dimensions and biomass of Didacna trigonoides (Pall.) in the Caspian Sea. GZ 16(2):53-55 (ES). The Veliger, Vol. 28, No. 3 [Biomass alters significantly with the seasons, being drastically reduced in April-June at the time of the release of gametes. ] Goromosova, S. A. & A. Z. SHAPIRO. 1979. Physiological and biochemical aspects of adaptations of mussels in normal and in extreme conditions. Promisl. Dvustvorchat. Mollyuski— Midii i ikh rol’ v ekosistemakh (Commercially important bi- valve mollusks—mussels and their role in ecosystems). Len- ingrad, pp. 45-47. [Under hypoxic conditions, the oxidized NAD necessary for gly- colysis is produced by malate dehydrogenase. | Goromosova, S. A. & V. A. TAMOZHNYAYA. 1980. Seasonal variation of transaminases in tissues of Black Sea mussels. BMV, No. 2, pp. 67-68 (ES). [Intracellular localization and seasonal variation in activity of alanine aminotransferase and aspartate aminotransferase in the tissues of Mytilus galloprovincialis were studied. The intracellular distribution of aminotransferases depends on the function of the tissue, being mainly cytoplasmic in muscles and gills and mito- chondrial in the hepatopancreas and gonads. Two peaks in ac- tivity occur: autumn and spring, both declining during active gametogenesis. ] GREENBERG, M. J. & L. I. Ditton. 1979. Salinity adaptation and probable interdependence between heart muscle physiol- ogy, phylogeny, and biogeography of bivalve mollusks: basic directions for future research. 14th PSC, Sect. F, pp. 14-15. [In the bivalve heart, the auricles appear especially to be the primary filter of urea. The subclasses Pteriomorpha, Hetero- donta, and Paleoheterodonta are distinguished by the following physiological characters: the form of the action potential, ionic dependence, excitability, cholinergic systems of the myocardia, as well as by larger structural differences. ] IGNATEV, A. V. & E. V. Krasnov. 1980. Isotopic oxygen composition of water and the growth temperatures of Recent and Quaternary mollusks of the Chukotsk Sea. PMIN, pp. 56-60. [Basing their analysis on living and fossil bivalves from the shores of Wrangel Island in the Chukotsk Sea, the authors show that temperature changes of marine waters in the Northern Hemi- sphere during Pliocene-Quaternary time can be adequately doc- umented by oxygen isotope paleothermometry.] IGNAT’EV, A. V. & I. M. ROMANENKO. 1980. Correlation of magnesium content of mussel shells with their mineral com- position, structure, and growth temperatures. PMIN, pp. 85- ile [In mussels from Peter the Great Bay, Mg levels increase on- togenetically, show seasonal fluctuations coincident with changes in water temperature, and exhibit sharp increases not correlated with seasonal events. | IZZATULLAEV, Z. 1980. Bivalve mollusks of the family Cor- biculidae in central Asia. ZZ 59(8):1130-1136 (ES). [Of five species of corbiculids found in central Asia, two, tabeten- sis and ferghanensis which are ovoviviparous, are allocated to Cor- biculina Dall, and three, cor, fluminalis, and purpurea which are presumed to be oviparous, to Corbicula Mihlfeld.] KaFanovy, A. I. 1980. On the nomenclature of the Cardiidae (Bivalvia) of the Sea of Azov and the Black Sea. ZZ 59(4): 623-626 (ES). [The nomenclature of three species and one subspecies of car- diids inhabiting the Sea of Azov-Black Sea basin as well as the Mediterranean is discussed. Cardium hystrix (Lightfoot, 1786) is considered a synonym of C. echinatum Linne, 1758. The follow- ing new names are proposed: C. ciliare L., 1758, for C. pauci- Notes, Information & News costatum Sowerby, 1834; Acanthocardia (Sphaerocardium) ciliaris milaschewitschi Kafanov, nom. n., for C. paucicostatum var. impedita Milaschewitsch, 1909, non C. impeditum Deshayes, 1860; Didacna (Pontalmyra) kamyshburunensis Kafanov, nom. n., for C. paucicostatum Deshayes 1838, non Sowerby 1834. The divi- sion of Cerastoderma glaucum (Poiret, 1789) into four species by Skarlato and Starobogatov (1972, “Guide to the fauna of the Black Sea and the Sea of Azov,” pp. 178-249, Kiev) is regarded as correct. ] KaRPENKO, A. A. 1980. Avoidance reaction to living starfish in the marine scallop Patinopecten yessoensis (Mollusca: Bi- valvia). ZZ 59(1):146-149 (ES). [The avoidance reaction, which changes with age of scallops, is a compound, unconditioned reflex, composed of three reactions: an “alert phase,” a “response phase,” and a “swimming phase.”’] KARTAVTZEV, YU. F. 1979. Possible determination of a bal- anced polymorphism in loci coding for isoenzymes. BPGF, pp. 36-40 (ES). [Either an increase or a decrease in heterozygosity with age was observed in the majority of loci (approx. 70%) of five species of mussels. It is interpreted as being due to some form of balancing selection and indicative of the selective nature of isozyme poly- morphisms. | Krasnov, E. V., N. A. SIn’kov, V. O. KHUDOLOZHKIN, A. V. IGNAT’EV, A. A. KARABTZOV & O. I. NEDAVA. 1980. Com- plex studies of the shell material in fossil and Recent speci- mens of Arctica islandica L. PMIN, pp. 73-80. [X-ray, spectrophotometric and mass-spectroscopic analyses of Plio-Pleistocene and Recent Arctica islandica from eastern Ice- land showed that concentrations of Mg, Sr, Na, Fe, and Mn in aragonitic shells increased with geological age. Growth temper- atures were investigated by '*O/'°O ratios in glacial and inter- glacial periods.] Krasnov, E. V., V. A. ZAIkKO & N. N. Zaiko. 1979. Biogeo- chemical indicators of adaptations of marine mollusks to changes in salinity. 14th PSC, Sect. F, p. 27. [Ontogenetic variation in the incorporation of chlorine into the shells of pectinids was shown, with maximum levels occurring during the autumnal period of rapid growth; sculptural features such as the number of ribs in Patinopecten yessoensis and Swifto- pecten swifti as well as in Anadara broughtoni vary with salinity and temperature. | KUZNETZOV, A. P., M. Kozaka & I. Isipasi. 1980. Dimen- sional characteristics of gills and labial palps of several marine mollusks. ZZ 59(2):175-180 (ES). [Dimensions of gills and palps were measured in Moerella je- doensis, a deposit feeder, and in Ruditapes philippinarum and Mytilus edulis, both suspension feeders; in M. jedoensis, about 40% of the total gill-palp area was taken up by the gill and 60% by the palp, while in the other species, over 90% is gill and less than 10% is palp. Thus, the Tellinacea (deposit feeders) should be considered an independent ecological group. Arguments are advanced supporting the origin of the Eulamellibranchia, Pseu- dolamellibranchia, and Filibranchia from the Protobranchia.] Lukanin, V. V. 1979. Roles of cellular and organismic reac- tions in the accommodation of mussels to changes in salinity. Promisl. Dvustvorchat. Mollyuski—Midii i ikh rol’ v ekosis- temakh. (Commercially important bivalve mollusks—mussels and their role in ecosystems). Leningrad, pp. 82-83. [Mussels have the ability to undergo adaptive changes in func- tion at cellular and organismic levels during seasonal changes in salinity. Evolutionary pathways of adaptation to low salinities are considered. | Page 335 MILEIKOvskKII, S. A. 1979. On the maintenance of the structure and recruitment of spat into the druzes [mats] in the mussel Crenomytilus grayanus. BMV, No. 5, pp. 39-43 (ES). [Young larval spat are recruited into the adult attached masses, called druz in Russian, of these mussels; such ‘“‘nursery-like” behavior is apparently caused by an attraction to the byssal strands of adults and also protects the tiny spat.] NIKIFOROV, S. M. 1979. Genetic and morphometric variability of the far eastern oyster (C7vassotrea gigas). BPGF, pp. 134- 138 (ES). [Electrophoretic study of 46 loci in five populations in Peter the Great Bay showed polymorphism at more than 30% of the loci, an average heterozygosity of 0.07-0.08, and heterozygote defi- ciencies in most populations. | NIsTRATOVA, S. N., T. M. Turpayev, N. N. Gopovikov, M. N. Gopovikova & V. I. DaniLova. 1980. Analysis of the action of several organophosphate inhibitors of cholinesterase on the hearts of bivalve mollusks. ZEBF 16(1):30-38 (ES). [This study investigates the action of organophosphate inhibitors of cholinesterase on isolated ventricles from the hearts of the bivalve mollusks Crenomytilus grayanus, Spisula sachalinensis, and Anodonta complanata. | Popov, S. V. 1980. The formation and development of the hinge during the ontogeny of North Pacific bivalve mollusks of the family Carditidae. ZZ 59(6):945-948 (ES). [Hinge formation in six species (Cyclocardia ventricosa, C. cre- bricostata, C. rjabininae, C. isaotakii, Miodontiscus annakensis, and Crassicardia crassidens) originates in a similar manner. The car- dinal teeth of the right valve 3a 3b and the lower lateral tooth AIII appear from the lower primary plate III. The third car- dinal tooth appears as a raised edge in the larval stage (nymph). In the left valve, plate IV gives rise to teeth 4b and AIV, the anterior cardinal tooth 2 is newly formed. The complete formula of the hinge is: AV AIITI 3a 3b PIII AIV 2 4b PII PIV In species of both Cyclocardia and Crassicardia, the development of lateral teeth stops in the early stages; in the adult they are barely discernible. In Miodontiscus, these teeth are developed in all stages of growth. Crassicardia crassidens differs notably in morphology from species of the genus C'yclocardia at the early dissoconch stage, supporting the independence of the genus Cras- sicardia. | Popov, S. V. & O. A. SKARLATO. 1980. The bivalve mollusks of the family Carditidae in the North Pacific and adjacent seas. ZZ 59(7):996-1007 (ES). [Representatives of the family in the North Pacific, including the Sea of Japan, Okhotsk Sea, Bering Straits, and Chukotsk Sea are: Crassicardia crassidens, Cyclocardia crebricostata, C. rja- bininae, C. isaotaku, C. ferruginea, Miodontiscus annakensis, and M. prolongatus, the latter three being characteristic of the Bering Straits High Boreal Province, where they occur with Cyclocardia ventricosa ovata and C. ripensis. Diagnoses of genera and species are included. | Pozpnyakova, L. A. 1980. On the dynamics of the calcium/ magnesium ratio in calcitic shells of closely related species of bivalve mollusks in the Sea of Japan. PMIN, pp. 92-105. [Ontogenetic variations of Ca/Mg ratios in three species of pec- tinids (Patinopecten yessoensis, Chlamys swifti, and C. farreri) re- flect seasonal fluctuations of water temperature. ] PROSKURINA, E. S. 1979. On linear and weight growth of the Page 336 principle bivalve mollusks of the Aral Sea. GZ 15(5):105- 106. [Studies were conducted on the native Dreissena polymorpha var. aralensis and Cerastoderma lamarcku lamarcku and on the intro- duced Abra ovata collected in 1973-1974. Age was determined by analysis of growth lines; average annual growth was 1.7, 3.5, and 2.06 mm respectively. Abra may, in time, become one of the primary components of the benthos in the Aral Sea.] PRYADKO, V. P. & V.A. KRISAL’NYI. 1980. Histophysiological changes in tissues of several organs of Anodonta cygnea under the influence of different calcium concentrations. GZ 16(1): 56-59 (ES). {Entry of calcium ions into the organism causes a redistribution of the concentrations of K* and Na* ions in the cells of the glandular apparatus of the gills and in the foot muscle. The overall metabolism of calcium increases with the increase in ac- tivity of tissue enzymes. The gills of freshwater mollusks com- prise important depots of calcium salts. ] Rusakov, Yu. I. & V. K. Kazakov. 1979. Extraction of an insulin-like substance from mollusks and production of anti- sera to it. ZEBF 15(6):617-619 (ES). [An insulin-like substance was extracted from the visceral mass of the freshwater clams, Unio pictorum and Anodonta cygnea, antisera prepared, and their properties investigated. ] SELIN, N. I. 1980. Coordinating conference for the study of mussels (Mytilidae). Leningrad 12-14 Feb. 1979, BMV, No. 2, pp. 80-81. [45 papers were presented, including works on systematics, dis- tribution, morphology, ecology, growth, nutrition, and economic importance of mussels. } SkuL’skil, I. A., I. V. BurRovina & N. B. Pivovarova. 1979. Mechanisms of potassium homeostasis in mussels inhabiting seas of varying salinities. 14th PSC, Sect. F, pp. 42-43. [Two mechanisms are suggested for the maintenance of optimal intracellular concentrations of K in different environmental sa- linities. | STANKYAVICHYUS, A. B. 1979. Osmotic and ionic regulation in east Baltic mussels, Mytilus edulis, adapted to different salin- ities of water. Promisl. Dvustvorchat. Mollyuski—Midii i ikh rol’ v ekosistemakh (Commercially important bivalve mol- lusks—mussels and their role in ecosystems). Leningrad, pp. 114-115. [Under hypotonic conditions, mussels have the ability to main- tain elevated osmotic pressure due to the isolation of the mantle cavity from external surroundings. | YAVNoV, S. V. 1979. Second All-Union Symposium on the morphology, systematics, phylogeny and ecogenesis of bivalve mollusks. Tiraspol, 3-4 Oct. 1978. BMV, No. 5, pp. 93-94. [The symposium was dedicated to the morphology, taxonomy, paleo- and neo-ecology of oysters (suborder Ostreina) and mac- tras (superfamily Mactroida). Twenty-five papers were heard, among these: On the origin and phylogeny of oysters (O. A. Skarlato, Ya. I. Starobogatov & V. A. Sobetskii); Studies on Upper Cretaceous oysters and their habitats (L. A. Dorofeyeva, A. V. Khabakov & V. A. Sobetskii); Species structure, distri- bution and paleoecology of four subfamilies of Upper Cretaceous oysters (Z. N. Poyarkova); Genetic systematics of contemporary oysters of the southern shore (S. M. Nikiforov); Microstructure of mactrid shells and its implications for systematics (S. V. Yav- nov); Development of larval shells in mactrids (L. A. Medved- eva); and Trophic structure of populations (A. P. Kuznetsov).] The Veliger, Vol. 28, No. 3 Yavnov, S. V. 1980. Shell structure in mollusks of the family Mactridae. BMV, No. 3, pp. 62-66 (ES). [Three varieties of crossed-lamellar structures were discerned, with two sublayers in the external layer and a single internal layer, in seven species of this family from Japan, Okhotsk, Black and Barents seas. ] YAvNov, S. V. & A. V. IGNAT’EV. 1979. Shell structure and growth temperatures in mollusks of the family Mactridae. BMV, No. 5, pp. 44-48 (ES). [Using layered structures of the shells, the authors determined the maximum ages for three species in the Sea of Japan: Spisula sachalinensis, 55 years; S. voya, 52 years; and Mactra sulcatania, 12 years. Optimum growth temperatures were also determined. ] ZAIKO, V. A., N. N. ZAIkKO & E. V. KRAsNov. 1980. Shell sculptures of marine bivalve mollusks as an indicator of the salinity of their habitats. PMIN, pp. 106-112. [Salinity affects the number of ribs in Cardium edule; an equation is given correlating the relationship between the number of ribs and the average salinity.] ZOLOTAREV, V. N., D. M. PoLyAKov & N. A. SIN’Kov. 1980. Comparison of the chemical composition of the shells of several Recent and subfossil mollusks from the Sea of Japan. PMIN, pp. 61-72. [Incorporation of Mg, Sr, Fe, Mn, and Ba into the calcium carbonate matrix of shells decreases ontogenetically; Fe and Mn accumulations are initially greater while in larger annulations Ba, Mg, and Sr are found in higher concentrations. ] CEPHALOPODA Dusinina, T. S. 1980. On the finding of larvae of the squid Moroteuthis robsoni (Oegopsida: Onychoteuthidae) in the southwestern Atlantic. ZZ 59(7):1094-1096 (ES). [The late larval stage of Moroteuthis robsoni, a species widely distributed in the southern Atlantic, is described for the first time; its characters closely approach those of Onykia carriboea. | Fitippova, Yu. A & V. L. YUKHOov. 1979. Species composition and distribution of cephalopod mollusks in meso- and bathy- pelagic Antarctic waters. Antarktika (Moscow), No. 18, pp. 175-187. [Analysis of sperm whale gut contents and samples taken on various research vessels indicate a high degree of endemism in Antarctic cephalopods. ] KOTELEVTZEV, YU. V. 1980. Photoaffinity marking of the ace- tylcholine receptor from optic ganglia of squid. “Materials of the 11th Conference on molecular studies. Faculty of Biology, Moscow State University,” Moscow, pp. 103-109, figs. [Studies were conducted on the binding of the photoaffinity li- gand azidocytisine to the nicotine acetylcholine receptor in the optic ganglia of Loligo. | Nesis, K. N. 1979. A short note on the zoogeography of the pelagic fauna of the Australia-New Zealand region. TIO, 106:125-139. {Based on samples collected by the R/VS Mendeleev, Vityas, Ob1, the zoogeographic distributions of 66 species were characterized. ] Nests, K. N. 1980. Sepiids and loliginids: a comparative sur- vey of the distribution and evolution of neritic cephalopod mollusks. ZZ 59(5):677-688 (ES). [The horizontal and vertical distributions of the cuttlefish family Sepiidae and the squid family Loliginidae in the Pacific Ocean were analyzed. The greatest generic and species diversities of Notes, Information & News loliginids occur in the Indo-Malaysian Province of the Indo- West Pacific; aberrant forms tend to be tropical. Sepiids are absent from the New World; in the Old World, their distribu- tions practically coincide with those of loliginids. Sepiids are more diverse, with endemism and abundance of aberrant forms greatest not in the tropics but in the subtropics (South Africa, Japan, China, southern half of Australia). Species of both fam- ilies can be divided into upper sublittoral, eurybathic (entire shelf), and lower sublittoral-upper bathyal, but the fraction of “deep water” species is significantly higher in sepiids than in loliginids. The evolution and adaptive radiation of sepiids are discussed in connection with their dominance of the subtropics and relatively greater depths.] NEsis, K. N. 1980. On the systematic position of Chiroteuthis Jamelica Berry (Cephalopoda: Oegopsida). Byul. Mosk. Obshch. Isp. Prir. (Bulletin of the Moscow Naturalists Society), Bi- ology Series, No. 4, pp. 59-66 (ES). [Studies of a specimen intermediate in dimensions between the holotypes of Chiroteuthis famelica Berry, 1909 (postlarval), and Chiroteuthis acanthoderma Lu, 1977 (a grown but immature in- dividual), have shown that these taxa are synonymous. A new genus Asperoteuthis Nesis is proposed for C. famelica, an eastern central Pacific mesobathic species undergoing daily vertical mi- grations (day 600-1000 m, night 200-400 m). Nine genera are now known in the family Chiroteuthidae; a list of known species is provided. ] NIGMATULLIN, CH. M. 1979. Principle stages in the evolution of the squid family Ommastrephidae (Cephalopoda: Oegop- sidae). PEMZ, pp. 210-219. [Oegopsid squids have evolved as active swimmers (nektonic) with the ommastrephids representing the acme of the lineage. The earliest representatives of this family were probably unspe- cialized nekto-benthic forms in the transition zone between the shelf and continental slope (100-350 m). The three constituent subfamilies are discussed in terms of presumed evolutionary se- quences: the IIlicinae with J//ex, a nearshore nektonic form hav- ing originated in the western Atlantic and spread to the eastern Atlantic, with 7odaropsis diverging from the stem lineage early to occupy a nerito-oceanic niche: the Todarodinae occur over the continental slope and partly open ocean with basically neritic forms (e.g., Martialia and Nototodarus). Radiation of the todari- nines occurred in oceanic surface waters above the slope and adjacent parts of the open ocean, especially in high latitudes. The successful exploitation of the open ocean took place within two lineages—the ornithoteuthine and the ommastrephine. The former expanded into the bathyal and middle depths, the latter occupy the epipelagic niche with the larger species occurring in high latitudes. Only Dosidicus did not fully adapt to the oceanic epipelagic zone, staying predominantly in comparatively near- shore waters of high productivity. Parallelisms and convergences between the Ommastrephidae and the scombroid fishes in their adaptations to the nektonic oceanic niche are discussed. ] PincHukov, M. A. & Yu. V. KorzuN. 1979. On the discovery of a representative of the genus Nototodarus (Cephalopoda: Ommastrephidae) in the western portions of the Indian Ocean. Tr. 4th Konf. Mold. Uchenikh, pp. 144-146. (Transactions of the 4th conference of young scientists.) [This is a preliminary description of a squid of the genus No- totodarus first collected on the Saya de Malha Bank and on the southern shelf of Somalia. On the basis of important taxonomic features, these western Indian Ocean squid differ sharply from N. nipponicus and are close to N. sloani, representing a new subspecies of the latter, or possibly a full species in this genus. Page 337 ROZENGART, E. V., A. P. BRESTKIN & Yu. I. KAS’YANENKO. 1979. Specific differences in phosphatase activity in optic ganglia of Pacific squid. 4th Internat. Biochem. Meeting. Moscow, p. 166. [Nerve tissues of the squids, Berryteuthis magister, Ommastrephes bartrami, Todarodes pacificus, and Nototodarus sloani sloani, lack alkaline phosphatase and contain acid phosphatases that differ in molecular weight between species. | ZuEV, G. V., GH. M. NIGMATULLIN & V. N. NIKOL’ski. 1979. Growth and lifespan of the wing-armed squid Stenoteuthis pteropus in the eastern central Atlantic. ZZ 58(11):1632-1641 (ES). [Growth of linear dimensions and weight was studied. Life-span does not exceed 1-1.5 years. ] ZuEV, G. V., GH. M. NIGMATULLIN & V. N. NIKOL’SKII. 1980. A method for quantitatively surveying oceanic epipelagic squid. Kolichetsv. Metodi v ekol. Zhivotnich. (Quant. Methods in Animal Ecol.). Leningrad, pp. 57-59. [A method, utilizing the natural attraction of squids to light, was developed to study Stenoteuthis pteropus and S. oualaniensis. | Joseph Rosewater, 1929-1985 The malacological community suffered a great loss on March 22 with the passing of Dr. Joseph Rosewater. At the time of his death, Dr. Rosewater, an authority on the taxonomy and evolutionary biology of mollusks, was cu- rator of the mollusk division of the National Museum of Natural History, Smithsonian Institution. The author of more than 80 technical works, Dr. Rosewater was a val- ued contributor to the pages of our journal, both as an author and as a frequent reviewer of submitted manu- scripts. His talents, dedication, and generosity will be missed. California Malacozoological Society California Malacozoological Society, Inc., is a non-profit educational corporation (Articles of Incorporation No. 463389 were filed January 6, 1964 in the office of the Secretary of State). The Society publishes a scientific quarterly, The Veliger. Donations to the Society are used to pay a part of the production costs and thus to keep the subscription rate at a minimum. Donors may designate the Fund to which their contribution is to be credited: Operating Fund (available for current production); Sav- ings Fund (available only for specified purposes, such as publication of especially long and significant papers); En- dowment Fund (the income from which is available. The principal is irrevocably dedicated to scientific and educa- tional purposes). Unassigned donations will be used ac- cording to greatest need. Contributions to the C.M.S., Inc., are deductible by donors as provided in section 170 of the Internal Revenue Code (for Federal income tax purposes). Bequests, lega- cies, gifts, and devises are deductible for Federal estate and gift tax purposes under section 2055, 2106, and 2522 Page 338 of the Code. The Treasurer of the C.M.S., Inc., will issue suitable receipts which may be used by donors to substan- tiate their tax deductions. Page Charges Although we would like to publish papers without charge, high costs of publication require that we ask authors to defray a portion of the cost of publishing their papers in The Veliger. We wish, however, to avoid possible financial handicap to younger contributors, or others without fi- nancial means, and to have charges fall most heavily on those who can best afford them. Therefore, the following voluntary charges have been adopted by the Executive Board of the California Malacozoological Society: $30 per printed page for authors with grant or institutional sup- port and $10 per page for authors who must pay from personal funds (2.5 manuscript pages produce about 1 printed page). In addition to page charges, authors of papers containing an extraordinary number of tables and figures should expect to be billed for these excess tables and figures at cost. It should be noted that even at the highest rate of $30 per page the Society is subsidizing well over half of the publication cost of a paper. However, authors for whom the regular page charges would present a financial handicap should so state in a letter accompa- nying the original manuscript. The letter will be consid- ered an application to the Society for a grant to cover necessary publication costs. We emphasize that these are voluntary page charges and that they are unrelated to acceptance or rejection of manu- scripts for The Veliger. Acceptance is entirely on the basis of merit of the manuscript, and charges are to be paid after publication of the manuscript, if at all. Because these contributions are voluntary, they may be considered by authors as tax deductible donations to the Society. Such contributions are necessary, however, for the continued good financial health of the Society, and thus the contin- ued publication of The Veliger. Reprints While it was hoped at the “birth” of The Veliger that a modest number of reprints could be supplied to authors free of charge, this has not yet become possible. Reprints are supplied to authors at cost, and requests for reprints should be addressed directly to the authors concerned. The Society does not maintain stocks of reprints and also can- not undertake to forward requests for reprints to the au- thor(s) concerned. Patronage Groups Since the inception of The Veliger in 1958, many generous people, organizations, and institutions have given our journal substantial support in the form of monetary do- The Veliger, Vol. 28, No. 3 nations, either to The Veliger Endowment Fund, The Ve- lager Operating Fund, or to be used at our discretion. This help has been instrumental in maintaining the high qual- ity of the journal, especially in view of the rapidly rising costs of production. At a recent Executive Board Meeting, we felt we should find a way to give much-deserved recognition to those past and future donors who so evidently have our best interests at heart. At the same time, we wish to broaden the basis of financial support for The Veliger, and thus to serve our purpose of fostering malacological research and publica- tion. Accordingly, it was decided to publicly honor our friends and donors. Henceforth, donors of $1000.00 or more will automatically become known as Patrons of The Veliger, donors of $500.00 or more will be known as Sponsors of The Veliger, and those giving $100.00 or more will become Benefactors of The Veliger. Lesser donations are also sincerely encouraged, and those donors will be known as Friends of The Veliger. As a partial expression of our gratitude, the names of donors in these different categories will be listed in a regular issue of the journal. Of course, we will honor the wishes of any donor who would like to remain anonymous. The Treasurer of the California Malacozoological Society will provide each member of the new patronage groups with a receipt that may be used for tax purposes. We thank all past and future donors for their truly helpful support and interest in the Society and The Veli- ger. Through that support, donors participate directly and importantly in producing a journal of high quality, one of which we all can be proud. Notes to Prospective Authors The increasing use of computers to prepare manuscript copy prompts the following notes. We request that the right margin of submitted papers be prepared “ragged,” that is, not justified. Although right-justified margins on printed copy sometimes look “neater,” the irregular spac- ing that results between words makes the reviewer’s, ed- itor’s, and printer’s tasks more difficult and subject to error. Similarly, the automatic hyphenation capability of many machines makes for additional editorial work and potential confusion; it is best not to hyphenate words at the end of a line. Above all, manuscripts should be printed with a printer that yields unambiguous, high-quality copy. With some printers, especially some of the dot-matrix kinds, copy is generally difficult to read and, specifically, the letters “‘a, p, g, and q” are difficult to distinguish, especially when underlined as for scientific names; again, errors may result. Other reminders are (1) that three copies of everything (figures, tables, and text) should be submitted to speed the review process, and (2) absolutely everything should be double-spaced, including tables, references, and figure leg- ends. Because The Veliger is an international journal, we oc- Notes, Information & News casionally receive inquiries as to whether papers in lan- guages other than English are acceptable. Our policy is that manuscripts must be in English. In addition, authors whose first language is other than English should seek the assistance of a colleague who is fluent in English before submitting a manuscript. Subscription Rates and Membership Dues At its regular Annual Business Meeting on September 25, 1985, the Executive Board of the California Malacozoo- logical Society, Inc., set the subscription rates and mem- bership dues for Volume 29 of The Veliger. For affiliate members of the Society, the subscription rate for Volume 29 will be US$25.00; this now includes postage to domestic addresses. For libraries and nonmembers the subscription rate will be US$50.00, also now with postage to domestic addresses included. An additional US$3.50 is required for all subscriptions sent to foreign addresses, including Can- ada and Mexico. Affiliate membership in the California Malacozoologi- cal Society is open to persons (no institutional member- ships) interested in any aspect of malacology. There is a one-time membership fee of US$2.00, after payment of which, membership is maintained in good standing by the timely renewal of the subscription. Send all business correspondence, including subscription orders, membership applications, payments for them, and changes of address to C.M.S., Inc., P.O. Box 9977, Berke- ley, CA 94709. Moving? If your address is changed it will be important to notify us of the new address at least six weeks before the effective date and not less than sz7x weeks before our regular mailing dates. Send notification to C.M.S., Inc., P.O. Box 9977, Berkeley, CA 94709. Because of a number of drastic changes in the regula- tions affecting second class mailing, there is now a sizable charge to us on the returned copies as well as for our remailing to the new address. We are forced to ask our members and subscribers for reimbursement of these charges; further, because of increased costs in connection Page 339 with the new mailing plate, we also must ask for reim- bursement of that expense. The following charges must be made: change of address and re-mailing of a returned issue— $2.75 minimum, but not more than actual cost to us. We must emphasize that these charges cover only our actual expenses and do not include compensation for the extra work involved in re-packing and re-mailing re- turned copies. Sale of C. M. S. Publications All back volumes still in print, both paper-covered and cloth-bound, are available only through “The Shell Cab- inet,” 12991 Bristow Road, Nokesville, VA 22123. The same applies to the supplements still in print, with certain exceptions (see below). Prices of available items may be obtained by applying to Mr. Morgan Breeden at the above address. Volumes 1 through 13, 24, 26, and 27 are out of print. Supplements still available are: part 1 and part 2, sup- plement to Vol. 3, and supplements to Volumes 7, 11, 14, 15, and 16; these can be purchased from ‘““The Shell Cab- inet” only. Copies of the supplement to Volume 17 (“Growth rates, depth preference and ecological succes- sion of some sessile marine invertebrates in Monterey Harbor” by E. C. Haderlie) may be obtained by applying to Dr. E. C. Haderlie, U.S. Naval Post-Graduate School, Monterey, GA 93940; the supplement to Volume 18 (“Chitons”’) is available from ‘““The Secretary,” Hopkins Marine Station, Pacific Grove, CA 93950. Some out-of-print editions of the publications of C. M. S. are available as microfiche reproductions through Mr. Steven J. Long. The microfiches are available as negative films (printed matter appearing white on black back- ground), 105 mm Xx 148 mm, and can be supplied im- mediately. The following is a list of items now ready: Volumes 1-6: $9.95 each Volumes 7-12: $12.95 each Supplement to Volume 6: $3.95; to Volume 18, $6.95 Send orders to Mr. Steven J. Long, Shells and Sea Life, 1701 Hyland, Bayside, CA 95524. The Veliger 28(3):340 ( January 2, 1986) THE VELIGER © CMS, Inc., 1986 BOOKS, PERIODICALS & PAMPHLETS A Review of the Genera of the Rissoidae (Mollusca: Mesogastropoda: Rissoacea) by WINSTON F. PONDER. 1985. Records of the Australian Museum, Suppl. 4:221 pp. A Review of the Genera of the Barleeidae (Mollusca: Gastropoda: Rissoacea) by WINSTON F. PONDER. 1983. Records of the Australian Museum 35:231-281. These two major papers are a culmination of many years of work on the small marine gastropods of the su- perfamily Rissoacea. Ponder has added massively to our knowledge of this group by carefully examining the type specimens of the genera, and by detailed study and anal- ysis of living animals, anatomies, radulae, opercula, and shell morphologies. The illustrations are outstanding. Ponder also discusses in considerable detail taxa that have been previously misassigned to these families. Needless to say, shell morphology provides an imperfect clue to rela- tionships. These papers, and Ponder’s smaller papers on this su- perfamily, are relevant to students of the American faunas because he discusses and allocates many species from both the western Atlantic and the eastern Pacific. E. V. Coan A NEW JOURNAL FOR INDO-PACIFIC ZOOLOGY Indo-Malayan Zoology edited by JEAN BOUILLON & MICHEL JANGOUx. A. A. Balkema, Publishers. Lisplein 11, P.O. Box 1675, NL- 3000 BR Rotterdam, Netherlands. Annual Subscription price $25.00 + $3.50 postage. Although not a malacological journal, the first number of Volume 1 of Jndo-Malayan Zoology contains two mol- luscan papers and has a malacologist on its editorial board (J. L. van Goethem). The journal is published twice a year in two issues of about 160 pages each and accepts manuscripts in either English or French. The emphasis of the journal is strongly marine, and encompasses ecol- ogy, systematics, and biogeography of Indo-Malayan and Melanesian animals. C. S. Hickman Information for Contributors Manuscripts Manuscripts must be typed on white paper, 8%” by 11”, and double-spaced throughout (including references, figure legends, footnotes, and tables). If computer generated copy is to be submitted, margins should be ragged right (7.e., not justified). To facilitate the review process, manuscripts, including figures, should be submitted in triplicate. The first mention in the text of the scientific name of a species should be accompanied by the taxonomic authority, including the year, if possible. Underline scientific names and other words to be printed in italics. Metric and Celsius units are to be used. The sequence of manuscript components should be as follows in most cases: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, lit- erature cited, figure legends, figures, footnotes, and tables. The title page should be on a separate sheet and should include the title, author’s name, and address. The abstract should describe in the briefest possible way (normally less than 200 words) the scope, main results, and conclusions of the paper. Literature cited References in the text should be given by the name of the author(s) followed by the date of publication: for one author (Smith, 1951), for two authors (Smith & Jones, 1952), and for more than two (Smith et al., 1953). The “‘literature cited” section must include all (but not additional) references quoted in the text. References should be listed in alphabetical order and typed on sheets separate from the text. Each citation must be complete and in the following form: a) Periodicals Cate, J. M. 1962. On the identifications of five Pacific Mitra. Veliger 4:132-134. b) Books Yonge, C. M. & T. E. Thompson. 1976. Living marine molluscs. Collins: London. 288 pp. c) Composite works Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135. In: R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal invertebrates of California. Stanford Univ. Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Figures and plates Figures must be carefully prepared and should be submitted ready for publication. Each should have a short legend, listed on a sheet following the tables. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photographs for half-tone plates must be of good quality. They should be trimmed off squarely, arranged into plates, and mounted on suitable drawing board. Where necessary, a scale should be put on the actual figure. Preferably, photographs should be in the desired final size. It is the author’s responsibility that lettering is legible after final reduction (if any) and that lettering size is appropriate to the figure. Charges will be made for necessary alterations. Processing of manuscripts Upon receipt each manuscript is critically evaluated by at least two referees. Based on these evaluations the editor decides on acceptance or rejection. Acceptable manuscripts are returned to the author for consideration of comments and criticisms, and a finalized manuscript is sent to press. The author will receive from the printer two sets of proofs, which should be corrected carefully for printing errors. At this stage, stylistic changes are no longer appropriate, and changes other than the correction of printing errors will be charged to the author at cost. One set of corrected proofs should be returned to the editor. An order form for the purchase of reprints will accompany proofs. If reprints are desired, they are to be ordered directly from the printer. Send manuscripts, proofs, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. CONTENTS — Continued NOTES, INFORMATION & NEWS Consumption of pelagic red crabs by black abalone at San Nicolas and San Miguel Islands, California. GLENN R. VANBLARICOM AND BRENT S. STEWART ..............-.--.- 328 Soviet contributions to malacology in 1980. KENNETH J... Boss AND MiG: HIARASEWYCH 00s tiles) anaes 329 BOOKS, PERIODICALS & PAMPHLETS ISSN 0042-3211 THE VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 28 April 1, 1986 Number 4 CONTENTS Immunological detection of Mercenaria mercenaria in a predator and preparation of size-class specific antibodies. NOB ERGTan | ey RUBE RY vires adie aiarey ac icky 7 ive hierdie gies te contac Wek Ue et a sat toe 341 How a clam builds windows: shell microstructure in Corculum (Bivalvia: Car- diidae). MAR YaEaaVVADSON AND) PHILIP W. SIGNOR ..... 22.00.2002. oe ee, 348 Rates and processes of compensatory buoyancy change in Nautilus macromphalus. JPTBATRIBIRY: UD) IUNRRID) as 0 Nae Re Rea C a eS eicee UeRA 356 A model for shell patterns based on neural activity. BARD ERMENTROUT, JOHN CAMPBELL, AND GEORGE OSTER ............ 369, Predation-induced changes in growth form in a nudibranch-hydroid association. Gary GAULIN, LAUREN DILL, JULIE BEAULIEU, AND LARRY G. Harris... 389 Avoidance and escape responses of the gastropod Nucella emarginata (Deshayes, 1839) to the predatory seastar Pisaster ochraceus (Brandt, 1835). AYTBILTISISA, Ihr INA GIGI) A tee eae Sa eet OP PEST este Eee kacreieniotey MN ean 394 Oxygen production and consumption in the sacoglossan (=ascoglossan) Elysia chlorotica Gould. GLenys D. Gipson, DANIEL P. TOEWS, AND J. SHERMAN BLEAKNEY ..... BOF, Environmental perturbations reflected in internal shell growth patterns of Cor- bicula fluminea (Mollusca: Bivalvia). OWE VWeehREDZVAND RICHARDIYAG OWIRZ)) 2. ee. co cle! oglalueey aes at CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly on the first day of prpcdiBRre- January and April. Rates for Volume 28 are $22.00 for affiliate members (including domestic mailing charges) and $44.00 for libraries and nonmembers (including do- mestic mailing charges). An additional $3.00 is required for all subscriptions sent to foreign addresses, including Canada and Mexico. Further membership and subscrip- tion information appears on the inside cover. The Veliger is published by the Cali- fornia Malacozoological Society, Inc., % Department of Zoology, University of Cali- fornia, Berkeley, CA 94720. Second Class postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to C.M.S., Inc., P.O. Box 9977, Berkeley, CA 94709. THE VELIGER Scope of the journal The Veliger is open to original papers pertaining to any problem concerned with mol- lusks. This is meant to make facilities available for publication of original articles from a wide field of endeavor. Papers dealing with anatomical, cytological, distributional, eco- logical, histological, morphological, physiological, taxonomic, etc., aspects of marine, freshwater, or terrestrial mollusks from any region will be considered. Short articles containing descriptions of new species or lesser taxa will be given preferential treatment in the speed of publication provided that arrangements have been made by the author for depositing the holotype with a recognized public Museum. Museum numbers of the type specimen must be included in the manuscript. Type localities must be defined as accurately as possible, with geographical longitudes and latitudes added. Very short papers, generally not exceeding 500 words, will be published in a column entitled “NOTES, INFORMATION & NEWS”; in this column will also appear notices of meetings, as well as news items that are deemed of interest to our subscribers in general. Editor-in-Chief David W. Phillips, 2410 Oakenshield Road, Davis, CA 95616, USA Editorial Board Donald P. Abbott, Emeritus, Hopkins Marine Station of Stanford University Hans Bertsch, Universidad Autonoma de Baja California James T. Carlton, Williams College—Mystic Seaport Eugene V. Coan, Research Associate, California Academy of Sciences, San Francisco J. Wyatt Durham, University of California, Berkeley Cadet Hand, University of California, Berkeley Carole S. Hickman, University of California, Berkeley A. Myra Keen, Emerita, Stanford University David R. Lindberg, University of California, Berkeley James H. McLean, Los Angeles County Museum of Natural History Frank A. Pitelka, University of California, Berkeley Peter U. Rodda, California Academy of Sciences, San Francisco Clyde F. E. Roper, National Museum of Natural History, Washington Barry Roth, Santa Barbara Museum of Natural History Judith Terry Smith, Stanford University Ralph I. Smith, University of California, Berkeley Wayne P. Sousa, University of California, Berkeley T. E. Thompson, University of Bristol, England Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (no institutional memberships) interested in any aspect of malacology. As an afhliate member, a person may subscribe to The Veliger for US $22.00, which now includes mailing charges to domestic addresses. There is a one-time membership fee of US $2.00, after payment of which, membership is maintained in good standing by the timely renewal of the subscription; a reinstatement fee of US $1.00 will be required if membership renewals do not reach the Society on or before April 1 preceding the start of the new Volume. If a receipt is required, a self-addressed, stamped envelope (or in the case of foreign mem- bers, the envelope and two International Postal Reply coupons) should be included with the membership or subscription request. The annual subscription rate to The Veliger for libraries and nonmembers is US $44.00, which now includes mailing charges to domestic addresses. An additional US $3.00 is required for all subscriptions sent to foreign addresses, including Canada and Mexico. Memberships and subscriptions are by Volume only (July 1 to April 1) and are payable in advance to California Malacozoological Society, Inc. Single copies of an issue are US $30.00 plus postage. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: C.M.S., Inc., Post Office Box 9977, Berkeley, CA 94709. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. The Veliger 28(4):341-347 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 Immunological Detection of Mercenaria mercenaria in a Predator and Preparation of Size-Class Specific Antibodies by ROBERT J. FELLER Department of Biology, Marine Science Program, and the Belle W. Baruch Institute for Marine Biology and Coastal Research, University of South Carolina, Columbia, South Carolina 29208, U.S.A. Abstract. Successful culture of hard clams (Mercenaria mercenaria) requires high survivorship of seed stock during subtidal grow-out. This study was designed to identify natural predators of juvenile clams. Immunological techniques were used to identify M. mercenaria proteins in the guts of their natural invertebrate predators and to characterize antigen preparations (whole-organism extracts) of different size classes of clams. The grass shrimp Palaemonetes vulgaris was found to eat juvenile M. mercenaria. Immunoelectrophoretic separations and immunodiffusion tests of whole-organism extracts of M. mercenaria revealed unique antigens in the following size classes: veliger larvae, newly settled spat, juveniles, and adults. INTRODUCTION SHELLFISH PRODUCTION on the east coast of the United States is dominated by two commercially important bi- valves, the American oyster Crassostrea virginica (Gmelin, 1791), and the hard clam, Mercenaria mercenaria (Linné, 1758). Because favorable conditions for growth of the hard clam exist in the relatively warm estuarine waters of South Carolina, a pilot scale mariculture facility was established to estimate growth and survivorship of seed clams em- placed subtidally in cages (MANZI et al., 1980). Based on findings reported by numerous workers (e.g., MAc- KENZIE, 1970, 1977; KRANTZ & CHAMBERLIN, 1978; WHETSTONE & EVERSOLE, 1978, 1981; KRAEUTER & CasTaGNa, 1980), it was reasonable to expect that some seed stock would be lost to predators during grow-out. Unprotected juveniles are known to suffer tremendous mortality soon after settlement in the natural environment (MILEIKOVSKY, 1974; HIBBERT, 1977), but such losses of newly settled spat have never been successfully measured directly in the field. Besides the obvious difficulties of sort- ing, quantifying, and identifying very small (150-200 um) post-settlement individuals from sediment samples, the fragile young clams are also not easily identified in the guts of their potential predators. Loss of these small spat to their natural predators, then, is unlikely to be detected with conventional sampling and analytical techniques. This study was conceived to extend the use of an immunological method capable of detecting soluble proteins of M. mer- cenaria in stomachs of their natural marine predators, to identify heretofore suspected but undocumented predators upon this valuable species, and to characterize antigenic changes in the soluble proteins of M. mercenaria during its growth to marketable size. MATERIALS ann METHODS Preparation and Characterization of Antibodies Target specimens (seed clams) were procured from Tri- dent Seafarms Co., Charleston, SC, in eight size classes ranging from 0.6 to 16 mm total shell length (BROWN et al., 1983). Clams that had been growing in tray culture in natural seawater for periods up to several weeks were sorted from their culture debris, separated by size, and immersed live in filtered seawater. The seawater was treated with antibiotics to reduce bacterial contamination, and the clams allowed to empty their stomachs of ingesta for 3 or 4 days at room temperature (20-22°C). Animals from each size class were then solubilized in ice-cold, buff- ered TES-saline and centrifuged at 3000 x g for 15 min to remove particulates. The whole-organism-extract su- pernates of each size class served as antigens for the prep- aration of antisera in New Zealand white female rabbits, according to the protocol of FELLER et al. (1979). Antibodies were harvested from the rabbits by cardiac Page 342 puncture and assayed for titer and specificity in replicate using the double immunodiffusion micro-Ouchterlony technique in agarose gels (OUCHTERLONY, 1968). Further characterization of the antigen-antibody specificities was established using rocket-line and two-dimensional (crossed) immunoelectrophoresis separations in agarose (AXELSEN et al., 1973). In the two-dimensional technique, antigenic components of a given size class are separated according to their electrophoretic mobilities prior to precipitin line formation in the second dimension with all mobilities rel- ative to a bovine serum albumin (BSA) standard (AXELSEN & Bock, 1972). These methods allow visualization of an- tigen-antibody precipitin line patterns shared by each size class of clams when antigens are reacted with either their homologous or heterologous antibodies. Immunological distances among the eight size classes of Mercenaria mer- cenaria were assessed by computing a matrix of cross- reactions of each antiserum based upon their relative sim- ilarities with the homologous reaction. A hierarchical clustering algorithm (BMDP2M; Dixon, 1981) was used to construct a dendrogram as in FELLER & GALLAGHER (1982). Detection of Potential Predators Experimental grow-out cages belonging to Trident Sea- farms Co. were examined for macrobenthic predators at their high subtidal (—0.1 m, MLW) sites on Oak Island on 29 October 1980. The cages had been in place for five days and contained juvenile Mercenaria mercenaria in the size range 14-18 mm, a size well known to be essentially immune to predation even in open, unprotected sediments. No potential predators were found either on or within the pea gravel of any of the newly emplaced cages. Some other cages in the vicinity of the new ones (in place for approx- imately one year—Dr. J. Manzi, personal communica- tion) were also examined for potential clam predators by divers. A variety of invertebrates was taken from the cages which, at this time, no longer contained any clams, nor were their containment screens intact. Sediment samples taken in replicate with 2.5-cm diameter cores to a depth of 5 cm were collected at random from the area surround- ing both the old and new Oak Island cages and screened through a 250-um mesh. Several small grow-out cages (1 x 1 Xx 0.3 m) were emplaced subtidally in February, 1981 by Dr. J. Manzi near the Trident Seafarms Co. shore facility on Folly Beach, SC. They contained Mercenaria mercenaria rang- ing in size from 3.0 to 7.0 mm. A single cage containing clams was examined on 23 March 1981 for potential predators; the restraining mesh was not completely intact, and the cage contained numerous invertebrates, including the grass shrimp Palaemonetes vulgaris (Say, 1818) and the mud snail //yanassa obsoleta (Say, 1822). All organisms collected from the cage were frozen on dry ice immediately after collection. Immunological analysis of the stomach contents of sus- The Veliger, Vol. 28, No. 4 pected predators involved dissection of the gut from indi- vidual specimens, microscopic examination for visually identifiable remains, and solubilization of the gut mass in TES-saline using a chilled mortar and pestle (FELLER et al., 1979; FELLER, 1984). The solubilized proteins from within the gut were then analyzed by double immunodif- fusion on 25 xX 75 mm glass slides coated with agarose. The fluids (15 uL) were placed in a central well sur- rounded by antibody wells, each containing an antiserum to suspected prey—in this case different size classes of Mercenaria mercenaria. Antiserum to the predator itself, if available, was also used as a test control to ensure that any precipitin lines formed were due to proteins from the gut contents rather than sloughing of the predator’s gut wall proteins. When control antisera were not available, an antiserum to the most closely related taxon was used. If none of these were available, it was assumed that gut wall proteins did not mask the observed reactions. This typically did not cause problems, because very few pre- cipitin lines due to prey were observed in any of the pred- ator guts tested. RESULTS Characterization of Mercenaria mercenaria Antigens by Immunodiffusion Antisera were successfully prepared to antigens from eight size classes of Mercenaria mercenaria (Table 1). An- tibody titer, the reciprocal of the highest dilution of an antiserum that gives a detectable reaction with its homol- ogous antigen, was between 128 and 512 for size classes B and D-H, but only 64 for the two antisera prepared using antigens with the lowest protein concentrations, A and C. The number of precipitin lines produced by double immunodiffusion in replicated homologous double im- munodiffusion self-reactions increased with increasing clam shell length, but there was considerable overlap among the size classes in the numbers of self-reaction precipitin lines formed (Figure 1). The veligers (size class A) and spat (B and C) had similar and significantly fewer num- bers of homologous precipitin lines than the other size classes of older clams (D-H) which had similar numbers of lines (d.f. = 7,122; P < 0.001 by single-classification ANOVA). This immunological overlap reflects the amount of overlap in shell length among the size classes themselves and was entirely expected despite differences in mean weight per individual or protein concentrations among the eight size classes (Table 1). Antisera were unique to the extent that most of the cross-reactions between a given antiserum and antigens from heterologous size classes were not as extensive (did not produce as many precipitin lines) as the identity or self-reaction from that antiserum (Table 2). Exceptions to this include size classes E, F, and G; both E and F contained 6.0-mm individuals, and group G contained individuals similar in size to group F (Table 1). The cross-reactions involving antisera to these three Reap keller 986 Page 343 Table 1 Size-class composition of Mercenaria mercenaria antigen preparations. Shell length size Total wet wt. TES-buffer Protein conc. Sizeclass! No. indiv. range (mm) (mg) (mL) (mg/mL) Comments A? 10,000 0.2-0.3 300.0 10.0 1.4 veligers B 400 2.0-5.0 100.0 6.0 3.7 starved 3 days Cc 1466 0.6-3.5 4.3 6.0 eS, ground whole D 484 3.0-5.0 1285 10.0 2.8 ground whole E 320 5.0-6.0 11.9 985 Sell ground whole lf 197 6.0-9.9 5.6 9.5 4.1 ground whole G 53 10.0-16.6 il) 5.5 4.1 foot muscle H? 3 70.0-75.0 7.6 23.0 3.5 foot muscle ' A = veligers; B, C = newly set spat; D, E, F, G = juveniles; H = adult. 2 Provided by J. W. Ewart, Hatchery Manager, University of Delaware; these veligers had been fed Jsochrysis aff. galbana (T ISO) and Thalassiosira pseudonana 3H. >From North Inlet, South Carolina. groups were strong enough with antigens from adults (size class H) that neither D, E, F, nor G antisera could be considered size class specific. Cross-reactions among the A, B, C, and H size classes were all lower than the ho- mologous reactions based upon maximum numbers of lines observed (Table 2). Thus, it was possible, using double immunodiffusion tests, to distinguish among the following four size classes of M. mercenaria: veliger larvae (A), new- ly settled spat (B, C), juveniles (D, E, F, G), and adults (H) or “chowders.” A dendrogram of immunological sim- ilarity based on data in the cross-reaction matrix of Table 1S £ = 10 | ; E Ghee O 2 MBC. De EB. ik GH Size Class Figure 1 Mean number (+95% confidence limits) of precipitin lines ob- served in homologous antigen-antibody reactions for each size class of Mercenaria mercenaria defined in Table 1. 2 also reflects these differences between size classes (Fig- ure 2). The presence of cross-reactions among antisera to, and antigens from, Mercenaria mercenaria indicated that many of the size classes shared common antigenic proteins. To visualize these antigenic similarities, rocket-line electro- phorograms were prepared in 1% agarose gels. These tests essentially confirmed the immunological identities of the size classes discussed above, and in nearly all cases the numbers of precipitin lines observed in double diffusion tests were the same as observed in the rocket-line com- parisons. As a further check on the specificity of the sep- arate antisera, two-dimensional (crossed) electrophoro- grams were prepared on which the precipitin line patterns of both self- and cross-reactions could be visualized. The existence of both common and unique antigenic components among each of the separate size classes of Table 2 Maximum number of precipitin lines observed in micro- Ouchterlony double immunodiffusion tests using homol- ogous (on the diagonal) and heterologous reactions be- tween antibodies to and antigens of eight size classes of Mercenaria mercenaria (A-H defined in Table 1). Antigens (whole-organism extracts) A B Cc D E F G H Antisera (A) 9 qT 6 8 if 7 8 i (B) 7 10 6 8 10 9 5 9 (C) if 6 9 5 6 4 5 7 (D) 7 11 8 14 13 12 10 1 (E) 7 9 Tf 11 11 8 11 11 (F) 7 9 9 11 10 12 12 (G) 10 11 10 9 12 12 ils) 13 (H) 8 10 9 9 10 12 13 Page 344 Mercenaria mercenaria was demonstrated using the two- dimensional immunoelectrophoretic separation technique. All possible antigen-antibody reactions were compared, but to illustrate the basic principle, identity and cross- reactions involving the B and G size classes are shown in Figure 3. Not only do the precipitin-line peak patterns show which components are common to both antigens, but the heights of the peaks also reflect the different protein concentrations comprising each antigen-antibody complex (KENNY & Foy, 1975). Comparison of electrophoretic mobilities of each precipitin line relative to BSA’s migra- tion in the first dimension (where BSA’s migration dis- tance from the antigen well equals unity) also comple- mented the visual comparisons by establishing which peaks were unique or common to a given antigen-antibody re- action pair. For example, the antiserum for spat produced seven precipitin peaks of identical electrophoretic mobility with both spat and juvenile antigens, indicating that these antigenic components are common to both age classes (Figure 3). Detection of Potential Predators Having established the relative specificities and sensi- tivities of the antisera to individual size classes, it was possible to use them to detect the presence of specific Mer- cenaria mercenaria proteins in the stomachs of potential predators in the field. Amphipods of the genera Melita (n = 12) and Coro- phium (n = 15) collected from the year-old floating pens at Oak Island on 20 October 1980 did not contain any Mercenaria mercenaria protein, nor did any of the snap- ping shrimp Alpheus sp. (n = 8) or any of several nereid polychaetes examined. Apparently no clams were avail- able for ingestion in these old cages, and none were visible (as previously noted). Organisms collected by cores from around the newly placed pens (containing 18-mm clams in pea gravel) included a typical assemblage of low inter- tidal or high subtidal invertebrates in areas of compact oyster shell debris—nereid and phyllodocid polychaetes, turbellarians, nematodes, and a few harpacticoid cope- pods. The only member of this potential predator com- munity that contained bivalve protein was a single speci- men of Nereis sp. (12 mm total length), but it was not M. mercenaria protein; it was tentatively identified as C7ras- sostrea virginica protein. The newly emplaced clams were apparently not accessible to predators, nor were any pred- ators seen in, on, around, or under the new pens. The Folly Beach cage collections of predators on 23 March 1981 were examined for the presence of Mercenar- 1a mercenaria proteins in three specimens of the spionid polychaete Streblospio benedicti (Webster, 1879), six Ily- anassa obsoleta, two unknown errant polychaetes, and 22 specimens of Palaemonetes vulgaris. The P. vulgaris stom- achs were separated according to total shrimp length (from tip of rostrum to end of telson) into small (<20 cm), medium (20-25 cm), and large (26-27 cm) groups. Each of these groups of potential predators was homogenized The Veliger, Vol. 28, No. 4 60 50 DISTANCE 40 CBAHEGFD SIZE CLASS Figure 2 Dendrogram of immunological similarity (based on Euclidean distance) among antisera to the eight size classes of Mercenaria mercenaria defined in Table 1. in saline after visual analysis (dissecting microscope at 50) of individual organism’s gut smears revealed only the presence of amorphous material and fluids. Gut contents from the three groups of Palaemonetes vul- garis were tested for the presence of Mercenaria mercenaria proteins using antisera to four size classes, C-F inclusive. Because antiserum to P. vulgaris was unavailable for use as a control for these immunoassays, antiserum to P. pugio was used instead. The control reaction between antiserum to P. pugio and antigens from P. vulgaris produced eight distinct precipitin lines, whereas the cross-reaction be- tween any one of the four M. mercenaria antisera and either P. pugio or P. vulgaris antigens produced a maxi- mum of only two lines. Presence of M. mercenaria in P. vulgaris gut contents would thus be indicated by existence of more than two precipitin lines in the immunoassays, as an empty gut would produce the same number of lines as occur in the control cross-reaction. The combined gut contents of five small Palaemonetes vulgaris produced four precipitin lines with anti-D, the antiserum to juvenile Mercenaria mercenaria in size class D. The combined gut contents of 13 medium shrimp pro- duced three lines with anti-C, seven lines with anti-D, and three lines with anti-E. No more than two precipitin lines were produced in tests of the four large P. vulgaris whose gut contents were combined into one group for analysis. As an additional check that the lines observed were due to the presence of M. mercenaria in the shrimp guts, immunoassays were run with antiserum to size class D and known antigens of the same M. mercenaria size class adjacent to wells containing the shrimp gut contents. These tests produced lines of identity between the gut contents and the M. mercenaria antigens, thus confirming that the grass shrimp had indeed eaten M. mercenaria. Reva eller s986 Page 345 Se 0 EN ———————————————— a Two-dimensional Electrophoresis antigen well Figure 3 Schematics of two-dimensional immunoelectrophorograms for the homologous (1 and 4) and heterologous (2 and 3) reactions between antigens and antibodies to Mercenaria mercenaria size class B (spat) and G (juveniles). Antigens were separated for 1.25 h in the first (horizontal) dimension and then rocketed into a 20% (vol./vol.) antibody gel bed for 6.0 h, all at 2 Volts/cm with constant power at room temperature. Bovine serum albumin (BSA) was used as a marker to which all first dimension migration distances from the antigen well were referenced. Two microliters of anti-BSA (Miles Laboratories) were present in each antibody bed. (1) Spat antigen into anti-spat bed (the homologous reaction). (2) Spat antigen into anti-juvenile bed (the heterologous reaction). (3) Juvenile antigen into anti-spat bed (the heterologous reaction). (4) Juvenile antigen into anti-juvenile bed (the homologous reaction). DISCUSSION sede : immunodiffusion and immunoelectrophoresis (either Because separations of antigenic proteins based on Fickian rocket-line or two-dimensional) are independent meth- diffusion in agarose gels by double immunodiffusion are odologies suitable for establishing unique immunological not strictly comparable to separations based on electro- specificities of antisera. phoretic mobility, one cannot explicitly equate cross-re- Increasing immunogenicity of Mercenaria mercenaria actions observed with these two techniques. However, both antigens as a function of size reflects the more complex Page 346 nature of its antigenic components with increasing age (Figure 1). This ontogenetic phenomenon is reasonably well established for a variety of invertebrate taxa wherein the existence of unique developmental stage-specific or age-specific antigenic components allows immunochemi- cal detection of these taxa in predators (BOREHAM & OHIAGU, 1978). Unique two-dimensional separation pat- terns seen for antigens from the different size classes of M. mercenaria (e.g., Figure 3) point to potential use of the antibodies to detect age-specific predatory mortalities. If antigenic components are also a function of local food resources, then it may even be possible to develop habitat- specific antisera for the veliger age class of M. mercenaria and detect the routes of larval dispersal for this species. Notwithstanding the polymorphic traits of natural popu- lations (e.g., PESCH, 1974), such an approach has already been suggested by MENZIES & KERRIGAN (1978) for trac- ing routes of spiny lobster recruitment on the basis of their biochemical genetics. The development of antisera capable of detecting mi- nute quantities of Mercenaria mercenaria tissue proteins in the predator gut environment is a prerequisite for the use of immunological methods to detect otherwise un- known predators. The technique has been used success- fully in both terrestrial and aquatic habitats, and previ- ously unknown predator-prey linkages have usually been identified (BOREHAM & OHIAGU, 1978; CALVER, 1984; FELLER et al., 1985). The immunological study of gastro- pod predation on oysters by MARSHALL (1977) also attests to the power of this technique for identifying previously unknown predatory species. Finding that Palaemonetes vulgaris had eaten juvenile Mercenaria mercenaria was not particularly surprising, as these small shrimp are generalist feeders that typically tear and shred their food upon ingestion, rendering clam tissue visually unidentifiable. A more serious question is whether losses from such a small predator are potentially as great as those posed by other well known predators (drills, xanthiid crabs, asteroids, blue crabs, etc.). The Folly Beach cages were not sampled on any other dates; hence, it is unknown whether there were other predators present that could have ingested M. mercenaria from them. Casual observations of fauna in the area revealed the pres- ence of several known predators on bivalves (e.g., Calli- nectes sapidus {Rathbun, 1896], Urosalpinx cinerea [Say, 1822], birds, and xanthiid crabs), so the potential for ad- ditional losses via predation from those small cages did exist. The impact of large predators (whelks, drills, rays, and crabs) is known to be destructive on Mercenaria mercenaria populations (WALKER et al., 1980), and preventive mea- sures may be successful in restricting their access to cul- tures. A small motile predator such as Palaemonetes vul- garis, however, will be much more difficult to exclude, especially if it is small enough to go through protective meshes or screens. It is conceivable that the young of such a predator might gain access to a culture tray and grow The Veliger, Vol. 28, No. 4 amidst an unlimited food supply. Successful bivalve cul- ture requires not only rapid growth at high stocking den- sities and absence of pathogens, but also high survivorship and favorable socioeconomic conditions. Most efforts to reduce predatory mortality have been basically physical in nature (mesh cages, gravel burial, etc.), but such meth- ods have, in the past, been directed at preventing known predators from gaining access to cultures (e.g., MENZEL & Sims, 1964; ELDRIDGE et al., 1976; CASTAGNA & KRAEUTER, 1977). Losses that occur in physically pro- tected culture trays are typically assumed to be a result of mechanical damage, handling artifacts, innate morbidity, parasites, disease, or environmental stresses—little consid- eration had been given to previously unknown predators. Reasons for this are logical and obvious, for if it is not known what all the predators are, it is not possible to design protection from all of them. This is evidenced by the sporadic success of cages in protecting desirable or- ganisms (MENZEL et al., 1976; VIRNSTEIN, 1978). Iden- tification of previously unknown predators enhances the probability that preventive measures can be taken to avoid them, either by emplacing grow-out trays in areas having low predator abundance, by removing specific predators, or by designing more effective exclosure devices. Preventive measures employed by the Trident Seafarms Co. (mesh and aggregate protection) coupled with em- placement of relatively large individuals for grow-out ap- pears to be effective in reducing predatory losses. Whether optimal growth can occur under these grow-out conditions is still a question (HADLEY & MANz1I, 1984). ACKNOWLEDGMENTS I would like to express my gratitude to Dr. J. Manzi, Marine Resources Research Institute, Charleston, SC, and Mr. H. Clawson, Trident Seafarms Co., Charleston, SC, for their cooperation in this study. The able field assis- tance of M. Maddox and M. Luckenbach was essential, and laboratory work could not have been completed with- out the expertise of C. McIlvaine and J. Dorsch. This work was funded by the Office of Sea Grant, N.O.A.A., U.S. Department of Commerce, under Grant No. NA- 80-AA-D-105, and the South Carolina Sea Grant Con- sortium. A part of the work was also supported by Grant No. OCE-7919473 from the National Science Founda- tion, Biological Oceanography Section. Dr. P. A. Jumars made especially helpful comments on the manuscript for which I am grateful. LITERATURE CITED AXELSEN, N. H. & E. Bock. 1972. Identification and quan- titation of antigens and antibodies by means of quantitative immunoelectrophoresis. A survey of methods. J. Immun. Meth. 1:109-121. AXELSEN, N. H., J. KRoLL & B. WEEKE. 1973. A manual of quantitative immunoelectrophoresis. Methods and applica- tions. Scandinavian Journal of Immunology. Vol. 2 (suppl. no. 1): 169 pp. Blackwell Sci. Publ.: Oxford. Reale Feller, 1986 BorenHAM, P. F. L. & C. E. OnIAGU. 1978. The use of serology in evaluating invertebrate prey-predator relationships: a re- view. Bull. Ent. Res. 68:171-194. Brown, J. W., J. J. Manzi, H. Q. M. CLawson & F. S. STEVENS. 1983. Moving out the learning curve: an anal- ysis of hard clam, Mercenaria mercenaria, nursery operations in South Carolina. Mar. Fish. Rev. 45:10-15. CALVER, M. C. 1984. A review of ecological applications of immunological techniques for diet analysis. Aust. J. Ecol. 9:19-25. CastaGna, M. & J. N. KRAEUTER. 1977. Mercenaria culture using stone aggregate for predator protection. Proc. Natl. Shellfish Assoc. 67:1-6. Dixon, W. J. (ed.). 1981. Statistical software. BMDP. Univ. of Calif. Press: Berkeley. 725 pp. ELDRIDGE, P. J.. W. WaALtTz, R. C. Gracy & H. H. Hunt. 1976. Growth and mortality rates of hatchery seed clams, Mercenaria mercenaria, in protected trays in waters of South Carolina. Proc. Natl. Shellfish Assoc. 66:13-20. FELLER, R. J. 1984. Dietary immunoassay of J/yanassa obso- leta, the eastern mud snail. Biol. Bull. 166:96-102. FELLER, R. J. & E. D. GALLAGHER. 1982. Antigenic similar- ities among estuarine soft-bottom benthic taxa. Oecologia 52:305-310. FELLER, R. J., G. L. TaGHon, E. D. GALLAGHER, G. E. KENNY & P. A. JumMars. 1979. Immunological methods for food web analysis in a soft-bottom benthic community. Mar. Biol. 54:61-74. FELLER, R. J., G. ZAGURSKY & E. A. Day. 1985. Deep-sea food web analysis using cross-reacting antisera. Deep-Sea Res. 32:485-497. Hapbtey, N. H. & J. J. MANzI. 1984. Growth of seed clams, Mercenaria mercenaria, at various densities in a commercial scale nursery system. Aquaculture 36:369-378. HIBBERT, C. J. 1977. Growth and survivorship in a tidal-flat population of the bivalve Mercenaria mercenaria from South- ampton water. Mar. Biol. 44:77-84. Kenny, G. E. & H. M. Foy. 1975. Detection and quantitation of circulating polysaccharide in pneumococcal pneumonia by immunoelectroosmophoresis (counterelectrophoresis) and rocket electrophoresis. Microbiology (Amer. Soc. for Micro- biol.) 1:97-102. KRAEUTER, J. N. & M. Castacna. 1980. Effects of large predators on the field culture of the hard clam, Mercenaria mercenaria. Fish. Bull. 78:538-541. Krantz, G. E. & J. V. CHAMBERLIN. 1978. Blue crab pre- dation on cultchless oyster spat. Proc. Natl. Shellfish Assoc. 68:38-41. MacKenzig, C. L., JR. 1970. Causes of oyster spat mortality, Page 347 conditions of oyster setting beds, and recommendations for oyster bed management. Proc. Natl. Shellfish Assoc. 60:59- 67. MacKenzig, C. L., JR. 1977. Predation on hard clam (Mer- cenaria mercenaria) populations. Trans. Amer. Fish. Soc. 106:530-537. Manzl, J. J.. V. G. BURRELL, JR. & W. Z. Carson. 1980. A mariculture demonstration project for an alternative hard clam fishery in South Carolina: preliminary results. Proc. World Maricul. Soc. 11:79-89. MarsHAaLL, M. J. 1977. Serologically detected patterns of gas- tropod predation on an intertidal oyster bar. M.S. thesis, Univ. of Florida, Gainesville. 50 pp. MENZEL, R. W. & H. W. Sims. 1964. Experimental farming of hard clams, Mercenaria mercenaria, in Florida. Proc. Natl. Shellfish Assoc. 53:103-109. MENZEL, R. W., E. W. Cake, M. L. HaINnes, R. E. MARTIN & L. A. OLSEN. 1976. Clam mariculture in northwest Florida: field study on predation. Proc. Natl. Shellfish As- soc. 65:59-62. Menzies, R. A. & J. M. KERRIGAN. 1978. Implications of spiny lobster recruitment patterns of the Caribbean—a bio- chemical genetic approach. Proc. 31st Ann. Gulf and Ca- ribbean Fish. Inst., pp. 164-178. MILeEIkovsky, S. A. 1974. On predation of pelagic larvae and early juveniles of marine bottom invertebrates by adult ben- thic invertebrates and their passing alive through their pred- ators. Mar. Biol. 26:303-311. OUCHTERLONY, O. 1968. Diffusion-in-gel methods for im- munological analysis. Ann Arbor Science Publ., Inc.: Ann Arbor. 215 pp. PrscH, G. 1974. Protein polymorphisms in the hard clams (Mercenaria mercenaria and Mercenaria campechiensis ). Biol. Bull. 146:393-403. VIRNSTEIN, R. W. 1978. Predator caging experiments in soft sediments: caution advised. Pp. 261-273. In: M. L. Wiley (ed.): Estuarine interactions. Academic Press: New York. WALKER, R. L., M. A. FLEETWooD & K. R. TENORE. 1980. The distribution of the hard clam Mercenaria mercenaria (Linne) and clam predators in Wassaw Sound, Georgia. Georgia Marine Science Center, University System of Geor- gia, Skidaway Island, Georgia, Tech. Rep. Ser. No. 80-8, 59 pp. (unpublished manuscript). WHETSTONE, J. M. & A. G. EVERSOLE. 1978. Predation on hard clams, Mercenaria mercenaria, by mud crabs, Panopeus herbstu. Proc. Natl. Shellfish Assoc. 68:42-48. WHETSTONE, J. M. & A. G. EVERSOLE. 1981. Effects of size and temperature on mud crab, Panopeus herbsti, predation on hard clams, Mercenaria mercenaria. Estuaries 4:153-156. The Veliger 28(4):348-355 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 How a Clam Builds Windows: Shell Microstructure in Corculum (Bivalvia: Cardiidae) by MARY E. WATSON anp PHILIP W. SIGNOR Department of Geology, University of California, Davis, California 95616, U.S.A. Abstract. Corculum is unusual among bivalves because it, like the closely related genus Fragum and their distant relative 77:dacna, possesses endosymbiotic dinoflagellates. But Corculum does not expose its algae-laden tissues directly to the sunlight. Instead, the shell of Corculum incorporates several unique features that have been interpreted as adaptations to permit passage of incident light through the shell to algae-bearing mantle and gill tissue encased within. These shell modifications in Corculum shell include: posterio-anterior compression (resulting in a prominently keeled shell with a flattened upper surface), posterior thinning of the shell, and unique transparent “windows” radially arrayed on the shell posterior. The windows are present only on the posterior (upper) surface, are triangular in outline, and are constructed by expansion and elaboration of the outermost, fibrous prismatic layer of the shell micro- structure. A concomitant reduction in pigmentation enhances transparency. Direct measurements of light transmission through the Corculum shell show the windows transmit an order of magnitude more light than either the other portions of the shell posterior or the shell’s anterior. INTRODUCTION Two BIVALVE lineages, the tridacnids and the Fraginae (including Corculum, the heart cockle, and the closely re- lated genus Fragum), have independently evolved sym- biotic relationships with dinoflagellates, or zooxanthellae (KAwaGutTl, 1950, 1983). Although both bivalve groups are assigned to the Cardiacea, or cockles, they have evolved very different sets of adaptations to accommodate their symbionts. Unlike its distant cousin 77idacna, Corculum has received little scientific attention. Consequently, its biology is not well known. Unlike tridacnids, Corculum does not gape to expose its zooxanthellae to light. In fact, it cannot employ gaping to expose the algae because the shell’s umbos restrict the bivalve’s gape to only a few degrees. RAuP (1966) noted the severe valve overlap in Corculum, and observed that the two umbos are slightly offset to permit the clam to open. The zooxanthellae are situated within the mantle and gills and are densely packed in the anterior mantle epithelia (KAWAGUTI, 1950, 1966). Under the posterior (upper) valve regions, the algae are stacked in two layers: a thin covering in the posterior mantle and a dense pop- ulation in the directly subjacent gills (KAWAGUTI, 1966). Except for the narrow peripheral mantle fringe, the gills and mantle tissue containing the zooxanthellae are always covered by the shell. Instead of direct exposure, Corculum appears to employ a “windows” strategy to cul- ture its symbionts, utilizing a unique set of shell modifi- cations to enhance light penetration through the posterior shell surface (KAWAGUTI, 1950; SEILACHER, 1972, 1973). The animals are noticeably compressed in the anterio- posterior direction (Figures 1, 2) and the posterior region of the shell is thin. Located in the posterior region are numerous transparent areas arranged in radial rows. SEI- LACHER (1972) and VOGEL (1975) suggested that the roughly triangular to dendritic clear areas function not only as transparent windows, but also as optical lenses providing maximum light diffusion. The purpose of this investigation was to examine the structure and properties of these putative windows and to learn how they might have evolved through modification of pre-existing shell microstructure. Preliminary investi- gation suggested two possible origins for the windows. Modifications of the aragonite microstructure that consti- tutes the shells might account for the shape and nature of the windows. Secondly, concomitant changes in shell pig- mentation may enhance the transparency of the windows. We also examined the light transmission properties of the shells in order to determine whether the windows afforded any significant increase in light transmitted to the shell interior. M. E. Watson & P. W. Signor, 1986 MATERIALS anpD METHODS Specimens of Corculum cardissa (Linnaeus, 1758) and Fragum fragum (Linnaeus, 1758) were provided by the California Academy of Sciences. Additional specimens of Corculum were collected live at the Motupore Island Re- search Center (9°32'S; 147°16’E) in Papua New Guinea. The shells were in good condition, with clean, unmarred surfaces. For preliminary observations, petrographic slides of radial sections of C. cardissa and F. fragum were pre- pared and studied under a polarizing microscope. For scanning electron microscopy (SEM), shells were sec- tioned and fractured in radial, oblique, and tranverse ori- entations. Sectioned specimens were then polished and etched in dilute hydrochloric acid. Specimens were mount- ed, coated with gold/palladium, and examined with a Hi- tachi S-450 Scanning Electron Microscope housed in the Geology Department at the University of California, Da- vis. Further SEM work was done in the laboratory of J. G. Carter at the University of North Carolina, Chapel Hill. Two additional techniques were utilized to examine shell microstructure. Pieces of shell were embedded in lucite plugs and sectioned in radial, oblique, and trans- verse orientations. Acetate peels were then prepared from the polished and etched sections and examined under a light microscope. Embedded sections were also mounted, coated and examined under SEM. These techniques are detailed in CARTER (1980a). Measurements of light transmission characteristics of different parts of the Corculum shell were obtained using a microspectrophotometer. Pieces of shell were mounted on glass slides, with cover slips, in an embedding wax. The percent light transmission was then measured for the anterior region of the shell, the posterior non-window por- tion, and the windows. Measurements were taken every 10 nm at wavelengths from 420 to 700 nm. Measurements are given as percent of incident light at the specified wave- length transmitted through the shell, relative to that trans- mitted through the glass slide, cover slip, and mounting medium. RESULTS Anterior Shell Microstructure of Corculum The region of the Corculum shell anterior to the pro- nounced medial keel is composed of three types of micro- structure arranged in layers. (Terminology for micro- structure employed here follows CARTER, 1980a.) The relatively thin outer layer of the shell consists of fibrous prismatic crystals (Figure 3). Previous workers (TAYLOR et al., 1973) have found cardiacean shells to be entirely aragonitic, and the morphology of the fibrous prismatic crystals and of crystals forming the other shell layers is consistent with that observation (for comparisons, see CarTER, 1980a). The prisms are oriented in a plumose Page 349 pattern, running parallel to the shell surface and then radiating away from the central axis toward the interior and exterior surfaces of the shell. The central plane of this layer tends to form a natural breakage plane parallel to the surface of the shell. Throughout the layer, the prisms have identical ratios of length to width. Directly underlying the fibrous prismatic layer is a thick layer of irregular complex crossed-lamellar shell material. This microstructure is an irregular three-dimensional ar- rangement of aragonite crystals with three or more crystal orientations. In radial section, the pattern produced by the different orientations resembles dendritic striping (Figure 4). The innermost shell layer consists of an interdigitating cone-complex crossed-lamellar structure (Figure 5). Ara- gonite crystals are arranged in a pattern of stacked cones, with the crystals radiating downward from the apices of the cones. This pattern presents nearly identical appear- ances when viewed in any direction normal to the shell surface, thus distinguishing it from the irregular complex crossed-lamellar structure. In some places, especially where the shell is flexed by formation of a plication, the cone- complex crossed-lamellar structure grades upward into the irregular complex crossed-lamellar structure. Posterior Shell Microstructure of Corculum The shell is thinner posterior to the medial keel than in the anterior portion. Also, the microstructure is strik- ingly different. The outer fibrous prismatic layer is in places greatly thickened, extending through the underly- ing layers to the interior surface of the shell. Where the fibrous prismatic layer penetrates to the interior shell sur- face, the crystals are greatly elongated and oriented per- pendicular to the shell surface. These elongated prisms constitute the features previously described as windows (Figure 6). Small topographic highs form on the interior shell surface where the fibrous prismatic layer reaches to the shell surface. Figure 8 presents a block diagram show- ing the geometric arrangement of fibrous prismatic crys- tals within the shell. The non-window matrix of the shell posterior is com- posed of irregular complex crossed-lamellar microstruc- ture and a very reduced layer of cone-complex crossed- lamellar structure. Our observations of petrographic slides of Corculum under polarized light indicate windows con- tain unpigmented growth lines. In contrast, non-window areas do contain pigment, suggesting pigmentation is sup- pressed during window formation. The combination of pigmentation and type of microstructure undoubtedly con- tributes to the visually obvious variation in translucency of the Corculum shell posterior. Interestingly, there are also relatively translucent areas in the shell anterior, though these areas are not as pronounced as the windows. There is no variation in microstructure associated with these patches, suggesting the effect is due entirely to pigmen- tation. Page 350 The Veliger, Vol. 28, No. 4 M. E. Watson & P. W. Signor, 1986 Page 351 Shell Microstructure of Fragum fragum There is no significant differentiation of the microstruc- ture in the anterior and posterior parts of the Fragum fragum shell. The shell is also less compressed, thicker, and visibly more opaque. The Fragum shell contains a more diverse suite of microstructure; four types of micro- structure are found in contrast to the three occurring in Corculum (Figure 7). The outermost shell layer is composed of relatively thin, fibrous prismatic aragonite. Underlying this layer is a thin, irregular complex crossed-lamellar layer. Subjacent to this is a relatively thick, extensive layer of simple crossed-lamellar microstructure. This micro- structure, not found in Corculum, consists of laths of ara- gonite arranged in two orientations within parallel pri- mary sheets, with adjacent sheets having alternate lath orientations. The innermost layer is a relatively thick lay- er of cone-complex crossed-lamellar structure that occurs dorsal to the pallial line and is separated from the over- lying layer by a thin pallial myostracum. Light Transmission Characteristics of the Corculum Shell Windows transmit as much as 40% of incident light, and a full order of magnitude more light than either pos- terior shell matrix or anterior portions of the shell. Light transmission is relatively poor in the shorter wavelengths, but increases rapidly through the greens, yellows, and reds (Figure 9). At the long end of the spectrum, transmission is again reduced. The windows provide significant in- creases in light transmission at wavelengths important for photosynthesis. Compared with the anterior part of the shell, the rel- atively thin posterior shell matrix transmits more light, but the maximum transmission is only a few percent of incident light (Figure 9). DISCUSSION Our results indicate that a combination of pigmentation and shell microstructure greatly increases the intensity of light transmitted to the shell interior of Corculum. Pre- sumably, this is an adaptation for the benefit of the sym- biotic dinoflagellates inhabiting the clam’s tissues. The evolutionary pathways followed in development of the unique windows of Corculum can be inferred from com- one BAKS IS AS 74) me fat ey N= S7=7=/ aris 2757] ati ( Wi Figure 8 Block diagrams showing the geometric arrangement of the three microstructural layers that constitute the posterior portion of the shell of Corculum. Expansion of the fibrous prismatic layer and penetration of the prisms to the inner shell wall form light- transmitting structures, or windows. Drawn approximately to scale. parison of character states present in Corculum and Fra- gum. The role of pigmentation in enhancing shell transpar- ency is not entirely clear from our results, nor is the gen- eral phenomenon of shell pigmentation well understood. Research to date (for review, see CRENSHAW, 1980) sug- Explanation of Figures 1 to 7 Figure 1. Ventral view of Corculum cardissa. Scale bar = 2 cm. Figure 2. Posterior view of C. cardissa. Scale bar = 2 cm. Figure 3. Outermost, fibrous prismatic microstructural layer of anterior shell of C. cardissa. Scanning electron micrograph of fractured radial section. Scale bar = 50 um. Figure 4. Irregular complex crossed-lamellar microstructure of anterior shell of C. cardissa. Scanning electron micrograph of polished and etched oblique section. Scale bar = 50 um. Figure 5. Innermost cone-complex crossed-lamellar microstruc- ture layer of posterior portion of C. cardissa shell. Scanning elec- tron micrograph of fractured radial section. Scale bar = 5 um. Figure 6. Expanded, window-forming fibrous prismatic micro- structure of posterior shell of C. cardissa. Scale bar = 5 um. Figure 7. Simple complex crossed-lamellar microstructure of Fragum fragum. Scanning electron micrograph of fractured transverse section. Scale bar = 50 um. Page 352 The Veliger, Vol. 28, No. 4 40+ BOlr : . | $ al t 7 & ty! a 5 Hg a ee ee ee ee ee ee ee ee ee ee eS ee ee ee ee Be io} ] i _—_§— 0.5 traders 4 aaAaAaaz~ A t Hl 1 a Pn nS) 400 450 500 550 600 650 700 WAVELENGTH (NM) Figure 9 Transmission characteristics of the shell of Corculum cardissa. Circles indicate mean value for four windows, lines indicate range of readings. Squares indicate values for four readings on the shell matrix of the shell posterior, and triangles indicate percent transmission for four areas of the anterior shell of Corculum. Lines above and below squares and triangles represent ranges of values obtained. gests changes in organic content and concentration are portions of Corculum and Fragum contain no pigmented linked to salinity fluctuations, mantle irritation, and an- growth lines. Thus, simple expansion of the prismatic aerobiosis during shell deposition. It is perhaps suggestive layer may be sufficient to cause an increase in light trans- that the prismatic layer of both the anterior and posterior mission. M. E. Watson & P. W. Signor, 1986 CORCULUM Posterior Anterior QQ WIN \' « eK y ICCL Page 353 FRAGUM Wy FP / XI ICCL Figure 10 Comparison of shell architecture of Fragum and Corculum. Layers indicated as follows: FP, fibrous prismatic; ICCL, irregular complex crossed-lamellar; CCCL, cone-complex crossed-lamellar; and SCL, simple crossed- lamellar. Not drawn to scale. On the other hand, several modifications of the Corcu- lum shell occurred, leading to the present combination of gross morphology and the presence of windows. These include anterio-posterior flattening of the shell (with a concomitant increase in the ratio of surface area to shell volume), thinning of the posterior portion of the shell through reduction in the number of shell layers and thin- ning of the remaining layers, and differential expansion of the fibrous prismatic layer resulting in triangular, clear windows in the posterior region of the shell. The peculiar, flattened shell of Corculum and the windows are unique among modern bivalves, but reduction in the numbers of microstructural layers is a common occurrence in the evo- lution of bivalve clades (TAYLOR et al., 1973; TAYLOR, 1973; CARTER, 1980b). Comparison of the shell microstructure of Fragum and Corculum reveals several evolutionary modifications (Fig- ure 10). Both Fragum and Corculum have an outermost fibrous prismatic layer. However, this layer is relatively thin in Fragum and well developed in Corculum. Under- lying the fibrous prismatic layer in both Corculum and Fragum is an irregular complex crossed-lamellar layer. In Fragum the irregular complex crossed-lamellar layer is thinner and more variable in extent than in Corculum, where it represents a major microstructural component of the shell. Whereas Corculum lacks a simple crossed-la- mellar structure, in Fragum it is a major component of the shell. The inner shell layer of cone-complex crossed- lamellar structure occurs in both Fragum and Corculum, but is much reduced in the posterior portion of the latter. In Fragum, a noticeable prismatic myostracum separates the inner cone-complex crossed-lamellar layer from the simple crossed-lamellar layer, but in Corculum the myos- tracum separating the inner cone layer from overlying irregular complex crossed-lamellar structure is prominent only in the anterior portion of the shell. Thus, the major evolutionary modifications apparent in the transition from the typical Fragum shell structure to that of Corculum are an elimination of the simple crossed-lamellar layer, an elaboration of the fibrous prismatic layer, and a differ- entiation of patterns of deposition between the anterior and posterior portions of the Corculum shell. The absence of a pallial myostracum and reduction in the extent of the cone-complex crossed-lamellar layer in the posterior region of the Corculum shell suggests a sig- nificant change in the pattern of shell deposition by the Page 354 mantle. Because the cone-complex crossed-lamellar layer lies below the myostracum, the region of the mantle de- positing this layer must be the mantle surface inside the pallial line. The lack of this layer and any appreciable myostracal prisms in the posterior region of the Corculum shell indicates that deposition by this part of the mantle must be suspended. The mantle edges of Corculum, which deposit the fibrous prismatic layer, must at times greatly increase the rate of deposition to form windows, while deposition continues normally at other parts of the mantle margin. This demonstrates an extreme flexibility in pat- terns of deposition and greatly increases the range of structures that potentially could be formed by the mantle. CARTER (1980b) has found that expansion or contrac- tion of shell microstructural layers during ontogeny occurs in at least three bivalve genera. Mercenaria, Cerastoderma, and Spisula show either occasional inter-tonguing of the outer and middle shell layers, or periodic expansion and contraction of the area of deposition of the outer shell layer in response to environmental stresses such as seasonal cli- matic changes and thermal shock or in response to spawn- ing. Observations of shell structures in Corculum cardissa and Fragum fragum reveal a greater diversity, both in variety and arrangement, of microstructure than described by earlier workers. TAYLOR et al. (1973) described car- diacean microstructure in general, and Fragum unedo (Linnaeus, 1758) in particular, as consisting of two con- centric layers: an outer aragonite crossed-lamellar struc- ture and an inner complex crossed-lamellar layer usually separated by a thin prismatic myostracum. The surprising diversity of shell microstructure now known to be present in Corculum and Fragum suggests the possibility of un- discovered diversity and variation in the microstructure of other molluscan taxa. Photosymbiosis in Fragum KAwWAGUTI (1983) has recently reported the presence of symbiotic dinoflagellates in Fragum fragum and F. unedo. In the latter species, the photosymbionts are primarily concentrated in the enlarged mantle tissues around the siphons and in the gills. Kawaguti reports that this species lives buried shallowly in the sediment, with enlarged man- tle tissues around the siphons spread over the sediment. In contrast, Kawaguti reports that F. fragum lives with the posterior side of the shell above the sediment. This species lacks the enlarged mantle tissues and does not gape widely. (PWS confirms these observations on F. fragum from field and laboratory work in Guam.) In Fragum fragum the zooxanthellae are distributed throughout the animal, and are concentrated in the gills and mantle. In this species, light transmitted through the shell apparently sustains the enclosed zooxanthellae, as in Corculum. Thus, these three species seem to constitute an evolutionary se- ries from Fragum unedo, which uses a Tridacna-like strat- egy of directly exposing zooxanthellae-laden tissue to light, The Veliger, Vol. 28, No. 4 to Corculum, where the shell encloses the animal at all times (KAWAGUTI, 1983). Life Habits of Corculum One remaining question is how Corculum keeps the posterior surface of its shell clear of fouling and boring organisms. Surfaces in shallow marine environments do not remain unfouled in the Indo-Pacific for long (more than a few days) unless there is some mechanism to deter organisms from settling or to remove organisms that have settled. Corculum occurs in shallow intertidal to subtidal areas throughout the Indo-Pacific region (BARTSCH, 1950). Ka- WAGUTI (1950) reported that Corculum occurs around reefs, often resting half obscured by filamentous algae and rub- ble. (This report would appear to contradict the sugges- tion that the posterior surface of the Corculum shell serves to conduct light to algae living within the clam’s tissues.) Other authors have reported Corculum occurring in shal- low sandy or muddy areas (Franco in BARTSCH, 1950; CERNOHORSKY, 1972). One of us (PWS) recently was able to observe living Corculum cardissa and C. monstrosum (Gmelin, 1791) at Motupore Island in Papua New Guinea. At Motupore Island, the two species occur inter- tidally. Corculum monstrosum was found living free on a sand and grass flat; C. cardissa was found living in a shel- tered area on sandy mud among intertidal rocks on the lee side of Motupore Island. In each case, the shell pos- terior was exposed to the sunlight and was free of algae or fouling organisms. Corculum maintained in aquaria will migrate within the tank from shady to sunny areas. If overturned, they will right themselves by planting the foot in the substrate and rotating the entire shell onto its anterior surface. The foot also occasionally sweeps the posterior surface of the shell, and may be the mechanism for keeping that area free of fouling organisms. Comparison of Corculum and Notoacmea There is an interesting morphological parallel between the Corculum windows and the limpet Notoacmea persona. LINDBERG e¢ al. (1975) noted the presence of light-trans- mitting spots in the anterior shell of N. persona, and dem- onstrated that these spots play a role in the limpet’s neg- atively phototropic reaction to light. The Corculum windows might also facilitate the bivalve’s positive re- sponse to light noted above, but that was not tested. Nevertheless, the structures documented in the limpet windows by LINDBERG e¢ al. (1975) and in Corculum (this paper) are quite distinct. Recognition of Photosymbiosis in Fossil Bivalvia Depending on the degree of preservation, shell modi- fications such as those seen in Corculum could be recog- nized in the fossil record. Unusual shell thinness, indica- M. E. Watson & P. W. Signor, 1986 Page 355 tive of translucency, would be especially obvious. Recognition of other modifications such as prismatic win- dows or layers would depend upon the degree of preser- vation, especially for aragonitic structures. Even though only two Recent bivalve lineages, tridacnids and the Fra- ginae, are known to maintain symbiotic algal associations, given the broad spectrum of Recent organisms that harbor algae we should expect to find other examples of paleo- photosymbiosis in the fossil record. For example, YANCEY (1982) has recently interpreted an unusual Permian my- alinid bivalve group, the Alatoconchidae, as the earliest photosymbiont-bearing bivalves. Alatoconchids resemble tridacnids in size and shell thickness, yet resemble Cor- culum in terms of anterio-posterior compression, life po- sition, and prismatic microstructure. COWEN (1983) has reviewed the evidence for photosymbiosis in bivalves and other fossil clades. ACKNOWLEDGMENTS We thank Kraig Derstler, Joseph G. Carter, Richard Cowen, and two reviewers for discussions and comments on this paper. Daphne Dunn and Peter Rodda, of the Department of Invertebrate Zoology at the California Academy of Sciences, provided us with some of the spec- imens used in this study. Ken Severin assisted with field work in Papua New Guinea and made some of the field observations cited here. T. Rost provided access to the microspectrophotometer and instructed us on its use. During the course of this investigation, we learned that an independent study of Corculum microstructure was being conducted by Mr. Clement Counts, III, at the Uni- versity of Delaware (Counts, 1981). We thank Mr. Counts for outlining his project for us, sharing his ideas on car- diacean microstructure and providing suggestions on tech- niques employed here. LITERATURE CITED BarTscH, P. 1950. The little hearts (Corculum) of the Pacific and Indian Oceans. Pacific Sci. 1:221-226. CarTER, J. G. 1980a. Guide to bivalve shell microstructures. Pp. 645-673. In: D. C. Rhoads & R. A. Lutz (eds.), Skeletal growth of aquatic organisms. Plenum Press: New York. CarTER, J. G. 1980b. Environmental and biological controls of bivalve shell mineralogy and microstructure. Pp. 69-113. In: D. C. Rhoads & R. A. Lutz (eds.), Skeletal growth of aquatic organisms. Plenum Press: New York. CERNOHORSKY, W. O. 1972. Marine shells of the Pacific. Vol- ume II. Pacific Publications: Sydney. 411 pp. + 68 pl. Counts, C. L., III. 1981. Shell ultrastructure of Corculum spp. (Roeding) (Bivalvia: Cardiacea). Bull. Amer. Malacol. Union Conv. Proc., Ft. Lauderdale, Florida. p. 35. Cowen, R. 1983. Algal symbiosis and its recognition in the fossil record. Pp. 431-478. In: M. J. S. Tevesz & P. I. McCall (eds.), Biotic interactions in the fossil record. Ple- num Press: New York. CRENSHAW, M. A. 1980. Mechanisms of shell formation and dissolution. Pp. 115-132. In: D. C. Rhoads & R. I. Lutz (eds.), Skeletal growth of aquatic organisms. Plenum Press: New York. KawacutTl, S. 1950. Observations on the heart shell, Corculum cardissa (L.) and its associated zooxanthellae. Pacific Sci. 4: 43-49. KawacGutTl, S. 1966. Electron microscopy on zooxanthellae in the mantle and gill of the heart shell. Biol. J. Okayama Univ. 141:81-92. KawacutTl, S. 1983. The third record of association between bivalve molluscs and zooxanthellae. Proc. Japan Acad., Ser. B, 59:17-20. LINDBERG, D. R., M. G. KELLOGG & W. E. HUGHEs. 1975. Evidence of light reception through the shell of Notoacmea persona (Rathke, 1833). Veliger 17:383-3806. Raup, D. M. 1966. Geometric analysis of shell coiling: prob- lems. J. Paleontol. 40:1178-1190. SEILACHER, A. 1972. Divaricate patterns in pelecypod shells. Lethaia 5:325-343. SEILACHER, A. 1973. Fabricational noise in adaptive mor- phology. System. Zool. 22:451-465. TayLor, J. D. 1973. The structural evolution of the bivalve shell. Palaeontology 16:519-534. TayLor, J. D., W. J. KENNEDY & A. HALL. 1973. The shell structure and mineralogy of the Bivalvia. II. Lucinacea- Clavagellacea, Conclusions. Bull. Brit. Mus. (Natur. Hist.) 22:253-294. VOGEL, K. 1975. Endosymbiotic algae in rudists? Palaeogeogr. Palaeoclimatol. Palaeoecol. 44:63-69. YANCEY, T. E. 1982. The alatoconchid bivalves: Permian an- alogs of modern tridacnid clams. Third North American Paleontological Convention, Proc. 2:589-592. The Veliger 28(4):356-368 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 Rates and Processes of Compensatory Buoyancy Change in Nautilus macromphalus by PETER D. WARD Department of Geological Sciences, University of Washington, Seattle, Washington 98195, U.S.A. Abstract When faced with sudden buoyancy gain or loss, specimens of Nautilus macromphalus undergo compensatory buoyancy change. Rates of compensatory buoyancy change (as measured by weight gain or loss in seawater) depend upon animal size and the amount of initial buoyancy change; the higher the initial buoyancy gain or loss, the higher the compensatory buoyancy change rates. For the experi- ments described here, the mean rate of compensatory buoyancy change during the first 10-h period following the initiation of the experiment was 0.15 g/h of weight gain for those mature N. macromphalus made suddenly more buoyant, and 0.10 g/h of weight loss for the nautiluses made suddenly less buoyant. During subsequent 10-h periods, the rates of weight gain for the initially buoyant animals dropped to less than 0.05 g/h, while rates for the initially heavy animals stayed approximately the same. The ultimate amount of weight change for highly buoyant nautiluses was limited to about 5 g of in-seawater weight increase, whereas the weight loss in the animals made artificially heavy was unlimited, as long as there was cameral liquid in the chambers to be removed. Positive buoyancy of as little as —5 g was sufficient to trap a mature N. macromphalus at the surface, so that no amount of swimming would allow resubmergence. INTRODUCTION THE PHRAGMOCONE of an ectocochliate cephalopod serves to reduce the overall density of shell and animal to ap- proximately that of seawater (DENTON & GILPIN-BROWN, 1966). It has been proposed that an additional function of the phragmocone is to produce buoyancy change on de- mand, either for vertical migration or for compensatory buoyancy change (HEPTONSTALL, 1970; MUTVEI & REYMENT, 1973). The latter, compensatory buoyancy change, can be defined as density change or buoyancy change brought about by the animal in response to some sudden addition or reduction in the animal’s specific grav- ity. HEPTONSTALL (1970) used, as an example, the case of the ammonoid Buchiceras bilobatum, which, during life, became covered with oysters (first described by SEILA- CHER, 1960). Heptonstall showed that the overgrowth of oysters on the shell of the living ammonite would have required compensatory action by the ammonite, in this case a reduction of overall shell density, to maintain neu- tral buoyancy. Other, perhaps more common examples requiring compensatory buoyancy change can be readily observed in living Nautilus. Nautilus shells of all species commonly exhibit healed breaks. In some cases, the scars of what must have been very large breaks are visible, indicating that at some time during the life of the nautilus, many grams of shell material were suddenly lost (Figure 1). Another type of buoyancy change common in nauti- luses would come from windfall feeding. Nautiluses ap- pear to be opportunistic feeders. In New Caledonia, spec- imens of Nautilus macromphalus are known to eat lobster molts (WARD & WICKSTEN, 1980). The ingestion of molt material makes the nautilus more dense. In both of these examples, the action produces rather sudden changes in the buoyancy of the animal, in the first place making it lighter or less dense, in the second, heavier or more dense. The purpose of this paper is to describe experiments con- ducted with specimens of Nautilus macromphalus, designed to test the potentiality and characteristics of compensatory buoyancy change in Nautilus under these types of condi- tions. MATERIALS anp METHODS Specimens of Nautilus macromphalus (the only species used in this study) were captured in baited traps at 150 to 400 m in New Caledonia and then immediately trans- P. D. Ward, 1986 Figure 1 Page 357 HAM Shell break in Nautilus that would have resulted in a sudden loss of buoyancy. Causes of such shell breaks are unknown, but probably result from mechanical impact, predatory attacks, or intraspecific competition. ported in cooled seawater to a nearby refrigerated sea- water system, maintained at 17-18°C. To determine weight in seawater (which will serve as a quantitative descriptor of buoyancy), the nautiluses were tightly wrapped in a piece of cheesecloth of known seawater weight, and while still being held underwater, transferred to a submerged rigid plastic box of known seawater weight, suspended by a wire from a modified, electronic top-loading balance (Ohaus Brainweight model 1500). The balance had a sen- sitivity of 0.1 g. Nautilus specimens with densities higher than that of seawater gave positive readings on the bal- ance, while those lighter than seawater gave negative readings. Neutral buoyancy (a nautilus with the density of seawater) zeroed the balance. Even using the wrapping procedure, the respiratory movements of the nautilus within the closed plastic box sometimes caused sufficient movement to produce a range of weight values. Under these conditions, the balance gave a stream of values, with those considered significant (by the balance) marked by a “‘g” after the reading. Each data point used in this study is the mean value of 10 stable ““g” readings recorded on the balance. The mean standard de- viation of 100 randomly selected data points used in this study (1000 balance measurements) was found to be 0.132 g. Prior to each experiment, the nautiluses were fed and radiographed. Except on very long-term experiments (more than 100 h), the nautiluses were not fed during the course of the experiments, because feeding significantly increased weight (both in air and in water). To observe rates of liquid movement into the phragmocone, the last three to five chambers in each specimen were drilled and the cam- eral liquid volume in each chamber measured or removed using methods described by WARD & GREENWALD (1982). The holes drilled in the chambers were resealed with ta- pered, hard-rubber stoppers, earlier determined to be leakproof using this procedure (WARD & GREENWALD, 1982). The buoyancy of the experimental animals was then suddenly either increased (by removing shell material from the apertural region with needle-nosed pliers, and in some cases by removing cameral liquid from the first three chambers as well) or decreased (by adding new liq- uid to chambers, or by adding weights cemented to the side of the shell). To ensure that the drilling procedure was not in some unknown way affecting experimental results, one animal (No. 10) was made more buoyant through shell breakage, but not drilled. This animal showed compensatory weight change similar to those in the drilled animals. The term buoyancy is used throughout this paper. However, the measures used in this study are weights in seawater and density (specific gravity, determined from weight in seawater and weight in air). In this paper, weight in seawater is used as a descriptor of buoyancy, for lack of a better method of attempting to quantify a qualitative term. A nautilus referred to as 5 g positively buoyant, or 20 g negatively buoyant, refers to specimens weighing —5 or +20 g respectively on the zeroed balance. Animals Page 358 showing increasing buoyancy were undergoing reductions of seawater weight and density. RESULTS Control Experiments: No Induced Buoyancy Change The weights in seawater of freshly captured nautiluses have been previously measured by DENTON & GIL- PIN-BROWN (1966), WARD et al. (1977), and WarRD & MarTIN (1978). In every case, the computed densities of the observed specimens have been equal to, or slightly higher than, seawater density (therefore, most display negative buoyancy). In the one study in which weights in seawater were followed through time in aquarium-main- tained animals (WARD & MarTIN, 1978), all animals eventually showed increasing buoyancy. In this latter study, however, the weighings were conducted on animals an- aesthetized in a 2% solution of urethane in seawater, and maintained in water temperatures of 23 to 26°C. The observations listed here are on unanaesthetized animals kept at cooler temperatures. The weights in seawater of four freshly captured Nau- tilus macromphalus used in this study are shown in Figure 2. These animals showed a seemingly random fluctuation near neutral buoyancy (0 g). Error bars showing the amount of experimental error (three standard deviations in each direction of a reading) are shown on this graph. No pattern of day versus nighttime weight patterns could be detected. These four animals, without experimentally produced buoyancy change, serve as control animals against which the following experiments and observations can be compared. Induced Buoyancy Change: Compensatory Buoyancy Change in Reaction to Increased Buoyancy To test for the possibility of compensatory change in nautiluses made artificially more buoyant, specimens were made suddenly more buoyant by either the removal of cameral liquid, the removal of shell material, or both. This latter procedure mimics the effect of shell breakage, which could occur through either mechanical action (such as impact in shallow water, high energy environments) or predation. Following the episode of increased buoyancy, the nautilus specimens were weighed periodically (gen- erally at 15-30-min intervals during the first 2 h, followed by every one to two hours). The results of these experiments are shown in Figure 3. In most cases significant weight increases occurred after the initiation of the experiment. Increase in weight (de- crease in buoyancy) was usually apparent within the first hour, and sometimes in as little as 30 min. The rate of weight increase was highest during the first 10 h following the initiation of the experiments, and then tapered off, so that most curves of buoyancy change can be seen to de- scend steeply during the first 10 to 20 h, and then level off after about 30 to 40 h. This suggests that the rate of The Veliger, Vol. 28, No. 4 “a 16 O ptt ty he +2 +4 WEIGHT IN SEAWATER (grams) | Oo nw ; 4 12 | aie +2 +4 -2 ff O Ps aes +4 O 20 40 60 80 HOURS Figure 2 Weight in seawater of four freshly captured Nautilus macrom- phalus (specimens No. 82-16, 15, 12, and 11). The sizes of these animals can be found in Table 2. None of these animals was manipulated in terms of its buoyancy; these measures can, there- fore, serve as controls against which the following experiments, involving sudden buoyancy change, can be compared. The ver- tical bars on the graphs refer to estimated experimental error (0.3 g). Experimental error comes from sensitivity of the balance (0.1 g) and the weighings themselves. Although no subsequent graphs show the error bars, all points listed on subsequent graphs have similar estimated error ranges. compensatory buoyancy change decreases with time, or has a limit to the amount of change. To test this, mean values for aggregate weight change during 10-h incre- ments following the initiation of the experiments were P. D. Ward, 1986 | camara Tauia Page 359 cameral liquid added Ch 1,2,3 added Ch 1,2 cameral liquid added Ch 1,2,3 ie -10 = — —— —~ Z cameral liquid added Ch 1,2,3 -8 =6 - 30 r a) 28 tl =| cameral liquid it - cameral liqui cameral liquid added Ch 1,2,3 [ Biola hee | added Ch P| a O 20 40 60 -22 O 20 40 60 cameral liquid _ Seeee Ch 1,2,3 [oi a aol (Ccamenal liquid added Ch 2,3 > ~~ cameral liquid added Ch 1,2,3 \ , SSS oe ee ee -2 rrrsiitt Cas ir cameral liquid || j ¢ 4 added Ch 1,2,3 O +2 O 20 40 60 80 O 20 40 60 80 HOURS HOURS Figure 3 Compensatory buoyancy change in nine Nautilus macromphalus made positively buoyant through cameral liquid removal, apertural shell removal, or both. The vertical axis in each graph refers to weight in seawater; the horizontal axes show the number of hours after the initiation of each experiment. The portion of each graph labeled “‘cameral liquid added” shows the amount of buoyancy change that can be attributed to measurable chamber refilling with cameral liquid. Page 360 9) BUOYANCY CHANGE grams/ hour 0 0-10 The Veliger, Vol. 28, No. 4 © Buoyant Nautilus e Heavy Nautilus lO-20 20-30 30-40 40-50 50-60 60-70 Figure 4 Mean seawater weight changes in 10-h increments for Nautilus macromphalus made suddenly positively buoyant (open circles) and negatively buoyant (closed circles). Error bars indicate 95% confidence limits using two-tailed tests for significance; ‘‘n” refers to the number of experiments used to compute means. Mean rates of compensatory buoyancy change are not significantly different for positive and negative animals during the first 10-h intervals, but then change. Animals made positively buoyant appear to have a limited compensatory response, as indicated by the significant drop in buoyancy change during the second and later 10-h increments. computed. These figures are shown in Figure 4 along with the number of observations for each value. The highest rate of weight change occurred during the first 10 h (0.15 g/h) for the 15 specimens used in the experiment. The following 10-h increments had significantly lowered rates of weight change; the second 10-h period showed rates of about 0.05 g/hour. Subsequent 10-h periods showed sim- ilarly lowered rates. These experiments suggest that the greatest amount of compensatory buoyancy change will occur within the first 10 h, and that subsequent change will be far less. A second group of experiments was designed to exam- ine the relationship between rates of compensatory buoy- ancy change in response to increased buoyancy and the amount of buoyancy change initiating the response: does an ever higher initial buoyancy change produce ever faster compensatory changes in response? To test this, the initial amount of induced weight change was plotted against P. D. Ward, 1986 AG ael?, ree slope = .0082 de int. = 043 j GRAMS 10 (20) 50 .40 WEIGHT INCREASE GRAMS/HOUR, FIRST 10 HOURS Figure 5 Relationship between rate of compensatory buoyancy change in animals made suddenly positive, against amount of positive buoyancy initiating the experiment. There does appear to be some positive correlation between the rate of compensatory re- filling response and the amount of buoyancy increase at the start of the experiment. Previously, GREENWALD et al. (1980) and WarD (1982) have shown that a similar relationship exists be- tween the opposite conditions, 7.e., liquid emptying rates and initial negative buoyancy. compensatory response for the first 10 h (Figure 5). The correlation coefficient for the linear regression is 0.738. Apparently the initial amount of buoyancy change does affect the rate of compensatory response. The ultimate amount of compensatory buoyancy change was limited. The maximum weight change observed was an increase of 6.5 g (No. 27), and the mean amount was slightly less than 4 g. From these experiments, it appears that the compensatory response to suddenly increased buoyancy is, at least at surface pressure, extremely lim- ited, and in this species probably never reaches as much as 10 g, regardless of the initiating buoyancy change. Induced Buoyancy Change: Decreased Buoyancy Experiments examining liquid emptying rates in re- sponse to buoyancy change have previously been made by GREENWALD et al. (1980) and WaRD (1982a). Compen- satory buoyancy change in nautilus specimens made heavy by the addition of cameral liquid to partially or completely emptied chambers, or through the addition of weights to the side of the shell, was monitored in a fashion similar to that used for nautiluses made artificially less dense. The heavy nautiluses were weighed in seawater at 30-min to 2-h intervals, and the aggregate amount of weight change Page 361 recorded. The results of these experiments are shown in Figures 6-7. As in the case of artificially increased buoyancy, the animals made suddenly heavier than seawater showed compensatory buoyancy change. The rates of weight change during the first 10 h for the nautiluses made heavy were not significantly different from the rates of decreas- ing buoyancy change for the nautiluses made light, as described above (—0.10 g/h for the first 10 h, as compared to 0.15 g/h for the nautiluses made light). Unlike the experiments with nautiluses made lighter, however, these specimens showed roughly similar rates of weight change during the following 10-h intervals (Figure 4). During the second 10-h period, rates were higher than during the first 10 h (0.13 g/h). Subsequent rates per 10-h incre- ments were variable, but never significantly different. Compensatory buoyancy change following sudden buoy- ancy decrease thus seems different from that following sudden buoyancy increase. The Mode of Buoyancy Change In both cases described above, the question of the mech- anism involved must be considered. It has been demon- strated that buoyancy change can be effected in a nautilus through the movement of liquid out of, as well as into, chambers (WARD & MarTIN, 1978; GREENWALD et al., 1980; WARD & GREENWALD, 1982). The nature of the weight-change curves in the experiments described here, however, suggests that the mechanism is more complex than the simple removal or addition of liquid from res- ervoirs of liquid pooled at the bottoms of the last several chambers. To examine the nature of the buoyancy change system, the amount of cameral liquid in each experimental animal was measured prior to the initiation of each experiment. Each nautilus was radiographed, and those specimens with visible pooled liquid within their chambers were drilled. As in previously reported cases with mature nautilus spec- imens, the presence of pooled water was usually found only in the last one or two chambers; it has been shown that mature nautiluses characteristically contain little or no cameral liquid (COLLINS et al., 1980). In the cases where nautilus specimens were made heavier through the addition of new cameral liquid into the last one or two chambers, the original chamber volume was also noted, so that the new chamber volume at the start of the exper- iment was known. Through repeated experiments of add- ing or removing liquid into chambers, it was found that experimental error in volume measurement using this technique was within 0.2 ml of the originally implaced volume. If 1 ml of cameral liquid or seawater is assumed to weigh 1 g, then the change of cameral liquid volume should be the same as the change of weight. It soon became ap- parent, during the course of the experiments, that the weight change of most of the experimental animals ex- Page 362 The Veliger, Vol. 28, No. 4 +2 +4 feeding +4 +6 lI writs me +8 rata Te 2\ liquid removed Gots +8 Chi+Ch2 +10 cameral liquid removed Ch 1,2,3 +10 O 20 40 60 80 0 20 +4 +6 +6 8 Heare F +8 iquid remove +8 me : Ch 1 22 cameral liquid [at others empty +10 _ removed +10 = | Ch 1,2,3 (0) 20 40 +12 +22 (0) 20 40 +24 WT. +26 +10 +28 no liquid removal Ch 1,2,3 cameral liquid +30 | removed fo) 20 40 60 | Ch 1,2,3 +6 4 40 +8 +10 +8 25 +12 cameral liquid removal +10 ai | =e es 29 cameral liquid +14 +12 removed feeding S45 sn 128 +16 +14 (0) 20 40 60 80 20 40 HOURS HOURS Figure 6 Compensatory buoyancy change in eight Nautilus macromphalus made artificially less buoyant by addition of seawater into chambers or addition of metal weights to the shell. The “liquid removed” part of the graph refers to the amount of compensatory buoyancy change that can be attributed to measurable cameral liquid removed. ceeded the measurable volume change of liquid moving into, or out of, chambers. This is best illustrated in the figures showing experimental results. In these figures, known weight change and volume change are shown on the same graph. The figures have been plotted so as to show the amount of weight change attributed to measur- able liquid removal or addition in the last two or three chambers. In almost every case, initial weight change was not caused by addition or removal of pooled liquid; in- stead, changes in the volumes of these liquids within the last two or three chambers were only seen after a duration of several hours. Only two possibilities exist to explain the observed buoyancy change. Either buoyancy change is being affected by density changes within the soft parts of the animal, or density change is occurring within the phragmocone, but in a manner not observable either with radiographs or in chamber volume determinations. To test for the possibility of density changes occurring in the body chamber, specimens of Nautilus macromphalus were made either lighter or heavier in a manner used in P. D. Ward, 1986 BUOYANCY Page 363 — 100 120 140 160 —— _ cameral liquid removed (ln) 2) 100 120 140 160 180 HOURS Figure 7 Compensatory buoyancy change in three Nautilus macromphalus made less buoyant. the previous experiments. After the confirmation of weight change (through balance measurements) these animals were sacrificed. Samples of blood, coelomic fluid (from the liver region), and a variety of tissue samples were taken (Table 1). In no case could accurate density measurements (+0.002) show tissue or fluid densities suggestive of soft- part compensatory response sufficient to account for the observed but unaccounted for weight change. For exam- ple, in those animals made heavy, the observed volumes and densities of blood, liver, and coelomic fluid do not appear to be agents of compensatory buoyancy change. Another possibility is that buoyancy change was occur- ring within the phragmocone, but not only through the addition or removal of pooled liquid (liquid observable on a radiograph as a distinct volume of liquid at the bottom of a chamber) from the last five or so chambers. This possibility is harder to test for, but is considered to be the cause of the buoyancy change. By breaking open individ- ual chambers, visual observations of the chamber walls indicated that significant volumes of non-pooled water were trapped within the nautilus shell. Most of this appears to be within the pellicle, a hydrophilic membrane that lines the inside of the chambers and covers the outer region of the siphuncle. The chalky layer of the siphuncle could also be a significant reserve of liquid. Also, the use of high-energy radiographic techniques (in contrast to the small, low kv portable machines used in my previous stud- ies) allowed the first observations of the interiors of cham- bers in the early whorls. Previously, chamber liquid could only be radiographically observed in the last-formed 10 or 11 chambers (chambers of the last whorl). By combin- ing high-energy exposures with high-contrast screens, Page 364 technicians at the Radiographic Facility of the Magenta Hospital, New Caledonia, succeeded in penetrating and observing earlier chambers. These chambers showed small but significant liquid volumes, usually trapped at the sep- tal wall-shell wall intersections (the sutures), rather than occurring as pools of liquid at the bottoms of the cham- bers. These early chambers thus show small volumes of water, long after emptying, in seemingly gravity-defying orientations (Figure 8). In the event of rapid buoyancy change, it appears that the liquid is moved to or from the pellicle and sutural regions (perhaps in concert with the reservoir contained within the chalky tube of the siphuncle). These volumes are so small (less than 0.2 ml) that movement would not be detected by changes in the volume of “pooled” liquid within the chamber, if any were present. By acting in concert over many chambers, however, this reservoir could produce a significant volume of cameral liquid. For in- stance, the movement of just 0.1 ml into, or out of, a single chamber would not be measurable by volume determi- nation methods. Produced over 32 chambers, however, over 3 g of buoyancy change would be effected. If this system acts in this way, it may answer the question posed by WARD et al. (1980) as to why the siphuncle remains unblocked in life long after earlier chambers have been emptied, and seemingly would be of no further use to the animal’s buoyancy system. The chambers, with their thin linings of hydrophilic membrane, may remain active and useful throughout life as a means of allowing “rapid” buoyancy change through admission or removal of small volumes of liquid. Size and Buoyancy Change If the pellicle system is indeed the source of the rapid, “unaccounted” buoyancy change observed in the experi- mental animals, it should become increasingly important in increasingly larger nautiluses. The amount of buoyancy change allowed by the pellicle system would be dependent on the surface area of the chambers, and thus would be dependent on the number of chambers present. Larger nautiluses should be expected to show larger amounts of buoyancy change. To examine the rates of compensatory buoyancy change The Veliger, Vol. 28, No. 4 Table 1 Soft-part densities for Nautilus macromphalus undergoing compensatory buoyancy change. Digestive | Coelomic Soft-part gland fluid Initial density density density Number buoyancy (+0.002) (+0.002) (+£0.01) 81 negative 1.064 1.073 1.06 21 negative 1.064 1.090 1.03 Wt negative 1.063 1.070 1.06 25 negative 1.067 1.069 1.06 20 positive 1.068 1.089 1.02 15 positive 1.059 1.081 — 83 positive 1.063 — — 24 positive 1.062 1.066 1.03 Fit positive 1.063 1.068 1.04 in differently sized specimens of Nautilus macromphalus, immature specimens of various sizes were made between 4 and 7 g positive, and the rate of buoyancy change then was monitored as in the previous experiments. Unfortu- nately, during these experiments it proved impossible to capture very small N. macromphalus (less than 15 cham- bers); almost all of the specimens used in this study were mature or within one or two chambers of final size. The list of animals and experiments in which liquid movement was monitored is shown in Table 2. Only three specimens, No. 82-7, 14, and 22, had fewer than 30 or 31 chambers. All three of these specimens showed lowered amounts of “unaccounted” weight change. The two small- est, No. 7 and 14, showed virtually no weight changes. However, it could be that smaller animals were more stressed by the experimental procedures. Buoyancy Change and Depth Equilibrium The last question examined here relates to the amount of positive buoyancy necessary to drive a nautilus to the surface and block it there, so that no amount of swimming effort allows resubmergence. Six mature or near mature specimens of Nautilus macromphalus were made between 6 to 8 g lighter than seawater. With these buoyancies, all Figure 8 Radiographs of four freshly captured specimens of Nautilus macromphalus. These specimens were radiographed with a high-energy hospital radiograph unit, with the use of high-contrast radiograph film in an image enhancement screen. These radiographs provide the first glimpse into the interior whorls. On normal radiographic exposures of nautilus shells, the interiors of only the last ten or so chambers can be observed; earlier chambers are screened from view by shell (whorl) overlap. The high energy radiographs penetrate two separate shell walls, and clearly show the presence of small volumes of liquid in early chambers. Previously, liquid in Nautilus was thought to be present in measurable volumes only in the last formed 4 or 5 chambers in juveniles, and one or two chambers in matures. Note the orientations of the liquid in these chambers, at the top of the chambers, caught between the shell and septal walls, rather than at the bottom of the chambers (the radiographs were taken in the living orientation of the specimen and rotated 90% counterclockwise in the figure, so that “up” is to the left). P. D. Ward, 1986 Page 365 Page 366 The Veliger, Vol. 28, No. 4 Table 2 Buoyancy change in Nautilus macromphalus. ““Unaccounted” buoyancy change refers to buoyancy change that cannot be attributed to water movement into or out of the last 2 chambers. Percentage “Unaccounted of buoyancy Specimen Total Number of Starting Total buoyancy for” buoyancy change due number weight (g) septa buoyancy change (g) change (g) to “unaccounted” 7 369 26 —4.5 0.5 0 0 9 665 30 +24.0 7.8 4.4 56 12 756 31 +17.0 10.0 3.6 36 14 219 25 —5.0 0.0 0 0 15 705 31 +11.0 8.2 325 43 15 — ~- —4.2 0.9 0) 0 15 -- — —8.5 2.0 1.0 50 16 569 30 = 5110) 1 1.0 59 16 a= — ile) 2.8 2.3 82 20 807 31 —Oif2) 1.0 0.8 80 21 658 31 +10.0 6.4 3.0 47 22 458 28 +10.5 Ce) 0.8 32 22 _ ~- = 1258 3.6 0.2 05 23 709 30 —10.0 2.9 0.4 14 24 688 30 —5.0 1.3 0.7 54 24 == — +14.0 2.4 1.2 50 27 739 31 —20.2 4.8 1.6 33 29 541 30 = 2930) 4.0 1.4 35) 29. — — stil SSL 395) EZ 34 81 780 30 +9.6 4.6 2.6 56 wt 708 31 seo TES 3.0 3.0 100 ! ) 1.15,SD=1.1,n= 2 Ee ele Excluding immature animals No. 7, 14, 22. were trapped at the surface. Each of these animals was then periodically weighed, and its position (on the surface, or submerged) noted. All showed buoyancy reduction. Each nautilus was weighed the first time it was found to be either attached to the wall or swimming so that the entire shell was submerged; the weights at first submergence varied between —3.5 and —5.0 g. It appears that more than about 5 g excess buoyancy is sufficient to isolate a mature Nautilus macromphalus on the surface. DISCUSSION The experiments and observations reported in this paper suggest that compensatory buoyancy change occurs in specimens of Nautilus macromphalus. The following points are also raised: (1) Rates of compensatory weight change for positive and negatively buoyant Nautilus macromphalus specimens (surface held animals) are not significantly different (al- though the directions of change are opposite, with one being an increase in weight, the other a decrease in weight) during the first 10 h, but then change. Those animals originally made negatively buoyant (heavier than seawa- ter) continue to reduce buoyancy at approximately con- stant rates. Those animals made positively buoyant (light- 7, for unaccounted buoyancy change, with starting buoyancy positive.* 2.81, SD = 1.1, n = 8, for unaccounted buoyancy change, with starting buoyancy negative.* er than seawater) show marked reduction in buoyancy change after the first 10-h period. (2) Because of the change in rates in the positively buoyant specimens, the potential for buoyancy change is limited in buoyant animals. For the nautiluses made more than 5 g positively buoyant, the range in total buoyancy change was found to be from 1.4 to 6.5 g + 0.3 g (K = 3.8 g). There was no limit of buoyancy change for animals made negatively buoyant. This indicates that the cameral liquid refilling system (in a compensatory response) al- lows replacement of a limited volume of water in emptied chambers (at the surface), whereas the cameral liquid emptying system has no limitation, as long as there is liquid within the chambers to remove. In mature animals, however, with small volumes of pooled cameral liquid, compensatory responses would be ultimately limited to liquid pooled and liquid tied up in the pellicle, and hence be quite limited as well. (3) Positive buoyancy of more than 5.0 g is sufficient to trap mature Nautilus macromphalus at the surface, so that no amount of swimming allows resubmergence. Negative buoyancy of 5 g, however, does not trap a mature N. macromphalus at the bottom. This is probably due to the position and anatomy of the hyponome, which produces water jet propulsion. The hyponome, located beneath the PD» Ward) 1986 Table 3 Liquid refilling rates in single chambers at the surface and at depth. Chamber Liquid refilling Specimen no. Depth (m) number rates uL/h 83-2 0 1 8.3 83-16 0 1 16.6 83-15 0 2 5.0 83-15 0 2 134/ 83-22 0) 1 125 83-27 0 1 93 83-24 0 1 12.5 (From WARD & GREENWALD, 1982) 81-5 0 1 63 81-5 250 1 100 81-5 250 1 21 81-10 0 1 70 81-10 250 1 75 81-20 0 2 54 81-20 250 2 38 tentacles and head region, is not long enough to direct jets of water directly upward, which would push the animal down. The hyponome is much more efficient at pushing the animal off the bottom, as it can jet directly downward. (4) Compensatory refilling or emptying appears to oc- cur over many chambers, not just the last two or three. There appear to be significant reserves of liquid within the chambers (perhaps mostly maintained in the pellicle) that allow for removal of liquid from chambers that do not have pooled liquid. Conversely, Nautilus specimens appear capable of replacing liquid into the phragmocone system, and thus increasing density, without accumulating volumes of “pooled” liquid at the bottoms of the chambers. Also, in contrast to previous observations, significant vol- umes of liquid exist in early-formed chambers. In some respects the experiments listed here are artifi- cial. For instance, the rates of liquid removal for nauti- luses held at the surface are always much faster than those for animals held at depth (WARD & MartTIN, 1978), and, hence, the rate figures found and listed above would prob- ably not be equivalent to those of a naturally occurring nautilus undergoing compensatory buoyancy change at depths greater than 50 to 100 m. The deeper the depth, the slower the emptying. On the other hand, specimens of Nautilus macromphalus are commonly encountered at near- surface depths (WARD, 1982b), and in these cases the surface rates found in this study would probably be quite similar. In the case of positively buoyant nautiluses, those animals having sufficient shell removed would be forced to the surface. In this case, the experiments performed here would directly model the case in nature. For those animals experiencing sudden positive buoyancy at depth, but still maintaining depth even though positively buoy- ant, it could be argued that the added force of ambient Page 367 pressure would force water into the chambers and, hence, allow more rapid compensatory buoyancy change than found in this study. Unfortunately, the logistics of pro- ducing in-water experiments on emptying and refilling are extremely difficult. No data are available about the rates of in-water buoyancy change. However, data about the amount (volume) of liquid volume change through time in positively buoyant nautiluses held at depth are avail- able (WARD & GREENWALD, 1982). In five specimens of Nautilus macromphalus held at a depth of 250 m for pe- riods of 4, 24, or 168 h after artificially induced positive buoyancy, rates of liquid refilling ranged between 21 and 100 wL/hour. Similar rates of refilling for single chambers at surface pressure in this study ranged between 5 and 125 uL/h, while rates listed by WARD & GREENWALD (1982) for surface-held specimens ranged between 54 and 70 wL/h (Table 3). Perhaps the most unexpected result of this study was the finding that variable but significant fractions of the ultimate amount of buoyancy change could not be attrib- uted to measurable liquid volume change within the last- formed two or three chambers in most specimens. Non- pooled liquid is that liquid within a chamber that is trapped by the pellicle and within porous calcareous layers of the siphuncular neck and connecting ring. It cannot be shown experimentally that the removal of this liquid (and also the addition of liquid into this system) is the cause of unexplained density change. However, because it can be demonstrated that density changes are not being produced from within the soft parts, there remain only the small- volume early chambers, and the pellicle and other porous regions within the chambers, that could conceivably be acting for liquid storage. During emptying (following ini- tiation of compensatory buoyancy change in response to increased density) liquid must first be removed from the chalky layers of the connecting ring and siphuncular neck, passing quickly and directly into the siphuncular epithe- lium. As these regions become emptied of liquid, more liquid will be drawn onto them from the contiguously attached pellicle of the septal face. The pellicle itself then draws up liquid from any pooled liquid volume present at the bottom of the chamber. In those chambers where no pooled liquid is still present, the pellicle will appar- ently be emptied until it is essentially dehydrated. At this time the chamber is no longer of any use in density change. Through simultaneous removal of liquid from the pellicles of many chambers, relatively rapid density change occurs prior to the observable removal of pooled liquid reserves, which in mature animals can only be found in the last one or two chambers if present at all. The pellicle system must have some equilibrium vol- ume. However, it appears to be able to take up and store additional liquid if necessary. In the experiment in which animals were made suddenly less dense, significant pro- portions of the density change observed could not be ac- counted for by the accumulation of pooled liquid in the chambers. Again, because density increase through soft- Page 368 part tissue change could not be demonstrated, the observed density change must have been through the movement of liquid from the siphuncle onto the pellicle, where it was stored. Continued addition of liquid onto the pellicle re- sults in saturation and the initiation of pooling at the base of the chamber. Apparently, as much as 2 g of weight increase can occur before measurable accumulation can be noted. ACKNOWLEDGMENTS This work was supported by NSF PCM 8202891, P. Ward and L. Greenwald, Co-P.I.’s. I would like to thank Dr. Y. Magnier, Aquarium de Noumea, and Dr. Paul de Boisseson, Director, O.R.S.T.O.M., Noumea, New Cal- edonia, for allowing me laboratory space. Dr. Marc, Ra- diography, Magenta Clinic, allowed me the use of his fine radiography facility. Mary Graziose did her usual fine job with the photographic plates. Ellen Bailey completed the drafting. Special thanks to C. Ward, M. Cala, and the Captain and crew of MV Vauban for help at sea. Dr. L. Greenwald thoughtfully reviewed this manuscript. LITERATURE CITED Co.uins, D., P. WARD & G. WESTERMANN. 1980. Function of cameral water in Nautilus. Paleobiology 6(2):168-172. The Veliger, Vol. 28, No. 4 DENTON, E. & J. GILPIN-BROWN. 1966. On the buoyancy of the pearly Nautilus. J. Mar. Biol. Assoc. U.K. 46:723-759. GREENWALD, L., P. WARD & O. GREENWALD. 1980. Cameral liquid transport and buoyancy control in chambered nau- tilus. (Nautilus macromphalus). Nature 286:55-56. HEPTONSTALL, B. 1970. Buoyancy control in ammonoids. Le- thaia 3:317-328. Mutvel, H. & R. REYMENT. 1973. Buoyancy control and siphuncle function in ammonoids. Palaeontology 16(3):623- 636. SEILACHER, A. 1960. Epizoans as a key to ammonoid ecology. J. Paleontol. 34:183-193. WarD, P. 1982a. The relationship of siphuncle size to emp- tying rates in chambered cephalopods: implications for cephalopod paleobiology. Paleobiology 8(4):426-433. WarD, P. 1982b. Have shell, will float. Natur. Hist. (Oct): 64-69. WarD, P. & L. GREENWALD. 1982. Chamber refilling in nau- tilus. J. Mar. Biol. Assoc. U.K. 62:469-475. Warp, P. & W. MarTIN. 1978. On the buoyancy of the pearly Nautilus. J. Exp. Zool. 205:5-12. WarbD, P. & A. MarTIN. 1980. Depth distributions of Nautilus pompilius in Fiji and Nautilus macromphalus in New Cale- donia. Veliger 22(3):259-264. Ward, P., R. STONE, G. WESTERMANN & A. MARTIN. 1977. Notes on animal weight, cameral fluids, swimming speed, and color polymorphism of the cephalopod Nautilus pom- pilius in the Fiji Islands. Paleobiology 3:377-388. WARD, P. & M. WICKSTEN. 1980. Food sources and feeding behavior of Nautilus macromphalus. Veliger 23(2):119-124. The Veliger 28(4):369-388 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 A Model for Shell Patterns Based on Neural Activity by BARD ERMENTROUT Department of Mathematics and Statistics, University of Pittsburgh, Pittsburgh, Pennsylvania 15260, U.S.A. JOHN CAMPBELL Department of Anatomy, University of California, Los Angeles, California 90024, U.S.A. AND GEORGE OSTER Departments of Biophysics and Entomology, University of California, Berkeley, California 94720, U.S.A. Abstract. The patterns of pigment on the shells of mollusks provide one of the most beautiful and complex examples of animal decoration. Recent evidence suggests that these patterns may arise from the stimulation of secretory cells in the mantle by the activity of the animal’s central nervous system. We present here a mathematical model based on this notion. A rather simple scheme of nervous activation and inhibition of secretory activity can reproduce a large number of the observed shell patterns. INTRODUCTION THE GEOMETRICAL patterns found on the shells of mol- lusks comprise some of the most intricate and colorful patterns found in the animal kingdom. Their variety is such that it is difficult to imagine that any single mecha- nism can be found. Adding to their mystery is the dis- turbing fact that, since many species hide their pattern in the bottom mud, or beneath an opaque outer layer, it is doubtful they could serve any adaptive function. Perhaps these wonderful patterns arise as an epiphenomenon of the shell secretion process. This may account for the ex- treme polymorphism exhibited by certain species—a phe- nomenon characteristic of traits shielded from selection. Several authors have attempted to reproduce some of these patterns using models that depend on some assumed behavior of the pigment cells in the mantle that secrete the color patterns (WADDINGTON & CoweE, 1969; CowE, 1971; WANSHER, 1972; HERMAN & LIU, 1973; HERMAN, 1975; LinDsay, 1982a, b; WOLFRAM, 1984; MEINHARDT, 1984). These models have generally been of the “cellular automata” variety, and the postulated rules were chosen so as to give interesting patterns, rather than to correspond to known physiological processes (WADDINGTON & COWE, 1969; LINDsay, 1982a, b; WOLFRAM, 1984). In the most recent attempt, MEINHARDT (1984) modeled the growing edge of the shell as a line of cells subject to activator- inhibitor kinetics and a refractory period. He was able to obtain a variety of shell-like patterns, suggesting that an activator-inhibitor mechanism is likely to be involved in the actual process. Recently, CAMPBELL (1982) proposed a novel expla- nation for the shell patterns. He reasoned that the pigment cells of the mantle behaved much like secretory cells in other organisms; that is, they secreted when stimulated by nervous impulses. Therefore, the shell patterns could be a recording of the nervous activity in the mantle. Because the phylogeny of mollusks is well represented in the fossil record, the implications of this view for the study of the evolution of a nervous system are obvious. Building on Campbell’s notion, and the suggestive sim- ulations of Lindsay, Meinhardt, and Wolfram, we have constructed a model for the shell patterns based on nerve- Page 370 Sf 4d é a b The Veliger, Vol. 28, No. 4 Figure 1 Three fundamental classes of shell pigment markings on Bankivia fasciata: a, longitudinal bands; b, incremental lines; c, oblique stripes. stimulated secretion of the mantle epithelial cells. This model differs from previous models in at least one impor- tant aspect: it depends on the “nonlocal” property of nerve nets. That is, because innervations may connect secreting cells that are not nearest neighbors, the possibility of co- operative, long-range interactions is present. This greatly enlarges the pattern-generating repertoire over nearest- neighbor models, and has the virtue of relating directly to the anatomy of the mantle. Despite its simplicity, the mod- el is remarkably successful in mimicking a wide variety of shell patterns. The paper is organized as follows. First, we catalog a number of regularities in the shell pattern that bear on the neural hypothesis. In particular, those phenomena that implicate a global organizer and preclude strictly local interactions. Second, we sketch the model equations and discuss their behavior. Third, we present patterns gener- ated by simulations of the model and compare them to actual shell patterns. Fourth, we discuss some experi- ments the model suggests and some generalizations of the model. The Appendices contain the mathematical details of the model and a discussion of how it relates to other models of shell patterns. OBSERVATIONS ON SHELL PATTERNS The variety of shell patterns is so enormous that it appears that any attempt to classify them will inevitably leave out many special cases. However, we do not hope to explain all of the patterns; rather, we seek to model the global features shared by all patterns in a restricted class. In particular, we shall focus mostly on the patterns exhibited by Nerita turrita and Bankiwia fasciata (Figures 1, 2, 3). These animals exhibit a representative variety of shell patterns from which we can draw some inferences. Many pigment patterns of gastropod shells are com- posites of three basic types of patterns: (a) longitudinal bands that run perpendicular to the lip of the shell, (b) incremental patterns arranged parallel to the growing shell edge, and (c) oblique patterns that run at an angle to the lines of the shell. Some mollusk taxa have more specialized types of patterns, such as the circular eye-spots on some cowry shells, or the intricate tent-like patterns on cone shells. Species differ in the categories of patterns that they display. Nerita turrita shells are always dominated by oblique patterns without longitudinal bands, whereas oth- er members of the genus have shells with bands as well B. Ermentrout et al., 1986 Bases! Figure 2 a, an incremental alternation in zebra stripes across the entire whorl of Bankivia fasciata; b, simultaneous termination of stripes in B. fasciata. as modified oblique lines. Shells of Bankivia fasciata are highly polymorphic, with various combinations of these pattern types, as illustrated in Figure 1. Longitudinal bands require only simple developmental controls. They could result from a mosaic mantle in which regions continuously deposit pigment, along with shell, separated by mantle zones that do not synthesize pigment. In general, the number and position of bands appears to be a genetic characteristic of the species, or of the individ- ual in a polymorphic species. A second possibility —which we Shall illustrate with the model—is that the band width and spacing are characteristic of the neural activity in the mantle. The two mechanisms are not mutually exclusive, as we Shall discuss. Banding indicates that variation can be a permanent (e.g., programmed) feature of the mantle edge. Incremental markings have several sources. Some ap- pear to result from haphazard physiological stresses or environmental factors that temporarily affect the activities of the mantle as a whole. In addition, some species of snails (other than the ones we shall consider here) show regular periodic incremental shell patterns, indicating that they are programmed in a deterministic and cyclic man- ner. One of the most important incremental features seen on shells of the two species we have chosen for analysis are varices: time periods during which shell synthesis was halted (Figures 2, 3). In general, mollusks do not produce shell continuously, but go through cyclic periods of shell building (producing about one-third to one-half whorl of shell in the case of Bankivia fasciata), followed by “rest” periods during which no shell is secreted. Shell patterns often are reorganized at these major interruptions in shell synthesis, and many sculptured shells produce flamboyant ridges or spines along varices. Oblique patterns are the most intricate, and have the most implications for our theoretical model. They imply that the activities involved in pigment secretion are coor- dinated laterally and proceed dynamically across the man- tle. For example, the oblique lines shown in most of the shell illustrations in this paper represent a patch or do- main of secretory activity that sweeps across the mantle, eventually migrating to its edge. These mobile domains of activity in the mantle behave in a variety of ways to pro- duce the diverse appearance of the patterns. EVIDENCE IN FAVOR OF LONG-RANGE COORDINATION OF PATTERNS The neural network model we propose here allows for interactions and coordination beyond nearest neighbors. As we shall demonstrate in the next section, this gener- alization enormously enlarges the possible types of pat- terns over previous models, which employ short range, or “nearest neighbor” interactions. What evidence do we have that pigment secretion is indeed a neurally controlled pro- cess? We can offer no direct experimental support, for we Page 372 The Veliger, Vol. 28, No. 4 b Figure 3 Abrupt reorganization of patterns on shells of Nerzta turrita (a), and after a break in a shell (b). have not been able to find any anatomical studies of man- tle innervation patterns nor of secretory cell physiology. Therefore, aside from the general observation that secre- tory cells in most organisms are influenced by neural ac- tivity, we can offer only the following indirect evidence in support of the neural activation hypothesis of shell pat- terns. Global reorganizations. At a varix, a shell pattern may become systematically and simultaneously reorga- nized across an entire shell (cf. Figures 2a, b), sometimes into an entirely different sort of pattern (Figure 3a). A variety of new patterns may arise in this manner, rather than arising locally and propagating as a wave across the shell. Such changes in the “state” of the pattern can also be initiated by a break in the shell (Figure 3b). It is hard to see how such local perturbations could have such global effects by means other than nervous activity. It should be noted that physiological and (or) environ- mental factors can influence the entire mantle simulta- neously. Indeed, it has been demonstrated that changes in diet can alter not only the color of the pattern, but the pattern itself (D. Lindberg, personal communication). This fact does not argue for or against the neural hypothesis, for it is relatively common for dietary factors to affect nervous activity, as well as other physiological systems. However, because diet and other environmental factors affect the pattern formed on a shell, there must be some physiological mechanism that relates the two. That is, there must be some mechanism whereby a systemic effect allows two separated regions of mantle tissue to manifest coincidental patterning. The two main avenues for trans- mitting stimuli from the environment to the mantle cells are soluble chemical factors (especially hormones) and nervous connections. Both may modulate patterns, but in- fluences that differentially affect discrete parts of the man- tle simultaneously seem more plausibly mediated by the nervous system. Entrainment of lines. Shells in which oblique lines become entrained in the middle of a longitudinal band also suggest coordination of pattern across sizable dis- tances, measured in cell diameters. A particular example of this is the shell in Figure 4a, on which a band appears spaced equidistant from the neighboring bands. Termination of lines. On the shell in Figure 4b three oblique lines terminated anomalously at about the same time. These events occurred in regions of shell separated by uninterrupted oblique bands. If these changes were due to a signal that propagated from one locale to another, that signal would have to have migrated cryptically past the unaffected domains in the mantle. The simpler inter- B. Ermentrout ef al., 1986 Page 373 Figure 4 a, appearance of a band spaced equidistantly between adjacent bands; b, simultaneous termination of several separated zebra stripes without noticeable concurrent alteration of the stripes in between. pretation is that the three separate areas were acted upon by a signal that could be conveyed to multiple local regions simultaneously. Blotching. A polymorphism (not otherwise described here) among Bankiwia fasciata shells is blotching (Figure 5). On blotched shells areas of pigmentation abruptly dis- appear or appear incrementally across large blocks of shell. Alternatively, various segments of the mantle can be af- fected simultaneously by blotching. Also, for some blotched shells the zone of pigment deposition did gradually spread along the mantle, indicating that blotching can be con- trolled in a variety of ways. Global appearance of patterns. On some shells a gen- eral type of pattern gradually develops across the entire mantle, but with no indication that the change sweeps across the mantle; the saw-tooth pattern in Figure 6 il- lustrates this phenomenon. Checkerboard patterns. (Figure 7) It is possible to create a checkerboard pattern from two sets of colliding waves that propagate by strictly local interactions. How- ever, it is remarkable that the checkerboard as a whole can stay in register without drifting in alignment. This synchronicity implies that a substantial segment of the mantle cycles back and forth between an active and in- active state in precise coordination. Adjacent subzones switch states of activity simultaneously, but in opposite directions. THE NEURAL MODEL In this section we present a qualitative description of the shell pattern model. The mathematical discussion is given in the Appendices. The model we shall present here is the simplest possible neural model, and we do not expect to reproduce every shell pattern, even those observed on the two species we have selected for study. However, the mod- el is capable of producing sufficiently diverse patterns that we consider it a reasonable first approximation; we shall suggest a number of improvements which will enlarge the class of patterns, but at the expense of computational sim- plicity. BIOLOGICAL ASSUMPTIONS OF THE MODEL The basic assumption of the model is that the secretory activity of the epithelial cells that generate shell patterns is regulated by nervous activity. Specifically, we assume that the secretory cells are enervated from the central gan- glion and secrete or not as they are activated and inhibited by the neural network that interconnects them with the ganglion. Although arguments in favor of this hypothesis were presented above, the issue can only be settled em- pirically, and experiments are under way to test the neu- ral hypothesis directly. Figure 8 shows a schematic of the Page 374 The Veliger, Vol. 28, No. 4 Explanation of Figures 5 to 7 Figure 5. Blotched patterns on Bankivia fasciata shells. Figure 7. Checkerboard patterns on Bankivia fasciata. Figure 6. Sawtooth patterns on Nerita turrita. mantle and the secreting cells (EMBERTON, 1963; KAPUR daily) bursts of activity. At the beginning of each & GIBSON, 1967; NEFF, 1972; KNIPRATH, 1977). session the mantle aligns with the previous pattern The specific assumptions that underlie the model are: and extends it by a small amount. This alignment process probably depends on the ability of the mantle (1.) Cells at the mantle edge secrete in intermittent (e.g., to sense (taste) the pigmented and (or) non-pigment- B. Ermentrout et al., 1986 periostracal groove py eeerae wm g periostracal cells ventral epithelium ou mantle edge gland Page 375 extrapallial space Sa dorsal epithelium Figure 8 Diagram of the anatomy of the mantle region. ed regions from the previous period of secretion. Equivalently, a section of pigmented shell laid down during the previous period will stimulate the mantle neurons locally to continue the pattern. (2.) The secretion during a given period depends on two factors: (a) the neural stimulation, S, from surrounding re- gions of the mantle. (b) the buildup of an inhibitory substance, R, within the secretory cell. (3.) The net neural stimulation of the secretory cells is the difference between excitatory and inhibitory in- puts from surrounding tissue. We incorporate these assumptions into the model as fol- lows. Secretion of Pigment Depends on Current Neural Activity Consider a line of secretory cells whose position along the mantle edge is located by the coordinate x (Figure 9). Let P(x) =the amount of pigment secreted by a cell at x during the time period t (e.g., one day). A(x) = the average activity of the mantle neural net at position x on the mantle edge during one secretion period, t. R(x) = the amount of inhibitory substance produced by cells at location x in day t. S[P] =the net neural stimulation at location x during period t. This will depend on sensing the pigment secreted during the previous period, P,_,(x). Then the equation governing the neural activity in the mantle during period t + 1 is related to the pigment se- cretion during period t by the equation A,44(%) = STP (x) ] aa, [1] Equation [1] says that the average neural activity, A,,,(x), at location x on the mantle during day t + 1 depends on the net neural stimulation at that location, which is stimulated by sensing the previous day’s pigment S{P,(x)]. In the absence of stimulation, this nervous activ- ity decays as the inhibitory substance R,(x) builds up. The inhibitory substance, R, builds up as pigment, P, is man- ufactured, and is degraded at a constant rate (6 < 1): Riyilx) = yPx) + 6R(x) [2] Finally, we assume that secretion of pigment will only occur if the mantle activity is above a threshold value, A*: P(x) = H(A — A*) [3] where H(A — A%*) is a threshold function for pigment secretion: it is zero for A < A*, and one for A > A*. Equations [1] and [2] describe how the activity, A, and refractory substance, R, evolve in time; having computed A, the actual pigment secretion is given by [3]. In the computer simulations we have simplified the model even further by incorporating equation [3] into equation [1], and writing equations for P directly (cf. Appendix A). This modification makes little difference in the computed Page 376 The Veliger, Vol. 28, No. 4 Figure 9 Diagram of the model: MPC, mantle pigment cells; PCN, pigment cell neurons; MNN, mantle neural net; CG, central ganglion; r, receptor cells sensing pigment laid down in time period t; P, pigment cells secreting pigment in time period t + 1; E, excitatory neurons; I, inhibitory neurons. patterns, but is somewhat simpler to simulate. Figure 9 shows a schematic of the model’s structure. Neural Activity Depends on the Difference Between Excitatory and Inhibitory Stimulation Next, we must model the process of neural stimulation that regulates the secretion of the pigment. We regard the net stimulation of a cell at x to be the difference between excitatory and inhibitory stimulations from nearby cells. The situation is illustrated in Figure 10a: a cell located at a position x on the mantle edge received excitatory inputs and inhibitory inputs. The inhibitory signals are generally more “long range” than the excitatory inputs; that is, the mantle edge exhibits the property of short- range excitation and long-range inhibition characteristic of neural nets (BERNE & LEvy, 1983; ERMENTROUT & Cowan, 1979). Moreover, we assume that the response of a nerve cell is a saturating function of its inputs; that is, both excitation and inhibition are sigmoidal functions of their arguments, as shown in Figure 10b. The mathe- matical form of the neural stimulation term we have em- ployed is given in the Appendix. When these assumptions are incorporated into the mod- el equations there results a set of functional difference equations that determines the pigment pattern, P,(x) (cf. equations [A1, 2]). In Appendix A we perform a linear analysis on these equations. This gives some idea of the repertoire of patterns the model can generate, and pro- vides a guide to the numerical simulations presented be- low. The Model Parameters Any model contains adjustable parameters, and equa- tions [1]-[3] contain several. These parameters fall into two categories: (A) those controlling the shape of the neu- ral stimulation function, and (B) the production and deg- radation rates of the inhibitory substance. Each parameter corresponds to a definite physiological quantity, and so is measurable, at least in principle. Neural parameters. The neural stimulation function, S, in equation [1] contains the curves for excitation, in- hibition, and firing threshold shown in Figure 10. Each of these functions must be described by formulae that con- tain parameters to control their shapes. The functions we have employed in our simulations are described in Ap- W,(x- x) (a) (b) Figure 10 Diagram of the neural influence function and _ threshold function. B. Ermentrout et al., 1986 prone bh shee mann aaa BY: pre preety SC ERTET Tecscf Sab vcssossbasssdbilesscslasaLineds bea uee suobeteh esa gve soe aula Page 377 Sse TR meme a Bad te ee ono a. EoEnEAT=O o rer a Figure 11 Simulations of: a, vertical stripes of constant width; b, vertical stripes of variable width; c, horizontal stripes. pendix A; however, experience has shown that the qual- itative predictions of the model depend only on the general shapes of the functions, not on their particular algebraic form. Cellular parameters. Each secretory cell is character- ized by its production rate of pigment under neural stim- ulation and its production and degradation of refractory substance, R. The production rate of pigment is controlled entirely by the neural stimulation, S, and so no new pa- rameters are required to describe it. The refractory sub- stance, however, requires the two parameters: y to regu- late the growth rate of R, and 6 to control the decay rate of R. Even though each of the model parameters has a direct physiological interpretation, with enough parameters one might feel that any variety of patterns is possible. How- ever, this is not true. For a fixed neural structure, there are but two adjustable parameters: y and 6. Varying the neural interactions involves changes in their shape-con- trolling parameters, and analysis and simulation studies Page 378 J) Degradation of R | il : production of “K Figure 12 0 y-6 parameter plane showing domain of stripes and obliques. show that the resulting patterns can be classified into a relatively small number of types. Within each distinct type, variations of the parameters merely alter the relative di- mensions of the pattern, and not its qualitative appear- ance. However, parameter variations that exceed certain thresholds, cause the pattern to shift not just its scale, but its qualitative type as well. This “bifurcation” behavior will be discussed further below. PATTERNS GENERATED BY THE MODEL In this section we describe the patterns generated by the neural model. We shall present numerical simulations of the neural model which mimic certain patterns observed on the shells of Bankivia fasciata and Nerita turrita. Basic Patterns Equations [1]-[3] constitute the simplest possible model for a neural net; consequently, we cannot hope to repro- duce all of the known shell patterns. However, we can reproduce all of the basic patterns; moreover, it is easy to see how the model can be elaborated to incorporate a wider variety of patterns. We shall briefly discuss these modifications here, and present a more detailed study in a subsequent paper. The three fundamental patterns exhibited by Nerita turrita and Bankivia fasciata are longitudinal bands, incre- mental lines, and oblique stripes (Figure 1). The param- eter values that realize these patterns are given in Table 2 in Appendix A. Qualitatively, the conditions that yield these patterns are as follows. Vertical stripes (Figures 1a, 11) occur when refrac- The Veliger, Vol. 28, No. 4 toriness is very low and the neural influence functions are strong and thresholds small. There are two mechanisms for producing stripes: one is similar to the Turing mech- anism in diffusion-reaction models. That is, short-range activation creates a laterally spreading zone of activity, which is eventually quenched by the longer range inhib- itory activity. This produces stripes whose width is con- stant, as shown in Figure 11a. The stripe width is a func- tion of the parameters (being roughly the width of the activation-inhibition zone), and the locations of the stripes are determined by the width of the domain (z.e., the size of the mantle). A different mechanism produces stripes of unequal widths, as shown in Figure 11b. It is also possible to produce vertical stripes by simply activating certain regions of the mantle permanently, so that secretion is always turned on. Only experiments can distinguish be- tween these two possibilities. Horizontal stripes, or incremental lines (Figures 1b, 11c), are produced when the refractory parameters are small and thresholds are high. This results from a syn- chronized, or homogeneous oscillation along the entire mantle (not to be confused with the incremental pattern associated with the episodic nature of shell deposition). Diagonal stripes, or zebra bands (Figures 1c, 2, 3, 4), are characterized by very low thresholds and gradual cut- offs. These arise as waves of activity propagate along the mantle. If the neural structure is constant, the presence of oblique stripes or vertical bands depends on the values of the two parameters controlling the refractoriness, y and 6. Figure 12 shows the parameter domain that character- izes each pattern type. The direction of the stripes produced by the model de- pends on the parameter values. However, downward ori- ented stripes (z.e., away from the apex of the shell) are more common in Bankivia fasciata and Nerita turrita and exhibit far fewer irregularities. Moreover, upward-di- rected stripes appear to be more unstable, reverting to downward stripes after a short progression. This points to a consistent inhomogeneity in the mantle. Indeed, su- perimposing a parameter gradient (e.g., in 6 and [or] y) on the model equations strongly biases the direction of striping in one direction. Interestingly, the direction of stable striping is in the same direction as the spiral of the shell. Because shell patterns are associated with shell con- struction, this could indicate a physiological (anatomical) correlation between the direction of shell growth and the pattern direction, such as an asymmetry in the muscle mass of the mantle. The direction of the zebra stripes can switch at certain times, especially—but not exclusively— at a varix. Divaricate patterns. Zebra patterns may reverse di- rections giving a herring-bone pattern. We have used the observation that synchronous switching of the direction of stripes indicates a global coordinating mechanism for the pattern. In terms of the model, switching of the direction of obliques involves a jump in a parameter value. The model does not address what the underlying signal for B. Ermentrout et al., 1986 AAARS; / AAAAAS: +WAAAARS, ~ SAAAAL, » SAARAAASt, « T8AAAAA, » SAAAAAAAS; q yt tdAAAAS AWSZ, AAAZ AAAHS; AAAAAS +AAAAAARS+S. «: ZAAAAA. » SARAAAK+ sSAAAAAA: = SAAAAAARS+, AAAARA: +. «i ZAAAAA, AA&Z, AAAS, AAAWZ: AAAAAS. s SAAAAARSH, > SAAAAS s SAARAAN+ sZHAAAAA: ~SAAAAAANS+, =) SAAAAA. +HAAAAARS+, «7 SAAAAS sBAARAAH+ » tZAAAAAA, ry SAAAAAANS;. . oy tSAAAAR AWSt. AAA+, AAARZ: AAAAAZ. +AAAAAARZ;. 1 aARAAt ,HAAAAAL; «i ZAAAAAA. sSAAAAAARTZ: y :Z&AAAAZ AuS; AAH+ AAASZ: AAAAAt. ZTAAAAANSt: »tRAAAA, - tAAAAAALS » =ZAARAAA ySARARAASH, HAAAAAA b Page 379 +WAAAAAA: AAAAAAA , SAAAARA. SAAAAAA SAAAAAAY AAAS. ,tSAAAAAR:, -AAAARAS: :AAAAAR AAAAAAAt °+AAAAARS ,AAAAARS, ,AAAAAZ:.. SZRAAAAAAR, :TWAAAAAT:. , SAAAARAA: :AAAAAAAS. LAAAAAT:. :SAAAAAA AAARAAAR , LAAAAAAAZ :AAAAAAS AAAAAAA TAAAAANS: : ,tLEAAAS — :AAAAAAS . PSAAAAAAS 5184 +AARAAAAS: )ZAAAAAAAARS: .SAAAAAS, ,+8AA , TARAAAA TAAAAAAAS +HARAAAAA, :SRAAA +RAAAAAAR? AAAAAAA AAAAAAL, TARARAA eETRAAAA SSAAAAN AAAAAAA+ . DLARAAAA+ +AAAAAAt:, = +23 SAAAAAS+. .AAAAS+,. ,SAAAAAAAt ZEAAAAAAAL. sRAAAAAASS. RAAAAAAAt AAAAAAR, ZAAAAAS:. :SAAAAAA AAAAAAAA +8AARAAAAS ,AAAAASt HAAAASZ:, JLRHAAA+, . rZAAAAAL ,IRAAAAAAA+ =: 23H8 SAAAAAAL, s SAAAAAAAN; *AAAAAS: , AAA sHAAAAAA. ;AAAAAAAS +RAAAAAAR ,+8AAAA ZAAAAAAAZ, ARAAAAA AAAAAAS: ,AAAAAK ,TSAAAAR SAAAAAS TAAAAAAAS TZAAAAARt RAAAAAS:.. 88S: SAAAARH: ,AAAAS;,.. :SAAAAAANS. +$AAAAAAAS +8AAAAAAS! SAAAAAAAt , AAAAAAAS RAAAAAS:. :SAAAAAA AAAAAAAA KAAAAAAAS -AAAAAA, -AAAAAAS SAAAAAS+S ' ,USSRAAt. ,TAAAAAA: STRAAAAAAR =. 5.188 ,WAAAAAAZ, ;HAAAAAAARS, «ARAAARS. +800 sWAAAAAA SAAAAAAAS +HAAAAAAA, sTRRAAA SAAAAAAAZ. AAAAAAA AAAAAAS: ,AAAAAA ,tEAAAAL SAAAAAR TAAAAAAAY T+HAAAAAZ EAAAAAS:. +HHE: LAAAARt:. HAAASt:.. :SAAAAAAS+. +$AAAAAAAL .+8AAAAAASt, HAAAAAAA+ .AAAAAAAS SAAAAAL:. :LAAAAAA AAAAAAAA THAAAAAAS ,ARAAAA, J AAAAAAS ,AAAAAAS+. : :S3HAAA: © +AAAAAAS TZRAAAAAAR. © 3288 sAAAAAAAP, .TSAAAAAAARS: -WAAAARS. s$AA SWAAAAAR +AAAAAAAS ;WAAAAAAA, :18HAAA SAAAAAAAZ. AAAAAAA AAAAAARS ,WAAAAA T+HAAAAY SAAAAAS ,WAAAAAAS .pBAAAAAS -ARAAAA+,. — ZAAN+ SAAAAAZS, TAAANZ:,. sSAAAAAAS:. sZRAAAAAAR, -t8AAAAAARY: -WAAAAAAAT -ARAAAAAS. ZAAAAAZ:. SHAAAAAA SAAAAAAA +2AAAAAAS +AAAAAA. AAAAAAS. AAAAAARZ, SSHHAAR, © :AAAAAAZ T+$AAAAAAAS = 488 T+8AAAAAAAASS. SAAAALS. 7HA L+HAAAAAS SAAAAAAAR TSAAAAAAA: 2ZRRAAA 3AAAAAAAt AAAAAAA, 3AAAAAH. HARAAA J rSAAAAA, SAAAAAK , BAAAAAAA SAAARAS SAAAAAAS. SAAAS TAAAARZ:, SAAAAZS:, -+8AAAAANTS. :+SAAAAAAR, . +8AAAAAANT: :WAAAAAAA RAAAAAAZ, ZAAAAAL:. sHAAAAAA ZAAAAAAA: SSAAAAAAS. AAAAAAL. AAAAAANL: ; +AAAAAART, Figure 13 a, divaricate patterns on Bankivia fasciata showing open and closed V’s; b, simulation of V’s. such an event is, but does provide a mechanism for gen- erating a coordinated reversal of the pattern orientation (Figure 13). Lines that converge as the shell grows will be called “closed V’s”; those that diverge as the shell grows are “open V’s.” Pattern reversals that produce a “closed V” frequently extend beyond the intersection a small amount, forming a “‘snout” on the V. This is also a feature of the simulations, because a collision of two obliques admits a small overlap of the activation region extending ° beyond the collision apex. Note also that the upward stripes are shorter than the downward stripes, suggesting a man- tle inhomogeneity. This has been suggested previously by WRIGLEY (1948). Wavy stripes (Figure 14). These are characterized by very sharp cutoffs of the excitatory and inhibitory thresh- olds, small thresholds, and large turnover of refractory substance (y, 6 © 1). Note the ‘‘shocklike” discontinuities in the stripes that the simulation reproduces. Streams (Figure 15) are irregular striped patterns that occur when the sharpness of the cutoff is quite large and refractoriness is persistent (6 ~ 0.8). Interaction Patterns In addition to the basic patterns, additional designs emerge from the interaction of the basic patterns. Typi- Page 380 a AF Figure 14 a, divaricate patterns (wavy bands) on Nerita turrita; b, simu- lation. cally, when two diagonals collide one of several things happen. Checks (Figures 7, 16) occur when the range of neural interaction is large. As the sharpness of the excitatory and inhibitory thresholds increases, the checks become more stable and persistent. On some shells, colliding diagonals pass through one another. This cannot happen in our two-variable model. In order to obtain this effect one must add a third variable; this implies that the secretory activity of the mantle is associated with more than one pigment, or that the mantle can sustain several coexisting and independent patterns of neural activity. We will deal with this phenomenon in a subsequent publication. Tents. These patterns are not observed on Nerita turrita or Bankivia fasciata, but are common on the cone shells. We include them here because the model also can produce a wide variety of tent patterns, examples of which are illustrated in Figure 17. These patterns most easily arise when the concentration of refractory substance, R, is very low (6, y < 1), the nonlinearities are extremely sharp, and the range of neural interaction small. In this limit the model resembles the “nearest neighbor” cellular automata models of WOLFRAM (1984) and others (cf. Appendix C). Indeed, the tent patterns appear to arise from more lo- calized interactions (“nearest neighbors” in the cellular automata models) than the other patterns described herein. In this regard, the models of Wolfram are able to mimic a remarkable variety of these kinds of “local” patterns, and the model presented here can do little better in pro- The Veliger, Vol. 28, No. 4 It, »tWAAR+; SAAS ZTAAAAAAAAA, LAAAtT SSL. .§ ytWAAA;, +8AA sAAR,, ;HAAAZS> +87, ... -$#AAA ytRAAAA; StAZZ:8HAa;, ys AAAAR, sists 72$AA vy tHAARAZ: SARAAAN TAAAAN, BRU IER RR Ca 98 BRAAAAZ: 794%: .AAAAA, 552 rtAAAW,., tESAAAMZAZZALSIZ AARAAZ: SSAAAAWS: +285: sSAAAR+ 2 AAAS tHAZZZAHAMARAAAN: 5 AAZ: TAAARAARAA AAAAAAAL; tAAAAAZTZ «+227 7ZZEARAASS = $+ sSAAAA; : &AAAt +ARAAARAAN 2HARAS, 9y4$Z%,,. ZLARARA sSAAAAt zAAA,. s:AA&.. ;ARAASS $$UZ.s25...5- :$AAA 2 SAAAAA. >» AARARAA, = ;AARAM: 97 shhh: 228A «,SAAAAAt, 50 AAAAAAA sAAAAAS ZAAALZ, ,++Z8. «+ +RAAAAt. ySAAZ; .AAAAS. +47 2 tAARAS AAAAAAS; SARARAS; AAAA+, +AAAAAA; ote sSAAAR+ sAAAA, Ws; :ZZRSHHAARAA,. AAR. = tAAAAAAAAR AAAAAAAL: «ARAAAS += $&$+,.5AAAAZ, h, ~tAAAA, ZAAA+ +AAARAAAAA:. »tAhAAA aan. yAAAAA » tAAAAA. oSAAZ; S+BAAt ZAAAR:, rSAA,5s5 42$,- 2$AAA +AAAAAS TAAAAAAA sAAAAAY yAAAASZ r++ 318 + SAAAAS. ys SAAAAAS »HAAAA+ AAAAAS. .ZAAS: SAAAAS. +AAASt+ ZSHS+ SAAS oSAAARS SHAAAA: ZAAAAA; AAAA. +AAAAAAS +AAAAAZ sAAAAN: =o -L; SRN ZAAAAASS . AN+ > HARASAAAAR AAAAAAAS; $AAAAR AAA. S&AAAAS Zz yHAAAR sAAAR. sAAASS SAAAARS yAARA AAW HAAAA sHAAAAS +AAARZAAAS +AAAASt sAAAS++. S5ty ytAAA ~TAAAAA; sAARAAAA sAAAAAZ yAAAAAt ZEWE ok » 8RAAAAS gAA, +AAAAA, AB sARAAR, AAAAAS AAAS AARAAS ZAAAAZ siyity SAAAA, yARRAt Z$AAA =6AARAA; AAAZ y SAAAAAAA ZAAARAA, ZTAAAAZ yAANZES SSAAR+:. AW. yLAAAS :AAAA, > AAAAAAAAt sAAAAM TARA: ~RAARAA y LAAAAZ ZAAA+,.5$AA: sAAAAZ. S$AAAZ. eAAA yARAA ZAAAAAt tAAAAAARAt eAARAAZ SAAA8+; 5% , 7 SAA yHAAAAZ ste yAARAAAA ~SAAAAR. sAAARA AAAH c yHAAAAZ AAA+ SAAAR, sBR, « $AAARS ZAAAA+ ZAAAAt AAAAZ »ARAAAM: ; yAAAAA; TAAAS; -;$AA AARAAA, AAA; +AAAARAAA: AAAAAAAt sAAAAA: AAAAA, +AARZ; A; +HAAA: = AAAAt TAAARSSARAt AAAAA. AAAAZ +AAAA +HAAAA: pAAAR+EZ$AA: AAAAMS yAAAA +AAA. HAAA yRAAAAA AAARAAAA, AAAAAR, yAAAARt 7 yAS. 529 » LAAAAAS s+ SAAAAAA SAAAAA, AAAAA AAAA, +HAAAAS tAAAS, pSAZZ. 5 ZAR+ pAAAAZ ~AAAA +AAAAA AAAA; +AAARAAt yAARAAZ sAAAAS, -79$+ AASHAAA, AAS « SAAARAAAAt AAAAAAAZ AAAAAt AAAAR AAAAR A; yZAAAA ZAAAE, «AAAS SAAAR. SAAAA; AARAA AAAA y LARAAA AAAAASSAAA ARAAAZ ZAAA. HAAA, sHAA +AAAAAZ HAAAAAARA HAAAAA+ AAAAAA; ZAAN aH yHAAAAZ $AA. sAAAAAS sAAAAA; AAARAA nARAS BAARAZ ZAAAAL «$82 AAAAZ »SAAAA. AAAA sARAAA; ARAA ZARAAAAZ WAAAAAZ sAAARAt «sAZtAA, ,8AR8;, AA; +ARAAtZAAAR, AAAAARAS AAARAAZ HAAAA sAAAA z +AAAAR AAAAt+, .BAS: ,AAAAZ, sAAAAZ AAAAA $AAA +HAAAAZ ZARAAAAAA, ZARAAR, tAAAZ s;AS$3. . 8 +8A »SAAAAA: . +AAAAAAA +AAAAAt SAAAAR: $AAAS SAAAAA: AAAZ SAAAA+ s+ » SAAAAZ AAAAA AAAAA. AAAAA; SAAAAZ ott ;AAAAZ sAAAAS, SAAS +AAAAAA. AAAZ » HAAAAAAR. AAAAAAW AARAAZ ZTAAARA = BAART Aa sTAAA; AAAA+ ~AAAARAAR, AAAAAS AAAAA AAAA 7 y BAAAAt TAAAAZ;::A&, ,AAAAA, y SAAAS AAAAA y AAA ZAARAA; sAAAAAARA sAAAAAt TAARAWZAA... AAH 158 +AAAAAS $+ SAAAAAS yAAAAAZ +AAAAA ZAAAA sAAAAAZ oAAAR «LEAAZS$8$; yAAAAS AAAAA AAAAA; AAAAZ yAAAAAR, tAAAAAZ = AAAAAZ U$. 5 .ZARSAAAS, AAg: sAAAAAAAAt ,AAAAAAA, ZAAAAA SAAAAZ = ,AAAA as SAAAA. SAAARS WZ+AAAAA: ZAAAAA AAAAA sAAA +HAAAA TAARAANZ+ SAAAASS s HAA; sAAAAZ #AA b sRAAAAA HAAAAAAt tAAAAAZ = AAAAAAZ AAAS y&AAAAA: $AA, y SAAAAR AAAAAR sAAAAA HAAAA; +AAAAA; +AAAA: «= ZAAAAASZ SAAAR SAAAA AAAAAR > > = ducing tents. However, where tent patterns are overlain with other patterns, which is frequently the case, then the local nature of the automata models is insufficient (cf. WRIGLEY, 1948). One point worth mentioning about the tent patterns is the apparent role that stochastic processes play in their evolution. In the neural model we have not included such stochastic features—although it would be trivial to do so— because we were primarily interested in the patterns that B. Ermentrout et al., 1986 Page 381 Figure 15 a, wandering stripes on Bankiwia fasciata; b, simulation. could be produced in a deterministic fashion. In a subse- quent study we shall demonstrate the role of stochastic influences on the structure of the patterns. DISCUSSION We have constructed a model for shell patterns based on the hypothesis that the secretion of pigment is stimulated by neural activity. Our model postulates the simplest pos- sible neural interactions: local activation and lateral in- hibition, such as is found in the retina. Despite its sim- plicity the model is able to reproduce a variety of observed shell patterns, such as bands, diagonal stripes, and various divaricate interference patterns that arise from the inter- action of propagating bands. The type of pattern generated by the model depends on the nature of the neural interaction, its range, persistence, WORMGRRMA, sssssssAAAMSSSSSEMAAAA,,,, ,AAAMSSSSSHASS 552s AAMNM ENUM WORRAMAAR, SSAARMSASSSSEHAAAAAANZAAANSESENAA. = SAANMOMORA WRRERAOAR,: SAAMMMSSSSTZZIZTISSSSSSSZZZZSSMHA. 555 scs SAAR RMR AME WHAMBRAAZ: +AMMARAHRHRAAAAAAAAAAAAARNAAAAS.., 355555 SANHMMOHNH HHH HAOM A+ MARSSSZLZZZ+ t+) ptt ttt ZtttZS BHA 2, 355555 AMHESSSN MN WRNHMHAA WHMWAAAS, yy o AARSSSSSSOH TWAAAAAAAL. cer y pyran ye AAy gens : S25 535 AAAASSSSSSAHA SAANRAAA,,,: WHUMAAAS. rss sAAAWHAAA.,: SSAAAAASSTZSS3NWA AWAHAAA:.,,,,SAARHHAHA. TAAAAAMSSZISS&MHAA WAAAAAM , 2:3 sAAAWHRARM, AAAAARSSSZSSLARNAH WHAAAAS ,,,: Peon EReLOeN SSAAAAHMESSSSESZENHHN WAAAAA ., sAAAWHSASSSSSLILLHENA AAAAAS.,: AAAAAA.:: Benenoue ses sues cacauaes : CTANNNNHNHAE SAAWHMUMAHAS: g :SAAAAWZERERHWUHARUMARAHA vAAAAMSSZZZ29SSSSSSSSS230H ESAAAAAWSESSNAAAAAARMMAMAARA VAAAWHRSL+ 444 22277444+44279988 GH TAAAAAAWANE --SAAAS. ANWR SHAHA: HAARAA: AWHRHSHAHA: AAAAAA: WAWHARMHHAS © 5 WAAAAA: GULL ED : Serr aaa HAAAAR: tf reAAAAASSSSSRA..,: WAAAAAss sis: SH8888 WHAAAA+s :23:. WAWHMHRAH: 11), ,AAAAASSSESKAS. . WHHHAAAS: oMUHRARHRH, ,,, AAAAWSSSSHAA...,: HHHGHMAN, ,,, WHRAMARH ,,,,AAAANSSSSHA+..,, , yi 5 WHHNHHAA,,,,..AAAAAA ,,,,AAANSSSHAA..., e233 yy 9 WHAWRWAAsS3:, AAAAA. WAHRHMAAS:,,, AAANT. WAHHRMAASssss,..... ails HHHHRWAAS S++ A nBRS y pAAAWAAAAZ., 3: HHAHSSWRANS 5 +5,,,,3 sHSSAMABEARNMNR, £25555 on . ’ . y . AWZESSSZAAAAZ:,,,,,2AAAAANSSSSNS .ss3secs AWHSSSZSHAAAA+....,AAAANSESEHA ..,sseees AAWHSSSSEHAAAZ.....AAAAMHWHA...,,s353333 WNNEESSSENARA,,,,,,BARAAAAA,.., 2 AWHHSSSZHAAAN,::,,,AAAAAAAA.,,: WHNRESSEENAAA,: rrr sAAAAAAAA. y 22555 AWANHSSESRRAAS : WHSSSSESEHMAS AMANSSESHRAA, y’ , WHRSSSSSHHAR. CANMMBOEA: : HHWSSSHRRAA,. TANKHHAMHZ: ; WUSSLZRMHAS ,s55 cc ARMHWHERA: : HUHSHUAWAA .sssocsSMNSSSSHAArss ee WWRHRWMAA,.,s3cc3 ’ ’ SIDES TTARMESSSSEAAAS See WHNHHAAA, , 22335 SAAWNSSSSSSAAAAES ©: WURAAAAS 2255 SAARWESZZZZSRHAAAAL: WHAAAAA TSAAAAREESSSSSEHHAAAAAL: HUAAAAN, re TAAAARRSSSSSSSLZSSENMARAAG SCs srpsees WHAAAAS..., ,AAAANSSSSSHHHUNSSESSHAAAAASS: 28 AAAAAN ..,, SAAANSSZZZZSSSSSSZZZZZSERAAAA, :: AAAAA ,, :::AAAAANNERUAAAAAAANWNSHRAAAAAAS: AAAA. LL ee na AE OO eo oo ee eo . y . y . y . y . yy js AanHNENG., a TTAAARRWNNS. socc,AAAARRR ae sis, AAAAAMAHA ., « “QAWMMNHA ,: MOREA ., tess 53553 AMUBA - Sc cass s re ey AaMAGHORM .,sscssssssiee 35 1) ANNRESSZELITISSSMAAAR: tt AAAAAAAA . sisialeteteieieye settee eeee ts: ARAAAAAAABRNMEHNAAAAAA: AAAAAAAY., poocacnae sone sr, HASSSSUZZIZZ+4447$ 22000 and threshold for activation. Very short-range interactions and strong nonlinearities produce tentlike patterns char- acteristic of the cone shells, and which resemble the pat- terns generated by the automata models of Lindsay and of Wolfram and his coworkers. Longer range interactions produce interference patterns, such as checks and wan- dering streams seen on Bankivia fasciata and other shells. Page 382 The Veliger, Vol. 28, No. 4 Senn S-sereemmeeeeeeeeseesemeee re Se St , SNO> Or AAR, , AAAAAALZ $, -RAAAAAAAS SAAAAAST+ ~TAAAARAAN sAAAAASS. , tAARAAAA T+AAAAAS. /SAAAL: T+AAAAAAAS SAAAAART: TAAAAAAAR: SAAAAAAN+ ,AAAAAAA, +3AAAAAA »tSSWAAAAAAA, ,,. AAAAAAAAS AAAAAA SAAAAL: a: :8AAA BAARA :+WAAAAAAAA: TRAAAAAAS .ZAAAAAAS: $aa TRAAAANSS: SAAAAAWE+. AAAAAAA WAAAAAAS. : AAAAAAAZ , TARAAAAAY SAAAAAAAt *AAAAAAAR AAAAAS: AAAAANEL: RAARSSEAAN +AAAAAAAA AAAAS ,Et+SHAAAAAAAt = STAAAAAAANSZ+3 > 2AAANS+ SHAAAAAY. nSt, ZAAAAA:. HARAAAZ SAAAAAS: -+2AAAAAL, : , ZAAAAAAA, ;AARAAAAL AAAAAAAZ ;AARAAARA -AAAAAAAK TAAAAAAAA HAAAAAAAS WAAAAAA - +AAAAAAAA -t+AAAAAAAA = .t,. .AAAAAAAAZ::. SAAAAA TAAAASt:, LAAART yrpet+ .SAAAA, HAA :BAAAAAAS . +8ARAAAAAL ZAAAAAAA -RAAAAAAL, +A SAAAAAAAR: LAAAAAAAY +ARAAAAR SAAAAAAAN AAAAAAAR . +RAAAAAAT AAAAAAAA yAAAAAAAA, AAANEZ: yAAAAREZ:: AAASZ+Z8AN, S+ZRAAAAA AAAAS «;KEAAAAAAAS ©. ZAAAAAARZ+35::ZAAAAAAR, TAAAAAAAt. AN+ TAAAAANZ. ZAAAAAA, ~AAAAARS SSARAAAAS. ; - +AAAAAAA, +AAAAAAAS -ARAAAAAT HAAARARA »AAAAARRZ: SAAAAAAAA BAAAAAAAA +3AAAAA - FEAAAAAAA, ,TtTRAAAAAAAS 3444+, ZHHAAAAAAZES. TAAAA SAAAARZ: +AAAAS, )AULUSZ, . SAAAAM SAA . +AARARAAY TRAAAAAAA. SAAAAAAR WAAAAAAZ, z SAAAAAARZ: AAAAAAAA: SAAAAAAR tSAAAAARA, AAAAAAA: ;SAAAAAAA AAAAAAAA: SAAAAAAAZ. AAARS+ AAAANS+,. TAASSIES$AAZ.. T+ SHAARAt AAAS .ELHAAAAAAA, , SAAAAAAZ.....28AAAAAAR, SAAAAAAL: As, SAAAAART. HAAAAAA. -HAAAAAA , SBAAAAAAT .SRAAAAAAA ZAAARAAA, AAAAAAAS, TAAAAAAA HAAAAHEZ: AAAAAAAAA sAAAAAAAA: »SRAAAA , ;AAAAAAAS -tZRHAAAAAANES 3$83%: LEHAAAAARTS, TAAAA ,+SHAAAA+ Figure 16 Checkerboard patterns. We have mapped out many, but not all, of the possible patterns that arise from the neural hypothesis. The model can be elaborated in several directions. For example, what is the effect of postulating a more complex neural struc- ture (such as long-range activation)? Many shells secrete several kinds of pigments; including more than one pig- ment into the neural model would increase enormously the possible patterns it could generate, including the char- acteristic of stripes passing through one another—a com- mon phenomenon that the simple neural model presented here cannot reproduce. It is clear from many studies (e.g., WRIGLEY, 1948) that the mantle is not a homogeneous tissue as we have assumed here. By adding to the model spatial gradients and periodic variations in the parameters (e.g., refractoriness or density of innervation) a far greater variety of patterns can be produced than from the homo- geneous mantle we have assumed here. We shall present simulations of more complex mantle structures elsewhere. In addition to spatial variations, a variety of transition patterns can be produced if parameter values evolve slow- ly as the simulation proceeds. These are distinct from the discontinuities and V-patterns that may involve a sudden, global perturbation of a system parameter. In particular, shell size is an important determinant of pattern. Small, or young animals will typically exhibit less complex de- signs, because fewer stripes will “‘fit” into a smaller do- main. Moreover, as shell size (1.e., domain size in the model) increases with growth, stripes widen until a threshold is reached, whereupon another stripe interca- lates, a phenomenon commonly observed, especially in Nerita turrita. Such sudden shifts in behavior triggered by smoothly varying a parameter are typical of models with strong nonlinear terms (MAy & OSTER, 1976; GUCKENHEIMER et al., 1976). The cowries have a mantle that imprints a pattern over a large expanse of shell, rather than just at the growing edge. To model this, one must employ a two-dimensional version of the neural model. Two-dimensional automata models with very local interactions can produce patterns that bear a striking resemblance to those found on the map cowrie (N. Packard & S. Wolfram, personal com- munication), and preliminary analysis of the neural model indicates that the eye-spot pattern found on many cowries can be easily obtained. The neural model also touches on the problem of shell construction, for as WRIGLEY (1948) and others have pointed out, there is a correlation between the color pat- terns and the geometrical features of the shell (e.g., pig- ments may concentrate in the grooves between ridges, and spines tend to be colorless). This is hardly surprising, because the same mantle that deposits the color is busy building the shell. However, this correlation between pat- tern and form suggests that the neural model might be extended to investigate the diversity of shell shapes and their mode of construction. If the neural hypothesis is correct, the shell is a hard- copy record of the neural activity in the mantle. The fossil record for these creatures is as complete as for any known lineage. What can such an electroencephalogram tell us about the evolution and ecology of mollusks? We shall not speculate here, but the model suggests an explanation for the diversity of patterns found on the same species in different environments, and the similarity of different species in the same environment. Moreover, the enormous diversity of pattern within certain species may reflect the fact that the patterns in those species are not visible during the animal’s life. Being invisible to selection generally leads to increased genotypic variance, and so we should expect the color patterns in such species to be highly polymor- phic. The usefulness of any model stems not only from its specific predictions and its ability to unify disparate ex- perimental observations, but also from its fertility in sug- gesting further experiments. If the neural hypothesis is correct experiments that intervene with mantle neural ac- tivity, without disrupting shell construction, need to be devised. Perhaps the topical application of neuroactive substances such as xylocaine, lysergic acid, or various kinds of neurotransmitters can provide information. Probably electrophysiological measurements will interrupt mantle activity, but perhaps the neural connections between pig- ment cells can be explicated in sufficient detail to deter- mine the range of neural interactions characteristic of each pattern type. It is a rich field for neurobiology and anat- omy which will have a direct impact on larger issues of evolution and adaptation. B. Ermentrout e¢ al., 1986 Page 383 Figure 17 a, tent patterns characteristic of olive snails (e.g., tent olive or royal purple olive (Oliva porphyria); b, tent patterns character- istic of the textile or courtly cones—these patterns differ from (a) by slightly longer range neural interactions. Finally, we should mention the issue of the uniqueness of the model. It would be gratifying if we could claim that our model can reproduce the observed patterns better than all competing models; however, this is not the case. Using a model based on diffusion and reaction of chemical mor- phogens, H. Meinhardt has produced simulations that are equally as convincing in reproducing the shell patterns as the neural model. The reason is clear: one can model the phenomenon of local activation and lateral inhibition characteristic of neural nets in a variety of ways. Any number of diffusion-reaction mechanisms can produce this effect by a slowly diffusing autocatalytic reaction that is quenched by a fast diffusing inhibitor molecule (MEIN- HARDT, 1982). Even the mechanical models that OSTER et al. (1985) have employed to model the regular patterns of microvilli on cells can be viewed as a mechanical im- plementation of this neural-like property. Therefore, we are left with the disappointing conclusion that it may be quite difficult to infer mechanism from pattern alone, be- cause several quite distinct cellular mechanisms can pro- duce identical patterns. Thus the issue of whether the patterns on mollusk shells arise from neural activity as we have suggested here will be settled only by experi- ments. Theory can provide only a shopping list of possible mechanisms. ACKNOWLEDGMENTS GBE was supported by NSF Grant # MCS 8300888 and the Sloan Foundation; JC by NSF Grant # PCM-20923; AAAAAAAAARARAAAAAAAA. AAAAAAARASARAAAAAAAARA SAAAAAAAAA = =ZAAAAAAAAAAAAA AAAAAAAAAAAAARAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAA AAAAAAA AAAAAAAA AAARAAAAAA AAAAAARAAA AAAAAAA AARAAAA AAAAAAAAAAAAAAAAAARAAAAA AAAAAAAAA AAAAAAA AARARAAAAA AAAAAAAAAA AAAAAAAAAAA AAAAAAA AAAAAAARAAAZ AAAAAAAAAAAAAA s AARAAAAAAAAAARA AAAAAAAAA AAAAARAAAAAAAAAAAARAAARAAAAAAAARAAAAAAAAAAAAAAAAAA AAARAAAAARA AAAAAAA AAAAAAARARAAAAAAAAAAAA ARAAAAA AAARAAA: AAAAAAA AAAAAAA AAAAAAA AAAAAAA AAAAAAA AAAAAA AAAAAA AAAAAAAAA AARAAAAA SAAAAAA AAAAAAAAAAA AAAAAAARAAAA ARAAAAAA AAAAAAAAAAAAA AAAAAAAAARARAAARAAAAARAAAAA AAAAAAAAAAAAAAA AAAAAAAA AAAAAA AAAARAAAAAAAAAAAA AAARAAAAAA ARAAAA AAAAAAAAAAARAAAAAAA AAAAAAARAAA AAAAA AAARARAAAA = AAAAAAAAAA AAAAAAARAAAAAA AAAARA AAAAARAAAAAAAAARAAAAAAA AAAAARAARAAAARAAA AAAAAAA AAAAAAAAAA AAAAAAARAARAAAAAAAAA: AAAAAAAAA AAAAAAAAA AAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAA AAAAAAAAAAA AAAAAAAAR. AAAAAAAAAA ARAAAAAAAAAAAAAAAAAAAAAAAAAAA AARAAAAAA = AAAAAAAAA AAAAAAA yARAAAA AAARAARAAAAAAAAAAAAA AAAAAAAA AAAAAA AAAAAAAAAASAAAAAARAAAA AAAAARAAA AAAAA AAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAA AAAAAA AAAAAAAAAAA SAAAAAAAARA = =©ARAAAAAAAAAAAAAA sAAAAAAA AARAAARAAAAAAAARARAAAAAAARAAAARARAAARAARAAAAAAA AAAAAAARA AAAAAAAAA AAAAAAAAAA ARAAAAAAARA AAARAAAAAA AAAAAAARAAAAAAAAAAAAARAAAAAA AAAARAAAAA AAAAAA +AARAAA AAAAAAAAAA. AAAAAAA AAAAAA AAAAAARAAARA AAAAAAAA AARAA AAAAARAAAAAAAA AAAAAAAAAAA AAAAAA AAAAAAAAAAAAAAAA AAAAAAAAAAAAAAA AARAAAAA AARAARAAAAAAAAAAAAA AAAAAARAAAAAAAAAAAA AAAARAAAAA AAAAAAARAAAAARAARAAA AAAAAAARA AAAAAARA AAAAAARAAAA AAAAAAAAA+ AAAAAARAAAA = AAAAAAAAAAAAAAAARAAAAAAAARAAAAAAAAAAAAAA AAAAAARAAAAZARAAAAAAAAAAAAAAAAAA AAAAARA AAAAAAAAAAAA. HAAAAAAAAAAA AAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAA AARAAAAAZ AAAAARAAA AAAAAA AAAAAAAAA AAAAAAAAARA AAAAAAAA ARARAAAAA AAAAAAAAAAAA AAAAAAAAAA ARARAAAAA AAAAAAAAAAAAAAA AAAAAAAARAAA if AAAAAAA AAAAAAAAAAARAAAAAAA sAAAAAAAAAAAAA h. ~AAAAAAAAAA AAAAAAAAAA AAAAAAAAA AAAAAAAAAAAAAAA AAAAARAAAAAAAAAAAAAAAAAAAAAARAAAAAARAAAAAAAA AAAAAAAARARARAAAA AAARAAA SARAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAA. AAAAAA AAAAAAAAA and GO by NSF Grant # MCS 8110557. Conversations with James Keener and Arthur Winfree were crucial to the completion of this work. Thanks also to E. R. Lewis and D. Lindberg for valuable comments and criticisms. The hospitality of James Murray and the Institute for Mathematical Biology, and Gerald Edelman and the Neurosciences Institute are gratefully acknowledged. The computer graphics were performed by Greg Kovacs. LITERATURE CITED BERNE, R. & M. Levy. 1983. Physiology. Pp. 116-118. Mos- by: St. Louis. CAMPBELL, J. 1982. Proposal submitted to the National Sci- ence Foundation (Grant No. PCM-20923). ComrorT, A. 1951. Pigmentation of molluscan shells. Biol. Rev. 26:285. Cowe, R. J. 1971. Simulation of seashell pigment patterns using an interactive graphics system. Computer Bull. 15: 290. EMBERTON, L. 1963. Relationships between pigmentation of Page 384 shell and of mantle in the snails Cepaea nemoralis (L.) and Cepaea hortensis (Mull). Proc. Zool. Soc. Lond. 140:273. ERMENTROUT, B. & J. Cowan. 1979. A mathematical theory of visual hallucination patterns. Biol. Cybernetics 34:137- 150. GUCKENHEIMER, J., G. OSTER & A. IPAKTCHI. 1976. Density dependent population models. J. Math. Biol. 4:101-147. HERMAN, G. T. & W. H. Liv. 1973. The daughter of Celia, the French flag, and the firing squad. Simulation 21:33. HERMAN, G. T. 1975. Developmental systems and language. North-Holland: Amsterdam. Kapur, S. & M. Gipson. 1967. A histological study of the development of the mantle-edged gland and shell in the freshwater gastropod Helisoma durvi endiscus. Can. J. Zool. 45:1169. KNIPRATH, E. 1977. Ontogeny of the shell field in Lymnaea stagnalis. Roux. Arch. Dev. Biol. 181:11. Linpsay, D. 1982a. Simulating molluscan shell pigment lines and states: implications for pattern diversity. Veliger 24: 297-299. Linpsay, D. 1982b. A new programmatic basis for shell pig- ment patterns in the bivalve mollusc Lionconcha castrensis (L.). Differentiation 21:32-36. May, R. & G. OsTER. 1976. Bifurcation and dynamic com- plexity in simple ecological models. Amer. Natur. 110:573- 599. MEINHARDT, H. 1984. A model for positional signalling, the threefold subdivision of segments and the pigmentation pat- terns of molluscs. J. Embryol. Exp. Morphol. 83(Suppl.): 289-311. NerFF, J. 1972. Ultrastructure of the outer epithelium of the mantle of the clam Mercenaria mercenaria in relation to cal- cification of the shell. Tissue Cell 4:591. OsTER, G., J. MuRRAY & G. ODELL. 1985. The formation of microvilli. Jn: G. Edelman (ed.), Molecular Determinants of Animal Form. UCLA Sympos. Molec. Cell. Biol. (in press). TIMMERMANS, L. 1969. Studies on shell formation in molluscs. Neth. J. Zool. 19:417. WADDINGTON, C. & R. Cowe. 1969. Computer simulation of a molluscan pigmentation pattern. J. Theo. Biol. 25:219- 225. WAUSHER, J. 1972. Considerations on phase-change and dec- orations in snail shells. Hereditas 71:75-94. WILBER, K. 1972. Shell formation in mollusks. Jn: M. Florkin & B. Scheer (eds.), Chemical zoology. Vol. 7. Academic Press: New York, p. 113. WOLFRAM, S. 1984. Cellular automata as models of complex- ity. Nature 311:419-424. WRIGLEY, A. 1948. The color patterns and sculpture of mol- luscan shells. Proc. Malacol. Soc. Lond. 27:206. APPENDICES A. The Model Equations In this Appendix we give the complete mathematical expression for the model equations given in the text, as well as the functional forms employed in the numerical simulations. The model consists of the three difference-integral equations A,4(x) = S[P,(x)] a7 Rk, [1] The Veliger, Vol. 28, No. 4 Rii(x) = yPx) + OR, (x) [2] P(x) = H(A — A*) [3] where 0 < vy < 1 is the rate R increases and 0 < 6 < 1 is its degradation rate. We can further simplify the model by assuming that the pigment secretion, P, is simply proportional to the activity, A, and let the function S take care of the threshold for secretion. This does not affect the patterns signifi- cantly, and is somewhat easier to treat numerically and theoretically. Thus the equations we shall deal with are Pri(x) a S[P(x)] in R, [4] Risi(x) = yP(x) + 6R,(x) [5] The neural stimulation function, S[P,(x)] in equation [4] is composed of excitatory and inhibitory effects. Note that the pigment secretion on day ¢ + 1 can depend only on the excitation during day ¢t + 1; however, according to the assumptions of the model, each day’s pattern of exci- tation is stimulated by “tasting” the previous day’s pig- ment pattern. We can safely assume that the time con- stants for neural interactions are much shorter than those of shell growth, so that we need deal only with the daily average, or steady state firing rate of the neurons in the mantle. Therefore, we define the following functionals: Excitation: E,44(%) = I W(x =x) Cc) idecg [6] Inhibition: Li41(x) = i Wx! — x)P,(x') dx! [7] Here the kernels W(x’ — x) and W,(x' — x) weight the effect of neural contacts between cells located at po- sition x’ and a cell at x; they effectively define the con- nectivity of the mantle neuron population. In general, the inhibitory kernel, W,(x’ — x) is broader than the exci- tatory kernel, W,(x' — x); 7.e., activation has a shorter range than inhibition. Q is the domain of the mantle; for most shells this is a finite interval, but may be circular in the case of mollusks such as limpets and planar in cowries. The particular connectivity functions we have em- ployed in our simulations are: WO) for? |x |"= chs ieee W, = q,[2* — (1 — cos(xx/a,)?] [8] if) Jay I for |x| = g,, where q; is chosen so that fl VAC) Cs SOR yf 18s I [9] B. Ermentrout et al., 1986 Page 385 Table 1 Neural influence function parameters a, = amplitude of the excitatory influence function. a, = amplitude of the inhibitory influence function. o, = range of the excitatory influence function. o, = range of the inhibitory influence function. Pe = sharpness of the excitatory influence cutoff, or the “flat- ness” of the influence function. p, = sharpness of the inhibitory influence cutoff. Firing threshold functions vy, = steepness of the excitatory cutoff (nonlinearity). y, = steepness of the inhibitory cutoff. 6, = location of the excitatory threshold; 7.e., the midpoint of the sigmoidal curve (threshold). 6, = location of the inhibitory threshold. Refractory parameters = production rate of refractory substance. 6 = decay rate of refractory substance. The shape of the connectivity functions is controlled by p: for p very small the W, are sharply peaked, for p large, the W, become nearly rectangular. In our simulations p is in the range of 4-8. The range for lateral inhibition is made greater than the excitation by choosing o, > o,, and since the local excitation strength is generally greater than the inhibition, we choose a; > a. The responses of the secretory cells to neural stimula- tion are assumed to be sigmoidal functions of their inputs: S[PAx)] = SA Evi)) — Sila) [10] For simulation purposes, we have employed the follow- ing function for both $; and S, S,(u) = j=E,1 [11] 1 eae)” The parameter v, controls the sharpness of the nonlin- earity, and 6; the location of the threshold. Thus the raw parameter list consists of the 12 quan- tities: [Qg, Qy, OE, OT, PE, Pi VE, Vj, Og, 6;, Y 6] This list can be reduced to nine because some parameters enter only as products, and some may be rescaled. Table 1 summarizes the model parameters. B. Analysis and Simulation of the Model A linear stability analysis of the model equations gives some idea of the patterns the model will generate. There- fore, we proceed as follows. The pair of equations [4, 5] are equivalent to the single second order equation long = Ween TP Olena] = Osy =e 6S[P,] [12] a c C B A b A (c) E 5 Figure Al a, the unit circle and the stability triangle on the coefficient (a, b) plane; b, dispersion relation \(k) for spatial instability; c, trajectories for each type of bifurcation. where we have suppressed the dependence on x for no- tational simplicity. Let P, be a homogeneous equilibrium, z.e., Je Se aian Ors = (brs a8 Cea) [13] or L=8 OA Nera [14] If we shift the sigmoid S so that $,(0) = S(O) = 0, then we can linearize about P, = O to obtain the linear differ- ence equation: Jepigy an Losey a Olen — ape ae WEI e =O) |S where L,(.) is the linear (convolution) operator L,[ul(x) = S’2(P,) { W(x’ — x)u(x’) dx’ — S"(P,) i Wi(x' — x)u(x’) dx’, [16] where S',(P.) are derivatives of S'. Page 386 (a) (b) Figure A2 a, the shapes of S(x) and L(k); b, the dispersion relation. On a periodic domain of length L (e.g., the limpet), the eigenfunctions for L, are exp(2minx/L), n =1,2,...30n a finite linear domain these are approximate eigenfunc- tions, since the domain size, L, is much greater than the range of the connectivity functions W. The characteristic equation for the spatially homoge- neous system is obtained by substituting P,(x) * Nexp(2mikx/L) into the linearized equation: dh? + (L*(k) + b)A — [y + 6L*(R)] =)? 4+ a(k)X + (Rk) = 0 [17] Here L*(k) = S',(P,) Welk) — S'(P,)Wk) [18] where the W, are (close to) the Fourier cosine transforms of the W;: ] W,(k) ® | cos(2mts/£) W(x) dx [19] The spatially homogeneous solution is stable if and only if the roots of the characteristic equation lie within the unit circle on the complex plane: |A| < 1 for every k = 0, 1, 2,.... This condition can be plotted on the coefficient plane (a, b), as shown in Figure Ala, where stability requires that a and b lie within the shaded triangle. Spatial instability requires that (i) the homogeneous solution be stable: |A|(k = 0) < 1, and (ii) there exists a The Veliger, Vol. 28, No. 4 finite range of unstable modes: |A|(k) > 1 for 0 (1 — y)/6 (path a in Figure Alc). This is a so-called “equilibrium” bifurcation because in the spatially homo- geneous case (k = 0) such a bifurcation creates a new equilibrium point (cf, May & OSTER, 1976; GUCKENHEI- MER et al., 1976). When k > 0 this creates a stationary spatial pattern of regularly spaced stripes as shown in Figure A3a. (b) Bifurcation through —1 will occur if L*(k) < —(1 + y/(1 + 6)) (path b in Figure Alc). From Figure A2b we see that this can occur only at k =0, so that homogeneous instability results. (This can only happen in this model for the kernel shown providing 6, > 6,, since we have assumed that a; > a;.) The pattern resulting from this bifurcation consists of fine horizontal stripes, as shown in Figure A2b. (c) Bifurcation through 4 =e” (6 # O, 7m) occurs if L*(k) >1+~7/(1 — 6) (path c in Figure Alc). This generates periodic spatio-temporal patterns, as shown in Figure A3c (e.g., stripes and checks). Note that (1 — y)/6 <1+-~¥/(1 — 6) if and only if 6>1— Vy. Thus +1 bifurcations occur first when 6 < 1 — \/y7; otherwise the bifurcation is via a complex ei- genvalue. When y =0, so that the refractory substance cannot build up, the model can take a particularly simple form. If we make p large, and o,, o, equal, and v large, then the model is approximated by the rule: P.4,(%) = 1 if Og < i Pix + x’) dx’ < 6, = 0 otherwise [20] This is essentially a continuous space analog of Wol- fram’s Class-3 cellular automata rule (WOLFRAM, 1984). This type of rule leads to “chaos” and the “‘tent”’ patterns. The linear analysis was employed to guide the numer- B. Ermentrout et al., 1986 AQAA MS AG A QQ AQAA M ( @) tH ) } UU, ) (b) Page 387 4 (c) Figure A3 a, spatial pattern arising from +1 bifurcation; b, spatial pattern arising from —1 bifurcation; c, spatial pattern arising from complex bifurcation. ical simulations. The model equations were converted to a single second order difference equation and the integrals approximated by flere ‘i W(x! — x)P(x') dx’ & i D>, Mesyl (2M) j=0 Generally, N was taken to be 64, although when un- usual patterns were encountered N was set to 128 or 256 to check that they were not numerical artifacts. Initial conditions were random, or small regions of the domain were excited. Typically, long transients generated com- plicated patterns which gradually simplified as the tran- sients damped out. Table 2 Fig. 6; 0, arp Qa, Or 0; OY 6 op lla 0.0 0 6 8 0.1 0.2 0.0 (OKO) il 1a 5.5) O22 5 O22 O71 C5 OM OW fo 1G 455 0) 15 0.5 0.1 0.12 0.05 0.6 8 145 1 100 5.0 4.0 0.05 0.2 0.8 0.4 2 1S be 425 OS2aaS 0.5 0.1 0.15 0.1 0.8 8 16 0) 0) 8.8 6.6 0.1 0.2 0.4 0.6 1 Nia 3 4 8.0 4.0 0.1 0.2 0 0 8 8 He Bed) 5.5 10 4 0.1 OA Os OF C. Alternative Formulations The behavior of the model equations can be illuminated by examining their continuous time limit. If we subtract P and R from both sides of [4] and [5], respectively, we obtain lene = Te SUA > hey de [22] Jig = dit, Oe ING ae OO, [23] By an appropriate choice of time scale, t, we can divide Figure A4 Phase plane for the differential equations [A24] and [A25]. Page 388 both sides by t and replace the differences by derivatives to obtain: pp lie = Ie [24] aR sO eae [25] Now let us examine the phase plane of this system at a fixed x = x,. The operator S is sigmoidal in P, and so the right-hand side is a cubic-shaped curve (a sigmoid minus a linear term). The (P, R) phase plane is shown in Figure A4; it is qualitatively similar to the FitzHugh- Nagumo model for excitable media. That is, each volume element is excitable, and the volume elements are spatially coupled by the activation-inhibition operator W. If only nearest neighbor cells interact inhibitorily, then W can be expanded in a Taylor series about x, and only lowest order terms retained. Then a familiar diffusion- reaction model emerges: AP &P — = D— + F(P 2 an Ae (Ee) [26] a “ = yP + (6—1)R [27] where D is a diffusion coefficient that can be expressed in terms of the expansion coefficients of the integrand. If activation-inhibition is to be retained in the model, then fourth order terms must be retained (odd order terms dropping out by symmetry), and we obtain the biharmonic diffusion-reaction system: oP oP 0 | 0°7P on a oe 24 | TEE Rates OR ay ea 1)R [29] Here the negative sign in D, corresponds to short-range activation, and the negative sign in D, corresponds to long- range inhibition. A model quite similar to this was arrived at by J. Keen- er (personal communication) by defining a net neural fir- ing rate, f(x, t) according to the equation of ) Hise La ap— y+ | we — x)f(x') dx [30] where W(x — x’) is the activation-inhibition kernel shown in Figure 9. Coupling to the secretion is obtained by de- fining the secretion rate to be a bistable function: oP Fin ae) [31] The Veliger, Vol. 28, No. 4 where F(P, f) is an S-shaped curve whose intercept is regulated by f/. By expanding the convolution to fourth order, this model can also be reduced to a biharmonic diffusion-reaction model: On ef ae wf ——— —., S ast. + — Fm mervremcrorriren 4) [32 aP Ne F(P, f) [33] Somewhat different approaches were employed by WADDINGTON & COwE (1969), MEINHARDT (1984), and WOLFRAM (1984). They modeled the shell patterns by an automata wherein the activation-inhibition effect was rep- resented by nearest neighbor interactions via diffusion. Meinhardt’s model employed two substances with differ- ent diffusion constants (D, > D,). He obtained some of the same patterns we obtain here by assuming that each cell of the automata could periodically fire and become refractory for a while. In a more recent simulation, Mein- hardt and Klingler (to appear) included longer range in- teractions by allowing morphogens to diffuse beyond near- est neighbors. These simulations resemble ours and it appears that most patterns can be created by either mech- anism. However, it is not clear how the diffusion-reaction model handles the problem of pattern alignment between episodes of shell secretion, whereas this is intrinsic to the neural model. Wolfram’s simulations mimic to a remark- able extent the “tent” patterns observed on many cone shells. However, his rules were rather arbitrary, and have no obvious physiological interpretation. The neural net model, in the limit of short-range interactions and sharp threshold functions, reduces to the automata model, and can also reproduce the tent patterns. All of these models have a similar structure: a locally excitable activator-inhibitor system that is coupled spa- tially to nearby points. In order to obtain spatial patterns, the activation-inhibition is essential. Moreover, it appears that many of the patterns depend on the long-range (:.e., beyond nearest neighbor) interactions characteristic of neural nets. Also, the episodic nature of the secretion pro- cess dictated our choice of a discretized model in time; this feature also appears essential to the formation of certain pattern types. In a subsequent publication we shall inves- tigate a broader class of neural models, including kernels with long-range activation and two-dimensional mantles, such as are found in cowries. The Veliger 28(4):389-393 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 Predation-Induced Changes in Growth Form in a Nudibranch-Hydroid Association by GARY GAULIN, LAUREN DILL, JULIE BEAULIEU, anp LARRY G. HARRIS Department of Zoology, University of New Hampshire, Durham, New Hampshire 03824, U.S.A. Abstract. The estuarine hydroid Cordylophora lacustris was cultured under controlled laboratory conditions and several growth parameters were measured, including stolon growth rate, stolon budding rate, interstalk distance, and polyp budding rate. Young colonies of C. lacustris were subjected to experimental removal of polyps to determine whether physical and(or) chemical cues from predation by the nudibranch Tenellia fuscata induced any changes in growth form. Nudibranch predation and polyp removal with chemical stimuli from both direct and indirect exposure to nudibranch mucus caused decreased stolon growth rate and increased stolon budding. The result of these changes in growth should be a hydroid colony of greater density than without predation. Polyp removal without a chemical stimulus inhibited stolon budding, which would cause a more dispersed colony form. The results suggest that the hydroid responds differently to physical damage alone than to physical damage combined with the chemical stimulus of the mucus of its predator. INTRODUCTION NUDIBRANCH PREDATION on cnidarians has been docu- mented numerous times and the feeding behavior of many nudibranchs described (see TODD, 1981). Less well known are defensive mechanisms utilized by their cnidarian prey. The defensive behaviors of a number of anemone species have been reported (ROSIN, 1969; Harris, 1973; EDMUNDS et al., 1976; Topp, 1981), but little is known about pos- sible responses by hydroids to nudibranch predation. Cer- tainly nematocysts are released and polyps may contract, but are there any more subtle responses such as changes in growth form? Growth forms for a number of hydroid species have been described (see BRAVERMAN, 1974), but no studies to date have discussed the effect of predation on growth rate and form in a hydroid. Herbivore attacks have been shown to induce changes in growth in plants (BOSCHER, 1979). HARVELL (1984) reported that nudibranch predation stimulated spine for- mation in the bryozoan Membranipora membranacea (Lin- naeus, 1767), but no changes in growth form were ob- served. An arborescent hydroid might be expected to respond in one of three ways to a loss of tissues from nudibranch predation: (1) to grow slower with no change in colony form; (2) to grow more densely, paralleling the response of trees to pruning; and (3) to spread out the colony by elongation of stolons to “run away” from the predator. The purpose of the present study is to determine wheth- er nudibranch predation on the hydroid Cordylophora la- custris (Allman, 1844) would induce changes in growth form. Cordylophora lacustris is common in the euryhaline portions of rivers emptying into the Great Bay Estuary, and it is easily cultured under controlled laboratory con- ditions. The dendronotacean nudibranch Tenellia fuscata (Gould, 1870) is a natural predator of C. lacustris in the Great Bay Estuary and adapts well to laboratory condi- tions (HARRIS et al., 1980). This study describes the re- sults of short-term experiments testing for the effect of nudibranch predation on hydroid growth. MATERIALS anpD METHODS The animals used in this study were collected from the Great Bay Estuary in New Hampshire and maintained in laboratory cultures in the Zoology Department of the University of New Hampshire. Colonies of the hydroid Cordylophora lacustris were collected from floats and pil- ings in the Lamprey River at Newmarket, New Hamp- shire, and both hydroids and the nudibranch Tenellia fus- cata were obtained from floats in the Salmon Falls River at Stratham, New Hampshire. Page 390 The Veliger, Vol. 28, No. 4 Table 1 A summary of results of growth rate parameters measured under a series of control and experimental treatments testing for the effects of physical and chemical stimuli associated with polyp removal on growth rate in Cordylophora lacustris. The parameters measured were stolon growth (mm per day), stolon budding (buds per stolon per day), polyp budding (polyps per stolon per day), and interstalk distance (mm). The experimental treatments of polyp removal were (1) direct predation by TJenellia fuscata for 6 h, (2) removal of three polyps by forceps (no chemical stimulus), (3) removal of three polyps by forceps and rubbing a nudibranch over the wounds (direct chemical stimulus), and (4) removal of three polyps by forceps and presence of a nudibranch nearby but not in contact with the hydroid (indirect chemical stimulus). * Each experimental treatment differed from the control value to at least P = 0.05 level using ANOVA. Direct Forceps Growth parameters Control feeding only Stolon growth* (mm-day ') 1.06 0.61 0.88 Stolon budding* (buds: stolon”!-day~') 0.1 0.37 0.00 Polyp budding (polyp: stolon!-day~') 0.32 0.23 0.691 Interstalk distance (mm) 3.25 3.04 3.0 The hydroids were maintained in aerated aquaria at a constant temperature of 23°C and salinity of 20%c. Salinity was checked daily to eliminate the possibility that any changes in growth form were induced by abiotic factors. Experimental colonies were established by attaching frag- ments of colony containing stolon and one or two polyps to glass slides with nylon fishing line. The slides were suspended in an aquarium to which freshly hatched brine shrimp nauplii were added daily in order to maintain a constant food supply. Each slide was examined daily. Once a colony had attached and begun growing, it was utilized for one of several experiments. Growth parameters in each colony were determined by placing the glass slide onto a plastic coated sheet of graph paper with a 1-mm? grid in a water-filled dish. The col- ony was then observed under a dissecting microscope and the colony was mapped on another piece of graph paper. All measurements were then made from the daily se- quence of diagrams for each colony. Four growth param- eters were consistently measured: (1) stolon growth rate; (2) stolon budding rate; (3) interstalk distance per stolon; and (4) polyp budding rate per stolon. Tenelha fuscata was maintained in aquaria at the same temperature and salinity as the hydroids and fed colonies of Cordylophora lacustris. There were periods when a sec- ond hydroid, Bougainvillea sp., was used as food due to decreased availability of C. lacustris. In order to determine the feeding rate of adult nudi- branchs, specimens of Tenellia fuscata were allowed to feed on hydroid colonies of known size for 24 h. The number of polyps consumed was counted and an hourly polyp consumption rate was calculated. Nudibranchs were also allowed to feed on colonies for 6 h to verify the feeding rate. All experiments for growth responses to nudibranch predation in Cordylophora lacustris were based on this 6-h time period. Forceps _ Forceps & mucus & mucus (direct) (indirect) df F-ratio Significance 0.42 0.07 4/103 5.89 P < 0.001 0.46 0.23 4/45 5.19 P < 0.005 0.21 0.44 4/56 2.36 NS 2.61 2.76 4/42 0.43 NS The impact of polyp removal on hydroid growth was tested using four variations. (1) A nudibranch was al- lowed to feed on a colony for 6 h and then removed. (2) Polyps were removed by forceps to mimic predation with- out a chemical stimulus. (3) Polyps were removed by for- ceps and then a nudibranch was rubbed over the wounds to mimic predation including a direct chemical stimulus. (4) Polyps were removed by forceps and the colonies were maintained in an aquarium where a nudibranch was held in a mesh container to provide only an indirect chemical stimulus. Regeneration rates of removed polyps were also monitored using the same four treatments. All experiments involved three or more replicates per treatment and each experiment was run at least two times. Colonies were followed for approximately a week after the treatments. Analysis of variance was used to test for the statistical significance of various results in each ex- periment. RESULTS Growth rate measurements, made during the early phases of colony formation by newly attached fragments of col- onies, are given in Table 1. Colony formation during the early stages of establishment on glass slides involves ex- tensive stolon growth and limited upward growth of stalks, so that most stalks contain only one or two polyps. This pattern appears to change only after stolons begin to reg- ularly overlap each other and most of the slide is colo- nized; then growth shifts to production of branching stalks containing many polyps. Tenellia fuscata is an active and fast growing nudi- branch (HARRIS e¢ al., 1980). Individual nudibranchs were seen feeding in two ways during observations to determine feeding rates. The most common method was for Tenellia to crawl up a hydroid stalk to the base of the polyp, grasp G. Gaulin et al., 1986 it in the jaws, and rasp away tissue until the hydranth was consumed. Nudibranchs were also observed to make a hole in the perisarc of a stalk and to consume tissue and fluids by a combination of rasping and pumping move- ments of the radula and buccal mass. Direct consumption of polyps was the primary feeding method observed when nudibranchs were on growing hydroid colonies with an abundance of polyps. The mean feeding rate determined for 6- and 24-h tests was 0.48 polyp per hour or 11.52 polyps per day. In all experiments in which forceps were used to mimic predation, three polyps were removed to represent approximately 6 h of feeding by a nudibranch. All hydroid colonies used had been recently established on slides. Three polyps constituted a loss of no more than 10% of the polyps in a colony. The most obvious change in growth following polyp removal was a decrease in stolon growth rate and a change in stolon budding rate (Table 1). Stolon growth rate might be expected to decrease along a stolon where polyps and, therefore, feeding capability were reduced. The increase in stolon budding may have accounted for the slowdown in stolon growth rate and represented a shift in growth form. The decrease in stolon growth was least where no chemical stimulus from a nudibranch was present; stolon budding actually ceased in the portions of those colonies that were monitored. Stolon growth was least in colonies that were exposed to only indirect chemical cues, but the stolon budding rate did not change as dramatically as in the cases where mucus directly contacted the colonies. A major difference between treatments, though, was that in cases of direct mucus contact, the contact lasted no more than 6 h (direct predation) and often only a few minutes (forceps and mucus), while the colonies used in the indi- rect chemical stimulus tests remained in culture with a nudibranch for several days and were, therefore, presum- ably exposed to low levels of chemical stimulus for the duration of the experiment. The different treatments caused decreased or increased rates of polyp budding. In the two cases where nudibranch mucus directly contacted the damaged portion of the col- onies, polyp production decreased. In contrast, polyp pro- duction increased in the two tests where mucus was not present or was not in direct contact with the hydroid. Slight decreases in interstalk distance occurred along sto- lons where polyps were removed, but such changes did not correlate with decreases in stolon growth rates. The net result of the observed responses to polyp re- moval was predicted to be a denser colony form with more stolon branching and decreased interstalk distance. The one exception may be where no chemical stimulus was present, because a continuation of the observed sharp de- crease in stolon budding would produce a very diffuse colony form. Cordylophora lacustris showed a consistent pattern of polyp regeneration following removal of several polyps (Table 2). Polyps regenerated in a sequential fashion along Page 391 Table 2 A summary of polyp regeneration rates and totals follow- ing the four methods of removal. Polyps were removed using the following treatments: (1) direct predation by Tenellia fuscata, (2) forceps alone, (3) forceps and appli- cation of nudibranch mucus, and (4) forceps and nudi- branch in the same bow! but no direct contact. Regeneration Percent Treatment rate (day ') regenerated Nudibranch predation 0.8 12.5 Forceps only 0.734 64.0 Forceps and mucus (direct) 0.429 68.0 Forceps and mucus 0.482 68.0 the stolon, beginning with the proximal or oldest polyp and progressing distally along the stolon. In all treat- ments, the regeneration rate of removed polyps was less than one polyp per day (Table 2), but this was faster than the polyp production rate for new polyps on a stolon. The quickest regeneration occurred in those polyps removed by nudibranch predation (Table 2), but the number of polyps that regenerated was small. The results were not statistically significant so they are at best suggestive that actual nudibranch predation of a polyp may affect the hydroid differently from removal by forceps. DISCUSSION The growth form and rates of colony growth for Cordy- lophora lacustris are well studied (FULTON, 1961, 1962, 1963; OVERTON, 1963). BRAVERMAN (1974) has clearly documented the value of using hydroids for modeling growth in colonial organisms, plant and animal. Although it has been shown that herbivory can alter growth form in plants (BOSCHER, 1979), no similar work has been con- ducted on hydroids. The results of this study suggest that nudibranch predation does induce changes in rates of sto- lon growth and budding, which may alter growth form, and that there may be a chemical component involved in the induction process. The colony morphology described in this study is sim- ilar to that reported for laboratory cultures of Cordylo- phora lacustris by previous workers (FULTON, 1961; OVERTON, 1963). Our stolon growth rate of 1.06 mm/ day is slower than the approximately 3 mm/day reported by FULTON (1963), but the interstalk distances of about 3 mm were the same, suggesting that while growth form was similar, growth rate was about one-third the maxi- mum rate achieved by FULTON (1963). The culture tech- niques we used could certainly have been refined, but our primary concern was in determining whether nudibranch predation affected growth rates and(or) form. Therefore, once we determined that our culture techniques produced healthy colonies with a consistent growth form similar to that described in the literature, we focused our efforts on the experiments described. Page 392 The experiments showed that predation either by Te- nellia fuscata or by polyp removal by forceps, accompanied by direct or indirect exposure to 7. fuscata mucus, induced increased stolon budding rates. Enhanced stolon budding results in a denser colony form. This change could be caused by removal of three polyps, using forceps, and an application of mucus to the wounded portion of the colony lasting only a few minutes. In fact, removal by forceps and direct contact with mucus stimulated a greater re- sponse, lower stolon growth rate, and higher stolon bud- ding rate than actual predation by a nudibranch. The greater impact on the hydroid may have resulted from the removal of three polyps in a minute or two rather than from a nudibranch feeding on the same number of polyps over a 6-h period. Harris & Howe (1979) found that mucus from the nudibranch Aeolidia papillosa (Linnaeus, 1767) induced a behavioral response in its prey, the anem- one Anthopleura elegantissima (Brandt, 1835), but this ap- pears to be the first evidence of predator-induced growth changes in a cnidarian. It is interesting that removal of polyps independent of a chemical stimulus had a negative impact on stolon bud- ding and thus alters colony morphology in an opposite way to that induced by polyp removal accompanied by a chemical cue from the nudibranch predator. Polyp bud- ding increased when forceps were used alone, but without stolon budding there would be only a higher density of polyps along non-branching stolons growing away from the point of colony initiation—similar to a hedgerow in- stead of a grove. One of us (Harris) had hypothesized that a hydroid might respond to predation by a nudibranch by sending out stolons to disperse the colony as occurred when forceps were used alone. However, the opposite occurred, suggesting that C. lacustris grows away from areas of abiotically caused injury, whereas predation induces a growth response that leads to a denser colony. Tenellia fuscata has a short life-span of about 30 days (HARRIS et al., 1980). A single individual, even if it were able to feed at the maximum rate for its entire life-span, would consume less than 360 polyps. FULTON (1962) re- ported that a Cordylophora lacustris polyp might live more than 100 days and his colonies, in limited space, produced well over 2000 polyps. Therefore, a healthy colony could easily survive limited nudibranch predation, but would be swamped by numerous individuals as described by CHAM- BERS (1943). A denser colony form induced by nudibranch predation may benefit the hydroid by making it more dif- ficult for larvae, including the veligers of 7. fuscata, to pass through the canopy of predatory polyps to settle within the colony. STANDING (1976) reported that Obelia sp. in- hibited barnacle settlement by eating settling cyprids, thereby prolonging the persistence of the colony in the community. However, limited predation might have a positive effect on a hydroid as has been suggested for some host-parasite associations (CHENG, 1971; LINCICOME, 1971). In conclusion, the predator-prey association between The Veliger, Vol. 28, No. 4 Tenellia fuscata and Cordylophora lacustris is more complex than the simple physical act of polyp removal by the nu- dibranch. It is a dynamic process involving rates of polyp addition and replacement to counter their loss through predation and changes in growth form induced by both physical damage as well as the combination of physical and chemical! stimuli. The fact that both species adapt well to laboratory culture suggests that this system could be useful for studies relating to (1) chemical induction and (2) the dynamics of predator-prey associations. ACKNOWLEDGMENTS Support was provided for this study through a grant from the Sea Grant Program to Dr. Joseph Murdoch for the Ocean Projects course at the University of New Hamp- shire. Ted Donn and Wayne Lord provided assistance with statistical analyses. Regina Levine was particularly helpful as a source of constructive and insightful advice during all phases of the study. LITERATURE CITED BoscHER, J. 1979. Modified reproduction strategy of leek Al- lium porrum in response to a phytophagous insect, Acrole- piopsis assectella. Oikos 33:451-456. BRAVERMAN, M. 1974. The cellular basis of morphogenesis and morphostasis in hydroids. Oceanogr. Mar. Biol. Ann. Rev. 12:129-221. CHAMBERS, L. A. 1943. Studies on the organs of reproduction in the nudibranchiate mollusks, with special reference to Embletonia fuscata Gould. Bull. Amer. Mus. Natur. Hist. 66:599-641. CHENG, T. C. 1971. Enhanced growth as a manifestation of parasitism and shell deposition in parasitized molluscs. Pp. 103-138. In: T. C. Cheng (ed.), Aspects of the biology of symbiosis. University Park Press: Baltimore, MD. EDMUNDS, M., G. W. Potts, R. C. SWINFIN & V. L. WATER. 1976. Defensive behavior of sea anemones in response to predation by the opisthobranch mollusc Aeolidia papillosa (L.). J. Mar. Biol. Assoc. U.K. 56:65-83. FULTON, C. 1961. The development of Cordylophora. Pp. 287- 295. In: H. M. Lenhoff and L. Loomis (eds.), The biology of Hydra and of some other coelenterates. University of Miami Press: Coral Gables, FL. FULTON, C. 1962. Environmental factors influencing the growth of Cordylophora. J. Exp. Zool. 151:61-78. FULTON, C. 1963. The development of a hydroid colony. De- vel. Biol. 6:333-369. Harris, L. G. 1973. Nudibranch associations. Pp. 213-314. In: T. C. Cheng (ed.), Current topics in comparative patho- biology. Academic Press. Harris, L. G. & N. R. Howe. 1979. An analysis of the de- fensive mechanisms observed in the anemone Anthopleura elegantissima in response to its nudibranch predator Aeolidia papillosa. Biol. Bull. 157:138-152. Harris, L. G., M. Powers & J. Ryan. 1980. Life history studies of the estuarine nudibranch Tenellia fuscata (Gould, 1870). Veliger 23:70-74. HarVvELL, C. D. 1984. Predator-induced defense in a marine bryozoan. Science 224:1357-1359. LincicoME, D. R. 1971. The goodness of parasitism: a new G. Gaulin et al., 1986 hypothesis. Pp. 139-228. In: T. C. Cheng (ed.), Aspect of the biology of symbiosis. University Park Press: Baltimore, OVERTON, J. 1963. Intercellular connections in the outgrowing stolons of Cordylophora. J. Cell Biol. 17:661-671. Rosin, R. 1969. Escape responses of the anemone Anthopleura nigrescens (Kerrill) to its predatory aeolid nudibranch Her- viella (Baba). Veliger 12:74-77. Page 393 STANDING, J. D. 1976. Fouling community structure: effects of the hydroid, Obelia dichotoma, on larval recruitment. Pp. 155-164. In: G. O. Mackie (ed.), Coelenterate ecology and behavior. Plenum Publishing Corp.: New York. Topp, C. D. 1981. The ecology of nudibranch molluscs. Oceanogr. Mar. Biol. Ann. Rev. 19:141-234. The Veliger 28(4):394-396 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 Avoidance and Escape Responses of the Gastropod Nucella emarginata (Deshayes, 1839) to the Predatory Seastar Pisaster ochraceus (Brandt, 1835) MELISSA L. MILLER Bodega Marine Laboratory, Bodega Bay, California 94923, U.S.A. and University of California, San Diego, La Jolla, California 92093, U.S.A.! Abstract. The gastropod Nucella emarginata (Deshayes, 1839) exhibits both avoidance and escape responses to the predatory seastar Pisaster ochraceus (Brandt, 1835). When exposed to water “scented” by P. ochraceus, N. emarginata demonstrated a strong avoidance response that was absent when exposed to normal (control) seawater. Nucella emarginata also responded rapidly to the touch of a tube foot from P. ochraceus by changing direction and increasing mobility. Little or no response was elicited by the touch of a glass rod. INTRODUCTION MANY GASTROPOD mollusks exhibit defensive behaviors in response to predatory animals. These have been consid- ered to be of two types: avoidance and escape behaviors (PHILLIPS, 1977). Avoidance behavior is exhibited when a prey species, responding to substances that have diffused from a predator through the water, reacts to the presence of the still distant predator (FEDER & ARVIDSSON, 1967; MackIE, 1970). Escape behavior is a response to actual contact with the predator. There are many documented examples in the literature of escape responses and chem- ically mediated avoidance responses by marine gastropods to predatory asteroids, crabs, and gastropods (BULLOCK, 1953; EDWARDS, 1969; FEDER, 1963, 1967; GELLER, 1982; Gonor, 1965; MACKIE, 1970; MARGOLIN, 1964; MENGE, 1972; PHILLIPS, 1975, 1976, 1977, 1978). Nucella emarginata (Deshayes, 1839) and Pisaster ochraceus (Brandt, 1835) are two species that overlap both in spatial distribution and dietary composition. They are also linked in prey-predator interactions, as N. emarginata is one of the species preyed upon by P. ochraceus (BERTNESS, 1977; FEDER, 1959). Nucella emarginata is found on rocky shores ranging from Alaska to Mexico (RICKETTS & CALVIN, 1968) and has been previously ' Present address: 6086 48th Street, San Diego, California 92120, U.S.A. shown to demonstrate a weak defensive response to the predatory seastar Leptasterias hexactis (MENGE, 1972). The purpose of the present study is to determine if Nucella emarginata exhibits avoidance and escape re- sponses to the predatory seastar Pisaster ochraceus. (Henceforth Nucella emarginata and Pisaster ochraceus will be referred to by their generic names only.) MATERIALS anp METHODS Study Sites Specimens of Nucella were collected from Dillon Beach and Doran Rocks, two locations near Bodega Bay, So- noma County, California. Both sites are boulder-strewn beaches supporting abundant populations of this gastro- pod. Experimental Methods Snails from both sites were maintained at the nearby Bodega Marine Laboratory (BML) in running seawater in a partitioned aquarium. A separate aquarium housed a seastar, collected from the coast adjacent to the BML. The apparatus for testing both escape and avoidance responses consisted of four open-top boxes constructed of 0.32-cm thick plexiglas with dimensions 30.5 x 30.5 x 2.54 cm. A grid of 28 x 28 cm in 1-cm gradations was drawn on the base. Gradations were identified along one side by numbers and along the perpendicular side by let- ters of the alphabet in order to record the displacement of the snail without disturbance. M. L. Miller, 1986 Table 1 Avoidance response of Nucella emarginata from two sites, Dillon Beach and Doran Rocks, and the combined results from both locations. Fifty snails were used in each trial. The test criteria for a positive response were shell lifting behavior, changing direction, or a combination of both behaviors. A response was scored as negative if there was no apparent change in activity. x* values resulted from comparing the control with the seastar run by means of a 2 x 2 contingency table corrected for continuity. ** = P < 0.001. Posi- Nega- tive re- tive re- Site sponses sponses x? Dillon Beach Control 1 49 Pisaster-scented water 49 1 88.36** Doran Rocks Control 5 45 Pisaster-scented water 47 3} O7/s Dillon Beach and Doran Rocks Control 6 94 Pisaster-scented water 96 4 158.48** In the avoidance response test procedure, the seastar was weighed and placed in a 4-L beaker containing 3 L of seawater. After 2 h water from the beaker containing the seastar was poured into a 250-mL beaker for use in the experiments. The plexiglas boxes were filled to a depth of 1 cm with fresh seawater. Snails were measured with calipers (apex to end of siphonal canal), and placed, one per box, at the center of the grid. Once the snail’s foot was extended, 1 mL of fresh seawater or 1 mL of scented water was allowed to flow freely from a pipette 2 cm from the anterior end of the snail. Responses were closely ob- served and recorded for 5 min. The position of the snail, based upon the location of the apex of the shell, was re- corded at the end of 5, 10, and 15 min. Fifty snails were tested from each site. Control and experimental trials were conducted on consecutive days. In the escape response test procedure, the plexiglas box- es were each filled to a depth of 1 cm with fresh seawater. Snails were measured and placed, one per box, at the center of the grid. Once the foot was extended, the snail was touched anteriorly either with the tip of a glass pipette or a tube foot that had been removed from a seastar and was held with forceps. Responses were closely observed and recorded for 5 min. Displacement of the snails was recorded at the end of 5, 10, and 15 min. Fifty snails were tested from each site. Control and experimental trials were conducted on consecutive days. Between each trial the boxes were rinsed in running seawater. The pipette was also rinsed between each trial in the escape response procedure. A tube foot was used Page 395 Table 2 Escape response of Nucella emarginata from two sites, Dil- lon Beach and Doran Rocks, and the combined results from both beaches. Fifty snails were used in each trial, and the test criteria were as in Table 1. x? values resulted from a comparison of control with seastar trial data by means of a 2 X 2 contingency table corrected for conti- nuity. ** = P < 0.001. Posi- Nega- tive re- tive re- Site sponses sponses x? Dillon Beach Control 5 45 Pisaster tube foot touch 42 8 5202"* Doran Rocks Control 4 46 Pisaster tube foot touch 44 6 60.93** Dillon Beach and Doran Rocks Control 9 91 Pisaster tube foot touch 86 14 115.80** four times before being discarded and replaced. In all trials, a positive response was defined as a change in be- havior from that observed prior to the stimulus. RESULTS Nucella responded to the scent of Pisaster (Table 1). A total of 96 of 100 snails responded to the Pisaster-scented water, while only 6 of 100 responded to fresh seawater, a highly significant difference (x? = 158.5, P < 0.001). The behavior exhibited by Nucella when exposed to the Prsas- ter-scented water was remarkably different from activity prior to exposure. Some snails (2.1%) were observed only to lift their shells repeatedly and extend upward the body mass between the head and foot. They appeared to be rocking back and forth. Others moved in circles (21.9%), while many others combined both responses (76.0%). Nucella exhibited an escape response when contacted by the tube foot of Pisaster (Table 2). A total of 86 of 100 snails responded to the tube foot touch compared to only 9 of 100 responding to the touch of the tip of a glass pipette. Trials involving tube foot touch were significantly different from control trials (x? = 115.8, P < 0.001). The escape response by Nucella was similar to the avoidance response, but it was not as pronounced. Snails were ob- served to lift up only initially, and then make 180° change of direction from the point of stimulus. Snails were ob- served to lift up (2.3%), change direction (64.0%), and to change direction and lift up (33.7%). DISCUSSION The present study confirms that the gastropod Nucella emarginata does exhibit both avoidance and escape re- Page 396 sponses to the predatory seastar Pisaster ochraceus. The primary response of the snail was to change direction and increase its activity. It must be kept in mind, however, that laboratory studies, such as the test procedure used here for avoidance responses, are subject to the criticism that laboratory conditions do not duplicate those normally found in the natural habitat. For example, high concen- tration of seastar scent, minimal water disturbance, and smooth substrate are not naturally found in the intertidal zone. Assuming that the behaviors elicited in the labora- tory also occur in the field, they would be advantageous to an organism such as Nucella whose habitat overlaps that of a predator, in this case, Pisaster. ACKNOWLEDGMENTS I express my gratitude to the entire teaching staff at the Bodega Marine Laboratory for all of the assistance I re- ceived throughout the quarter. I wish to thank V. Connor for her insight and encouragement, and for the use of her laboratory space. I also thank G. Adest and D. Phillips for valuable discussions. My special thanks go to G. Adest for accepting me into this undergraduate program. LITERATURE CITED BerTNESs, M. D. 1977. Behavioral and ecological aspects of shore-level size gradients in Thais lamellosa and Thais emar- ginata. Ecology 58:86-97. Buttock, T. H. 1953. Predator recognition and escape re- sponses of some intertidal gastropods in the presence of star- fish. Behavior 53:130-140. Epwarps, D.C. 1969. Predators on Olivella biplicata, includ- ing a species-specific predator avoidance response. Veliger 11:326-333. FEDER, H. M. 1959. Food of the starfish Pisaster ochraceus along the California coast. Ecology 40:721-724. FEDER, H. M. 1963. Gastropod defensive responses and their The Veliger, Vol. 28, No. 4 effectiveness in reducing predation by starfishes. Ecology 44: 505-512. FEDER, H. M. 1967. Organisms responsive to predatory sea- stars. Sarsia 29:371-394. FEDER, H. M. & J. ARvIDssON. 1967. Studies on a seastar (Marthasterias glacialis) extract responsible for avoidance re- actions in a gastropod (Buccinum undatum). Arkiv For Zoo- logi 19:369-379. GELLER, J. B. 1982. Chemically mediated avoidance response of a gastropod, Tegula funebralis (A. Adams), to a predatory crab, Cancer antennarius (Stimpson). J. Exp. Mar. Biol. Ecol. 65:19-27. Gonor, J. J. 1965. Predator-prey relations between two ma- rine prosobranch gastropods. Veliger 7:228-232. Mackig, A. M. 1970. The escape reactions of marine inver- tebrates to predatory starfish. J. Exp. Mar. Biol. Ecol. 5: 63-69. MarGOoLin, A. S. 1964. A running response of Acmaea to sea- stars. Ecology 45:191-193. MENGE, B. A. 1972. Foraging strategy of a starfish in relation to actual prey availability and environmental predictability. Ecol. Monogr. 42:25-50. PaInE, R. T. 1969. The Pisaster-Tegula interaction: prey patches, predator food preference, and intertidal community structure. Ecology 50:950-961. PHILLIPS, D. W. 1975. Distance chemoreception-triggered avoidance behavior of the limpets Acmaea (Collisella) lima- tula and Acmaea (Notoacmea) scutum to the predatory star- fish Pisaster ochraceus. J. Exp. Zool. 191:359-368. PHILLIPS, D. W. 1976. The effect of species-specific avoidance response to predatory starfish on the intertidal distribution of gastropods. Oecologia (Berlin) 23:83-94. PHILLIPS, D. W. 1977. Avoidance and escape responses of the gastropod mollusc Olivella biplicata (Sowerby) to predatory asteroids. J. Exp. Mar. Ecol. 28:77-86. PHILLIPS, D. W. 1978. Chemical mediation of invertebrate behaviors and the ability to distinguish between foraging and inactive predators. Mar. Biol. 49:237-243. RICKETTS, E. F. & J. CALVIN. 1968. Between Pacific Tides. 4th ed., rev. by J. W. HEDGEPETH. Stanford University Press: Stanford, California. 614 pp. The Veliger 28(4):397-400 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 Oxygen Production and Consumption in the Sacoglossan (=Ascoglossan) Elysia chlorotica Gould by GLENYS D. GIBSON, DANIEL P. TOEWS, ano J. SHERMAN BLEAKNEY Biology Department, Acadia University, Wolfville, Nova Scotia BOP 1X0, Canada Abstract. A naturally occurring population of Elysia chlorotica Gould (Opisthobranchia: Sacoglossa) composed of a mixture of individuals ranging from dark green to non-green in color was found in the Minas Basin, Nova Scotia, Canada. This species is usually dark green in color because of endosymbiotic chloroplasts derived from their food alga Vaucheria sp. Individuals from this population were examined for O, production and consumption. A correlation between chlorophyll content and O, production was found. INTRODUCTION SINCE THE FIRST identification of sacoglossan endosym- bionts as chloroplasts (KAWAGUTI & YAMASU, 1965), many such associations have been described, especially in ref- erence to plastid origin (TAYLOR, 1968; TRENCH et al., 1969; GREENE, 1970a; TRENCH, 1975) and functional ca- pacity, including carbon fixation (GREENE, 1970b; TRENCH, 1973; STIRTS & CLARK, 1980; CLARK ef al., 1981) and oxygen production (Brandt, 1883, zn TAYLorR, 1968; KAWAGUTI & YAMASU, 1965; TRENCH et al., 1969; TRENCH, 1975; GRAVES et al., 1979). These authors have demonstrated a net O, production under illumination and attributed it to the presence of endosymbiotic chloroplasts. Sacoglossans are herbivorous and feed by slitting or piercing a food plant with their radula and suctorially removing plant sap and chloroplasts (TRENCH e¢ al., 1969; JENSEN, 1983). The chloroplasts are phagocytized by the slug’s digestive cells (MUSCATINE et al., 1975; MCLEAN, 1976) in which they are maintained for variable periods of time depending upon the sacoglossan and algal species (TAYLOR, 1968; TRENCH, ef al., 1969; HINDE & SMITH, 1972; CLARK et al., 1981). The presence of chloroplasts in the digestive diverticula typically colors the slug identically to the plastid source, and comparison of pigment spectra of the slug and pos- sible food choices is commonly used to determine the ac- tual food plant (TayLor, 1968; TRENCH et al., 1969; GREENE, 1970a; TRENCH, 1975). Chlorophyll content has been shown to be indicative of chloroplast functional ca- pacity, often decreasing with starvation of the sacoglossan (GREENE, 1970b; CLARKE & BUSACCA, 1978). In the summers of 1983 and 1984, three salt marshes in the Minas Basin, Nova Scotia, contained populations of Elysia chlorotica Gould, 1870 (Opisthobranchia: Saco- glossa) ranging from light green to non-green in color, instead of the usual rich dark green that results from the presence of endosymbiotic chloroplasts derived from the food alga Vaucheria sp. (BAILEY & BLEAKNEY, 1967; GRAVES et al., 1979). A survey of the literature indicated that a naturally occurring population of symbiotic elysiid sacoglossans not strongly pigmented by chloroplasts had never been reported. The aim of this study was to use the above population as the basis for an examination of sacoglossan color and chloroplast function as related to chlorophyll content, oxy- gen production, and oxygen consumption. MATERIALS anon METHODS Collection and Maintenance of Animals Elysia chlorotica individuals were collected from salt marshes located at Kingsport, Pickett’s Wharf, and Por- ter’s Point in the Minas Basin. They were usually found on exposed mats of an undetermined Vaucheria species along the edges of pools and creeks, on pool bottom sedi- ments, or on an assortment of submersed algae, including Rhizoclonium, Cladophora, and Ectocarpus species. The slugs were divided into three study groups by com- parison with MUNSELL (1977) color charts. Color group- ings used were as follows: Group 1 (dark green) ranged in color from 7.5GY 4/4, 4/6 to 5GY 7/6, 7/8, Group 2 (light green) from 5GY 7/6, 7/8 to 2.5GY 7/4, 7/6, and Group 3 (non-green) from 2.5GY 7/4, 7/6 to 5Y 8/4, 8/6. All specimens were used as soon as possible after Page 398 0.4 0.2 ; -1 mq 0, g( fw.) The Veliger, Vol. 28, No. 4 60 90 120 TIME (min) Figure 1 Gross O, production of each Elysza chlorotica study group. Mean values are plotted, and SE bars included where the standard error is greater than the area covered by the mean symbol. n= 6 for each group. a, Group 1; b, Group 2; c, Group 3. collection and were maintained in the dark at 9°C and at 28%o0 salinity without food between tests. Measurements of Oxygen Production and Consumption Photosynthetic and respiratory activity were compared between study groups by measuring light-dark O, pro- duction and consumption. Control tests without slugs were also conducted. Slugs were gently blotted to remove excess water, weighed, and placed in a stoppered 12-mL flask filled with O,-saturated (PO, = 155 mm Hg) seawater (28%o salinity). Changes in oxygen pressure were mea- sured at 15-min intervals for 2 h with a Radiometer O,- electrode system. IJlumination was provided by a 120-Volt American Optical fluorescence lamp with two bulbs po- sitioned at right angles to each other and at a 1-cm dis- tance from the flask. During the dark trials, the flask was wrapped in aluminum foil and continuous illumination was maintained to eliminate the possibility of light-in- duced temperature changes (1.2-2.0°C) affecting the elec- trode characteristics and O, solubility. The water was mixed with a small magnetic stirrer and the electrode allowed to stabilize before each reading was taken. PO, was converted to mg O, with the following equation (adapted from Hoar & HICKMAN, 1975): ae OrdGpeie = PO, mm Hg: 1000a- 1.43 BP-g(fw)-h where g(fw) = fresh weight of the slug in grams, h = time (in hours), a =the appropriate O, solubility coefficient alpha for a specific temperature, 1.43 = a conversion fac- tor to change mL O, to mg O,, and BP = barometric pres- sure. Gross O, production was determined by adding the amount of O, consumed in the dark to that produced in the light. Rates of respiration (O, consumed in the dark) were compared using a Mann-Whitney U test. Measurement of Chlorophyll Content Animals were gently blotted to remove excess water, weighed, anaesthetized at —9°C for 2-3 min, and homog- enized in 2 mL of absolute methanol. The suspension was centrifuged in a IECHT centrifuge at 8000 rpm for 10 min. The centrifuged pellet was washed twice with ab- solute methanol and the extracts combined for a total vol- ume of 5 mL. Samples were stored temporarily (less than 1 h) in the dark at 9°C to prevent bleaching. Chlorophyll content (chlorophylls @ and c inclusive) was determined with a Varian Techron model 635 spectrophotometer us- ing the following equation (after MACLACHLAN & ZAH- LIK, 1963): mg chl g(fw)-! = 25.5(Agoo) + 4.0(Agss)° V/(g(fw)- 1000) where g(fw) = fresh weight of slug in grams, A = absorp- tion at the indicated wavelength, and V = total volume (5 mL) of methanol extract. OBSERVATIONS The three study groups showed some gross O, production (Figure 1). Group 1 (dark green slugs) have both the greatest O, production and the highest chlorophyll content (Table 1), Group 2 (light green slugs) showed less O, production and a lower chlorophyll content, while Group 3 (non-green slugs) showed almost no O, production and a very low chlorophyll content. Groups 2 and 3 consumed O, at a greater rate than O, was produced, in contrast to G. D. Gibson et al., 1986 Page 399 Table 1 Chlorophyll content, O, production, and O, consumption for each Elysia chlorotica study group. Values given are means + SE. n = 6 for each group. Group Chlorophyll content mg chl g (fw)! 1.83 + 0.16 mg O, g(fw)7! h7! O, produced in light 0.0775 + 0.0095 O, consumed in dark —0.1109 + 0.0029 Gross O, production 0.1884 + 0.0119 mg O, g(fw)! g chi! h™! 0.1029 Group 1. O, produced per unit chlorophyll was greatest in Group 2, and lowest in Group 3. The three study groups displayed similar rates of respiration (O, con- sumption in the dark). DISCUSSION Although a naturally occurring non-green elysiid popu- lation has not previously been reported, the relationship between chlorophyll content, an indicator of chloroplast functional capacity (GREENE, 1970b), and sacoglossan starvation has been studied by several authors (TRENCH et al., 1969; CLARK et al., 1981). Results vary with species. Chlorophyll levels have been found to decrease after a 24-h starvation period in Elysia hedgpethi, accompanied by a parallel decrease in chloroplast functional capacity (GREENE, 1970b). In the same study, Greene observed that chlorophyll levels in Placobranchus ianthobapsus re- mained unaffected throughout a 27-day starvation period, while the photosynthetic ability of the chloroplasts de- creased. In Elysia viridis, chlorophyll levels increased over a 93-day starvation period and chloroplasts remained functional for 3 months (HINDE & SMITH, 1972). CLARK & Busacca (1978) found that chlorophyll content de- creased with starvation in four sacoglossan species: Elysza tuca, Tridachia crispata, Oxynoe antillarum, and Elysia cauze. It could not be determined whether or not starvation had occurred before collection of the naturally pale Elysza chlorotica specimens examined in this study, and if it did, for what length of time. However, there is a relationship between chlorophyll content and chloroplast functional capacity. Elysia chlorotica, from all three study groups, showed some gross O, production. Group 1 (dark green slugs) had both the greatest chlorophyll content and the largest O, production. Although less O, was produced by the animals in Group 2 (light green slugs), each unit of chlorophyll produced 31% more O, than in Group 1. An- imals in Group 3 (non-green slugs) showed almost no gross O, production, the chlorophyll content decreased to 9% of that of Group 1, and each unit of chlorophyll pro- duced 26% less O, than in Group 1. The increased O, production per unit of chlorophyll in Group 2 is of inter- Group 1 (dark green) Group 2 (light green) Group 3 (non-green) 0.59 + 0.05 0.16 + 0.01 —0.0193 + 0.0073 —0.0828 + 0.0097 —0.1060 + 0.0082 —0.0913 + 0.0038 0.0883 + 0.0125 0.0122 + 0.0125 0.1500 0.0762 est, although difficult to explain. Possibly as chlorophyll is lost, more light is able to penetrate the sacoglossan tis- sue, or some means of regulating chlorophyll activity oc- curs. However, it appears that E. chlorotica can maintain a high level of O, production while chlorophyll levels are starting to decline. As chlorophyll levels continue to fall, this ability is lost, as shown in Group 3. All three study groups had equivalent rates of respi- ration (Groups 1 and 3 compared: Mann-Whitney U = 48.5, P > 0.05, n, =9, n» =9). This is indicative of the tissues being in a similar physiological condition, regard- less of color. Variations observed might have been influ- enced by several factors. Specimens used were selected from a certain size range (4-12 mm) but the variation within this range would result in differing rates of res- piration (SANDER & Moore, 1978). Respiration rates may also be affected by the small temperature changes that occurred (SANDER & Moore, 1978) as well as by the change in O, tension in the experimental chamber as O, was consumed during each test (MANDAN MOHAN Das & VENKATACHARI, 1984). The cause for the appearance of the naturally pale Ely- sta chlorotica population is not known. Field records for the salt marshes of the Minas Basin (Bleakney, 1966- 1982, unpublished data), report that only two light green E. chlorotica individuals have been previously collected (May 1, 1969). Field records for 1983 indicate that almost all of the algae in and around the marsh pools were dead by late May, probably as a result of heavy rainfall throughout the month. In 1984, Vaucheria mats did not appear until the end of June. Perhaps the loss of green pigment was related to the inability of the slugs to locate sufficient Vaucheria. This would either result in starvation or force the slugs to find another food source, presumably without compatible chloroplasts. However, starvation of this species does not usually produce a loss of green pig- ment. Dark green FE. chlorotica collected in other years have remained healthy and green for at least 4 months in a 9°C refrigerator. S. K. Pierce (1984, in litt.) was also unable to bleach out the chlorophyll through starvation of this species. Dark green slugs collected in the Minas Basin Page 400 in the summers of 1983 and 1984 did lose chlorophyll when starved for 2 to 3 wk. Non-green slugs fed freshly collected Vaucheria sp. turned green within 4 days. Per- haps the Vaucheria ingested during this period was in some way debilitated, causing the chloroplasts to bleach more rapidly. Examination of a naturally occurring green and non- green Elysia chlorotica population indicates that there is a relationship between declining chlorophyll content (re- flected in slug color) and the photosynthetic ability of the endosymbiotic chloroplasts. The functional capacity of the total amount of chlorophyll present seems to vary as pig- ment is lost. Regardless of color, the rate of O, consump- tion does not appear to change. ACKNOWLEDGMENTS This research was supported by a university research grant to J. S. Bleakney and by the National Sciences and En- gineering Research Council of Canada through an oper- ating grant to D. P. Toews. LITERATURE CITED BalLey, K. H. & J. S. BLEAKNEY. 1967. First Canadian report of the sacoglossan Elysia chlorotica Gould. Veliger 9:353- 354. Cxiark, K. B. & M. Busacca. 1978. Feeding specificity and chloroplast retention in four tropical Ascoglossa, with a dis- cussion of the extent of chloroplast symbiosis and the evo- lution of the order. J. Moll. Stud. 44:272-282. Cxiark, K. B., K. R. JENSEN, H. M. StTirRTs & C. FERMIN. 1981. Chloroplast symbiosis in a non-elysiid mollusc, Cos- tastella lilianae (Marcus) (Hermaidae: Ascoglossa) (=Saco- glossa): effects of temperature, light intensity, and starvation on carbon fixation rate. Biol. Bull. 160:43-54. GRAVES, D. A., M. A. GIBSON & J. S. BLEAKNEY. 1979. The digestive diverticula of Alderia modesta and Elysia chlorotica (Opisthobranchia: Sacoglossa). Veliger 21:415-422. GREENE, R. W. 1970a. Symbiosis in sacoglossan opistho- branchs: symbiosis with algal chloroplasts. Malacologia 10: 357-368. GREENE, R. W. 1970b. Symbiosis in sacoglossan opistho- branchs: functional capacity of symbiotic chloroplasts. Mar. Biol. 7:138-142. HinpeE, R. & D. C. SmirH. 1972. Persistence of functional The Veliger, Vol. 28, No. 4 chloroplasts in Elysia viridis (Opisthobranchia, Sacoglossa). Nature New Biology 239:30-31. Hoar, W.S. & C. P. HICKMAN. 1975. A laboratory compan- ion for general and comparative physiology. 2nd ed. Pren- tice-Hall Inc.: New Jersey. JENSEN, K. R. 1983. Factors affecting feeding selectivity in herbivorous Ascoglossa (Mollusca: Opisthobranchia). J. Exp. Mar. Biol. Ecol. 66:135-148. KawacutTl, S. & T. YAMASU. 1965. Electron microscopy on the symbiosis between an elysioid gastropod and chloro- plasts of a green alga. Biol. J. Okayama Univ. 11:57-65. MacLacu.an, S. & S. ZAHLIK. 1963. Chlorophyll mutant of barley. Can. J. Bot. 41:1053-1062. MaDAN MOHAN Das, V. & S. A. T. VANKATACHARI. 1984. Influence of varying oxygen tension on the oxygen con- sumption of the freshwater mussel Lamellidens marginalis (Lamarck) and its relation to body size. Veliger 26:305- 310. McLEAN, N. 1976. Phagocytosis of chloroplasts in Placida den- dritica (Gastropoda: Sacoglossa). J. Exp. Zool. 197:321- 329. MUNSELL. 1977. Colour charts for plants tissues. 2nd ed. Koll- morgen Corporation: Baltimore. MuscaTINE, L., R. R. Poor & R. K. TRENCH. 1975. Sym- biosis of algae and invertebrates: aspects of the symbiont surface and the host-symbiont interface. Trans. Amer. Mi- crosc. Soc. 94:450-469. SANDER, F. & E. A. Moore. 1978. Comparative respiration in the gastropods Murex pomum and Strombus pugilis at different temperatures and salinities. Comp. Biochem. Physiol. 60:99-105. SkirTS, H. M. & K. B. CLark. 1980. Effects of temperature on products of symbiotic chloroplasts in Elysia tuca Marcus (Opisthobranchia: Ascoglossa). J. Exp. Mar. Biol. Ecol. 43: 39-47. TayLor, D. L. 1968. Chloroplasts as symbiotic organelles in the digestive gland of Elysia viridis (Gastropoda: Opistho- branchia). J. Mar. Bio. Ass. U.K. 48:1-15. TRENCH, R. K. 1973. Further studies on the mucopolysac- charide secreted by the pedal gland of the marine slug, 777- dachia crispata (Opisthobranchia, Sacoglossa). Bull. Mar. Sci. 23:299-312. TRENCH, R. K. 1975. Of ‘leaves that crawl’: functional chlo- roplasts in animal cells. Soc. Exp. Biol. Cambridge Sym- posium 29:229-266. TRENCH, R. K., R. W. GREENE & B. G. BysTROM. 1969. Chloroplasts as functional organelles in animal tissues. J. Cell Biol. 42:404-417. THE VELIGER © CMS, Inc., 1986 The Veliger 28(4):401-417 (April 1, 1986) Environmental Perturbations Reflected in Internal Shell Growth Patterns of Corbicula fluminea (Mollusca: Bivalvia)! by LOWELL W. FRITZ? anpD RICHARD A. LUTZ Department of Oyster Culture, New Jersey Agricultural Experiment Station and Center for Coastal and Environmental Studies, Rutgers University, New Brunswick, New Jersey 08903, U.S.A. Abstract. Anthropogenic and natural seasonal environmental perturbations were reflected in shell growth patterns of specimens of Corbicula fluminea living at the northernmost extent of their range along the east coast of North America (Raritan River, New Jersey). Growth of organisms in experi- mental cages was monitored from August 1981 to January 1982 and from July to December 1982 at stations located upstream (controls: 2 stations) and immediately downstream (perturbed: 1 station) from a combined industrial-sewage effluent. In 1981, the growing shell margin of each clam was notched with a small drill before each was placed in a cage; these marked organisms were sacrificed after various lengths of time. In 1982, specimens were not notched, but a growth cessation mark in the shell micro- structure of all caged organisms marked the beginning of the monitored growth period. Growth patterns in shell microstructure were examined in acetate peels and polished thin sections. Microgrowth incre- ments in the outer crossed-lamellar layer were deposited at an average rate of approximately one increment per day. A growth cessation mark found in all specimens sampled in 1981 (n = 53) was dated to within two days of a major storm using increment counts, revealing the accuracy of their use to date shell regions. Lack of growth in winter resulted in a growth discontinuity in the inner complex crossed-lamellar layer and an associated growth cessation mark in the outer layer. Increment counts suggested that growth resumed in late March or early April each year as water temperatures rose above approximately 10°C. Growth rates of 1+ year old individuals during spring and early summer (before entering experimental cages) averaged 65 and 45 wm/increment in 1981 and 1982 respectively. In 1981, growth rates at each site were significantly slower during the monitored growth period than before it, which was probably due to injury inflicted by notching the ventral shell margin. In 1982, growth rates of unnotched clams at the control sites were similar before and after entering the exper- imental cages (after an initial two-week decrease in growth rates). However, unnotched specimens moved to the perturbed site in 1982 subsequently grew at significantly slower rates and had fewer increments during the monitored period than those collected from cages at control sites. INTRODUCTION AQUATIC ECOLOGISTS are frequently concerned with as- sessing effects of environmental events, such as chronic, periodic, or accidental additions of a pollutant, on growth 'NJAES Publication No. D-32507-1-85 supported by state funds and New Jersey Department of Environmental Protec- tion. 2 Mail reprint requests to: L. W. Fritz, Rutgers University Shellfish Lab., P.O. Box 687, Port Norris, New Jersey 08349. of organisms after the event has occurred. In the absence of information about pre-disturbance growth rates, the ecologist, like the paleontologist, is confronted with the problem of after-the-fact data acquisition. Detailed anal- yses of growth patterns in molluscan shell structure pro- vide a tool for addressing this problem. Records of an organism’s dynamic environment are capable of being pre- served as structural, morphological, or chemical changes in the shell. Research into these relationships has been largely to reconstruct paleoenvironments (PANNELLA & MacCLinTock, 1968; RHOADS & PANNELLA, 1970; BERRY Page 402 & BARKER, 1975; PANNELLA, 1976; JONES, 1980; RYE & SOMMER, 1980; Lutz, 1981; Dopp & Crisp, 1982). However, implications of this approach for ecological work have also interested many neontologists (RHOADS & PANNELLA, 1970; KENNISH & OLSSON, 1975; KENNISH, 1980). Present knowledge of the relationship between the en- vironment and bivalve shell structure is limited. However, studies of shell growth patterns of Mercenaria mercenaria (Linné) by KENNISH & OLSSON (1975) and KENNISH (1980) have demonstrated the application of shell analyt- ical techniques to im situ environmental monitoring stud- ies. In the present study, microstructural shell growth pat- terns of the freshwater Asiatic clam Corbicula fluminea (Miller) were analyzed to assess effects on clam growth of the combined effluents of an organic chemical plant and a sewage treatment facility in a freshwater, non-tidal seg- ment of the Raritan River, New Jersey. Corbicula flumi- nea was chosen as the test organism because it maintains a large population within the river (TRAMA, 1982), has rapid rates of shell growth (BRITTON et al., 1979; BRITTON & Morton, 1982), and because it is an opportunistic, pest species (BRITTON & Morton, 1982). In order to discern the effects of environmental perturbations in shell microstructure, it was necessary to first document the nat- ural, seasonal shell growth pattern of the species. The shell of Corbicula fluminea is composed of three calcareous layers when viewed in radial section. These are, from exterior to interior: the outer fine crossed-la- mellar layer, a pallial myostracum, and the inner complex crossed-lamellar layer (TAYLOR et al., 1973; GOUNTS & PREZANT, 1982; PREZANT & TAN-TIU, 1985). Cursory examination of thin radial sections (or acetate peels) of the aragonitic outer and inner layers reveals bands or lines within the shell (Figure 1A, B). A single pair of dark lines within the outer layer that delineates a lighter band of shell between them is defined as a microgrowth incre- ment. A growth cessation mark is an irregularity in the periodic deposition of microgrowth increments, often re- sulting in a V-shaped notch in the shell exterior, an un- usually thick microgrowth increment boundary, or an abrupt change in the depositional surface, caused by a loss of mantle attachment to the ventral margin (KENNISH & OLSSON, 1975; RICHARDSON et al., 1980). Furthermore, growth lines within the inner layer, which are often as- sociated with microgrowth increments or growth cessation marks in the outer layer, may also reflect the growth his- tory of the animal. TRAMA (1982) first reported the occurrence of Corbicula fluminea in the Raritan River from collections made in March 1981. He estimated the year of introduction as not later than 1978 based on an age of between 3 and 4 years for the largest specimen collected, which had a shell length of 25 mm. Little is known about the population ecology of C. fluminea in the Raritan River. Based on other studies in North America and Asia (see reviews of BRITTON & The Veliger, Vol. 28, No. 4 Morton, 1979, 1982), Corbicula can aptly be described as opportunistic, but also well adapted, especially in its reproductive biology, to lotic environments. Density-in- dependent factors, such as weather and reservoir draw- down, have been shown to cause catastrophic mortalities of Corbicula in streams, lakes, and reservoirs, but due to early maturation and high fecundity, the organism has been able to maintain populations in many such systems (BRITTON & MorTON, 1979). The temperature ranges for survival and growth of C. fluminea are not precisely known, but RODGERS e¢ al. (1979) reported a minimum of 10°C for growth. Low winter temperatures and their duration are suspected of limiting the spread of the species to generally south of latitude 40°N on the North American continent. Maximum shell lengths of Corbicula in North America vary from a low of 18 mm in San Luis Reservoir, California, to between 40 and 50 mm in other systems in Texas and California, suggesting that factors other than low winter temperatures are involved in controlling ulti- mate shell size (BRITTON & MorTon, 1982). The rela- tively small maximum size of Corbicula attained in the Raritan River, New Jersey, then may indicate that this population is “dwarfed,” and possibly stressed by factors other than low winter temperatures. To study shell growth during known periods, caged populations of Corbicula fluminea were established near the efuent discharge and at two control stations in the Raritan River in summer 1981 and sampled, along with the natural population, during the next one and one-half years (Figure 2). Through comparisons of microstructural shell growth of caged C. fluminea at the three locations and uncaged specimens at one site, we (1) analyzed effects of chronic exposure to the effluent on growth, (2) deter- mined effects of notching the ventral shell margin and caging on growth, (3) documented the natural annual pat- tern of shell growth (from which changes caused by the effluent, notching and(or) caging had to be distinguished), and (4) determined the periodicity of microgrowth incre- ment formation. MATERIALS anpD METHODS Sampling Methods Caged populations of Corbicula fluminea were estab- lished at three stations in the Raritan River, New Jersey (Figure 2). The experimental station, AC, received the combined effluents of the American Cyanamid Organic Chemical Plant and the Somerset-Raritan Valley Sewage Authority. Composition of the effluent varied considerably on a daily basis (B. Ruppel, NJ Department of Environ- mental Protection, personal communication). Conse- quently, we did not determine the specific effects on shell growth of high nutrient loadings or organo-chlorine com- pounds in the effluent, but rather the integrated results of chronic exposure to a broad spectrum, sublethal alteration of water quality. Only a small population of C. fluminea Paw Etzice RevAS utz, 1986 Page 403 Figure 1 A (above). Photographic enlargement of a thin shell section of a specimen of Corbicula fluminea collected on 13 July 1981 from the natural population at station C (see Figure 2). Print was made by placing thin shell section in an enlarger and exposing photographic paper directly; thus, the enlargement is a negative image of the section. The winter 1980-1981 growth cessation (W1) and the growth cessation mark (gcm) in the outer fine crossed- lamellar layer, and the growth discontinuity (gd) in the inner complex crossed-lamellar layer are labelled. Total shell height is 18.1 mm and growth is to the right. B (below). Light micrograph of the outer fine crossed-lamellar layer of a specimen of Corbicula fluminea showing a series of microgrowth increments. Arrows delineate one increment. Micrograph is a positive image of the section. Growth is to the right and the horizontal field width is 0.7 mm. was present at station AC, but a large population, with shell lengths ranging from 2 to 20 mm, was located at the confluence of the Millstone and Raritan rivers (TRAMA, 1982; this study), approximately 1.6 km upstream from station AC. This site was one of the controls (station C) and was the source of all animals placed in cages at the three stations. The other control station, DI, located 12 km upstream from station AC, was selected because there were no direct inputs to the site from sewage treatment plants, industries, or landfills. Station DI in 1981 did not have a natural population of C. fluminea associated with it. In 1982, the station was moved approximately 500 m upstream where clams were found in numbers similar to those at station AC. Page 404 The Veliger, Vol. 28, No. 4 40° 40'N 36’ 32’ 28' 74°44'W 32’ Figure 2 Map of New Jersey, U.S.A. (right); shaded area is enlarged on the left. Stations: AC, experimental site; C, control site and location of natural population; DI, control site; G, United States Geological Survey (USGS) gauging station at Manville, NJ. Two groups of caged specimens of Corbicula fluminea (shell lengths ranging from 8 to 20 mm) were established and sampled in 1981. The first group, Notch I, consisted of animals collected on 17 August 1981. The ventral mar- gin of each animal was notched with a drill (bit size of 1.6 mm) to provide a reference point on the shell prior to planting in cages at stations DI and AC (one cage per station). Cages were 0.3 x 0.3 x 0.15 m open wood frames lined with 1 mm galvanized steel mesh. Substratum from each site was placed in the cage to a depth of approxi- mately 10 cm and 50 specimens of C. fluminea were plant- ed in each cage, a density of 530 clams per m~’. The second group, Notch II, was composed of animals collect- ed on 16 September 1981 and notched as described for Notch I. For this group, cages consisted of plastic mesh (1.6 mm) bags filled with substratum from each station and anchored to the bottom. Fifty clams were placed in each bag, yielding the same cage densities as the Notch I group. One cage with Notch II specimens was placed at the two control stations (DI and C) and one perturbed (AC) station. Samples of Notch I and II animals were collected and sacrificed on five and three occasions, re- spectively, through January 1982. There were massive mortalities of C. fluminea in spring 1982 in all cages, as well as in the natural population at station C. To continue the experiment, new animals were collected from the re- juvenated population at station C. In 1982, cages were constructed and sampled by Dr. Angela Cristini as part of a study of the use of adenylate energy charges to detect stress (CANTELMO-CRISTINI et al., 1983). Shells of caged and uncaged specimens were sup- plied to us by Dr. Cristini. Cages were 0.9 x 0.9 x 0.25 m wood frames, with the top, bottom, and a portion of two opposing sides (0.9 x 0.1 m openings) composed of 1.6-mm mesh cloth. Two cages were placed at each site. Substratum from station C was placed in each cage to a depth of 0.2 m. On 19 July 1982, approximately 200 clams were placed in each cage, a density of 250 clams per m~*. The ventral shell margin of these animals was not notched. Three animals were sampled from the cages at each station and the natural population at station C on six dates through December 1982. Examination of Shell Growth Patterns A single valve from each specimen was embedded in epoxy resin and radially sectioned (along the height axis from umbo to ventral margin). The two cross-sectional surfaces were finely ground with 600-grit carborundum powder, polished with diamond compounds on lapidary wheels, etched for 30 secs in 0.9 N HCl, rinsed in distilled water, and air-dried; one was used to prepare an acetate peel replica of the microstructural growth patterns (KENNISH et al., 1980) while the other was used to prepare a thin shell section (CLARK, 1980). To determine the periodicity of microgrowth increment formation, the number of microgrowth increments (mean of three counts) in specified regions of the shell was de- DOW. Eritz é& Ro A; Lutz, 1986 Page 405 Figure 3 Photographic enlargement of a thin shell section of a specimen of Corbicula fluminea from the Notch I group collected from station DI on 19 September 1981. Clam was notched on 17 August 1981 (NI). Total shell height is 19.6 mm. All other features are as in Figure 1A. termined using the acetate peel under a compound micro- scope at 100 magnification. Counts were made in shell regions in which the dates at the beginning or end of the count, or both, were known. The distance (in um) between the two points was measured along the shell exterior sur- face, which is the surface of maximum growth (SMG) (PANNELLA & MACCLINTOCK, 1968) in radial shell sec- tions. Measurements were made from either the peel or thin section using a compound microscope (at 40x) equipped with a calibrated ocular reticle. The shell height corresponding with the beginning and(or) endpoints of an increment count was measured to the nearest 0.1 mm us- ing either (1) a pair of calipers using the unembedded valve of each specimen, or (2) an ocular reticle at 40x using the acetate peel or thin section. Height measure- ments were chord distances from the umbo to the shell exterior surface along a straight line between the two points and not along the shell exterior surface. Statistical pro- cedures used (analysis of variance, least-squares linear regression, Student’s t-test and Kruskal-Wallis test) were those of SOKAL & ROHLF (1969). RESULTS Natural Shell Growth Patterns Collections in 1981 Notching the ventral shell margins of clams on 17 Au- gust (Notch I) or 16 September 1981 (Notch II) produced a growth cessation mark, NI or NII, in each radial section (Figures 3-5). The notch divided each shell section into regions deposited before (toward the umbo, or dorsal) and after (toward the shell margin, or ventral) it. Exact lo- cation of the growth cessation mark caused by notching was aided by the location of the V-shaped notch on the exterior shell surface. In 25 of 44 Notch I and 28 of 32 Notch II clams (or 53 of 76 clams sectioned in 1981), there was a recogniz- able series of microgrowth increments in the outer layer and two growth lines within the inner layer (Figure 3; see also Figure 1A). A growth cessation mark (W1) formed the ventral boundary of the group of microgrowth incre- ments, and was associated with one growth line in the inner layer. Another growth cessation mark (GCM) in the outer layer, located between 1.2 and 3.7 mm ventral from W1, was associated with the other growth line. The growth lines within the inner layer will hereafter be re- ferred to as discontinuities, because, as it will be shown, each resulted from a period of little or no shell growth. There are two lines of evidence supporting the hypoth- esis that the series of microstructures described above was formed between late fall 1980 and spring 1981. First, Notch I and II clams without growth discontinuities in the inner layer were an average of 6.9 and 4.2 mm smaller in shell height, respectively, at the time of notching than those with discontinuities (Table 1A). Mean shell heights of clams without discontinuities, 9.1 mm in mid-August (Notch I) and 12.4 mm in mid-September (Notch II), could be attained by young-of-the-year, or the spring 1981 brood (BRITTON & Morton, 1979). These shell heights were similar to those at mark W1 in clams with discon- tinuities (Table 1B). These data suggest that clams with the series of microgrowth increments in the outer layer and growth discontinuities in the inner layer were mem- bers of the 1980 year-class. Winter 1980-1981 was rep- resented as growth cessation mark W1 and an associated discontinuity in the inner layer. No clams in either the Notch I or II groups had other growth discontinuities in the inner layer or growth cessation marks in the outer layer that would correspond with winter 1979-1980; thus, all clams were most likely either members of the 1980 (1 Page 406 The Veliger, Vol. 28, No. 4 Figure 4 Photographic enlargement of a thin shell section of a specimen of Corbicula fluminea from the Notch I group collected from station DI on 14 November 1981. Clam was notched on 17 August 1981 (NI). Total shell height is 10.2 mm. Dashed lines mark portion enlarged in Figure 5. All other features are as in Figure 1A. L. W. Fritz & R. A. Lutz, 1986 Table 1 Shell height (mm) at the notch (Table 1A) and at the winter 1980-1981 growth cessation (W1; Table 1B) in Notch I (notched on 17 August 1981) and Notch II (notched on 16 September 1981) groups of Corbicula flu- minea with and without growth discontinuities (GD) in the inner shell layer. Collections from all dates and sta- tions were pooled. With GD Without GD Group N Mean Range N Mean Range A. Shell height (mm) at notches Notch I 25) NOW: WAS ily 9.1 6.3-12.3 Notch II 28 16.6 13.1-19.2 4 12.4 11.1-13.5 B. Shell height (mm) at W1 Notch I 25 8.8 4.2-12.3 Notch II 28 9.0 S0S1ie3 Total 53 8.9 4.2-12.3 year old; Figure 3) or 1981 year-class (young-of-the-year; Figure 4). The second group of data that aids in dating the mi- crostructures described above was microgrowth increment counts and measurements of shell growth between W1 and NI or NII. Data obtained from these shell regions (and from clams sampled in 1982) suggest that micro- growth increments were periodically deposited, because the number of increments was independent of the amount of shell deposited by both Notch I (7? = 0.02) and Notch II (7? =0.18) clams (Figure 6). The mean number of increments from W1 to NI or NII in each group suggested an average deposition rate of approximately one incre- ment per day. Data obtained from clams moved to the three stations and notched on the same day were pooled because there were no significant differences in shell growth (Kruskal-Wallis test: Notch I: H = 0.71, P > 0.1; Notch II: H = 4.18, P > 0.1) nor in the mean number of incre- ments counted (Notch I: t = 0.01, P > 0.9; Notch II: F = 2.35, P > 0.1) among clams in each group (Table 2). In Notch I clams, the grand mean (+95% confidence inter- val) number of increments from W1 to the notch was 136.3 (+6.2). If one increment were formed each day, the “mean” date of growth resumption after W1 would be 3 April, or 136 days before 17 August. Similarly, the grand mean number of increments in Notch II clams was 160.5 Page 407 (+5.5), placing the ““mean” date of growth resumption on 9 April, or 160 days before 16 September. These two independent estimates of the date of growth resumption are quite similar, and for purposes of this discussion, the grand “mean” date of growth resumption in spring 1981 is 6 April, or halfway between the two dates. Further support for an average daily deposition rate of micro- growth increments was seen in the difference in mean number of increments from W1 to each notch (24.2), which was similar to the number of days between notch dates (30; Figure 6). Resumption of growth in early April 1981 might also have been predicted on the basis of the water temperature record for the Raritan River and the reported temperature tolerances of Corbicula fluminea (RODGERS et al., 1979). From late November 1980 to late March 1981, water temperatures near station C were 7°C or below (U.S. Geol. Survey Water-Data Report NJ-81-1, 1982). A prolonged period of valve closure and inactivity could be reflected in the shell as a growth discontinuity and cessation mark as in Figures 1A and 3 (see LuTz & RHOADS, 1977). Be- tween 25 March and 7 April 1981, water temperatures near station C increased from 7 to 12°C (Figure 7), which could have stimulated shell deposition. The close agree- ment between the estimated date of growth resumption from increment counts and the water temperature record supports both of the following hypotheses: (1) discontin- uance of growth in winter 1980-1981 was reflected in shell microstructure as a discontinuity within the inner layer and W1 in the outer layer, which followed deposi- tion of a recognizable series of microgrowth increments in fall 1980, and (2) microgrowth increments in the outer layer were formed at an average rate of one per day from W1 to each notch. Assuming that winter 1980-1981 was reflected in shell microstructure as in hypothesis (1) above, then GCM was formed subsequently, possibly during spring 1981 (Fig- ures 1A, 3). As can be seen in Figure 7, the mean daily discharge of the Raritan River near station C increased over 50-fold, from 6.3 to 342.6 m*:sec™'!, from 10 to 12 May 1981 as a result of 10 cm (4 inches) of rain in the Raritan River watershed. This was the highest mean daily flow recorded during the two-year study period. Based on increment counts from W1, GCM could have resulted from the increase in turbidity and high flow rates asso- ciated with this storm. As in the shell region from W1 to each notch, the number of increments from W1 to GCM was independent of the amount of shell deposited (7? = Figure 5 Light micrograph of the portion of outer fine crossed-lamellar layer outlined in Figure 4 showing the growth cessation mark resulting from notching on 17 August 1981 (NI). Note the narrow microgrowth increments deposited after (to the right of) the growth cessation mark, as well as the greater proportion of “crossed-lamellar” micro- structures. Growth is to the right and the horizontal field width is 0.7 mm. Page 408 200 aA 180 4 ” bas 4A Cc is A g 160 s , 8 oO Aa A 4 : ee ray e@ e = 440 ® ee ( ) = oo. eA G e@ 1120 € — z 100 80 6 8 10 The Veliger, Vol. 28, No. 4 ay A is ba Notch ll iN mean = 160.5 ‘ i Ps 24.2 A | Notch | mean =136.3 e e e 12 14 16 Shell Growth (mm) Figure 6 Number of microgrowth increments in the outer layer of Notch I (solid circles) and II (open triangles) groups of Corbicula fluminea, from W1 (see Figure 1) to the notch, as a function of shell growth (see Table 2). The mean numbers of increments in each group are shown, along with the difference between the two means. 0.0003; Figure 8). There were no significant differences in the mean number of increments from W1 to GCM in both of the following groups of tests (Table 3): (1) among clams collected from different stations within each notched group (Notch I: ¢=1.02, P > 0.2; Notch II: F = 0.09, P > 0.75), and (2) between separately pooled Notch I and II clams (¢t = 1.82, P > 0.05). Pooling increment counts from the 53 Notch I and II clams resulted in a grand mean of 34.3 (40.9) increments, which placed the date of GCM formation on 10 May 1981, or only two days before the date of highest mean flow. Average growth rates along the SMG (shell height axis) from W1 to Notch I and II were 69 and 62 wm/increment (day), respectively, with a total range of 44-108 wm/in- crement. Individual shell length increases from W1 to Notch I and II, when divided by the mean number of increments in each group (136 and 160 respectively) yield- ed mean daily growth rates of 63 and 55 wm/day along the length axis, respectively, with a total range of 41-87 um/day. These rates were calculated from clams with initial (at W1) shell heights ranging from 4.2 to 12.3 mm (Table 1), and lengths ranging from 5.6 to 13.6 mm. Post- Table 2 Shell growth and number of microgrowth increments from W1 to the notch in Notch I and II groups of Corbicula fluminea (see Table 1). Collections from all dates were pooled. Shell growth (um) Group Station N Median Notch I DI 12 9500 AC 1 8830 Total 25 9350 Notch II DI 12 9840 C 5 8960 AC 11 10,400 Total 28 10,000 Range 8300-12,200 7000-14,150 7000-14,150 7150-14,920 8140-10,240 7040-12,800 7040-14,920 Number of microgrowth increments Mean +95% CI Range 136.3 129.7-142.8 118.0-154.0 136.2 125.1-147.3 93.7-160.7 136.3 130.1-142.5 93.7-160.7 166.5 157.5-175.5 135.7-185.0 152.3 134.2-170.4 129.3-168.7 EV ot 149.4-166.0 138.7-178.3 160.5 155.0-166.0 129.3-185.0 Wr britz 6 RaAe Wutz, 1986 Page 409 12 May 350 300 250 200 150 100 Mean Daily Discharge (m3/sec) 50 1981 (Do) BAnjesodwa] 1982 Figure 7 Mean daily discharge (solid line) and water temperature (symbols and dashed line) of the Raritan River at the gauging station at Manville, NJ (see Figure 2). Data from U.S. Geological Survey Water-Data Reports NJ-81-1 and NJ-82-1, and from USGS-WRD, 418 Federal Building, 402 E. State St., Trenton, NJ 08608. Some water temperatures were measured at station C during this study. notch growth rates along the SMG of Notch I clams de- clined to a mean of 25 wm/increment (range of 14-44 ym /increment), a decline of over 60% from pre-notch rates regardless of the station to which each clam was moved or the date of collection. Post-notch growth rates of Notch II clams were negligible at all stations and from all col- lection dates. As will be shown with reference to collec- tions of unnotched clams in 1982, the large decline in post- notch growth rates in 1981 was most likely a result of notching and not an effect of the cage or station to which clams were moved. Collections in 1982 The growth disturbance mark in shell microstructure caused by moving clams from the natural population at station C to cages at the three stations on 19 July 1982 was less distinct than that caused by notching the ventral margin in 1981. The move was reflected in microstructure as a growth cessation mark (M) that was translucent in thin section. In all specimens moved to control stations DI and C, an opaque region in the outer shell layer was deposited ventral to M (Figure 9). This opaque region was generally not observed in post-move shell growth of clams moved to station AC (Figure 10). Identification of the move disturbance in clams moved to stations DI, C, and AC was based on shell growth measurements and counts of microgrowth increments in shell regions ventral and dorsal to M, as well as its absence in clams collected from the wild population at station C (Figure 11). Analyses of shell dorsal to M revealed the presence of a single discontinuity in the inner shell layer associated with a recognizable series of two or three growth cessation marks in the outer shell layer of 45 of 50 clams moved to experimental or control stations (Figures 9, 10). This growth pattern was also observed in 14 of 17 clams sam- pled from the natural population at station C (Figure 11). As with the Notch I and II clams collected in 1981, it will be shown that this series of microstructures was caused by a growth discontinuance in the winter of 1981-1982; clams without this series of microstructures were members of the spring brood of 1982. The ventral-most growth cessation mark in the series of two or three will hereafter be referred to as W2. Page 410 The Veliger, Vol. 28, No. 4 45 40 A A sve a e@ 4 CN S oO A e @a = Om AR Notch | and Il ‘ 35 Boe oe ge e e@ 4 mean = 34.3 £ O o 2 = =) z 1.0 1.5 2.0 3.0 3.5 4.0 Shell Growth (mm) Figure 8 Number of microgrowth increments in the outer layer of Notch I (solid circles) and II (open triangles) groups of Corbicula fluminea, from W1 to the growth cessation mark (gcm; see Figures 1B and 3), as a function of shell growth (see Table 3). The mean number of increments for the two groups combined is shown. Measurements of shell growth and counts of micro- growth increments from W2 to M yielded results similar to those from collections in 1981: (1) there was no cor- relation between the number of increments and the amount of shell growth (7? = 0.21; similar in pattern to Figure 6), suggesting that microgrowth increments were periodically deposited, and (2) the mean number of increments from W2 to M suggested an average deposition rate of one increment per day. Data obtained from clams moved to the three stations were pooled because there were no sig- nificant differences in shell growth (Kruskal-Wallis test: H = 1.06, P > 0.5) nor in the mean number of increments counted (F = 0.07, P > 0.75) among clams in each group (Table 4). Thus, prior to entering the period of monitored growth (or before 19 July 1982), there were no significant differences in growth among the one experimental and two control groups of clams. Using the grand mean of 113.9 (+5.6) increments from W2 to M, placed the “mean” date of growth resumption after winter on 27 March 1982 (114 days before 19 July). Resumption of Shell growth and number of microgrowth increments from W1 to growth cessation mark, GCM (see text), in Notch I Table 3 and II groups of Corbicula fluminea (see Table 1). Collections from all dates were pooled. Group Notch I Notch II Shell growth (um) Number of microgrowth increments Station N Median Range Mean +95% CI Range DI 12 2000 1700-2600 34.1 32.1-36.1 26.3-38.7 AC 13 2000 1350-3700 32.8 30.8-34.8 26.7-38.7 Total 25 2000 1350-3700 33.4 32.1-34.7 26.3-38.7 DI 12 2280 1250-3200 35.4 33.1-37.7 27.0-40.7 Cc 5 1660 1300-2350 34.7 31.8-37.6 33.0-38.7 AC 11 2130 1350-3200 34.9 32.4-37.4 27.3-40.0 Total 28 2140 1250-3200 35.0 33.7-36.3 27.0-40.7 Grand total 53 2000 1250-3700 34.3 33.4-35.2 26.3-40.7 L. W. Fritz & R. A. Lutz, 1986 Page 411 Figure 9 Photographic enlargement of a thin shell section of a specimen of Corbicula fluminea collected from the cage at station C on 20 October 1982. Growth cessation caused by moving clam from the natural population to the cage on 19 July 1982 is labelled (M), as is the winter 1981-1982 growth cessation (W2). Total shell height is 14.9 mm; growth is to the right. growth in late March could also have been predicted from the water temperature record (Figure 7). The average growth rate along the SMG from W2 to M for the 45 clams was 45 um/increment (day), with a range of 25-94 wm/increment, or approximately 20 um/ increment less than in 1981. Dividing individual increases in shell length from W2 to M by the mean number of increments (114) yielded a mean daily shell length in- crease of 46 wm/day, with a range of 28-78 um/day. These rates were calculated from clams with initial (at W2) shell heights ranging from 4.8 to 10.4 mm and lengths ranging from 6.2 to 11.7 mm. Shell Growth during Monitored Periods in 1982 Because growth of clams during spring and early sum- mer 1982 was similar in the groups moved to the three Figure 10 Photographic enlargement of a thin shell section of a specimen of Corbicula fluminea collected from the cage at station AC on 20 October 1982. Growth cessation marks are labelled as in Figure 9. Note differences in post-move shell growth between this clam and the one collected from the control station (Figure 9). Total shell height is 15.7 mm; growth is to the right. Page 412 The Veliger, Vol. 28, No. 4 Figure 11 Photographic enlargement of a thin shell section of a specimen of Corbicula fluminea collected from the natural population at station C on 20 October 1982. Growth cessation mark is labelled as in Figure 9. Total shell height is 15.5 mm; growth is to the right. stations, any differences in shell growth between groups subsequent to the move would most likely be due to site- specific factors. As previously stated, the growth distur- bance associated with the move and placement in cages in 1982 had a much less deleterious effect on subsequent growth than did notching the ventral shell margins in 1981 (compare Figure 9 with Figures 3-5). Shell depos- ited after the notch resembled the immediate post-move shell in that both were relatively opaque in thin section and growth rates (um/increment) were depressed. Me- dian shell growth from W2 to the shell margin in three of the six collections (30 July, 27 August, and 17 Novem- ber) was considerably greater in wild clams than those from control stations, but this was most likely a result of the smaller mean shell height at W2 in the wild clams collected on these dates (Table 5). Counts of microgrowth increments from W2 to the shell margin in clams from control stations DI and C were not significantly different from counts in the same shell region in clams from the natural population (F = 1.11, P > 0.25; Table 5; Figure 12). Furthermore, the number of increments counted from W2 to the shell margin in the four collections through October (or those collections before which water temper- atures were greater than 10°C) suggests an average de- position rate of one increment per day. A linear regression of the pooled counts from wild and control clams against days since 19 July 1982 resulted in a slope (1.18) that was not significantly different from 1.00 (t, = 1.52, P> 0.2; n = 28, r? = 0.86; see Figure 12). From 19 July 1982 to the date of collection, clams moved to experimental station AC grew slower and deposited fewer increments than caged clams at control stations (Ta- ble 6). A linear regression of increments from M to the shell margin against days since 19 July in clams collected from the control stations through October resulted in a slope (0.97) which was not significantly different from Table 4 Shell growth and number of microgrowth increments from the winter 1981-1982 growth cessation mark (W2) to the growth cessation mark caused by the move (M) on 19 July 1982 in specimens of Corbicula fluminea. Collections from all dates were pooled. Shell growth (um) Station N Median Range DI 17 5390 3640-9290 Cc 15 5260 3210-6810 AC 13 3880 2015-9940 Total 45 5150 2015-9940 Number of microgrowth increments Mean +95% CI Range 115.0 108.6-121.4 99.3-139.3 112.6 101.1-124.1 79.7-162.3 114.1 100.0-128.2 81.7-150.3 113.9 108.3-119.5 79.7-162.3 eWeeknitzase Raye Lutz,, 1986 Page 413 Table 5 Shell growth and number of microgrowth increments from W2 (see Table 4) to the shell margin in specimens of Corbicula fluminea collected on six dates in 1982. Collections from stations DI and C were from cages, while those from NP were from the natural population at station C. Date Mean shell : : eae tied eh Shell growth (um) Number of microgrowth increments lection Station N W2 (mm) Median Range Mean Range 30 Jul DI 3 7.6 5310 4280-5680 125.1 112.0-137.3 Cc 3 8.7 3585 3560-5640 110.4 94.3-132.0 NP 2 6.1 7000 6310-7690 120.7 115.7-125.7 27 Aug DI 2 8.0 7300 5960-8650 147.5 142.7-152.3 Cc 2 Sari 5740 5350-6140 145.8 142.0-149.7 NP 2 2.9 12,530 11,310-13,750 149.0 142.3-155.7 23 Sep DI 3 8.8 8740 8440-8820 171.8 165.3-175.7 C 2 8.6 8390 7900-8880 186.0 176.7-195.3 NP 1 6.6 7620 138.7 20 Oct DI 3 8.1 10,280 7460-12,340 220.7 213.0-235.7 C 2 6.2 10,720 10,690-10,740 193.5 193.0-194.0 NP 3 6.9 10,060 9250-12,250 231.4 218.7-238.3 17 Nov DI 3 8.4 9430 9400-9450 222.9 214.3-230.3 Cc 3} 9.0 9380 8380-10,440 221.9 211.7-231.7 NP 3 5:9 12,000 10,000-13,120 234.6 215.7-271.7 15 Dec DI 3 6.6 11,010 10,320-14,150 246.5 234.7-268.7 C 3 8.0 9310 8750-11,120 256.1 217.3-290.3 NP 3 6.3 11,250 10,620-13,250 247.0 232.0-266.3 1.00 (t, = —1.36, P > 0.2; n = 24, r? = 0.98). However, 10, r? = 0.94; see Figure 13A). This strongly suggests that, a similar linear regression based on counts from clams on the average, clams at station AC were growing on collected from station AC had a slope (0.79) that was fewer days than those at stations DI and C. Furthermore, significantly lower than 1.00 (¢, = —3.03, P < 0.02; n = both the absolute amount of shell growth (Figure 13B) ‘Table 6 Shell growth and number of microgrowth increments from M (see Table 4) to the shell margin in specimens of Corbicula fluminea collected from cages at the three stations on six dates in 1982. Shell growth (um) Number of microgrowth increments Date of Se a ae ee a Ee ee ee collection Station N Median Range Mean Range 30 Jul DI 3 450 390-550 14.3 12.3-18.0 Cc 3 350 220-380 14.3 13.3-15.0 AC 3 230 170-310 Use 14.7-15.3 27 Aug DI 3 2320 2300-3220 41.3 40.0-42.3 C 3 2080 1800-5720 38.7 38.0-39.7 AC 2 440 160-570 29.5 22.7-36.3 23 Sep DI 3 3340 3120-3530 68.6 61.3-76.7 Cc 3 2750 2700-3560 64.2 63.3-65.0 AC D, 1520 1310-1740 60.0 56.3-63.7 20 Oct DI 3 3580 3320-4080 94.5 91.3-96.3 Cc 3 4620 4090-5810 94.0 90.7-96.0 AC 3 1530 820-2000 78.1 67.0-84.0 17 Nov DI 3 4060 3960-4390 99.4 91.3-105.3 C 3 3500 3370-3620 108.0 104.7-110.3 AC 1 1660 88.3 15 Dec DI 3 5220 4910-5330 129.6 128.0-131.0 C 3 3880 3500-4620 123.2 119.3-128.0 AC 3 950 870-1000 58.8 53.0-61.7 Page 414 300 Number of Increments oo 2. a > 80 — | 4 lee — 1 30 27 23 20 17 15 JUL AUG SEP OCT NOV DEC Date of Collection Figure 12 Mean (symbols) and range (vertical lines) of the number of microgrowth increments from the winter 1981-1982 growth ces- sation mark (W2; see Figures 9-11) to the shell margin in spec- imens of Corbicula fluminea collected from cages at stations DI (solid circles) and C (solid triangles) and from the natural pop- ulation at C (open triangles) in 1982. Date of collection is plotted three days before actual date for station DI and three days after for the natural population to permit plotting (see Table 5). and growth rate (Figure 13C) were lower in clams at AC than at the control stations. Post-move growth rates dur- ing the first 11 days (or through 30 July) at all stations were lower than pre-move rates. This may have been due to a period of acclimation to the site and(or) cage, and may also be related to deposition of the opaque region immediately ventral to M by control clams. Growth rates at control stations DI and C returned to pre-move levels by the August sample, while those at AC were lower than pre-move rates in all subsequent samples (Figure 13C). Evidence strongly suggests that the effluent discharged near station AC decreased the number of days of growth as well as growth rates of specimens of Corbicula fluminea relative to those at control stations. DISCUSSION In the present study, we have shown that microgrowth increments in the outer fine crossed-lamellar shell layer The Veliger, Vol. 28, No. 4 of Corbicula fluminea (described previously by PREZANT & Tan-Tiu [1985]) were formed at the rate of approxi- mately one per day and can be used to date shell regions. Furthermore, cessation of growth in winter resulted in a growth discontinuity within the inner complex crossed- lamellar layer and a growth cessation mark in the outer layer. Pericdically deposited growth patterns in the two shell layers and evidences of growth cessations were used to reconstruct the growth history of groups of C. fluminea exposed to natural and anthropogenic environmental per- turbations. Other methods, such as examination of exter- nal shell growth lines or serial measurements of shell axes (length or height) may not have been sensitive enough to document these changes in such short-term monitoring studies. The periodicity of formation of microgrowth incre- ments in outer prismatic or crossed-lamellar layers of sev- eral other bivalve species has been investigated previously by a number of researchers (see LUTZ & RHOADS, 1980). In the most commonly analyzed species, Mercenaria mer- cenaria, several investigators have documented a solar dai- ly periodicity of formation in subtidal populations (PANNELLA & MACCLINTOCK, 1968; KENNISH & OLSSON, 1975; THOMPSON, 1975; KENNISH, 1980; FRITZ & Ha- VEN, 1983), but there is also evidence that suggests a closer correlation with the lunar day in intertidal specimens (PANNELLA, 1976). Tidally deposited growth increments have also been observed in intertidal populations of Cer- astoderma edule (RICHARDSON et al., 1979) and Clinocar- dium nuttalli (EVANS, 1972). Longer cycles, such as sea- sonal changes in temperature, are often reflected (and most easily discerned) in middle and inner shell layers, such as those of M. mercenaria (FRITZ & HAVEN, 1983), Mya arenaria (MACDONALD & THOMAS, 1980), and Mytilus edulis (LUTZ, 1976). Similarly, in Corbicula fluminea, short cycles (days) are reflected in outer layer microgrowth in- crements and long cycles (seasons) in inner layer growth discontinuities. Caution must be exercised in using growth patterns to reconstruct life histories of individual specimens. This is due to both subjectivity in the method of detecting and counting increments (CRABTREE et al., 1979/1980; HUGHES & CLAUSEN, 1980) and natural variability with- in a bivalve population in growth rate (7.e., number of increments deposited and their width in a specified time) due to age and individual differences in sensitivity to en- vironmental stresses (KENNISH & OLSSON, 1975; CRAB- TREE et al., 1979/1980; RICHARDSON et al., 1980; FRITZ & HAVEN, 1983). This does not imply, however, that at- tempts to interpret growth patterns in bivalve shell struc- tures should be avoided. On the contrary, use of growth patterns in carbonate and proteinaceous secretions to de- termine age, growth rates, and aspects of life history of both vertebrates and invertebrates is well founded in stud- ies of population dynamics and ecology (see RICKER, 1975). One strives to be as accurate as possible by analyzing large numbers of shells, applying objective criteria to the defi- L. W. Fritz & R. A. Lutz, 1986 140 120 100 Number of Increments 30 27 23 20 17 15 JUL AUG SEP OCT NOV DEC Date of Collection Shell Growth (mm) 30 27 23 20 17 15 JUL AUG SEP OCT NOV DEC Date of Collection Growth Rate (um/inc) W2 to M 30 27 23 20 17 15 JUL AUG SEP OCT NOV DEC Date of Collection Figure 13 Analyses of shell growth of specimens of Corbicula fluminea at stations DI (circles and long dashed lines), C (triangles and short dashed lines), and AC (squares and solid lines) from the growth disturbance mark caused by the move on 19 July 1982 (M; see Figures 9 and 10) to the shell margin (date of collection). Offset on X-axis is the same as in Figure 12 (see Table 6). A. Mean Page 415 nition of increments and patterns in shell sections (such as those of CRABTREE et al., 1979/1980), and being con- sistent in interpretation and analysis. KENNISH & OLSSON (1975) were the first to use bivalve shell growth patterns to monitor environmental pertur- bations. Growth cessation marks, a decrease in micro- growth increment width, and “replacement” of prismatic with “crossed-lamellar” microstructures in the outer layer of specimens of Mercenaria mercenaria were directly cor- related with increased exposure to elevated water temper- atures from a nuclear power plant. In the present study, the number of days of growth and shell growth rates of specimens of Corbicula fluminea, as measured through analyses of microstructural banding patterns, decreased (relative to controls) with exposure to the combined ef- fluents from chemical and sewage treatment plants. Re- sults of a concurrent im situ study of physiological re- sponses to the efHuent (CANTELMO-CRISTINI et al., 1983) support these observations. In 1982, CANTELMO-CRISTINI et al. (1983) measured total adenylates and calculated ad- enylate energy charge of each of the specimens whose shells were sectioned for growth analysis in the present study. Energy charge of specimens was lower at station AC than at the control sites on 7 of 9 sampling dates during the monitored period, an indication of a site-spe- cific stress at AC which was, most probably, chronic ex- posure to the effluent. It is common in shell growth studies of this nature (e.g., RICHARDSON et al., 1980; FRITZ & HAVEN, 1983) to notch the ventral shell margin to induce a size-time benchmark in shell microstructure. As shown in the present study, notching should be avoided because it caused a decrease in growth rate and alteration of shell microstructure in the 1981 group. Alternative methods of inducing a growth cessation mark in shell microstructure, such as thermal shock (RICHARDSON et al., 1979; FRITZ & HAVEN, 1983) or moving the animal to a new location (FRITZ & HAVEN, 1983; this study) cause less alteration to shell microstruc- ture and apparently less damage to mantle tissue. Size-specific growth rates of Corbicula fluminea in the Raritan River, New Jersey, were slower than those mea- sured in Lake Benbrook, Texas, for periods of similar duration (BRITTON et al., 1979). In the present study, mean growth rates along the length axis in spring and summer of 1981 and 1982 (periods of 136 and 160 days in 1981 and 114 days in 1982) were 63 and 55 um/day, and 46 uwm/day, respectively, for uncaged clams with ini- _— (symbols) and range (vertical lines) of number of microgrowth increments. B. Median (symbols) and range (vertical lines) in shell growth. C. Median (symbols) and range (thin vertical lines) in growth rate (um/increment). Also shown are distributions of growth rates from W2 to M (or prior to the monitored growth phase) in each group. Top and bottom of heavy bar are 75th and 25th percentiles, respectively, of the distributions of pre- move growth rates. Page 416 tial shell lengths ranging from 4.2 to 12.3 mm. In Lake Benbrook, the mean growth rate of specimens held in containers for 107 days was 54 wm/day, but initial shell lengths ranged from 10 to 25 mm. Measured growth rates in the two studies were similar, but the larger initial shell lengths of clams in the Texas study, and the fact that smaller clams tend to grow faster than larger clams (BRITTON et al., 1979) indicated that size-specific growth rates were slower in the Raritan River. This difference may actually be greater because container-held specimens tend to grow slower than uncaged individuals (BRITTON et al., 1979). The largest specimen of Corbicula fluminea collected to date from the Raritan River is 25 mm in shell length and was found dead in March 1981 (TRAMA, 1982). The larg- est specimen analyzed in the present study was 20.8 mm in shell length at the end of its second growing season (age 1+ years). Consequently, it is likely that the 25 mm spec- imen was between 2 and 3 years old at the time of death, and not 3 to 4 years old as concluded by TRAMa (1982), which changes the latest possible year in which the species became established in the Raritan River system to 1979 instead of 1978. The relatively small maximum size and slow size-specific growth rates may result from low tem- peratures and(or) their duration in winter, because the Raritan River is the northernmost extension of the re- corded range of C. fluminea along the Atlantic seaboard (TRAMA, 1982). However, other factors such as food sup- ply and quality, bottom type, and flow regime may also be involved in creating a “dwarfed” population. Causes of the catastrophic mortalities of C. fluminea observed in the Raritan River (and in other lotic systems [BRITTON & MortTon, 1982]) are also not known. Research into causes of “dwarfism” and mortalities would be a logical direction for future studies of the population dynamics of C. fluminea in the Raritan River. ACKNOWLEDGMENTS We are grateful to Ellen Suchow for initiating the project and to Hank Garie and Bruce Ruppel of New Jersey Department of Environmental Protection for support and guidance. Sincere thanks to Drs. Angela Cantelmo, Susan Ford, and Robert Prezant for critical reviews of the manu- script; Louis Chiarella for technical and photographic as- sistance; Rutgers University Cartography Laboratory for drafting figures; and Beverly Webster for final production of the manuscript. Special appreciation to Alyce Fritz. LITERATURE CITED Berry, W. B. N. & R. M. BaRKER. 1975. Growth increments in fossil and modern bivalves. Pp. 9-25. In: G. D. Rosen- berg & S. K. Runcorn (eds.), Growth rhythms and the his- tory of the earth’s rotation. John Wiley & Sons: London. BRITTON, J. C., D. R. Coxpiron, L. P. Evans, Jr., C. GOLIGHTLY, K. D. O’KANE & J. R. TENEYCK. 1979. Re- evaluation of the growth pattern in Corbicula fluminea The Veliger, Vol. 28, No. 4 (Miller). Pp. 177-192. In: Proceedings of the First Inter- national Corbicula Symposium, Fort Worth, Texas. BRITTON, J. C. & B. Morton. 1979. Corbicula in North America: the evidence reviewed and evaluated. Pp. 249-287. In: Proceedings of the First International Corbicula Sym- posium, Fort Worth, Texas. BRITTON, J. C. & B. MorTON. 1982. A dissection guide, field and laboratory manual for the introduced bivalve Corbicula fluminea. Malacol. Rev., Suppl. No. 3. 82 pp. CANTELMO-CRISTINI, A., F. Hospop & R. LAZELL. 1983. In situ studies on the adenylate energy charge of populations of Corbicula fluminea in a freshwater system. International Council for the Exploration of the Sea Publication No. CM 1983/E:53. 18 pp. CxiarK, G.R., I. 1980. Study of molluscan shell structure and growth lines using thin sections. Pp. 603-606. Jn: D. C. Rhoads & R. A. Lutz (eds.), Skeletal growth of aquatic organisms: biological records of environmental change, Vol. 1, Geobiology Series. Plenum Press: New York. Counts, C. L., III & R. S. PREZANT. 1982. Shell microstruc- ture of Corbicula fluminea (Bivalvia: Corbiculidae). Nautilus 96:25-30. CRABTREE, D. M., C. D. CLAUSEN & A. A. ROTH. 1979/1980. Consistency in growth line counts in bivalve specimens. Pa- laeogeography, Palaeoclimatology, Palaeoecology 29:323- 340. Dopp, J. R. & E. L. Crisp. 1982. Non-linear variation with salinity of Sr/Ca and Mg/Ca ratios in water and aragonitic bivalve shells and implications for paleosalinity studies. Pa- laeogeography, Palaeoclimatology, Palaeoecology 38:45-56. Evans, J. W. 1972. Tidal growth increments in the cockle, Clinocardium nuttalli. Science 176:416-417. Fritz, L. W. & D. S. HAVEN. 1983. Hard clam, Mercenaria mercenaria: shell growth patterns in Chesapeake Bay. Fish. Bull. 81:697-708. HuGHEs, W. W. & C. D. CLAUSEN. 1980. Variability in the formation and detection of growth increments in bivalve shells. Paleobiology 6:503-511. Jones, D. S. 1980. Annual cycle of shell growth increment formation in two continental shelf bivalves and its paleoeco- logical significance. Paleobiology 6:331-340. KENNISH, M. J. 1980. Shell microgrowth analysis: Mercenaria mercenaria as a type example for research in population dynamics. Pp. 255-294. In: D. C. Rhoads & R. A. Lutz (eds.), Skeletal growth of aquatic organisms: biological rec- ords of environmental change, Vol. 1, Geobiology Series. Plenum Press: New York. KENNISH, M. J., R. A. Lutz & D. C. RHoapDs. 1980. Prep- aration of acetate peels and fractured sections for observa- tions of growth patterns within the bivalve shell. Pp. 597- 601. In: D. C. Rhoads & R. A. Lutz (eds.), Skeletal growth of aquatic organisms: biological records of environmental change, Vol. 1, Geobiology Series. Plenum Press: New York. KENNISH, M. J. & R. K. OLsson. 1975. Effects of thermal discharges on the microstructural growth of Mercenaria mer- cenaria. Environ. Geol. 1:41-64. Lutz, R. A. 1976. Annual growth patterns in the inner shell layer of Mytilus edulis (L.). J. Mar. Biol. Assoc. U.K. 56: 723-731. Lutz, R. A. 1981. Electron probe analysis of strontium in mussel (Bivalvia, Mytilidae) shells: feasibility of estimating water temperature. Hydrobiologia 83:377-382. Lutz, R. A. & D. C. RHoaps. 1977. Anaerobiosis and a theory of growth line formation: micro- and ultrastructural growth patterns within the molluscan shell reflect periodic respiratory changes. Science 198:1222-1227. Peer bnitzGa Re A Lutz. O86 Lutz, R. A. & D. C. RHoaDs. 1980. Growth patterns within the molluscan shell: an overview. Pp. 203-254. In: D. C. Rhoads & R. A. Lutz (eds.), Skeletal growth of aquatic organisms: biological records of environmental change, Vol. 1, Geobiology Series. Plenum Press: New York. MacDonaLp, B. A. & M. L. H. THomMas. 1980. Age deter- mination of the soft-shell clam Mya arenaria using shell internal growth lines. Mar. Biol. 58:105-109. PANNELLA, G. 1976. Tidal growth patterns in Recent and fossil mollusc bivalve shells: a tool for the reconstruction of paleotides. Naturwissenschaften 63:539-543. PANNELLA, G. & C. MacCLintTock. 1968. Biological and en- vironmental rhythms reflected in molluscan shell growth. J. Paleontol. 42:64-80. PREZANT, R. S. & A. Tan-Tiu. 1985. Comparative shell mi- crostructure of North American Corbicula (Bivalvia: Sphaeriacea). Veliger 27:312-319. Ruoaps, D. C. & G. PANNELLA. 1970. The use of molluscan shell growth patterns in ecology and paleoecology. Lethaia 3:143-161. RICHARDSON, C. A., D. J. Crisp & N. W. RUNHAM. 1979. Tidally deposited growth bands in the shell of the common cockle, Cerastoderma edule (L.). Malacologia 18:277-290. RICHARDSON, C. A., D. J. Crisp & N. W. RUNHAM. 1980. Factors influencing shell growth in Cerastoderma edule. Proc. Roy. Soc. Lond. 210:513-531. RICKER, W. E. 1975. Computation and interpretation of bio- Page 417 logical statistics of fish populations. Fish. Res. Bd. Canada, Bull. No. 191: 382 pp. Ropcers, J. H., Jk., D. S. CHERRY, K. L. Dickson & J. CAIRNS. 1979. Invasion, population dynamics, and elemental ac- cumulation of Corbicula fluminea in the New River, VA. Pp. 99-110. In: Proceedings of the First International Corbicula Symposium, Fort Worth, Texas. Rye, D. M. & M. A. Sommer, II. 1980. Reconstructing pa- leotemperature and paleosalinity regimes with oxygen iso- topes. Pp. 169-202. In: D. C. Rhoads & R. A. Lutz (eds.), Skeletal growth of aquatic organisms: biological records of environmental change, Vol. 1, Geobiology Series. Plenum Press: New York. SoOKAL, R. R. & F. J. ROHLF. 1969. Biometry. The principles and practice of statistics in biological research. W. H. Free- man, Co.: San Francisco. 776 pp. Taytor, J. D., W. I. KENNEDY & A. HALL. 1973. The shell structure and mineralogy of the Bivalvia. II. Lucinacea- Clavagellacea. Conclusions. Bull. British Museum (Natur. Hist.) 22:255-294. TuHompson, I. 1975. Biological clocks and shell growth in bi- valves. Pp. 149-162. In: G. D. Rosenberg & S. K. Runcorn (eds.), Growth rhythms and history of the earth’s rotation. John Wiley & Sons: London. TRAMA, F. B. 1982. Occurrence of the Asiatic clam Corbicula fluminea in the Raritan River, New Jersey. Nautilus 96: 6-8. The Veliger 28(4):418-425 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 Reproductive Cycle in the Freshwater Mussel Diplodon chilensis chilensis (Mollusca: Bivalvia) SANTIAGO PEREDO and ESPERANZA PARADA Department of Natural Sciences, Catholic University of Chile-Temuco, Casilla 15-D, Temuco-Chile Abstract. The reproductive cycle of the freshwater mussel Diplodon chilensis chilensis (Gray, 1828) was investigated between March 1982 and April 1983 by monthly examination of gonad sections and inspection of demibranchs in the females. The gonad is active throughout the year in both sexes, suggesting the existence of an annual cycle with continuous gametogenesis. A spawning phase occurs during the spring-summer months. Spawning is partial and asynchronous in the specimens of the Chilean population studied from Lake Villarrica. No regression in gametogenic activity or rest period was observed. INTRODUCTION SEVERAL ASPECTS of reproduction in freshwater mussels show a number of marked differences and specializations among the species that have been studied. Much of the available information concerns unionids of North America and Europe. Virtually all members of the Sphaeriidae are hermaphroditic (ZUMOFF, 1973), whereas the Unionidae has been reported to be generally comprised of dioecious species (VAN DER SCHALIE, 1969). Some members of the Margaritiferidae have been reported to be dioecious, with occasional hermaphroditic specimens being recorded (HENDELBERG, 1960; HEARD, 1970; VAN DER SCHALIE, 1970). The life histories of numerous unionacean species have shown marked differences. These differences include the relative life-span of the species, the size and number of broods produced each year, the season in which gameto- gensis is most active, and the period and number of spawnings within a year. Freshwater mussels from Chile are representative of the Hyriidae, a family that has re- ceived little attention concerning its life history. The pres- ent study was undertaken to elucidate the annual repro- ductive cycle of Diplodon chilensis chilensis, a hyriid abundant in lakes and rivers of Chile. This species may prove useful as an indicator of environmental changes due to freshwater pollution considering the value of freshwater mussels as qualitative indicators of pesticides, radionu- clides, and other substances as has been reported for sev- eral Unionidae (NELSON, 1962; MILLER et al., 1956; LEE & WILSON, 1969; BEDFORD et al., 1968) and Margariti- feridae (HENDELBERG, 1960; BJORK, 1962; McMILLAN, 1966). Explanation of Figures 1 to 7 Figure 1. Male gonad of Diplodon chilensis chilensis at the be- ginning of the fall-winter period (April). Gonadal follicles have abundant spermatozoa (S) in the lumen and immature gametes located at the periphery of the follicles (arrow heads). x20. Figure 2. Male gonad of Diplodon ch. chilensis at the end of the winter (August). Spermatozoa (S) fill all the intrafollicular spaces, and immature gametes are confined to the periphery of the fol- licles (arrow heads). x 20. Figures 3 and 4. Male gonads of Diplodon ch. chilensis in the spring-summer period (September and December, respectively). Spermatozoa (S) are less abundant than in preceding months S. Peredo & E. Parada, 1986 Page 419 p 4 Naa tue Sal * ad . (Figures 1 and 2) and immature gametes occupy larger areas foliicles (S) and immature gametes to the periphery close to the within the follicles (arrow heads) which are partially full at this follicle walls (IG). x20. time (spawning period). x20. Figures 6 and 7. Female gonad of Diplodon ch. chilensis in the Figure 5. At the end of the spring-summer period (March) male fall-winter period (April and June, respectively). Female folli- follicles are partially full, with sperm located at the center of the cles contain abundant oocytes (arrow heads). x 20. Page 420 MATERIALS anpD METHODS In March 1982, specimens of Diplodon chilensis chilensis were randomly collected from shallow waters (20-60 cm depth) of Lake Villarrica (39°17’'S, 72°13’W). The indi- viduals (564 in total) were placed in eight size classes (5-mm intervals) ranging from 16 to 65 mm in total length. The individuals of each size class were placed in wire cages (25 x 25 x 25 cm, 5-mm mesh). The cages were kept in a stream flowing from the Cautin River in the trout hatchery of Lautaro (39°17'S, 72°30'W). Five individuals representing each size class were ran- domly selected each month from March 1982 to April 1983. The viscera were fixed in aqueous Bouin’s fixative. After embedding in paraffin, 7-um serial sections were cut and stained with hematoxylin and eosin. Ten to 15 sec- tions through different regions of the gonads of each spec- imen were examined under the light microscope to deter- mine the type and abundance of germinal cells present in the gonads during the study period. Chi-square (x?) analysis was used for sex-ratio deter- minations and Student’s f-tests were employed to deter- mine eventual differences in follicular areas occupied by spermatozoa and by clusters of immature germ cells throughout the year. The water temperature was recorded daily and the monthly mean temperature was calculated. RESULTS Diplodon chilensis chilensis is dioecious and its gonads are ramified organs bearing numerous follicles closely packed among the intestinal coils (PEREDO & PARADA, 1984). Males Inspections of gonadal sections revealed the presence of abundant gametes at various stages of maturation and abundant spermatozoa. These cells could be recognized by the morphological features described by PEREDO & PARADA (1984) in a study of gametogenesis in Dzplodon chilensis chilensis. There were no major changes throughout the year either in the type or in the abundance of germinal cells present in the gonadal follicles. In April, the gonadal follicles contained abundant sper- matozoa in the lumen, and clusters of germinal cells at various stages of maturation were located at the periphery of the follicles (Figure 1). The same features were appar- ent in gonadal sections during the fall-winter months (April-August). At the end of winter (August) there was a slight increase in the number of spermatozoa, which by then almost filled all of the intrafollicular spaces; the cell clusters of immature gametes were confined to the pe- riphery of the follicles (Figure 2). From September to February there was a decrease in the number of spermatozoa and, conversely, an increase in the size of the cell clusters within the follicles. These changes were especially evident during September and November (Figures 3, 4). At the end of the spring-sum- The Veliger, Vol. 28, No. 4 Table 1 Diplodon chilensis chilensis: follicular areas occupied by spermatozoa and immature germ cell clusters. Percentage of area occupied Percentage of by immature area occupied germ Month (1982-1983) by spermatozoa cell clusters March 61.69 32.67 April 67.72 26.52 May 70.10 20.52 June 61.96 32.55 July 58.99 DES) August 60.87 33.44 Fall—winter mean area (X) 63.53 28.83 September STESI SEZ October 42.66 49.82 November 23.85 68.08 December 42.15 55.81 January 37.78 S)2) February 52.30 43.21 Spring-summer mean area (X) 39.38 54.56 mer months (February—mid-March) the follicles were partially full, with sperm located at the center of the fol- licles and with the cell clusters at the periphery, close to the follicular walls (Figure 5). The cell population dynamics within the follicles, mea- sured as the area of the follicle occupied by the sperma- tozoa versus that occupied by immature germinal cell clus- ters, showed that from March through August, the follicular area occupied by spermatozoa was larger than the area occupied by the cell clusters. From September until February, the converse was observed: the immature cell clusters occupied a larger intrafollicular area than did sperm (Table 1). Student’s t-tests showed that the differ- ences observed in the areas occupied by spermatozoa and by cell clusters in the different months throughout the year are Statistically significant (P = 0.05). The interstitial tissue did not show differences in the number or type of cells present throughout the year. Cells similar to the amoebocytes described by TRANTER (1958) were abundant in the tissue surrounding the follicles, es- pecially in the areas close to the walls. These amoebocyte- like cells were not seen except occasionally within the follicles, either free or in close contact to germinal cells. No differences in the histological features of the gonads throughout the year were found in the specimens of the different size classes studied, except, as expected, in the size and number of follicles. Females Gonadal follicles in the females showed few changes in the germ cell populations related to the types and numbers S. Peredo & E. Parada, 1986 Page 421 Explanation of Figures 8 to 11, 14 Figure 8. Female follicle in the fall-winter period, with oogonia embedded in the walls and previtellogenic (PV), vitellogenic (V), and full-grown oocytes (F). x50. Figure 9. At the end of the fall-winter period (August) increases in vitellogenic (V) and full-grown (F) oocytes are seen in the female follicles. x20. Figure 10a. Gravid female in November with developing em- bryos (E) (blastulae) in the inner demibranchs (ID). Numerous oocytes can be seen in the gonadal follicles (F) at this time. x 20. Figure 10b. Gravid female in the same month (November) with more advanced embryos (E) in the inner demibranch (ID). Go- nadal follicles (F) contain oocytes. x20. Figure 11. Glochidia in the inner demibranch of a gravid female in November. Water tubes (W) of the gill can be seen. x 20. Figure 14. Male gonadal follicle showing tissue degeneration. Interstitial cells (1) can be seen within the follicle in contact with gametes. x50. Page 422 Table 2 Diplodon chilensis chilensis: monthly mean temperature, percentage of gravid females, and total females exam- ined. Percentage Total Monthly mean _ of gravid females Month (1982-1983) temperature females examined March 15.19 — 46 April 12.85 — 48 May 10.14 -- 59 June WES? _ 55 July 7.96 — 56 August 8.12 — 36 September 9.86 12 25 October 9.94 68 22 November We) 60 52 December 16.05 64 33 January 16.89 36 28 February 15.69 4 45 March 14.28 29 14 of oogonia and oocytes in the sections examined through- out the year. During the fall-winter period, the female follicles con- tained abundant oocytes (Figures 6, 7) which, according to their morphology, correspond to previtellogenic, vitel- logenic, and even vitelline (full-grown) oocytes (Figure 8). Embedded in the follicle walls were oogonia (Figure 8). During August increases in vitellogenic and vitelline oo- cytes were seen (Figure 9). During the spring-summer period (September through February) similar characteristics were seen in the sections, with gonadal follicles containing various types of oocytes. No empty follicles were observed throughout the year. During the spring-summer season a slight decrease was observed in the number of oocytes within the follicles. Gravid females containing embryos in the inner demi- branchs were observed from September until March. This condition was also verified by examination of gonad smears. October, November, and December were the months with the highest percentages of gravid females, with 68, 64, and 60% respectively (Table 2). In the same months, em- bryos in various developmental stages were present, in- cluding glochidia (Figures 10a, b, 11). The demibranchs of both sides of a gravid female contained embryos, which were all at the same stage of development; that is, the embryos housed in the inner demibranchs of a female were either zygotes, blastulae, gastrulae, or glochidia. As in males, no empty follicles were observed in the sections examined throughout the year, including the pe- riod in which gravid females were present. No differences in the histological characteristics of the gonads were observed throughout the year in females of the different size classes studied. The Veliger, Vol. 28, No. 4 DISCUSSION Examination of the gonadal sections of Diplodon chilensis chilensis reveals that in this species the gonad is continu- ously active throughout the year. All types of gametogenic cells, including mature gametes, can be seen in the go- nadal follicles at the same time throughout most of the year. However, slight variations in the relative quantity of gametes within the follicles observed during the seasons of the year suggest the existence, in males and in females, of a continuous single reproductive cycle. In this cycle of Diplodon ch. chilensis a proliferation and maturation phase can be recognized that occurs throughout the year, but more intensively during the fall-winter months, and is characterized by an intensive maturation process; at the end of the winter (August) the follicles are packed with mature gametes. The spawning phase occurs only during the spring-summer months and is characterized by a par- tial evacuation of mature gametes; the gonadal sections indicated that the gonadal tissue was active, with the fol- licles producing early stages of gametogenesis. Continuous gametogenic activity reported in the present paper has been described in the reproductive cycle of sev- eral freshwater bivalves, both dioecious (VAN DER SCHALIE & VAN DER SCHALIE, 1963; GHOSH & GHOSE, 1972; SMITH, 1979) and hermaphroditic (HEARD, 1965, 1975; ZUMOFF, 1973), although there are variations in the dif- ferent species studied. The proliferation and maturation phase observed in Di- plodon ch. chilensis is more evident in males than in fe- males, probably due to the accumulation of food reserves in the oocytes; consequently, this phase of the reproductive cycle is less noticeable. The proliferation and maturation of gametes in males is distinguishable from spawning by means of differences in the intrafolicular areas occupied by mature gametes (spermatozoa) versus immature ga- metes (cell clusters) during the seasons in which these processes occur. Thus, during the fall-winter season the area occupied by spermatozoa is larger than that occupied by cell clusters in the follicles. This difference becomes greater at the end of the winter (August) as a result of the intensive maturation of gametes at this time (Table 1). Conversely, during the spring-summer seasons the in- trafollicular area occupied by immature gametes (cell clusters) is larger than that occupied by spermatozoa, and larger than that occupied by the cell clusters during the previous season, thus indicating evacuation of mature ga- metes and proliferation of immature gametes (Table 1, Figure 12). Spawning is characterized as partial because during this period (September-March) the gonadal follicles of males and females are not depleted of gametes. On the contrary, they are partially full with growing and mature gametes (Figures 3-5, 10, 11). Spawning is also asyn- chronous in the individuals of the population studied, as shown by the occurrence of embryos at different stages of S. Peredo & E. Parada, 1986 Page 423 % 20 immature germ cell clusters A % 40 sk x ages ate * Ky, Sarat a % EY Se es > LY Kh 60 x tv py 30 spermatozoa ay x 80 10 M A M J J A S (e) N D J F M A 1982 1983 Figure 12 Follicular areas (in %) occupied by spermatozoa and cell clusters (immature gametes) in the reproductive cycle of Diplodon chilensis chilensis during the study period (t-test, P < 0.05). development housed in the inner demibranchs of the fe- males examined in the same month. This partial and asynchronous spawning explains the prolonged spawning period, extending from September until March, observed in Diplodon ch. chilensis. The continuous gonadal activity and the almost year-round occurrence of mature gametes in the gonads of Diplodon ch. chilensis could indicate the existence of two or more spawning periods within the year in this species as described for several freshwater bivalves (ZUMOFF, 1973; HEARD, 1975; CHUNG, 1980; KENMUIR, 1981). However, this situation seems to be unlikely in Diplodon ch. chilensis because gravid females were not present in the fall-winter months (April-August) (Table 2). Based on these observations it can be concluded that Diplodon ch. chilensis is a spring-summer breeder, unless gametes released during the fall-winter months for some reason were not fertile, an unlikely situation from the point of view of the reproductive effectiveness of this species. As with other reports on the reproductive cycles of sev- eral mollusks, the spawning period observed in Diplodon ch, chilensis coincided with increasing water temperatures in the study area (Figure 13). The role of temperature in the reproductive cycle of mollusks has been reviewed by several authors (GIESE, 1959; FRETTER & GRAHAM, 1964; GIESE & PEARSE, 1974; SAsTRY, 1977), and water tem- perature could be the factor that induces spawning in Diplodon ch. chilensis. Specifically, it would be triggered when the water temperature reached values equal to or higher than those registered from September on. Lower temperatures would in turn determine the interruption of gamete emission during the fall-winter months, despite the presence of mature gametes (at least morphologically) in the gonadal follicles of males and females. Consequent- ly, in Diplodon ch. chilensis abundant mature gametes are kept for months in the gonads before their release, a sit- uation also reported by HEARD (1970) in the freshwater mussel Margaritifera falcata. According to the results obtained in the present study, Diplodon ch. chilensis is a seasonal breeder with continu- ous gonadal activity, being coincident in this respect with other freshwater mussels. HEARD (1965) observed mature Page 424 The Veliger, Vol. 28, No. 4 spawning a T°c 19 gametogenesis Figure 13 Monthly means and average ranges of water temperature of the stream flow in the trout hatchery of Lautaro during the study period. gametes throughout the year in populations of Anodonta umbecilis and Anodonta peggyae. GHOSH & GHOSE (1972) reported a similar situation in the spermatogenesis of Lamellidens marginalis, and NAGABHUSHANAM & LOH- GAONKER (1978) reported continuous gonadal activity and seasonal breeding in Lamellidens corrianus. Diplodon ch. chilensis also coincides in this aspect of reproduction with freshwater clams. Woops (1931) observed mature ga- metes throughout the year in Sphaerium striatum, as did ZUMOFF (1973) in Sphaerium simile, a freshwater clam of the Northern Hemisphere. In the latter species fertiliza- tion occurs throughout the year, resulting in embryos housed in the brood sacs year round and differing from Diplodon ch. chilensis in which gravid females are found in the spring~summer months only. The reproductive cycle of Diplodon ch. chilensis also shows differences from those reported for other freshwater bivalves. For instance within the Sphaeriidae, Musculium securis is a hermaphroditic species in which the gonads are spent during the winter months; it has been reported as having one fertilization period during its life-span, which is only one year (MACKIE et al., 1976). Lamellidens cor- rianus is a functional hermaphrodite that has a gonadal cycle that includes spent and recovery stages in addition to growing, maturing, and spawning stages (NAGAB- HASHANAM & LOHGAONKER, 1978). Spent and recovery stages were not observed in the present study. Within the Margaritiferidae, SMITH (1978) reported two gametogen- ic cycles per year with only one spawning period in a population of Margaritifera margaritifera from northeast- ern North America. In New England Margaritifera mar- garitifera has been reported as dioecious, with a gameto- genic period from mid-May to mid-August (spring- summer), mature gametes in the gonadal follicles in July, and spawning in August. Following evacuation of mature gametes a few remain in the gonads, being resorbed after a few weeks (SMITH, 1979). Cells of the interstitial tissue have been reported to have an active role in the gonadal cycle of several bivalves, phagocytizing residual gametes or providing nutrients to developing gametes. In such cases, amoebocytes or phago- cytes have been observed within gonadal follicles in close relation to gametes. In the present study, such a situation was not observed. Interstitial tissue cells similar to those described by ‘TRANTER (1958) were always seen in the interstitial spaces or close to the follicle walls. In only one or two of the specimens examined, interstitial cells of the type mentioned above were observed within the gonadal follicles, providing an anomalous situation of gonadal tis- sue degeneration of unknown origin (Figure 14). This evidence suggests that interstitial cells do not have an ac- tive role in the normal gonadal cycle in Diplodon ch. chi- lensis. The differences observed in the reproductive cycle of Diplodon ch. chilensis compared with other freshwater mussels point out the notable variations existing among S. Peredo & E. Parada, 1986 freshwater mollusks with respect to various reproductive aspects including life-span, sexuality, gonadal activity, number of spawning and breeding periods, and number of births in the year. HEARD (1965) points out that these differences are present in the various species of a genus and occasionally in different populations of a species that inhabit different latitudes (intraspecific variations). These considerations should be kept in mind to temper gener- alizations or extrapolations of results obtained in the study of the reproductive behavior of one species to another species or related groups. These considerations should also be kept in mind in the study of reproduction in popula- tions of the same species occupying distant geographical subranges or markedly different environmental condi- tions. ACKNOWLEDGMENTS This work was supported with funds from the Direccién de Investigacion, Catholic University of Chile-Santiago and Comision de Investigacion, Catholic University of Chile-Temuco. The authors are indebted to and wish to thank the trout hatchery of Lautaro for the use of facilities during the execution of this investigation and Carlos Jara of Universidad Austral de Chile for his critical reading of the manuscript. LITERATURE CITED BEDFORD, J. W., E. W. ROELEFS & W. J. ZABIK. 1968. The freshwater mussel as a biological monitor of pesticide con- centrations in a lotic environment. Limnol. Oceanogr. 13: 118-126. Byork, S. 1962. Investigations on Margartifera margaritifera and Unio crasus. Acta Limnol. 4:1-109. CHUNG, E. Y. 1980. Reproductive cycle and breeding season of the freshwater clam, Anodonta (Sinanodonta) woodiana (Lea). Bull. Korean Fish Soc. 13(4):135-144. FRETTER, V. & A. GRAHAM. 1964. Reproduction. Pp. 127- 163. In: K. M. Wilbur and C. M. Yonge (eds.), Physiology of Mollusca. Academic Press: New York. Guosu, C. & K. C. GHosE. 1972. Reproductive system and gonadal activities in Lamellidens marginalis (Simpson, 1900). Veliger 14(3):283-288. GigEsE, A. C. 1959. Comparative physiology: annual repro- ductive cycles of marine invertebrates. Ann. Rev. Phys. 21: 547-576. GigEsE, A. C. & J. S. PEARSE. 1974. General principles. Pp. 1-49. In: A. C. Giese and J. S. Pearse (eds.), Reproduction of marine invertebrates. Vol 1. Academic Press: New York. HEARD, W. H. 1965. Comparative life histories of North American pill clams (Sphaeriidae: Pistdium). Malacologica 2:381-411. HEARD, W.H. 1970. Hermaphroditism in Margaritifera falcata Page 425 (Gould) (Pelecypoda: Margaritiferiidae). Nautilus 83:113- 114. HeEarD, W. H. 1975. Sexuality and other aspects of repro- duction in Anodonta (Pelecypoda: Unionidae). Malacologia 15:81-103. HENDELBERG, J. 1960. The freshwater pearl mussel: Margar- itifera margaritifera (L.). Rep. Inst. Freshw. Res. Dottningh. No. 41:149-171. KENMUIR, D. H. S. 1981. Repetitive spawning behaviour in two species of freshwater mussels (Lamellibranchiata: Unionacea) in Lake Kariba. Trans. Zimbabwe Sci. Assoc. 60(8):49-56. LEE, G. F. & W. WILSON. 1969. Use of chemical composition of freshwater clam shells as indicators of paleohydrologic conditions. Ecology 50:990-997. Mackig, G. L., S. U. Qapri & A. H. CLaRKE. 1976. Devel- opment of brood sacs in Musculium securis. Bivalvia: Sphae- riidae. Nautilus 88(4):109-111. McMILLAN, N. F. 1966. Margaritifera margantifera (L.) in hard water in Scotland. J. Conchol. 26:69-70. MILLER, C. W., B. M. ZUCKERMAN & A. J. CHARIG. 1966. Water translocation of diazinom-C' and parathion-S* off a model cranberry bog and subsequent occurrence in fish and mussels. Trans. Amer. Fish. Soc. 95:345-349. NAGABHUSHANAM, R. & A. L. LOHGAONKER. 1978. Seasonal reproductive cycle in the mussel, Lamellidens corrianus. Hy- drobiologia 61:9-14. NELSON, D. J. 1962. Clams as indicators of Strontium 90. Science 137:38-39. PEREDO, S. & E. PARADA. 1984. Gonadal organization and gametogenesis in the freshwater mussel Diplodon chilensis chilensis (Mollusca: Bivalvia). Veliger 27(2):127-134. Sastry, A. N. 1977. Reproduction of pelecypods and lesser classes. Pp. 137-151. In: A. C. Giese and J. S. Pearse (eds.), Reproduction of marine invertebrates. Academic Press: New York. SmitTH, D. G. 1978. Biannual gametogenesis in Margaritifera margaritifera in northeastern North America. American Malacological Union (1978):49-53. SmiTH, D. G. 1979. Sexual characteristics of Margaritifera margaritifera (Linnaeus) populations in central New En- gland. Veliger 21(3):381-383. TRANTER, D. J. 1958. Reproduction in Australian pearl oys- ters (Lamellibranchia). II. Pinctada albina (Lamark): ga- metogenesis. Aust. J. Freshw. Res. 9:144-158. VAN DER SCHALIE, H. 1969. Two unusual unionid hermaph- rodites. Science 163:1333-1334. VAN DER SCHALIE, H. 1970. Hermaphroditism among North American freshwater mussels. Malacologia 10(1):93-112. VAN DER SCHALIE, H. & A. VAN DER SCHALIE. 1963. The distribution, ecology and life history of the mussel, Actino- naias ellipsiformis (Conrad), in Michigan. Occ. Pap. Mus. Zool. Univ. Mich. 633:1-17. Woops, F. 1931. The history of the germ cells in Sphaeri1um striatum. J. Morphol. 51:545-595. ZumorF, C. H. 1973. The reproductive cycle of Sphaerium simile. Biol. Bull. 144:212-228. The Veliger 28(4):426-428 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 Egg Capsule and Young of the Gastropod Beringius (Neoberingius) friele (Dall) (Neptuneidae) by RICHARD A. MacINTOSH Northwest and Alaska Fisheries Center, National Marine Fisheries Service, National Oceanic and Atmospheric Administration, 7600 Sand Point Way N.E., Seattle, Washington 98115, U.S.A. Abstract. The egg capsule and capsule young of the gastropod Beringius frielei are described for the first time. Two egg capsules, one containing two young, were collected in the eastern Bering Sea. The young closely resembled adults of the species, and the capsules were similar to those of other eastern Bering Sea members of the genus Beringius. Beringius (Neoberingius) frielei Dall, 1895, occurs in the eastern Bering Sea (Dall, 1895), off the east coast of Sa- khalin Island and in the northern part of the Okhotsk Sea (HaBE & ITO, 1972) and off Hokkaido Island (Pilsbry, 1907). The nominate race, B. (Neoberingius) frielei frieler, is found in the eastern Bering Sea from Unimak Pass to the Pribilof Islands at depths of 121 to 350 m (DALL, 1895; author’s unpublished data). Although Beringius friele: is common in the eastern Be- ring Sea, the egg capsule and young have not been de- scribed. On 11 July 1977, several specimens and two egg capsules of B. friele: were collected from a trawl haul made at a depth of 300 m north of Unimak Pass (55°24'N, 168°08'W). The adult snails (Figure 1) were cleaned and stored dry. The egg capsules, which were attached to an empty shell of Fusztriton oregonensis (Redfield, 1848), were preserved in alcohol. Although one capsule was empty and open along its distal perimeter, the other contained two well developed young that were easily recognized as B. frieler (Figure 2). Explanation Figure 1. Adult Beringius frielei taken at 300 m on the eastern Bering Sea shelf in same trawl haul with egg capsules. Figure 2. Well developed young of Beringius frielei taken from an egg capsule. Figure 3. Two egg capsules of Beringius frielei on empty shell of Fusitriton oregonensis. Each capsule was pouchlike with a single internal chamber. The two capsules were 18 and 21 mm high. Both were 27 mm wide and 7 mm thick. Their width decreased to 16 mm above the point of attachment. They were firmly cemented to the Fusztriton oregonensis shell by a flat expanded base measuring 18 x 25 mm. The com- mon base shared by both capsules (Figure 3) indicates that they were laid by a single female. As in other members of the genus, the capsule of Be- ringwus friele. was a complete envelope within an envelope (Cowan, 1964; MacINTosH, 1979) (Figure 4). Outer and inner layers were 0.15 and 0.10 mm thick, respectively. The outer surface of each capsule was pale yellow, smooth, and rubberlike, while the interior surface of the outer envelope was covered with numerous fine lamellae run- ning approximately parallel to the capsule base. These lamellae were 0.1-0.2 mm high and numbered 4-6 per mm. The outer surface of the inner envelope was circum- scribed with similar fine lamellae. The lining of the brood chamber was smooth and without macroscopic structural of Figures 1 to 4 Figure 4. Diagram showing capsule wall of Beringius frieler egg capsule. A, outer layer showing fine lamellae on inner surface; B, layer of slender yellow fibers; C, inner layer showing fine lamellae on outer surface. De Page 4 R. A. MacIntosh, 1986 Page 428 detail. Between the two envelopes was a layer of slender 20-25 mm long yellow fibers parallel to the lamellae. They were mostly unattached and loosely packed, allow- ing easy separation of inner and outer layers. Some fibers were partially attached to the inner wall of the outer en- velope. The two capsule young were 16.2 and 15.0 mm in length (Figure 2). One shell was broken near the anterior canal while the other was whole. The shells were elon- gate, acute, and consisted of 4% well-rounded whorls with a deep suture. The three unsculptured nuclear whorls were pink, whereas the post-nuclear whorls were white. The first nuclear whorl was covered by a thin, parch- mentlike film that made a shriveled apical cap. MacIntosh (1979) found similar caps on capsule young of Beringius beringi (Middendorff, 1849). The conspicuous sculptur- ing of the anterior quarter of the body whorl faded grad- ually towards the nuclear whorls. Spiral sculpture con- sisted of 34 evenly spaced, flattened close-set cords and an axial sculpture of fine but distinct incremental lines. Adults from the same area had this pattern in early whorls, but in the second to fourth post-nuclear whorls, the spiral cords became medially grooved or paired. The overall shape of the capsule young was similar to that of adults. The egg capsules of Beringius (Neoberingius) friele: bear a striking resemblance to those of B. (Neoberingius) turtoni (Bean, 1834) from the North Sea and Skagerrak and B. (Neoberingius) oassianus from Norway (THORSON, 1940). Even the capsules of the more distantly related B. eyer- dami Smith, 1959, and B. beringu are essentially identical in gross form and structure to those of B. frieler (GOWAN, The Veliger, Vol. 28, No. 4 1964; MacINTosH, 1979), suggesting a greater degree of affinity among these species than might be presumed from studies of comparative shell morphology. ACKNOWLEDGMENTS The late Robert R. Talmadge gave advice and enthusiasm to the project. Doyne W. Kessler and John H. Bowerman took the photographs. Dr. James H. McLean kindly re- viewed the manuscript. LITERATURE CITED Cowan, I. M. 1964. The egg capsule and young of Beringius eyerdami Smith. (Neptuneidae). Veliger 7(1):43-44, pl. 7. Da.L, W. H. 1895. Scientific results of explorations by the U.S. Fish Commission steamer Albatross. No. XXXIV.— Report on Mollusca Brachiopoda dredged in deep water, chiefly near the Hawaiian Islands, with illustrations of hith- erto unfigured species from northwest America. Proc. U.S. Natl. Mus. 17:675-733, pl. 23-32. Habe, T. & K. Iro. 1972. A new subspecies of Beringius (Neoberingius) frielei (Dall) from Okhotsk Sea. Venus 31(3): 113-114. MaclIntTosn, R. A. 1979. Egg capsule and young of the gas- tropod Beringius beringi (Middendorff) (Neptuneidae). Ve- liger 21(4):439-441, 1 pl. Pitspry, H. A. 1907. New and little-known whelks from northern Japan and the Kuril Islands. Proc. Acad. Natl. Sci. Phila. (1907):243-246, pls. 19, 20. THORSON, G. 1940. Notes on the egg-capsules of some North- Atlantic prosobranchs of the genus T7roschelia, Chrysodomus, Volutopsius, Sipho, and Trophon. Vidensk. Medd. Dan. Na- turhist. Foren. 104:251-266. The Veliger 28(4):429-435 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 The Systematic Position of Royella sinon (Bayle) (Prosobranchia: Cerithiidae) by RICHARD S. HOUBRICK Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract. Royella sinon (Bayle), a marine prosobranch representing a monotypic genus, is herein assigned to the family Cerithiidae Fleming on the basis of characters derived from the shell, soft anatomy, operculum, and radula. A white shell sculptured with two nodular spiral cords per whorl, a short shallow anterior siphon, and the concave base of the body whorl are distinctive. The corneous operculum is circular—-ovate, having few spirals and a central nucleus. The taenioglossate radula is typically cerithioid. Digitate papillae fringe the mantle edge, and within the mantle cavity a long monopectinate osphradium, a ctenidium comprised of long triangular filaments, and a thick, wide, open pallial oviduct with a large albumen gland are notable features. A pair of salivary glands that pass through the nerve ring, a midesophageal gland, and a large stomach with a short style sac, large cuticularized gastric shield, and complex sorting area indicate herbivory. The habitat is subtidal rubble bottoms. Development is inferred to be planktotrophic on the basis of the sculptured protoconch and distinct sinusigera notch. INTRODUCTION Royella sinon (Bayle) is an uncommon marine cerithiacean prosobranch of uncertain familial assignment that has a white shell about 20 mm long with rugose cancellate sculpture. It has an extremely wide geographic distribu- tion throughout the Indo-Pacific region. Its exact affinity to other cerithiacean groups has long been unknown due to lack of information about the soft parts and radula. Royella is a monotypic genus that has been assigned to the Potamididae H. and A. Adams, 1854, due to shell characters shared in common with some potamidid species such as Pirenella conica (Blainville), but this allocation is unsatisfactory. To resolve this problem I examined vir- tually all specimens of Royella in major national and in- ternational institutions and museums, but not a single pre- served animal or shell with a dried animal was found. In addition, requests in The Hawauan Shell News (HOUBRICK, 1984:12) for live-collected, preserved specimens were not successful, although some new locality records were ob- tained. Recently, Mr. Gustav Paulay collected Royella si- non alive in the Cook Islands and, through the kindness of Dr. Anders Warén, I was able to obtain the preserved specimen and study the soft anatomy and radula. The results are presented below and include a historical ac- count, synonymy, description of the shell, animal, and radula, and the new family allocation for this taxon. MATERIALS anp METHODS Specimens of this uncommon snail were examined from major museums and from private collections throughout the world for shell measurements. Variables included total shell length and width, aperture length and width, and length of the penultimate whorl. The single available pre- served specimen, a female, was dissected using methylene blue solution under a Wild M-5 dissecting microscope. The radula was removed, measured, and prepared for scanning electron microscope (SEM) study. Superficial and internal anatomy were studied, but swelling of the albumen gland due to the aqueous dissecting solution un- fortunately obscured and partially destroyed details of the pallial oviduct. Scanning electron micrographs of the rad- ula and operculum were made on a Cambridge 250 Mark II Stereoscan microscope. The following abbreviations have been used in this pa- per: AMNH—American Museum of Natural History; AMS—Australian Museum, Sydney; ANSP—Academy of Natural Sciences, Philadelphia; BMNH—British Mu- seum (Natural History); CAS—California Academy of Sciences; DMNH—Delaware Museum of Natural His- tory; HUJ—Hebrew University of Jerusalem; LACM— Los Angeles County Museum of Natural History; MNHNP—Museum National d’Histoire Naturelle, Paris;s NMNZ—National Museum of New Zealand; Page 430 NMV—National Museum, Victoria; USNM—National Museum of Natural History; WAM—Western Austra- lian Museum. Material examined and literature records RED SEA: Aqaba, Jordan (HUJ 21.311/9); Elat, Israel (HUJ 21.312/8). INDIAN OCEAN ISLANDS: Anse Boileau, Mahe, Seychelles (BMNH); Mauritius (BMNH). AUSTRA- LIA: North Australia (BMNH); outer reef, West Gun Id., Abrolhos Islands, Western Australia (WAM); Lodestone Reef, N of Townsville, Queensland (AMS); S Side Beach, York Id., Torres Strait, Queensland (AMS); Four Mile Beach, Port Douglas, Queensland (AMS); Saxon Reef, off Cairns, Queens- land (Thora Whitehead coll.); Norfolk Id. (AMS); 44 m off Lord Howe Id., 31°38’25”S, 159°03'W (AMS); Lord Howe Id. (AMS, NMV). KERMADEC ISLANDS: Raoul (Sunday Id.) (AMS, NMNZ MF 141616, USNM 214757); 29°17.2’S, 177°57.2'W, 27-29 m, E end of Denham Bay, Raoul Id. (NMNZ MEF26957); 29°15’S, 177°50.9'W, 31-45 m between Dayrell and Chanter Is., Herald Islets (NMNZ MF27068). JAPAN: Nada, Kii, Honshu (ANSP 224908); Shionomizaki, Kii, Honshu (ANSP 224769); Hachijojima Izu (ANSP 86166). RYUKYU IS- LANDS: Ryukyus (ANSP 243288, DMNH 80887, USNM 666629); Kikai, Osumi (MNHNP, USNM 273329, 175588, CAS, AMS); Osuma, Osumi (USNM 343916, MNHNP). PHILIPPINES: Baclayon Id., Bohol (A. Adams, 1855); Cebu (LACM 25166, MNHNP). TAIWAN; (Kuroda, 1941). PA- LAU: SW tip, Ngatpaet Passage, E Babelthuap (ANSP 202742). MARSHALL ISLANDS: Taka Atoll (USNM 615494); Bock Id., Rongerik Atoll (USNM 594667); Majuro Id., Majuro Atoll (Bob Purtymun coll.); lagoon side, Edgigen Id., Kwajalein (DMNH 93854); Enewetak Atoll (LACM 70-72, USNM 821778). NEW CALEDONIA: (MNHNP). LOYATY IS- LANDS: (BMNH); Lifu (ANSP 132658, 196063, ANS, AMNH, MNHNP). SAMOA ISLANDS: Poloa Bay, Tutuila (Bob Purtymun coll.). COOK ISLANDS: (MNHNP); off Kim- iangatau, Mauke Id. (USNM 842296); RAPA: mouth of Ahurei Bay (USNM 725617); E side of Tematapu Point (USNM 725691). SOCIETY ISLANDS: Papeete, Tahiti (Trondle Coll.); Papara, Tahiti (Trondle coll.). PITCAIRN ISLAND: off NW corner, Pitcairn (USNM 789325). HENDERSON ISLAND: (BMNH 1913.7.28.85.6). DESCRIPTION Family CERITHIIDAE Fleming, 1822 Royella Iredale, 1912 Royella IREDALE, 1912:219. Type-species, by monotypy: Cerithium clathratum Sowerby, 1855; WENZ, 1940:739, fig. 2140; THIELE, 1931:205. Diagnosis: Shell elongate, turreted, multi-whorled, hav- ing angulate whorls sculptured with two nodulose spiral cords and weaker axial riblets. Sculptured protoconch with sinusigera notch; early whorls cancellate; suture deep- ly impressed. Aperture circular, with short shallow ante- rior canal. Operculum corneous, circular, moderately spiral, with central nucleus. Radula_taenioglossate (2+1+1+1+2). Mantle edge fringed, osphradium mono- pectinate. Pallial gonoduct open. Paired salivary glands, esophageal gland, and large stomach with style sac, sort- ing area, and gastric shield present. The Veliger, Vol. 28, No. 4 Remarks: This monotypic genus is not well known in the literature. It was allocated to the Potamididae H. and A. Adams, 1854, by THIELE (1929:205) and WENz (1940: 739) on the basis of shell sculpture, but as this family comprises an intertidal estuarine group, it seems unlikely that Royella belongs here. The type-species was first as- signed to Cerithium Bruguiére, 1789, by SOWERBY (1855: 883) and later allocated to Pirenella Gray, 1847, by KoBELT (1895:173) and TRYON (1887:165) and to Ceri- thiopsis Forbes & Hanley, 1850, by MELVILL & STANDEN (1895:116). IREDALE (1911:320) pointed out its distinct- ness from these genera and proposed Royella to accom- modate it. IREDALE (1912:219) suggested that he had seen ““... other forms which appear to be congeneric .. .” and figured an undescribed species (pl. 9, fig. 3), but his il- lustration is poor and does not allow critical comparison with Royella sinon. The figured shell does not appear con- generic. Royella is herein assigned to the Cerithiidae on the basis of the radular and anatomical characters described in more detail below and in the discussion. Royella sinon (Bayle, 1880) Figures 1a-i, 2a, b Cerithium clathratum A. Adams (Cerithiopsis) in SOWERBY, 1855:883, pl. 185, fig. 258 (Holotype: BMNH; Type- locality: Baclayon Id., Bohol, Philippines; not De- shayes, 1833 nor Menke, 1828, nor Grateloup, 1832, nor Roemer, 1841.) SOWERBY, 1865:pl. 20, fig. 147. Cerithium (Pirenella) clathratum A. Adams: Kobelt in Mar- TINI-CHEMNITZ, 1895:173-174, pl. 32, fig. 13. Cerithtum sinon BAYLE, 1880:245 (new name for clathratum A. Adams, 1880); IREDALE, 1911:320. Cerithiopsis sinon (Bayle): MELVILL & STANDEN, 1895:116, pl. 1, fig. 3. Royella sinon (Bayle): IREDALE, 1912:219; H1RASE, 1936:54, pl. 84, fig. 18; Kira, 1962:26, pl. 13, fig. 13. Description: Shell (Figure 1a—h; Table 1). Shell elon- gate, turreted, reaching 29 mm in length and consisting of angulate whorls sculptured with two strong spiral, nod- ulose cords crossed by weak axial riblets. Numerous mi- croscopic incised spiral lines give silky appearance to shell. Wide anterior and posterior sutural ramps present on each whorl due to deeply impressed suture. Nodules formed where axial riblets cross spiral cords and tend to be point- ed; 23 axial riblets on penultimate whorl. Axial ribs more defined on early whorls, which have cancellate, pitted ap- pearance where axial and spiral elements cross. Proto- conch pink, with sinusigeral notch (Figure 1b). Body whorl sculptured with two major nodulose, spiral cords in mid- dle and with two, closely spaced, smooth, spiral cords above siphonal constriction. Base of body whorl concave. Ante- rior siphonal canal short, shallow, and slightly reflected upwards and to the left. Outer lip thin, nearly straight, but wavy where spiral cords end. Shell color white, but light tan maculations may be present on spiral cords be- tween nodules. Operculum corneous, thin, circular—ovate, with few spirals and central nucleus. R. S. Houbrick, 1986 Page 431 Figure 1 Shell and operculum of Royella sinon. a, Kikai, Osumi, Japan, 19.4 mm long (USNM 175588). b, protoconch showing larval sculpture and sinusigeral notch; 29°15'S, 177°50.9’N between Dayrell and Chanter Ids., Herald Islets, Kermadecs, New Zealand (NMNZ MF27068). c, Enewetak Atoll, Marshall Islands, 14.3 mm long (USNM 821778). d, Kikai, Osumi, Japan, 16.9 mm long. e, enlarged view of base of shell. f, Kikai, Osumi, Japan, 17 mm long (USNM 273329). g, enlargement of aperture. h, detail of sculpture on middle whorls. i, operculum, Mauke Id., Cook Ids., 1.8 mm diameter (USNM 842296). Page 432 The Veliger, Vol. 28, No. 4 Figure 2 Radula from Royella sinon, Mauke Id., Cook Ids. (USNM 842296). a, general view of radula. b, half row showing details of central, lateral, and marginal teeth. Radula (Figure 2a-b). Radula typically cerithioid. Central tooth ovate, wider than tall, having a short, point- ed central cusp flanked on each side by 3, sometimes 2, small denticles. Lateral tooth trapezoid with long lateral extension and buttressed central plate with short ventral extension. Marginals long, moderately curved at tips. Central pointed cusp of inner marginal flanked with 5 short, pointed inner denticles and 2 or 3 outer denticles. Outer marginal same, only smooth on outer surface. Animal. (This description is based on a single pre- served animal from Mauke Id., Cook Ids., collected under a rock on a rubble bottom in 20-24 m depth. Removed from the shell, the body comprised about 5.5 whorls.) The animal is cream colored with darker gray lines on the head-snout. The head is broad and has a short wide snout with a bilobed tip. The cephalic tentacles are thick, each with a small dark eye at the outer base of the peduncular stalk. The left tentacle is shorter than the right. The foot is long and narrow and has a deep crescent-shaped pro- podial groove (gland). Behind the anterior pedal gland the sole of the foot stains heavily with methylene blue indi- cating a mucus-secreting area. The end of the foot is pointed. The dorsal edge of the mantle is fringed with many digitate papillae, but is smooth ventrally. There is no obvious siphonal fold or indentation. The mantle cavity is deep, extending about 2.5 whorls. Within the mantle cavity, a long, dark brown, monopectinate osphradium extends the length of the ctenidium. Individual osphradial filaments are rectangular. The ctenidium is white, about 1.5 mm wide and extends the length of the mantle cavity. It is composed of long triangular filaments. The pallial oviduct is open, but was largely destroyed in the preserved specimen due to swelling of the albumen gland by water absorption. The buccal mass is elongate and relatively large and there is a pair of small jaws just inside the oral cavity. A short, wide, robust radular ribbon, 2.5 mm long and 0.5 mm wide, had 63 rows of teeth and was over one- seventh of the shell length (18.7 mm). A pair of large white, loosely coiled, salivary glands is present behind the nerve ring, and partially extends through it. The mid- esophagus is wide and has a large brown glandular area on its dorsal surface, which is the esophageal gland. The stomach is very large, about 2.5 whorls in length and comprises an extensive complex sorting area, a large cu- Table 1 Summary of shell measurements of Royella sinon (in mm). Statistic n Range D4 SD Length 45 11.1-28.6 17.91 4.44 Width 45 3.2-9.7 6.59 1.50 Aperture length 36 1.3-6.2 3.59 0.97 Aperture width 36 1.1-4.8 2.46 0.78 Body whorl length 25 3.5-10.4 6.86 1.50 Figure 3 Geographic distribution of Royella sinon (Bayle). 002 00+ 002 DSO/OY) O/1OpP + “SI S¥SaNOUWW %G ‘St ANTANOL: SOU S/AYD by ONI1Bay - 90302 ‘S| SAV1aHIAasS “yoy osnforr 2 yt 8/jom e ‘s! eictecn 6 eodos « aAnNINOUYVDS “apn «+ ‘ VNVISVA FY 42/9 « © 9 ay ‘Ss! ANIddITIHd 6 0u/ sy g /1227 a a | ascaiaos Or Sewn men emweetira ra Shiono tat or =e x aie pum J a - Page 434 ticular gastric shield and a short style sac. The intestine and rectum are large and contain transversely oriented, ovate fecal pellets. The nervous system is epiathroid and typically cerithioid in layout. DISCUSSION Although Royella sinon has a wide geographic distribution it is not a common species in museum collections and has not been well known to malacologists or conchologists. Most specimens are empty shells when collected, but liv- ing animals may be common in the proper habitat. The few records that cite details about collecting sites mention offshore, subtidal, coral-rubble habitats. The live-collect- ed specimen described herein was taken in a similar hab- itat by SCUBA. On the basis of the new data supplied by characters derived from analysis of the preserved specimen, I consid- er Royella to be a distinctive monotypic genus and assign it to the family Cerithiidae Fleming, 1822. The proto- conch and shell sculpture resemble those of some members of the genus Cerithium Bruguiére, 1798, but most Cerithi- um species are sculptured with three spiral cords per whorl and have an operculum with an eccentric nucleus; more- over, the short, shallow, anterior siphonal canal is atypical of cerithiids. The circular, spiral, corneous operculum with a central nucleus is much like those of some cerithiid gen- era such as Argyropeza Melvill & Standen, 1901, Brttiwm Gray, 1847, and Varicopeza Grundel, 1976, whereas the wide short snout, fringed mantle edge, and radula of Roy- ella most closely resemble those of Cerithium species. The monopectinate osphradium differs from the bipectinate condition in Cerithium species and is a unique, distinctive anatomical character of Royella. The radular morphology, stomach contents, fecal pellet composition, and the elab- orate stomach all indicate herbivory. A few sponge spic- ules were found in the stomach, but these are to be ex- pected in any algal-detritus feeding cerithiid. The subtidal habitat on rubble is similar to that of many cerithiids. Royella has been placed in the Potamididae, but there are no compelling reasons for this assignment. The fringed, digitate mantle edge is unlike that of potamidids, which is smooth, and most potamidids have long tapering ten- tacles and relatively extensible snouts; moreover, the os- phradium in all potamidids is a simple ridge. The radula of Royella is very unlike that of any potamidid species I have seen: members of the Batillariinae have distinctive cusps on the basal plate of the central tooth while the Potamidinae usually have narrow central teeth with long ventral extensions and marginal teeth with spatulate ser- rated tips and lateral flanges. They also have long style sacs and well developed crystalline styles. Nothing like these features are found in Royella sinon. The subtidal, purely marine habitat of Royella is also distinctly different from that of any potamidid. I had initially suspected that Royella might be a very large cerithiopsid and a sponge feeder, but it does not have The Veliger, Vol. 28, No. 4 an acrembolic proboscis, and is clearly a herbivore. The radula and protoconch are totally unlike those of ce- rithiopsid species (see MARSHALL, 1978). The shell of Royella sinon superficially resembles some triphorid shells such as Metaxia Monterosato, 1884, especially in the con- cavity of the base of the body whorl. Royella, however, has a much larger, bulkier shell than any cerithiopsid or triphorid species, attaining a length of 26 mm and a width of 9.4 mm. The uniquely sculptured shell does not resemble that of any other cerithiacean snail with the exception of Ce- rithium excavatum Sowerby, 1865, a species known only from SOWERBY’s figures (1865, 1866). I have not been able to find the holotype of C. excavatum, nor have I seen any specimens so labeled. The pictures in SOWERBY (1865, 1866) show that C. excavatum does not have the two spiral nodulose cords. It is thus best to regard C. excavatum as a nomen dubwum. The two nodulose spiral cords per whorl, deeply im- pressed sutural area, short shallow anterior siphon, and the strong, keel-like spiral cord on the body whor]l anterior to the siphonal constriction are the main distinguishing characters of this species. The range of variation in shell characters, such as the extent of pigmentation and node development on the spiral cords, as seen in Figure 1, is not great. Some specimens have distinct spots (Figure 1c) which others lack (Figure 1a, f). A specimen in the col- lection of J. Trondle, Papeete, Tahiti, had thin, brown spiral lines between the nodes. The largest specimens I have examined are from Norfolk Id. and Lord Howe Id., off the east coast of New South Wales, Australia, and from the Ryukyu Islands of Japan. IREDALE (1911:320) mentioned juveniles from dredg- ings as having a minute sinusigeral protoconch, and I have confirmed this by SEM studies of the protoconch. As may be seen in Figure 1b, the protoconch comprises 3.5 whorls and has a distinct, deep sinusigera notch. Thus, on the basis of protoconch morphology and the extensive geo- graphic range (JABLONSKI & LuTZ, 1980; JABLONSKI, 1982) it is reasonable to infer that Royella has a moderate to long planktotrophic larval stage. Geographic distribution (Figure 3). Royella sinon has a wide Indo-Pacific distribution ranging from the Red Sea and western Indian Ocean eastward to Pitcairn Island. Within the Pacific, it occurs from Japan south to Lord Howe Id., the Kermadecs, and Rapa. It probably occurs elsewhere throughout the Indo-Pacific in suitable habi- tats. ACKNOWLEDGMENTS I am indebted to Dr. Anders Warén, who, knowing of my interest in Royella, secured and sent me the preserved specimen collected by Mr. Gustav Paulay, who is also gratefully acknowledged for this contribution. I thank the following colleagues for information, loans R. S. Houbrick, 1986 of material, and assistance with collections under their charge: Philippe Bouchet (MNHNP), Sue Boyd (NMV), William K. Emerson (AMNH), Russell Jensen (DMNB), Ian Loch (AMS), Bruce Marshall (NMNZ), Henk Mienis (HUJ), James McLean (LACM), Winston F. Ponder (AMS), Robert Robertson (ANSP), Barry Roth (CAS), John Taylor (BMNH), and Fred Wells (WAM). I gratefully acknowledge the assistance of the following members of the National Museum of Natural History, Smithsonian Institution: Victor Krantz, Smithsonian Pho- tographic Services, for taking pictures of the specimens; Heidi Wolf, Smithsonian Scanning Electron Lab., for as- sistance on the SEM; Diane Bohmhauer, for assistance with the statistics. Travel expenses to various museums were partially funded by the Smithsonian Secretary’s Fluid Research Fund. LITERATURE CITED ApaMS, H. & A. ADAMS. 1813-1878. The genera of Recent Mollusca. 3 volumes, London. 389 pp., 138 pls. AzuMaA, M. 1960. A catalogue of the shell-bearing Mollusca of Okinoshima, Kashiwajima, and the adjacent area (Tosa Province) Shikoku, Japan. 102 pp., 5 pls. BayLeE, E. 1880. Liste rectificative de quelques noms de genres et despécies. J. Conchyl. 28(3):240-251. BRUGUIERE, J. G. 1789-1792. Encyclopédie méthodique, his- toire naturelle des vers. Paris. 1(1):1-344 (1889), 2(1):345- STI 92) DesHayeEs, G. P. 1824-1837. Description des coquilles des environs de Paris. 2 volumes + Atlas. 814 pp., 55 pls. (vol- ume 1); 106 pls. (volume 2). FLEMING, J. 1822. The philosophy of zoology or a general view of the structures, functions and classifications of ani- mals, etc. Edinburg. 2 volumes. Forses, E. & S. HANLEY. 1850-1851. A history of British Mollusca and their shells. Volume 3. Van Voorst: London. 133 pls. GRATELOUP, J. P. S. 1832. Tableau (suite de) des coquilles fossiles qu’on réncontré de Dax, department des Landes: par M. Grateloup, membre honoraire, 5eme Article. Actes Société Linnéenne de Bordeaux 5(29):263-282. Gray, J. E. 1847a. The classification of the British Mollusca by N. E. Leach, M.D. Ann. Mag. Natur. Hist. 20:267- De Gray, J. E. 1847b. A list of the genera of Recent Mollusca, their synonyma and types. Proc. Zool. Soc. Lond. 15:129- 219. GRUNDEL, J. 1976. Zur Taxonomie und Phylogenie der Bit- tium-gruppe (Gastropoda, Cerithiacea). Malakologische Abhandlungen Staatliches Museum fur Tierkunde in Dres- den 5(3):33-59, 2 pls., 17 figs. Hirase, S. 1936. A collection of Japanese shells. 5th ed. To- kyo. 217 pp., 128 pls. Houpsrick, R. S. 1984. Going collecting? Look for a live Roy- ella sinon. Hawaiian Shell News 32(4):12. Page 435 TREDALE, T. 1911. On the value of the gastropod apex in classification. Proc. Malacol. Soc. Lond. 9:319-323. IREDALE, T. 1912. New generic names and new species of marine mollusca. Proc. Malacol. Soc. Lond. 10:217-228, pl. 9. JABLONSKI, D. 1982. Evolutionary rates and modes in late Cretaceous gastropods: role of larval ecology. Proc. Third North Amer. Paleontol. Conv. 1:257-262. JABLONSKI, D. & R. A. Lutz. 1980. Molluscan larval shell morphology. Ecological and paleontological applications. Pp. 323-377. In: D. C. Rhoads & R. Lutz (eds.), Skeletal growth of aquatic organisms. Plenum: New York. Kira, T. 1962. Shells of the western Pacific in color. Osaka. 224 pp., 72 pls. KosBELT, W. 1888-1898. Die gattung Cerithium. 297 pp., 47 pls. In: F. H. W. Martini & J. H. Chemnitz, Neues sys- tematisches Conchylien-Cabinet .... 1(26). Nurnberg. Kuropa, T. 1941 A catalogue of molluscan shells from Tai- wan (Formosa), with descriptions of new species. Mem. Fac. Sci. Agri., Taihoku Imperial Univ. 22(4):Geology no. 17:65-216, pls. 8-14. Kuropa, T. 1960. A catalogue of molluscan fauna of the Oki- nawa Islands. 106 pp., 3 pls. MarRSHALL, B. A. 1978. Cerithiopsidae (Mollusca: Gastropo- da) of New Zealand, and a provisional classification of the family. New Zealand J. Zool. 5:47-120. MELVILL, J. & M. STANDEN. 1895. Notes on a collection of shells from Lifu and Uvea, Loyalty Islands, formed by the Rev. James and Mrs. Hadfield, with list of species. J. Con- chol. 8:84-132. MELVILL, J. & M. STANDEN. 1901. The Mollusca of the Per- sian Gulf, Gulf of Oman and Arabian Sea, as evidenced mainly through the collections of Mr. F. W. Townsend, 1893-1900, with descriptions of new species. Proc. Zool. Soc. Lond. 2:327-400, pls. 21-24. MENKE, C. T. 1828. Synopsis Methodica Molluscorum Ge- nerum Omnium et Specerum earum .... Pyrmont. Pp. i- xii, 1-91. MONTEROSATO, T.A., DI. 1884. Nomenclatura Generica e Specifica di Alcune Conchiglie Mediterranee. Palermo. 152 PP- RoEMER, F. A. 1840-1841. Versteinerungen des norddeutschen Kreidegebirges. Hannover. Pp. i-iv, 1-145, 16 pls. Sowerby, G. B. 1855. Monograph of the genus Cerithium, Adanson. Thesaurus Conchyliorum 2:847-899, pls. 176- 290. Sowerby, G. B. 1865. Monograph of the genus Cerithium. Conchologia Iconica 15:20 pls. Sowerby, G. B. 1866. (Supplementary plate) Thesaurus Con- chyliorum. Volume 3(24-25); pl. 290 [pl. 12]. THIELE, J. 1929-1931. Handbuch der systematischen Weich- tierkunde 1:1-376 (1929), 377-778 (1931). Berlin. Tryon, G. W. 1887. Manual of conchology; structural and systematic; with illustrations of the species. First series; 9: Cerithium. Pp. 127-149, pls. 20-29. Philadelphia. WENZ, W. 1938-1944. Gastropoda Allgemeiner Teil und Prosobranchia. Handbuch der Palaozoologie 6(1):1-1639, 4211 figs. The Veliger 28(4):436-443 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 New Philippine Cancellariidae (Gastropoda: Cancellariacea), with Notes on the Fine Structure and Function of the Nematoglossan Radula RICHARD E. PETIT anp M. G. HARASEWYCH Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract. Three new species of Cancellariidae, Cancellaria boucheti, C. atopodonta, and C. aqual- ica, are described, all from deep water off the western Philippines. The radula of two of these, C. boucheti and C. atopodonta, are figured, and a mechanism by which the cusps are interlocked is described. On the basis of mechanical considerations, we suggest that the function of the nematoglossan radula is limited to the penetration of tissues of prey organisms, in order to reach the internal fluids on which the cancellariids then feed suctorially. INTRODUCTION THE THREE SPECIES of Cancellaria described herein were brought to our attention by Dr. Philippe Bouchet, Cu- rator of Marine Mollusca, Muséum National d’ Histoire Naturelle, Paris. Specimens were collected during the MUSORSTOM.-2 cruise in 1980, aboard the R/V Co- riolis, with Dr. Bouchet as expedition malacologist. Ad- ditional specimens of one species were located in the col- lections of the National Museum of Natural History, Smithsonian Institution. That three rather large, unde- scribed species could be collected near the Philippines is indicative of our incomplete knowledge of the deeper water faunas. In keeping with our recent work, both published and unpublished, on cancellariid anatomy, we have here adopted a conservative stand by placing these new species in the genus Cancellaria, even though they represent rath- er distinct morphological forms previously placed in dif- ferent genera or subgenera. Placement of these species into more appropriate genera must await our proposed revi- sion of cancellariid supraspecific taxa after additional study of soft parts is possible. Abbreviations for museum collections cited are: MNHN, Muséum National d’Histoire Naturelle, Paris; USNM, National Museum of Natural History, Smithsonian In- stitution, Washington, D.C. SYSTEMATICS Genus Cancellaria Lamarck, 1799 Cancellaria boucheti Petit & Harasewych, spec. nov. (Figures 1-4, 9-14; Table 1) Description: Shell large, reaching 47 mm, moderately thin, elongate-oval, with conical spire and rounded ante- rior. Protoconch of about 1% inflated, glassy, amber-col- ored whorls, deviated from shell axis by 10-15°. Transi- tion to teleoconch marked by rapid development of spiral sculpture (Figure 3). Teleoconch with up to 6% strongly convex whorls. Suture deeply impressed. Surface sculp- ture of both spiral and axial elements, axial dominant in majority of specimens examined. Axial sculpture of 12- 14 prosocline ribs on early whorls, 16-18 on body whorl of larger specimens. Ribs strong, evenly spaced, with nar- row areas of weak, numerous ribs every 2 to % whorl indicating position of internal varices. Spiral sculpture of 15-18 major cords on body whorl, 7-8 on penultimate whorl, with 3-4 fine threads between adjacent cords. Ap- erture large, hemi-elliptical, deflected from coiling axis by 13-18°. Outer lip with shallow indentation at juncture of body whorl and short siphonal canal, containing 19-22 R. E. Petit & M. G. Harasewych, 1986 weak, recessed lirae that extend backward for several mil- limeters. Inner lip adpressed posteriorly, with two simple columellar folds and siphonal fold. Base color white, often with ginger along crests of axial and spiral sculpture. Some young specimens with broad patches of light ginger inter- rupted by white spiral bands along periphery and siphon- al juncture. Aperture white. No specimens fractured or sectioned, as presence of internal varices (HARASEWYCH & PETIT, 1982) at intervals of 120-180° apparent in intact shells. Critical-point dried jaw (Figure 9) 2.5 mm long (4.8% shell length), furrowed along dorsal midline. Left and right margins overlapping ventrally along anterior portion of jaw (Figure 10), forming tube (about 75 um in diameter) leading from oral tube to buccal mass. Posterior portion of jaw broad, covering the buccal mass dorsally and laterally. Radular teeth long, ribbonlike, tricusped. Central cusp simple, smooth, with ventrally recurved rim (Figures 11-13) and laterally expanded basal areas (Fig- ure 14). Lateral cusps each with 4 complex, anteriorly directed secondary cusps (Figures 11-14). Holotype: MNHN, 569-595 m, SE of Batangas, Luzon, Philippines (13°31'N, 121°24’E), Musorstom-2 sta. CP 36, L = 46.3 mm. 18 Paratypes: MNHN, 299-320 m, S of Batangas, Lu- zon, Philippines (13°49'N, 120°50’E), Musorstom-2 sta. CP 26, L = 34.4 mm; MNHN, 416-425 m, NW of Boac, Marinduque, Philippines (13°38’N, 121°43’E), Musor- stom-2 sta. CP 49, L = 24.7 mm; MNHN, 300-330 m, Page 437 off northwestern Mindoro, Philippines (13°51'N, 120°30’E), Musorstom-2 sta. CP 75, L = 26.5 mm, 36.6 mm, 39.2 mm; USNM 237060, 357 m, Tayabas Bay, off San Andreas, Philippines, U.S.B. Fish. sta. 5222, L= 37.7 mm; USNM 237580, 314 m, Batangas Bay, Luzon, Philippines U.S.B. Fish. sta. 5289, L = 32.0 mm; USNM 238191, 247 m, off Destacado Island, Philippines, U.S.B. Fish. sta. 5392, L = 46.0 mm; USNM 238675, off Opol, Mindanao, Philippines, U.S.B. Fish. sta. 5505, L = 29.1 mm; USNM 238902, 567 m, SE of Pt. Tanon, Cebu, Philippines, U.S.B. Fish. sta. 5535, L = 45.2 mm; USNM 242321, 247 m, off Destacado Island, Philippines, U.S.B. Fish sta. 5392, L = 20.0 mm, 34.6 mm, 37.8 mm; USNM 242323, 247 mm, off Destacado Island, Philippines, U.S.B. Fish. sta. 5392, L=19.5 mm; USNM 278521, 311 m, off Matocot Pt., W. Luzon, Philippines, U.S.B. Fish. sta. 5268, L = 22.4 mm; USNM 281805, 247 m, off Adyagan Island, E. Masbate, Philippines, Bur. Fish. sta. 5392, L = 20.2 mm, 28.5 mm, 30.4 mm. Comparisons: Cancellaria boucheti lacks the angled, no- dose shoulder of C. spengleriana Deshayes, 1830, to which it is most closely related. Cancellaria jonkeri Koperberg, 1931, from the Tertiary of Timor, is also similar to this Recent species, but has much finer spiral and axial sculp- ture that does not form nodules at intersections. Remarks: Cancellariid radular teeth become highly coiled during critical-point drying, the radular ribbon taking on the appearance of a plate of spaghetti. This process aids Explanation of Figures 1 to 8 Figure 1. Cancellaria boucheti spec. nov., holotype, MNHN, taken in 569-595 m, SE of Batangas, Luzon, Philippines (13°31'N, 121°24’E) Musorstom-2 sta. CP 36. 1.5x. Figure 2. Cancellaria boucheti spec. nov., paratype, MNHN, taken in 299-320 m, S of Batangas, Luzon, Philippines (13°49’N, 120°50’E) Musorstom-2 sta. CP 26. 1.5. Figure 3. Cancellaria boucheti spec. nov., protoconch of para- type, USNM 242323, taken in 247 m, SW of Destacado Island, Philippines. U.S.B. Fish. sta. 5392. 70x. Figure 4. Cancellaria boucheti spec. nov., paratype, USNM 238902, taken in 567 m, SE of Pt. Tanon, Cebu, Philippines. U.S.B. Fish. sta. 5535. 1.5x. Figure 5. Cancellaria atopodonta spec. nov., holotype, MNHN, taken in 441-510 m, SSW of Batangas, Luzon, Philippines (13°49'N, 120°28’E) Musorstom-2 sta. CP 78. 2.5*x. Figure 6. Cancellaria atopodonta spec. nov., paratype, MNHN, taken in 300-330 m, off northwestern Mindoro, Philippines (13°51'N, 120°30’E) Musorstom-2 sta. CP 75, 2.5%. Figure 7. Cancellaria aqualica spec. nov., holotype, MNHN, taken in 299-320 m, S of Batangas, Luzon, Philippines (13°49'N, 120°50’E) Musorstom-2 sta. CP 26, 1.5x. Figure 8. Cancellaria aqualica spec. nov., paratype, MNHN, taken in 170-187 m, off the northwestern tip of Mindoro, Phil- ippines (14°00’N, 120°17’E) Musorstom-2 sta. CP 51. 2.0x. Explanation of Figures 9 to 14 Radula and jaw of Cancellaria boucheti spec. nov., taken from the specimen in Figure 2. Figure 9. Lateral view of critical-point dried jaw. Scale bar = 500 um. Figure 10. View of distal end of jaw. Scale bar = 25 um. Figure 11. Axial view of distal ends of radular teeth. Dorsal tooth with cusps in expanded position. Scale bar = 1 um. Figure 12. Detail of distal end of tooth. Stereo pair. The free end of the cusp on the left is locked under the rim of the central cusp. The cusp on the right is free. Scale bar = 1 um. Figure 13. Axial view of laterally expanded tooth. Both outer cusps are free. Scale bar = 1 wm. Figure 14. Lateral view of teeth in Figures 11 and 12. The outer cusps of the ventral tooth are in the interlocked position. The visible cusp on the dorsal tooth is in the expanded position. Scale bar = 5 wm. The Veliger, Vol. 28, No. 4 Page 438 R. E. Petit & M. G. Harasewych, 1986 Page 439 Page 440 Table 1 Cancellaria boucheti spec. nov. Measurements of shell characters. Linear measurements in mm. n = 10. Standard Character Mean deviation Range Shell length 35.6 10.2 19.5-46.5 Shell width 19.6 4.9 10.8-24.5 Aperture length 20.7 6.2 11.2-29.3 Aperture length 79 0.548_0.640 Shelluleneth oe Ce use No. whorls, protoconch 1.53 0.24 1.33-2.00 No. whorls, teleoconch 5.53 0.65 4.67-6.67 Spire angle 53.0° Baile 48.5-59.0° in the examination of the distal ends of the teeth, but makes accurate measurement of tooth length very difficult. Previous work has shown that the length of the tubular portion of the jaw approximates tooth length (OLSSON, 1970; HARASEWYCH & PETIT, 1982, 1984). Both tooth length (about 1.1 mm) and tooth width (about 16 wm) are in the range reported for other cancellariids. Etymology: This species honors Dr. Philippe Bouchet, of the Department of Malacology, Muséum National d’His- toire Naturelle, Paris, who brought this material to our attention. Cancellaria atopodonta Petit & Harasewych, spec. nov. (Figures 5, 6, 15, 16; Table 2) Description: Shell small, reaching 22 mm, heavy, conispi- ral, with rounded anterior. Protoconch of 11% glossy, pitted whorls, slightly deflected from coiling axis. Transition to teleoconch marked by appearance first of spiral then axial sculpture. Teleoconch with up to 5% convex whorls. Su- ture impressed. Axial ribs major sculptural feature, num- bering 13-16 on body whorl, 10-13 on early whorls. In- ternal varices, first detected on outer surface of shell between 3rd and 4th postnuclear whorls, occur every 120° thereafter. Spiral sculpture of 10-13 major cords on body whorl, 3-4 on penultimate whorl, with 3-5 fine spiral threads between adjacent cords. Aperture ovate, deflected from coiling axis by 20-25°. Lack of shallow indentation marking juncture of siphonal canal on outer lip may be artifact, as all specimens heavily scarred by predators. Inner surface of outer lip with 9 or 10 strong, short lirae lining last internal varix. Inner lip adpressed posteriorly, with 2 simple columellar folds and siphonal fold. Shell white within and without. Internal structure not studied. Periostracum thin, yellow, finely lamellose, forming fine hairs along spiral threads. Intact jaw not recovered. Rad- ular teeth long, ribbonlike. Central cusp with ventrally recurved rim and 2 dorsal, posteriorly recurved barbs (Figure 16). Thickening of basal areas (Figure 16) less The Veliger, Vol. 28, No. 4 pronounced than in preceding species. Lateral cusps dis- tally expanded, each with 4 secondary, anteriorly directed cusps (Figures 15, 16). Holotype: MNHN, 441-510 m, SSW of Batangas, Lu- zon, Philippines (13°49'N, 120°28’E) Musorstom-2 sta. CP 78, L = 21.5 mm. 2 Paratypes: MNHN, 300-330 m, off northwestern Mindoro, Philippines (13°51’N, 120°30'E) Musorstom-2 sta. CP 75, L = 20.6 mm, 20.8 mm. Comparisons: This species differs so markedly from other described taxa that it is difficult to make comparisons. Its columellar structure is similar to that of Cancellaria gar- rardi (PETIT, 1974), but the latter species has more round- ed whorls and lacks a deep suture and a distinct shoulder. Cancellaria atopodonta most closely resembles the Jap- anese Pliocene shell figured as “‘Cancellaria (Merica) reev- et laticostata (Lobbecke)” by SHUTO (1962:72, pl. 13, fig. 12), which also has rounded whorls and lacks a pro- nounced shoulder. The names “‘reever”’ and “‘laticostata”’ are incorrect spellings of reeveana Crosse, 1861, and Ja- ticosta Lobbecke, 1881, respectively, although neither of these names can, in our opinion, be correctly applied to the shell figured by Shuto. We have been unable to find an available name for this Japanese fossil. Remarks: The presence of recurved barbs on the central cusps of the radula of Cancellara atopodonta, a feature previously reported only from radulate species of Adme- tinae, suggests that such barbs are a primitive character, and were present in the common ancestor of all cancel- lariids. Such an interpretation implies that C. atopodonta is a primitive member of the lineage giving rise to the Cancellariinae and the Trigonostominae, as these barbs are absent in most species of these two “subfamilies.” Etymology: The specific name is derived from the Greek atopos, meaning anomalous or out of place, and the Greek odontos, meaning tooth, and refers to the unusual structure of the radular teeth. Cancellaria aqualica Petit & Harasewych, spec. nov. (Figures 7, 8; Table 3) Description: Shell of moderate size, reaching 39 mm, heavy, biconic, pseudoumbilicate. Protoconch of 1 to 1% whorls, smooth, glassy, slightly bulbous, deviated from shell axis by 10-15°. Transition to teleoconch evidenced by onset of spiral sculpture, followed within %4 whorl by first appearance of axial ribs. Teleoconch with up to 6 strongly convex, highly sculptured whorls. Suture deeply impressed. Shell surface strongly cancellated by intersect- ing axial ribs and spiral cords. Axial sculpture of 14-16 prosocline ribs on early whorls, up to 19 on body whorl. Ribs strong, evenly spaced, becoming weak and more nu- R. E. Petit & M. G. Harasewych, 1986 Page 441 Explanation of Figures 15 and 16 Radula of Cancellaria atopodonta spec. nov., taken from the holotype. Figure 15. Detail of distal end of tooth. Stereo pair. The free ends of both outer cusps are locked under the rim of the central cusp. Scale bar = 1 um. merous every 120°, giving appearance of single, broad rib marking location of broad internal varices. Spiral sculp- ture of 13-17 major cords on body whorl, 6-8 on penul- timate whorl, with 1-3 fine threads between neighboring cords. Aperture large, hemi-elliptical, deflected from coil- ing axis by 18-25°. Outer lip with very shallow indenta- tion posterior to siphonal canal and 11-13 strong, slightly recessed lirae extending %4 whorl into aperture. Inner lip adpressed posteriorly, with 2 columellar and 1 siphonal fold. Folds simple, exhibiting periodic variation in size, reaching maximum extension into aperture in opposition to apertural lirae. Siphonal canal very short. Base color white, with light brown to ginger markings as in Cancel- laria reticulata (see HARASEWYCH & PETIT, 1982). Aper- ture white. Soft parts and periostracum unknown. Table 2 Cancellaria atopodonta spec. nov. Measurements of shell characters. Linear measurements in mm. n = 3. Standard Character Mean deviation Range Shell length 21.0 0.6 20.4-21.6 Shell width IB2 0.3 12.8-13.5 Aperture length 10.4 0.2 10.1-10.5 Aperture length 2 Shell length 0.494 0.016 0.485-0.512 No. whorls, protoconch 1.22 0.19 1.00-1.33 No. whorls, teleoconch 5.33 0.34 5.00-5.67 Spire angle 54.2° 0.8° 53.5-55.0° Figure 16. Lateral view of the same tooth as in Figure 15. The outer cusps are distally expanded. The central cusp has 2 dorsal barbs. Scale bar = 1 um. Holotype: MNHN, 299-320 m, S of Batangas, Luzon, Philippines (13°49’N, 120°50'E) Musorstom-2 sta. CP 26, L = 34.0 mm. 3 Paratypes: MNHN, 326-330 m, WSW of Batangas, Luzon, Philippines (13°55'N, 120°29'E) Musorstom-2 sta. CP 15, L = 38.3 mm; MNHN, 170-187 m, off the north- western tip of Mindoro, Philippines (14°00’N, 120°17’E) Musorstom-2 sta. CP 51, L = 15.7 mm, 19.5 mm. Comparisons: Cancellaria aqualica most closely resem- bles C. elegans Sowerby, 1822, as figured by GARRARD (1975: fig. 1[1]), from which it may be distinguished by its stronger axial and spiral sculpture, its lack of color bands, and by having a more swollen body whorl. Table 3 Cancellaria aqualica spec. nov. Measurements of shell characters. Linear measurements in mm. n = 4. Standard Character Mean deviation Range Shell length 26.9 11.0 15.7-38.2 Shell width 16.8 6.3 10.0-23.1 Aperture length 14.8 5.6 9.2-20.4 Aperture length a shelivlensth 0.557 0.022 0.535-0.587 No. whorls, protoconch 1.16 0.19 1.00-1.33 No. whorls, teleoconch 5.08 0.92 4.00-6.00 Spire angle 62.0° 1.4° 60.0-63.0° Page 442 Remarks: Our concept of Cancellaria elegans does not agree with Garrard’s interpretation. As mentioned by GARRARD (1975:4), the type lot (British Museum [Natural History] 1968387) contains 3 specimens. This lot is labelled “C. elegans Baclayon, Bohol. Id., Philippines” on the back of an old board. It is possible that two specimens with lo- cality data were added later, as only one specimen has an old label with the number ‘‘4,” the same sort as used by Sowerby, glued inside the aperture. In any event, the spec- imens in this lot are brown, with a brown protoconch, lack the white band present in many related species, and agree in all respects with the type figure. These specimens and the type figure have a finely cancellate sculpture that is quite distinct from the Australian species. Although a color photograph of the holotype of Can- cellaria asprella Lamarck, 1822, is available to us, we are unable to determine if C. asprella and C. elegans are con- specific, and this determination must await the opportu- nity to physically examine Lamarck’s type. In any event, the specimens that we consider to represent C. elegans as well as the holotype of C. asprella have apertures that are *% the total length of the shell, while C. aqualica has an aperture that is only 2 as long as the shell. Etymology: From the Latin aqualicus, meaning belly or paunch, referring to the swollen body whorl of this species. FUNCTIONAL MORPHOLOGY OF THE NEMATOGLOSSAN RADULA Other than one anecdotal report of Cancellaria crawfordi- ana feeding on pieces of fish and egg capsules of squid and whelks (TALMADGE, 1972), there have been no pub- lished reports on the food or feeding of any cancellariid. Although examinations of the gut contents of a number of species have failed to uncover any identifiable traces of solid food (GRAHAM, 1966; HARASEWYCH & PETIT, 1982, 1984), several speculations as to the diet of cancellariids have been based on the unusual morphology of the ante- rior alimentary system, especially the radula (GRAHAM, 1966; OLSSON, 1970; OLIVER, 1982; HARASEWYCH & PETIT, 1982, 1984). Data presented in this paper on the morphology of the jaws and radulae of Cancellaria bou- cheti and C. atopodonta, as well as figures of these organs of other cancellariids (OLSSON, 1970; OLIVER, 1982; Har- asewych & PETIT, 1982, 1984), permit the following ob- servations and inferences regarding their functional mor- phology. Although the length of the teeth is extreme, the cancel- lariid radula is formed and functions as a normal radular ribbon, with teeth produced posteriorly and migrating an- teriorly, contrary to OLSSON’s (1970:21) suggestion that teeth are either added from the center of the ribbon and directed anteriorly or posteriorly, or are no longer added once the radula is fully formed. HARASEWYCH & PETIT (1982) observed several teeth in the process of being re- directed from posterior to anterior. GRAHAM (1966) sug- The Veliger, Vol. 28, No. 4 gested that such redirection is accomplished by a change in the thickness and tension of the subradular membrane. The majority of the cusps, both primary and secondary, on each tooth are directed anteriorly, parallel to the line of motion of the tooth rather than perpendicular to it, indicating a piercing or grappling rather than a rasping function for the radula. The extreme flexibility of the radular teeth limits their ability to transmit force. Like ribbons, they only transmit tensile and not compressive force. Figure 12 reveals a mechanism by which the distal ends of the lateral cusps interlock under the recurved rim of the central cusp. The left cusp is shown in a locked po- sition, the right cusp is free. Thickened areas on either side of the central cusp may serve to buttress the lateral cusps when they are in a locked position. When the radula is protruded, the distal-most tooth is in the compact, in- terlocked form. During the rasping motion, as each tooth slides posteriorly over the tooth ventral to it, its lateral cusps are pushed into the unlocked, laterally expanded position (Figures 11, 14) by the distal end of the tooth below. In the reverse action, the free distal end of each tooth passes under the next dorsal tooth, and the cusps are again compressed into an interlocked position. We suggest that this interlocking mechanism functions in the following manner. Each radular tooth is applied to the prey tissue in the compact, interlocked position, the an- teriorly directed cusps impaling or entangling the tissue. The tissue is spread laterally when the cusps are pushed into the unlocked position by the next tooth, which repeats the action, penetrating deeper into the tissue. The length of the teeth and the tubular nature of the jaw make it unlikely that the cusped distal ends of the teeth can be retracted sufficiently to convey food to the esophageal opening, but instead suggest that their function is limited to the penetration of the tissues of prey organisms, or as suggested by TALMADGE (1972), of the walls of molluscan oothecae, in order to reach the internal fluids, on which the cancellariids then feed suctorially. ACKNOWLEDGMENTS We are indebted to Dr. Philippe Bouchet, Museum Na- tional d’Histoire Naturelle, Paris, for making available most of the material on which this paper is based. We also wish to recognize the work of CENTOB (Centre National de Tri d’Oceanographie Biologique) Brest, in sorting the material collected by MUSORSTOM-2. W. O. Cernohorsky, Auckland Institute and Museum, furnished us with a color photograph of the type of Can- cellaria asprella Lamarck. LITERATURE CITED CrossE, J. C. H. 1861. Etude sur le genre cancellaire, suivie du catalogue des espéces vivantes et fossiles actuellement connues. J. Conchyl. 9:220-256. R. E. Petit & M. G. Harasewych, 1986 Page 443 DeEsHaYES, G. P. 1830. Encyclopédie méthodique (Vers.) 2(1): 1-256. Paris. GARRARD, T. A. 1975. A revision of the Australian Cancel- lariidae (Gastropoda: Mollusca). Rec. Aust. Mus. 30(1):1- 62. GRAHAM, A. 1966. Fore-gut of marginellid and cancellariid prosobranchs. Stud. Trop. Oceanogr. Miami 4(1):134-151. HaARASEWYCH, M. G. & R. E. PETIT. 1982. Notes on the morphology of Cancellaria reticulata (Gastropoda: Cancel- lariidae). Nautilus 96(3):104-113. HaARASEWYCH, M. G. & R. E. PETirT. 1984. Notes on the morphology of Olssonella smithu (Gastropoda: Cancellari- idae). Nautilus 98(1):37-44. KOPERBERG, E. J. 1931. Jungtertidre und quartare Mollusken von Timor. Jaarb. Mijnwezen Ned.-Indie. Verh. I, 1-165, pls. 1-3. Lamarck, J. B. P. A. 1822. Histoire naturelle des animaux sans vertébres. Vol. 7. Paris. LOBBECKE, T. 1881-87. Das Genus Cancellaria. Syst. Conch. Cab. 4:1-96, pls. 1-23. OLIVER, P. G. 1982. A new species of cancellariid gastropod from Antarctica with a description of the radula. Brit. Ant- arct. Surv. Bull. 57:15-20. Oxsson, A. A. 1970. The cancellariid radula and its interpre- tation. Palaeontogr. Amer. 7(43):19-27. Petit, R. E. 1974. Notes on Japanese Cancellariidae. Venus 33(3):109-115. SHuTo, T. 1962. Buccinacean and volutacean gastropods from the Miyazaki Group. Mem. Fac. Sci. Kyushu Univ., Ser. D, Geol. 12(1):27-85, pls. 6-13. Sowerby, G. B. 1821-1825. The genera of Recent and fossil shells. Vol. 1: pls. 1-126 and text (pages not numbered). London. TALMADGE, R. 1972. “Pinky.” Of Sea and Shore 3(4):189, 200. The Veliger 28(4):444-447 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 A New Species of Helminthoglypta (Gastropoda: Pulmonata: Helminthoglyptidae) from the Cuyamaca Mountains of Southern California by RICHARD L. REEDER Faculty of Biological Science, University of Tulsa, Tulsa, Oklahoma 74104, U.S.A. Abstract. A new species of land snail, Helminthoglypta milleri Reeder, is described from the Cuy- amaca Mountains of San Diego County, California, and its relationships are discussed. IN THE LATE 1800’s, Henry Hemphill collected snails from the Cuyamaca Mountains, in San Diego County, which he labeled Epiphragmophora trasku v. cuyamacensis. He sent specimens to H. A. Pilsbry at the Academy of Nat- ural Sciences of Philadelphia as well as to Paul Bartsch at the U.S. National Museum. However, he separated the shells from the preserved bodies, sending the shells to Bartsch in 1890 and the preserved bodies to Pilsbry in 1893, thereby raising an element of doubt that the bodies may not have come from the same population as the shells. PitsBry duly figured the reproductive system in the Manual of Conchology (1895:pl. 59, fig. 87) and cited it as Epiphragmophora (Helminthoglypta) traski v. cayamacen- sis Hemph. He later stated (PILSBRY, 1939:145) that he withheld description of the shells because he did not wish to trespass on Hemphill’s field. He also stated: “with the locality, which I gave in 1897, there would be little doubt of what was intended.” This reference (PILSBRY, 1897), however, merely repeated, with correct spelling: “Ep- phragmophora traski cuyamacensis Hemp. Cuyamaca Mt., San Diego Co.” Two decades later, P. BARTSCH (1916) published a description of the shell as Epiphragmophora cuyamacensis cuyamacensis Bartsch, 1916, and considered Pilsbry’s ear- lier name to be a nomen nudum. In 1933, Pilsbry, accompanied by Joshua L. Bailey, went to the Cuyamaca Mountains and collected snails at a locality “% of a mile above Cuyamaca Lake, ¥2 mile south of road at the ‘beef pasture’ under heaps of rotten wood in mixed evergreen and deciduous woods.” There was no doubt in his mind that he had obtained H. cuya- macensis cuyamacensis and he figured the reproductive anatomy of this snail as fig. 71B (PILsBRy, 1939), labeled H. cuyamacensis, Cuyamaca Mountains. This anatomy, however, was considerably different from the one pub- lished in the Manual of Conchology in 1895 and because he was uncertain about the origin of the bodies sent by Hemphill in 1893, he again figured an anatomy of the 1893 shipment as fig. 73 (PILsBRY, 1939), labeling it “Genitalia of a Helminthoglypta of uncertain status.” It can now be ascertained, from fig. 73 (PILSBRY, 1939), that the genitalia are typical of the subgenus Rothelix Miller, 1985, and therefore are probably those of H. c. cwyama- censis Bartsch, 1916 (MILLER, 1985). On the other hand, the genitalia of fig. 71B (PILsBRy, 1939) are typical of the nominate subgenus; they appear to be referable to those of H. thermimontis Berry, 1953, but a firm identification will require a precise pinpointing of Pilsbry’s exact lo- cality and its population and a careful, detailed, micro- scopic examination of the reproductive anatomy. During the 1960’s, W. O. Gregg and W. B. Miller collected extensively at several localities in the Cuyamaca Mountains. They found populations of snails along the main highway in the Cuyamacas that correspond precisely with size, sculpture, description, and illustration of Bartsch’s specimens as well as PILSBRY’s original figure of the genitalia (1895:pl. 59, fig. 87; 1939:fig. 73). This locality can now be more precisely identified as “Cuya- maca Mountains, along Cold Stream in the vicinity of Cold Spring, 32°56.5'N, 116°33.8’W, elev. 4400 ft. (1350 m), in rotting logs.” It can now be considered the more detailed type locality of H. cwyamacensis cuyamacensis Bartsch, 1916. In the process of exploring the Cuyamaca Mountains, several populations of other species of Helminthoglypta were also discovered. Most of these populations live in rotting logs, as does H. c. cuyamacensis. One population, however, differs from all the others in that it is a rock Rev ieweedens 1986 dweller, located in a large rock pile just below the summit of Cuyamaca Peak. Furthermore, this population does not belong in the subgenus Rothelix Miller, 1985, as does H. c. cuyamacensis, but rather in the nominate subgenus. Its anatomy clearly indicates that it is a new species, de- scribed below. Helminthoglypta (Helminthoglypta) milleri R. L. Reeder, spec. nov. (Figures 1-4) Diagnosis: A large sized, depressed Helminthoglypta with radial growth wrinkles and moderately papillose sculp- ture. Description of shell of holotype: Shell (Figures 2-4) large, depressed, with conic spire, helicoid, umbilicate, the umbilicus contained about 7 times in the diameter of the shell. Color light brown, glossy, with a darker reddish- brown band on the round shoulder. Aperture broad, near- ly round, with peristome moderately reflected, expanded slightly more at columellar junction. Embryonic shell of 1%4 whorls with faint radial wrinkles and minute papillae. Post-embryonic whorls with increasingly coarse radial growth wrinkles, superimposed with moderately dense papillae. Papillae weaker on base of shell, becoming prominent again within the umbilicus. Spiral sculpture wanting. Diameter 26.6 mm, height 12.3 mm, diameter of umbilicus 3.8 mm, number of whorls 5%. Reproductive anatomy of holotype: The reproductive system (Figure 1) is typical of the genus, having a large atrial sac with a dart sac at its proximal end. There are two mucus glands with mucus bulbs, the ducts of which form a common duct before entering the upper portion of the atrial sac. The spherical spermatheca is large with a long duct bearing a spermathecal diverticulum about mid- way along its length. The penis and epiphallus form a continuous duct with the epiphallus bearing a relatively long epiphallic caecum at its proximal end. The penis is divided into a short lower penis and a long upper penis, the latter being a double-walled tube. The upper penis is a cylindrical duct of nearly uniform diameter. The lower penis is initially as wide as the upper penis at their junc- tion, then tapering into a venturi-like constriction, the throat of which is moderately narrow. There is no verge. The vas deferens passes around the dart apparatus and the penial retractor muscle inserts on the epiphallus. Measurements of distinctive structures are as follows: Penis 17.7 mm Epiphallus 27.4 mm Epiphallic caecum 17.1 mm Spermathecal duct 33.7 mm Spermathecal diverticulum 23.1 mm Variations in paratypes: A total of 10 adult and 3 im- mature shells was examined. The largest adult paratype Page 445 lOmm Figure 1 Portion of reproductive system of Helminthoglypta milleri Reed- er, spec. nov., prepared from projection of stained whole mount of holotype; as, atrial sac; ds, dart sac; ec, epiphallic caecum; ep, epiphallus; go, genital orifice; lp, lower part of penis; mb, mucus gland bulbs; mg, mucus gland membranes; ov, oviduct; pr, penial retractor muscle; pt, prostate; sd, spermathecal duct; sp, spermatheca; sv, spermathecal diverticulum; up, upper part of penis; ut, uterus; va, vagina; vd, vas deferens. measures 26.7 mm in diameter and 15.2 mm in height, and the smallest measures 22.2 mm and 12.8 mm respec- tively. All of the specimens demonstrate the radial wrin- kles and papillae as described and two of the adult shells show a few faint spiral incised lines above the shoulder on the body whorl just behind the aperture. Disposition of types: Holotype: Santa Barbara Museum of Natural History, No. 33915. Paratypes: The Academy Page 446 The Veliger, Vol. 28, No. 4 Explanation of Figures 2 to 4 Helminthoglypta milleri Reeder, spec. nov. Shell of holotype, SBMNH No. 33915; diameter 26.6 mm. Figure 2. Aperture view. of Natural Sciences of Philadelphia, No. 359212; U.S. National Museum, No. 842307; W. B. Miller collection, No. 2596, 7118, and 7440; R. L. Reeder collection, No. 574. Type locality: San Diego County, California; Cuyamaca Peak, Cuyamaca Mountains; in rocks at junction of Burnt Pine Fire Road and Cuyamaca Peak Road; 32°56.9'N, 116°36.3'W; elevation 1900 m (6200 ft.). Discussion: The long, cylindrical, double-tubed portion of the penis, and the short, thin, saccular, lower part of the penis clearly establish that Helminthoglypta milleri Figure 3. Apical view. Figure 4. Umbilical view. belongs in the nominate subgenus. In that subgenus, its nearest geographical relatives are H. tudiculata (Binney, 1843) and H. waltoni Gregg & Miller, 1976. Helmintho- glypta milleri can be separated from H. tudiculata by shell characters alone in that H. tudiculata and its subspecies have a strongly malleated shell while H. milleri has a papillose shell. From H. waltoni it can be separated by its reproductive anatomy in that H. milleri has a long, stout, cylindrical upper part of the penis while H. waltoni has a decidedly club-shaped upper part of the penis. Farther to the north, in the Hot Springs Mountains, another species of Helminthoglypta with papillose shells, H. thermimontis Re eeReeden | 986 Page 447 Berry, 1953, appears to be related to H. milleri, but it can be separated by its reproductive anatomy in that H. thermimontis has an upper penis whose diameter decreases markedly before it joins the narrower lower penis; in turn the venturi-shaped lower penis has a very constricted throat, whereas that of H. milleri is only moderately con- stricted. Moreover, the shell of H. thermimontis is consid- erably more papillose than that of H. milleri. It is prob- able that H. milleri, H. waltoni, and H. thermimontis evolved from a common papillose ancestor. Helminthog- lypta thermimontis speciated and adapted to an environ- ment of fallen logs and humus in the Hot Springs Moun- tains, H. waltoni to a rock-dwelling existence in the Laguna Mountains, and H. milleri to its isolated rock pile on top of Cuyamaca Peak. Distribution and habitat: Helminthoglypta milleri is currently known only from the type locality on Cuyamaca Peak. Vegetation at this locality consists principally of Quercus kellogi, Quercus chrysolepis, Pinus lambertiana, Abies concolor, Libocedrus decurrens, and Arctostaphylos sp. Etymology: This species is named for Walter B. Miller, friend and mentor, who has provided me the opportunity to study the Helminthoglypta of southern California. ACKNOWLEDGMENTS I wish to express my gratitude to the late Wendell O. Gregg for providing material used in preparation of this manuscript and to my long-standing colleague and teach- er, Walter B. Miller, for material and critical information. Thanks also to my colleagues Noorullah Babrakzai for help in collecting material and Susan J. McKee for pho- tographs of the holotype. Thanks also to the University of Tulsa for kindly providing funds for field work and to the University of Arizona for providing laboratory facili- ties. LITERATURE CITED BartTscH, P. 1916. The Californian land shells of the Epr- phragmophora trasku group. Proc. U.S. Natl. Mus. 51 (2170): 609-619. MILLER, W. B. 1985. A new subgenus of Helminthoglypta (Gastropoda: Pulmonata: Helminthoglyptidae). Veliger 28(1):94-98. Pitssry, H. A. 1895. Guide to the study of helices. Manual of Conchology, Series 2, 9:i-xlvii + 1-366 pp.; 71 plates. Pitsspry, H. A. 1897. A classified catalogue of American land snails, with localities. Nautilus 11(5):59-60. Pitssry, H. A. 1939. Land Mollusca of North America (north of Mexico). Acad. Natur. Sci. Phila. Monogr. (3)I(1):i- xvili + 1-573 + i-ix; figs. A, B, 1-377. THE VELIGER The Veliger 28(4):448-452 (April 1, 1986) ©) TOMAS) Une, 12186 A New Species of [schnochiton (Mollusca: Polyplacophora) from the Tropical Eastern Pacific by ANTONIO J. FERREIRA! Research Associate, Department of Invertebrate Zoology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118, U.S.A. Abstract. Ischnochiton skoglundi Ferreira, spec. nov., dredged from 8-15 m, Nayarit, Mexico, differs from other species of the genus in the area by its very small size (less than 5 mm long), broad, ovate body, sculptureless tegmentum, and girdle scales with spherules on the upper surface, riblets on the sides, and a round concavity on the insertion face. A brief historical account of the genus Jschnochiton Gray, 1847, is given. EXAMINATION OF A lot consisting of very small chitons, dredged off Nayarit, Mexico, revealed 13 minute speci- mens less than 5 mm long, which were dry and well pre- served. Of these, eight were firmly attached to fragments of old shells, still maintaining a “living” position. The species, not hitherto recognized, is here allocated to the genus Ischnochiton Gray, 1847a. Class Polyplacophora Gray, 1821 Order Neoloricata Bergenhayn, 1955 Suborder Ischnochitonina Bergenhayn, 1930 Family Ischnochitonidae Dall, 1889 Genus Ischnochiton Gray, 1847a Type species: Chiton textilis Gray, 1828, by subsequent designation (GRAY, 1847b). Remarks: Interpretations of the genus Jschnochiton are still in a state of flux. Although a full discussion of Jsch- nochiton is beyond the scope of this paper, some historical observations may better explain the allocation of skoglun- di to the genus. Ischnochiton was established by GRAY (1847a:126-127) for species characterized by “Valves thin; posterior valve entire; the plates of insertion very thin, smooth-edged, of the central valves each with a single notch [slit]; margin [girdle] covered with very small imbricated scales.” ' For reprints: 2060 Clarmar Way, San Jose, California 95128. The large number of species in Jschnochiton and their diverse characteristics led Carpenter (in DALL, 1879) to partition it into eight subgenera. Similarly, PILSBRY (1892a) divided Ischnochiton into seven subgenera, some further split into “sections.” Although several subgenera were eventually removed from Jschnochiton, either elevat- ed to genera or synonymized, its number soon grew to 18 in the conservative view of SMITH (1960). This unsatisfactory arrangement was further compli- cated by the finding (AsHBy, 1931:36; Allyn G. Smith in Kaas, 1974) that the intermediary valves of Chiton textilis, type species of Ischnochiton, are two-slitted and not one- slitted as assumed by Gray (1847a) and PILsBRY (1892b: 99). Because this finding made it appear that Ischnochiton was without a type, VAN BELLE (1974) erected Simplisch- nochiton (type species, Ischnochiton maorianus Iredale, 1914, new name for Chiton longicymba Quoy & Gaimard, 1835, not Blainville, 1825) for the one-slitted species, retaining Ischnochiton for two-slitted species. KAAS (1974), instead, proposed Chiton crispus Reeve, 1847, as a new type for Ischnochiton, a proposal that did not conform with Article 61 of the International Code of Zoological Nomenclature (1964). Later, Kaas (1979:856) suggested that GRay’s (1847a) notion of a single slit in Ischnochiton be modified to encompass species with one or two slits, a concept long in use by PILSBRY (1892a:53-54) and SMITH (1960:55). Thus, KAAS & VAN BELLE (1980) returned to the tra- ditional interpretation of Jschnochiton, this time divided into seven subgenera, with Simplischnochiton suppressed. But, in the most recent systematic classification of the A. J. Ferreira, 1986 Page 449 Figure 1 Ischnochiton skoglundi Ferreira, spec. nov.: Holotype (CAS 059841), girdle elements. A, dorsal surface scales, different views (arrow points to concavity on scale base); B, undersurface scales. Scale bar, 100 um. chitons, VAN BELLE (1983), dividing /schnochiton into eight subgenera, brought back Simplischnochiton for “‘Ischno- chitons with no more than one slit .. . in the intermediary valves,” and Jschnochiton s.s. for species with “two or more slits.” In my view of the systematics of Polyplacophora, subge- neric categories are neither necessary nor desirable. So, here as elsewhere (FERREIRA, 1983:311), Ischnochiton, a rather heterogeneous assemblage of species, is interpreted in accordance with the general characteristics outlined by SMITH (1960), z.e., accepting both one- and two-slitted species; and, as evidence may suggest, its well character- ized “subgenera” are elevated to generic status (FERREIRA, 1981, 1985). Simplischnochiton is suppressed as a syn- onym. Ischnochiton skoglundi Ferreira, spec. nov. (Figures 1-5) Diagnosis: Very small (up to 4.8 mm long), yellowish white chitons; shell wide, ovate; valves not beaked, cari- nate; tegmentum dull, sculptureless; lateral areas weakly elevated, hardly defined; mucro anterior. Slit formula 8-1-9. Girdle with imbricate, very small scales, with round spherules on upper surface, riblets on sides. Radula with unicuspid major lateral teeth. Type material: Holotype (CAS 059841) and paratypes (CAS 059842; CAS 060251; LACM 2119; USNM 859001; ANSP 360105; SDNH 87085; Skoglund Colln.; Ferreira Colln.). Figure 2 Ischnochiton skoglundi Ferreira, spec. nov.: Holotype (CAS 059841), radula. A, median and first lateral teeth; B, head of major lateral tooth; C, anterior end of spatulate tooth. Scale bar, 100 um. Page 450 The Veliger, Vol. 28, No. 4 Explanation of Figures 3 to 5 Figure 3. Ischnochiton skoglundi Ferreira, spec. nov.: Paratype, 2.4 mm long (CAS 060251), dorsal surface. SEM micrograph. Figure 4. Ischnochiton skoglundi Ferreira, spec. nov.: Paratype (CAS 059842) on dead shell. Type locality: Off Playa Novillero, Nayarit, Mexico (22°23'N, 105°45’W), dredged at 8-15 m (leg. Sally & Peter Bennett, Dec. 1975). Description: Holotype (CAS 059841), dry preserved, Figure 5. Ischnochiton skoglundi Ferreira, spec. nov.: Same paratype as in Figure 3, girdle upper surface scales. SEM mi- crograph. creamy white, about 4.8 mm long (largest specimen in lot), ovate, widest (2.3 mm) at valve v; valves thin, car- inate, not beaked, posterior edges straight. Tegmentum dull, with no noticeable sculpture; lateral areas hardly Mea lpeberreinas, 1286 Page 451 defined, very slightly elevated, with faintly distinguishable concentric rugosities; mucro anterior. Gills holobranchial (2). Articulamentum white; sutural laminae short, sub- rectangular; sinus shallow; width of valve i/width of valve vili, 1.1; on valve vill, width of sinus/width of sutural laminae, 0.5. Insertion teeth small, sharp; slit formula 8-1-9. Girdle dorsal surface black (an artifact) with trans- lucent, imbricated scales (Figure 1-A); scales up to 80 um long, upper surface covered with minute, round spherules, lateral surface with some 12 riblets, base with roughly round, sharply defined concavity, about 15 wm in diameter (feature never noticed in any other species); girdle ventral surface paved with transparent, rectangular scales (Figure 1-B) 40 x 12 um, arranged in columns. Radula 1.2 mm long, comprising 28 rows of mature teeth; median teeth (Figure 2-A) about 50 um long, 25 um wide at anterior blade, narrowing sharply posteriorly; first lateral teeth 50 um long, 15 wm wide at anterior blade; major lateral teeth with large, unicuspid head, with thin, long tubercle be- neath (Figure 2-B); spatulate teeth 22 wm wide anteriorly (Figure 2-C); outer marginal teeth 40 x 25 um (length/ width, 1.6). Paratypes (Figures 3, 4) very similar to holotype, 2.0- 4.1 mm long, width/length mean 0.76 (n = 10; SD= 0.04; range 0.71-0.85); ovate (mean width of vaives v + vi consistently greater than mean width of valves iii + iv); curvature index (width of widest valve/average width of end valves) 1.28 (n = 5, including holotype). Girdle black in most specimens due to extraneous fuliginous material; SEM micrograph of girdle scales (Figure 5) shows the same ornamentation of round spherules and riblets. Distribution: [schnochiton skoglundi is known only from the type lot. Remarks: Specimens of [schnochiton skoglundi are of un- usually small size, but they differ clearly from juveniles of any other known species in the eastern Pacific. It is quite distinct from any other Ischnochiton species in the area—I. muscarius (Reeve, 1847), J. rugulatus (Sowerby, 1832), and J. eucosmius Dall, 1919—in its ovate body- shape, carinate valves, sculptureless tegmentum, and in the girdle scales. The “ornamentation” seen on the girdle scales of I. skoglundi, consisting of minute spherules on the scale’s upper surface and riblets on the sides, has been seen and illustrated in three other species—Lepidozona allynsmithi Ferreira, 1974 (see FERREIRA, 1974: figs. 23 and 24), Callistochiton portobelensis Ferreira, 1976 (see FERREIRA, 1976: figs. 3-5), and C. periconis Dall, 1908 (see FERREIRA, 1979: figs. 22 and 23)—but not in Ischno- chiton. The sharply delineated concavity observed at the base (insertion surface) of the scales of /. skoglundi seems to be a unique feature inasmuch as it has not been re- ported or here noted in any other species. The species is named after Carol and Paul Skoglund, Phoenix, Arizona, who have generously provided these and many other specimens for study. ABBREVIATIONS USED IN THE TEXT ANSP—Academy of Natural Sciences, Philadelphia, Pennsylvania. CAS—California Academy of Sciences, San Francisco, California. Colln.—Private collection. LACM—Los Angeles County Museum of Natural His- tory, Los Angeles, California. SDNM-—San Diego Museum of Natural History, San Diego, California. USNM—USS. National Museum of Natural History, Washington, D.C. ACKNOWLEDGMENTS I thank Carol and Paul Skoglund, Phoenix, Arizona, who entrusted these specimens to my care and study, and Ter- rence M. Gosliner, Department of Invertebrate Zoology, California Academy of Sciences, for the SEM micro- graphs. LITERATURE CITED AsHBY, E. 1931. Monograph of the South African Polypla- cophora (Chitons). Ann. S. Africa Mus. 30(1):1-59, 2 text figs., 7 pls. BERGENHAYN, J. R. M. 1930. Kurze bemerkungen zur kennt- nis der schalenstruktur und systematik der Loricaten. Kungl. Svenska Vetensk. Handl. (3)9(3):3-54, 5 text figs., 10 pls. Uppsala. BERGENHAYN, J. R.M. 1955. Die fossilen schwedischen Lori- caten nebst einer vorlaufigen Revision des systems der gan- zen Klasse Loricata. Lunds Univ. Arsskrift. (Adv.2, N.S.) 51(8):1-43, 2 pls. Kungl. Fysiogr. Sallsk. Handl., N.F., 66 (8):3-42, 2 tables. BLAINVILLE, H. D. DE. 1825. Oscabrion, Chiton. In: Diction- naire des Sciences Naturelles, Paris 36:519-555. Da._, W. H. 1879. Report on the limpets and chitons of the Alaskan and Arctic regions, with descriptions of genera and species believed to be new. Proc. U.S. Natl. Mus. 1(for 1878):281-344, 5 pls. DALL, W. H. 1889. Preliminary catalogue of the shell-bearing marine mollusks and brachiopods of the southeastern Coast of the United States, with illustrations of many of the species. Bull. U.S. Natl. Mus. 37:3-221, 74 pls. DALL, W. H. 1908. Reports on the dredging operations off the west coast of Central America to the Galapagos, to the west coast of Mexico, and in the Gulf of California, in charge of Alexander Agassiz, carried on by the U. S. Fish Commission steamer “Albatross” during 1891, Lieut. Commander Z. L. Tanner, U.S. N., Commanding. XX XVIII. Reports on the scientific results of the expedition to the eastern tropical Pacific in charge of Alexander Agassiz, by the U. S. Fish Commission steamer “Albatross,” from October, 1904, to March, 1905, Lieut. Commander L. M. Garrett, U. S. N., Commanding. XIV. Reports on the Mollusca and Brachi- opoda. Bull. Mus. Comp. Zool. 43(6):205-487, pls. 1-22. DatL, W. H. 1919. Descriptions of new species of chitons from the Pacific coast of America. Proc. U.S. Natl. Mus. 55(2283):499-516. FERREIRA, A. J. 1974. The genus Lepidozona in the Panamic Province, with the description of two new species (Mollusca: Polyplacophora). Veliger 17(2):162-180, 6 pls. Page 452 FERREIRA, A. J. 1976. A new species of Callistochiton in the Caribbean. Nautilus 90(1):46-49, 5 figs. FERREIRA, A. J. 1979. The genus Callistochiton Dall, 1879 (Mollusca: Polyplacophora) in the eastern Pacific, with the description of a new species. Veliger 21(4):444-466, 9 text figs., 3 pls. FERREIRA, A. J. 1981. A new species of Stenosemus Midden- dorff, 1847 (Mollusca: Polyplacophora) in the abyssal northeastern Pacific. Veliger 23(4):325-328, 5 text figs., 1 pl. FERREIRA, A. J. 1983. The chiton fauna of the Revillagigedo Archipelago, Mexico. Veliger 25(4):307-322, 10 text figs., 2 pls. FERREIRA, A. J. 1985. Chiton (Mollusca: Polyplacophora) fauna of Barbados, West Indies, with the description of a new species. Bull. Marine Sci. 36(1):189-219. Gray, J. E. 1821. A natural arrangement of Mollusca, ac- cording to their internal structure. London Medic. Repos. 15:229-239. Gray, J. E. 1828. Spicilegia Zoologica; or Original figures and short systematic descriptions of new and unfigured an- imals. Part 1, 8 pp., 6 pls. British Museum. Gray, J. E. 1847a. Additional observations on Chitones. Proc. Zool. Soc. Lond. 15(178):126-127. Gray, J. E. 1847b. A list of the genera of recent Mollusca, their synonyma and types. Proc. Zool. Soc. Lond. 15(178): 129-219. IREDALE, T. 1914. The chiton fauna of the Kermadec Islands. Proc. Malacol. Soc. Lond. 11(1):25-51, pls. 1, 2. Kaas, P. 1974. Notes on Loricata. 7. On the type of the genus Ischnochiton Gray, 1847. Basteria 38:95-97. The Veliger, Vol. 28, No. 4 Kaas, P. 1979. The chitons (Mollusca: Polyplacophora) of Mozambique. Ann. Natal Mus. 23(3):855-879. Kaas, P. & R. A. VAN BELLE. 1980. Catalogue of living chi- tons. Dr. W. Backhuys, Publisher: Rotterdam. 144 pp. INTERNATIONAL CODE OF ZOOLOGICAL NOMENCLATURE, adopt- ed by the XV International Congress of Zoology, 1964. London. xix + 176 pp. PitsBry, H. A. 1892a. Polyplacophora. In: G. W. Tryon, Jr., Manual of conchology, 14:1-64, pls. 1-15. Pitsspry, H. A. 1892b. Polyplacophora. Jn: G. W. Tryon, Jr., Manual of conchology, 14:65-128, pls. 16-30. Quoy, J. R. C. & J. P. GaimarD. 1835. Voyage de decou- vertes de l’Astrolabe, executé par ordre du Roi, pendant les années 1826-1827-1828-1829, sous le commandement de M. J. Dumont D’Urville. Zoologies. Paris. 3:369-411. REEVE, L. A. 1847-1848. Monograph of the genus Chiton. In: Conchologia iconica, or Illustrations of the shells and mol- luscous animals. London. 4:28 pls., 194 figs. Smit, A. G. 1960. Amphineura. Jn: R. C. Moore (ed.), Trea- tise on invertebrate paleontology, Part I, Mollusca 1, pp. 41-76, figs. 31-45. Sowersy, G. B. 1832. Jn: A. J. Broderip & G. B. Sowerby, Characters of new species of Mollusca and Conchifera, col- lected by Mr. Cuming. Proc. Zool. Soc. Lond. 1832:50-61. VAN BELLE, R. A. 1974. A propos du genre Jschnochiton Gray, 1847 (Polyplacophora). Informations Soc. Belge Malacol. 3(2):27-29. Van BELLE, R. A. 1983. The systematic classification of the chitons (Mollusca: Polyplacophora). Informations Soc. Belge Malacol. 11(1-3):1-178, 13 pls. The Veliger 28(4):453-456 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 Three ‘Temperate-Water Species of South African Gastropods Recorded for the First Time in Southwestern Australia FRED E. WELLS Western Australian Museum, Perth, W.A. 6000, Australia RICHARD N. KILBURN Natal Museum, 237 Loop Street, Pietermaritzburg 3201, Natal, South Africa Abstract. Three temperate South African species of gastropods (Nassarius kraussianus, Bullia an- nulata, and Cymatium cutaceum africanum) are recorded for the first time in southwestern Australia. Possible mechanisms by which these species were able to transmigrate the Indian Ocean are discussed. INTRODUCTION THE TEMPERATE waters of southern Africa and southern Australia are in distinct biogeographic regions (BRIGGS, 1975). Aside from a few circumtemperate species there are few mollusk species in common between southern Af- rica and southern Australia, as a comparison of species included in WILSON & GILLETT (1979) and KILBURN & RIpPpEY (1982) shows. A number of mollusk species have been shown to have crossed the Pacific (EMERSON, 1967; VON COSEL, 1977) and Atlantic (SCHELTEMA, 1971) oceans. Only two temperate species of mollusks are known to have crossed the southern Indian Ocean. The southern African abalone Haliotis spadicea Donovan, 1808 (=H. sanguinea Hanley, 1840) has been collected in southern Western Australia at Cowaramup Bay, south of Cape Naturaliste (MACPHERSON, 1953). The southern Australian muricid Bedeva pawwae (Crosse, 1864) has recently established itself in East London Harbour (KILBURN & RIPPEY, 1982) and also in the Canary Islands (GOMEz, 1984). Three addi- tional southern African temperate species have now been recorded in southern Western Australia and are reported here. SOUTH AFRICAN SPECIES Nassarius kraussianus (Dunker, 1846) has been recorded as two lots. WAM 51-82 (Figure 1) is an adult shell 7.7 mm long, which was collected dead at Augusta, W.A., by W. Anson in January 1974. The shell has the thick, glossy callus that overlaps the sides and reaches the apex, as described by KILBURN & RippEy (1982). The shell is smooth dorsally, has three grayish-brown spiral bands on the body whorl separated by whitish bands. A thin brown line occurs along the suture. The outer shell color shows through the aperture. The callus is white, with a thin brown line going posteriorly from the posterior edge of the aperture. Two specimens, one an adult 7.8 mm long and the other a juvenile of 5.4 mm, were collected dead by G. Hansen at Flinder’s Bay, Augusta, W.A., on 2 July 1972 (WAM 2670-83). These shells closely resemble the specimen described above, except that the juvenile shell lacks the callus. A specimen of N. kraussianus from Dur- ban, South Africa, is shown (Figure 2) for comparison. A single beachworn specimen of Bullia annulata (La- marck, 1816) was collected dead by W. Anson at Flinder’s Bay, Augusta, W.A., on an unknown date, about the same time as the Nassarius kraussianus was collected. This spec- imen (WAM 52-82) is a juvenile shell that is 23.8 mm long, but the lower aperture is broken off (Figure 3). Despite being broken this specimen closely matches spec- imens from the Cape (NM and WAM collections; Figure 4). The Western Australian shell is not as heavy as the South African one but has the same stepped whorls, shal- low spiral grooves, and faint growth lines. The shell is Page 454 The Veliger, Vol. 28, No. 4 Explanation of Figures 1 to 6 Figure 1. Nassarius kraussianus (Dunker, 1846) from Augusta, W.A. WAM 51-82. Figure 2. Nassarius kraussianus (Dunker, 1846) from Durban, South Africa. WAM 50-82. Figure 3. Bullia annulata (Lamarck, 1816) from Flinder’s Bay, Augusta, W.A. WAM 52-82. buff colored, with distinct brown splotches just below the suture. The aperture is white. A single juvenile individual 19.7 mm long of Cymatium cutaceum africanum (A. Adams, 1854) was collected dead at Augusta, W.A., by W. Anson on 27 or 28 January 1979 (WAM 54-82). This species is discussed in detail by KILBURN & RippEy (1982) and is quite variable in South Africa, but the Western Australian specimen fits Figure 4. Bullia annulata (Lamarck, 1816) from False Bay, Mui- zenberg, South Africa. WAM 2672-83. Figure 5. Cymatium cutaceum africanum (A. Adams, 1854) from Augusta, W.A. WAM 54-82. Figure 6. Cymatium cutaceum africanum (A. Adams, 1854) from Nthlonyane, Transkei, South Africa. WAM 2671-83. easily into the range of variation observed in the species (Figures 5, 6). The Western Australian shell has a low spire, narrow umbilicus, and strong spiral cords—seven on the body whorl and two on the upper whorl. The spiral cords are crossed by several indistinct ribs and numerous fine growth lines. The spiral cords appear on the inside of the aperture as channels that extend onto the lip. The shell is a light brown and the aperture is whitish. F. E. Wells & R. N. Kilburn, 1986 DISCUSSION There are several points of similarity between the coastal environments of eastern South Africa and Western Aus- tralia. Both coasts show a parallel transition between a temperate-water fauna in the south and a tropical fauna of predominantly Indo-West Pacific incursives in the north (WELLS, 1980; KILBURN & RIPPEY, 1982). Although many such tropical species are common to both sides of the In- dian Ocean, the respective temperate-water molluscan faunas are very different, apart from certain tonnacean gastropods with teleplanic larvae (see BEU, 1976) which have been dispersed at various times since the Oligocene by the Westwind Drift, and circumtemperate species. En- vironmental factors of temperature, salinity, and topog- raphy are not dissimilar. For example, mean summer temperatures along most of the southern Cape coast (the center of distribution of all four species) are 19-20°C (CHRISTENSEN, 1980), which agrees with those of south- ern Western Australia (HODGKIN & PHILLIPS, 1969). Physical factors may thus support the colonization of the region by South African migrants. However, no direct evidence yet exists for the presence of established, viable populations of Nassarius kraussianus, Bullia annulata or Cymatium c. africanum in southern Western Australia. Nassarius kraussianus inhabits estu- aries and salt marshes in South Africa (KILBURN & RIPPEY, 1982). The site at Augusta, W.A., where the species was found is near the mouth of the Blackwood River, but two surveys of the estuary (WALLACE, 1975; WELLS & THRELFALL, 1981) did not record the species. Bullia an- nulata in South Africa is washed up in sheltered bays and lives in sand at low tide, but is most abundant subtidally at depths of up to 100 m, and C’ c. africanum lives among solitary ascidians offshore, under rocks at low tide or on sand near ascidians (KILBURN & RIPPEY, 1982). Halzotis spadicea was recorded by MACPHERSON (1953) as occur- ring alive near Cape Naturaliste in Western Australia. The Western Australian Museum conducted fieldwork in the Augusta to Cape Naturaliste area in January 1978 and April 1985 and failed to find living colonies of any of the South African species. Nor have local shell collectors reported additional finds of South African species, alive or dead, in Western Australia. Thus, the four species known to have crossed the southern Indian Ocean from South Africa to Western Australia appear to have arrived in small numbers and have not become established. The mechanism by which these species reached West- ern Australia is not known, but the literature suggests several possibilities: dispersal by pelagic larvae (SCHEL- TEMA, 1971), rafting on algae on the sides or in the ballast water of ships or on floating logs (SMITH, 1890; CLENCH, 1947), on the feet of birds (KEW, 1893), or in the gut of fishes. Although the reproductive mechanism of C. c. af- ricanum is not known, other cymatiids have long distance planktonic larvae that are able to cross open oceanic areas (SCHELTEMA, 1971). Nassarius kraussianus is ovovivipa- Page 455 rous with a planktonic veliger stage of a week or less (KILBURN & RIpPEY, 1982). Species of Bullia in which reproduction has been studied have either direct devel- opment (BROWN, 1982) or ovoviviparity (KILBURN, 1978). Halwtis have a planktonic stage of about one to two weeks (INO, 1952; LEIGHTON, 1972). Thus, none of these three species is likely to have arrived in southern Western Aus- tralia by means of a planktonic larval stage, but just how they arrived has not yet been determined. ACKNOWLEDGMENTS We sincerely thank Mrs. G. Hansen and Mrs. W. Anson for bringing their finds to our attention and donating the specimens to the Western Australian Museum. C. Bryce is thanked for taking the photographs. LITERATURE CITED Beu, A. G. 1976. Arrival of Semicassis pyrum (Lamarck) and other tonnacean gastropods in the southern ocean during Pleistocene time. J. Roy. Soc. N.Z. 6:413-432. Briccs, J.C. 1975. Marine zoogeography. McGraw-Hill: New York. Brown, A.C. 1982. The biology of sandy-beach whelks of the genus Bullia (Nassariidae). Oceanogr. Mar. Biol. Ann. Rev. 20:309-361. CHRISTENSEN, M.S. 1980. Sea-surface temperature charts for Southern Africa, south of 26°S. S. Afr. J. Sci. 76:541-546. CLENCH, W. J. 1947. The genera Purpura and Thais in the western Atlantic. Johnsonia 2:61-91. COSEL, R. vON. 1977. First record of Mitra mitra (Linnaeus, 1758) (Gastropoda: Prosobranchia) on the Pacific coast of Colombia, South America. Veliger 19:422-424. EMERSON, W. K. 1967. Indo-Pacific faunal elements in the tropical eastern Pacific, with special reference to the mol- lusks. Venus 25:85-93. GoMEZz, R. 1984. Primera cita para el Atlantico (Islas Canar- ias) de Bedeva paivae (Crosse, 1864). Bull. Malacologico 19: 249-252. Hopckin, E. P. & B. F. PHILLIPS. 1969. Sea temperatures on the coast of southwestern Australia. J. Proc. Roy. Soc. West. Austral. 53:59-62. Ino, T. 1952. Biological studies on the propagation of Jap- anese abalone (genus Haliotis). Bull. Tokai Reg. Fish. Res. Lab. 5:1-102. Kew, H. W. 1893. The dispersal of shells. Paul, Trench, Trubner & Co.: London. KiLvBurN, R. N. 1978. Four new Bullia species (Mollusca: Gastropoda: Nassariidae) from Kenya and Mozambique. Ann. Natal Mus. 23:297-303. KILBURN, R. N. & E. Rippry. 1982. Sea shells of southern Africa. Macmillan South Africa: Johannesburg. LEIGHTON, D. L. 1972. Laboratory observations on the early growth of the abalone Haliotis sorenseni, and the effect of temperature on larval development and settling success. Fish. Bull., Fish. Wildl. Serv. U.S. 70:7-19. MacPHERSON, J. H. 1953. Record of a South African mollusc from Australia (Haliotis sanguinea Hanley). Mem. Nat. Mus. Vic. 18:169. SCHELTEMA, R.S. 1971. Larval dispersal as a means of genetic exchange between geographically separated populations of shallow-water benthic marine gastropods. Biol. Bull. 140: 284-322. Page 456 The Veliger, Vol. 28, No. 4 SmiTH, E. A. 1890. Report on the marine molluscan fauna of St. Helena. Proc. Zool. Soc. Lond. 1890:247-317. WALLACE, J. 1975. The macroinvertebrate fauna of the Black- wood River estuary. West. Aust. Dept. Cons. Environ., Tech. Rept. 4. WELLS, F. E. 1980. The distribution of shallow-water marine prosobranch gastropod molluscs along the coastline of West- ern Australia. Veliger 22:232-247. WELLS, F. E. & T. J. THRELFALL. 1981. Molluscs of the Peel- Harvey estuarine system, with a comparison with other south-western Australian estuaries. J. Malacol. Soc. Aust. 5:101-111. WILSON, B. R. & K. GILLETT. lian shells. Reed: Sydney. 1979. A field guide to Austra- The Veliger 28(4):457-459 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 Indomya, a New Subgenus of Pholadomya from the Middle Jurassic of Kachchh, Western India (Bivalvia: Pholadomyidae) eke AT IAY. Department of Geology, Banaras Hindu University, Varanasi 221 005, India Abstract. A new subgenus of Pholadomya, Indomya, type Pholadomya (Indomya) rajnathi Jaitly, spec. nov., is described on the basis of four specimens from the Middle Bathonian (Middle Jurassic) of Kala Dongar, Pachchham Island, District Kachchh (Gujarat), Western India. Indomya differs from other members of Pholadomya by its faint vertical umbonal-ventral sulcus, an oblique posterior ridge, and surface ornamentation that consists of both concentric and radial ribs or threads. INTRODUCTION THE FAMILY Pholadomyidae Gray is represented in Kala Dongar by five genera: Pholadomya G. B. Sowerby, Homomya Agassiz, Oestomya Moesch, Pachymya J. Sow- erby, and Agrawalimya Singh, Jaitly & Pandey. Agra- walimya was created for specimens having a sulcus with asymmetrically inclined walls and extending from the umbo to just anterior to the middle of the ventral margin. The genus was tentatively referred to the Pholadomyidae because a sulcus was not previously considered to be of generic or subgeneric importance. However, this feature is frequently observed in many Middle Jurassic species of Pholadomya, and MOESCH (1878:58) and FURSICH (1982: 96) even mentioned the presence of a shallow sulcus be- tween the second and third anterior ribs in Pholadomya (Pholadomya) hemicardia Roemer. Subsequently, addi- tional specimens have been collected that have an outline similar to that of Pholadomya and Homomya, but which also possess a faint sulcus in the anterior third of the shell. To receive them, a new subgenus is created and tentatively assigned to the genus Pholadomya. A new subgenus, Indomya, with Pholadomya (Indo- mya) rajnathi Jaitly, spec. nov. as type, is described from Middle Bathonian rocks of Kala Dongar, Pachchham Is- land, Kachchh, India. The geology and stratigraphy of the area is described by JAITLY (1985); for location see JAITLY & SINGH (1983). SYSTEMATIC PALEONTOLOGY Class Bivalvia Order Pholadomyoida Suborder Pholadomyacea Family Pholadomyidae Pholadomya Sowerby, 1823 Type: Pholadomya candida G. B. Sowerby, 1823, by sub- sequent designation of Gray, 1847. (Indomya) Jaitly, subgen. nov. Etymology: Named after India. Type: Pholadomya (Indomya) rajnathi Jaitly, spec. nov., Middle Bathonian (Jurassic), Kachchh, India. Diagnosis: Shell sublunate with tapering posterior; sur- face with shallow, gradually downward widening sulcus extending vertically from anterior of umbo to ventral mar- gin and oblique posterior ridge; ornamentation of both concentric and radial ribs or threads. Remarks: In shape and size, Indomya is similar to Pho- ladomya s.s. and Homomya. The surface ornamentation, which consists of prominent radials that extend to the ventral margin, more closely resembles that of Pholadomya s.s. In Homomya, radial ornamentation is generally absent and, if present, is restricted to the umbonal region. Teto- rimya HayaMI (1959:151), from the Upper Jurassic of Japan, resembles Indomya in size and position of poste- rior gape, but lacks the vertical sulcus in the anterior region and differs in surface ornamentation. Agrawalimya differs in nature and position of its sulcus and the lack of radial ribs, which are prominent in Indomya. Page 458 The Veliger, Vol. 28, No. 4 Explanation of Figures 1 to 4 Pholadomya (Indomya) rajnathi Jaitly, subgen. et spec. nov. Figure 1. Holotype PK/139/3; Middle Bathonian, Kala Don- gar, Kachchh, India; exterior view of left valve. Figure 2. Holotype, dorsal view. Pholadomya (Indomya) rajnathi Jaitly, spec. nov. (Figures 1-4) Etymology: Named for the late Prof. Rajnath, an expert on the Kachchh Jura. Diagnosis: As for the genus. Types: Four paired specimens: holotype, PK/139/3, and three paratypes, PK/145/5, PK/145/3, and PK/141/11, deposited in the Invertebrate Paleontology Laboratory, Department of Geology, Banaras Hindu University, Var- anasi 221 005, India. Type locality: Middle Bathonian of Pachhmaipir, Kala Dongar (23°48'39"N, 69°50'E), Pachchham Island, Kachchh, India. Description: Shell medium sized (to 5 cm in length), highly inequilateral, moderately inflated and sublunate with ta- pering posterior end. Maximum inflation lies below um- bones, about one-third of distance to ventral margin. Um- bones orthogyrous, incurved, contiguous, and situated 6 to 8 mm from the anterior end. Lunule poorly defined, broadly ovate and small; escutcheon indistinct. Anterior margin broadly rounded, posterior margin acutely convex; ventral margin asymmetrically and gently convex, merg- ing with anterior and posterior in smooth curves. An ob- tusely rounded ridge, defined by the abruptly steeper slope Figure 3. Holotype, anterior view. Figure 4. Paratype PK/145/5; Middle Bathonian, Kala Don- gar, Kachchh, India; exterior view of left valve. of the surface posterior to it, extends obliquely from the umbo to the ventral margin slightly anterior to the pos- tero-ventral end of shell. A shallow, broadly rounded sul- cus extends vertically from the umbo downwards, grad- ually becoming shallower and wider. Surface sculpture consists of both concentric and radial ribs and (or) threads. The area anterior to the sulcus has only a few weak radial threads, but the area between the sulcus and the oblique ridge has narrow, widely spaced radial ribs. The area just posterior to the sulcus has four prominent ribs with a secondary riblet in each interspace. The secondaries gradually become stronger posteriorly and primaries become weaker, so that both are of equal strength and arranged in pairs. The area posterior to the oblique ridge is devoid of radial ornamentation and possesses only concentric ribs. Internal characters are unknown. Dimensions (mm): Specimen no. Length Height Inflation PK/139/3 (holotype) 49.5 335 24.5 PK/145/5 (paratype) 48.5 33 24 PK/141/11 (paratype) 54 38.5 31 PK/145/3 (paratype) 59 39 29 Remarks: The present specimens show some similarities in general outline and surface ornamentation to Pholado- mya inaequiplicata Stanton (IMLAY, 1964:C-36, pl. 4, figs. A. K. Jaitly, 1986 37-38) and Pholadomya ovalum Agassiz (LYCETT, 1863: 84, pl. 35, figs. 18, 18a). However, both P. inaequiplicata and P. ovalum lack the anterior sulcus and the posterior ridge. ACKNOWLEDGMENTS I am grateful to Prof. S. K. Agrawal, the late Dr. C. S. P. Singh, and Dr. S. Kanjilal, of the Department of Ge- ology, Banaras Hindu University, for their helpful dis- cussions and valuable suggestions. Thanks are also due to an anonymous reviewer for critically reviewing the manu- script, and to Dr. William J. Zinsmeister, Ohio State University, for providing a reprint of his pholadomyid paper. LITERATURE CITED FursicH, F. T. 1982. Upper Jurassic bivalves from Milne Land, East Greenland. Greenland Geologiske Undersee- gelse Bull. 144:1-126. Page 459 HayaMI, I. 1959. Late Jurassic isodont and myacid pelecypods from Makito, central Japan. Jap. J. Geol. Geogr. 30:151- 167. IMLay, R. W. 1964. Marine Jurassic pelecypods from central and southern Utah. J. S. Geol. Surv. Prof. Paper 483-C:1- 42. JaitLy, A. K. 1985. Note on the Middle Jurassic rocks of Kala Dongar, Pachchham Island, Diast. Kachchh, Gujarat. Proc. IV Indian Geol. Congr., Varanasi:55-62. JaitLy, A. K. & C. S. P. SINGH. 1983. Stratigraphy of the Bajocian sediments of Kachchh, Gujarat (W. India). Proc. Ind. Natl. Sci. Acad. 49(A):503-508. LycETT, J. 1863. Supplementary monograph on the Mollusca from Stonesfield Slate, Great Oolite, Forest Marble and Cornbrash. Paleo. Soc. London. 129 pp. MoescH, C. 1878. Monographie der Pholadomyen. Ach. Schweiz. Paléont. Ges. 1:135 pp. The Veliger 28(4):460-465 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 NOTES, INFORMATION & NEWS Retention of Nassarius corpulentus (C. B. Adams, 1852) in West American Nassariid Nomenclature by Walter O. Cernohorsky Auckland Institute and Museum, Private Bag, Auckland 1, New Zealand PETIT (1984) drew attention to the existence of the taxon Cancellaria nassiformis Lesson, 1842, as being an earlier name for Nassarius corpulentus (C. B. Adams, 1852) from the west coast of America. He advocated either the accep- tance of Cancellaria nassiformis Lesson, which on exami- nation of the type specimens proved to be conspecific with Nassarius corpulentus (C. B. Adams), or a rejection of Les- son’s name by action of the International Commission on Zoological Nomenclature. The publication of Declaration 43 (INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLATURE, 1970) has established binding rules for the treatment of unused names in zoology. Under these emended rules of Article 23(b), paragraph 2(b)(ii), a se- nior synonym is considered to be a nomen oblitum if during the immediately preceding 50 years (7.e., 1934-1984) it has not once been applied to a particular taxon as its presumably valid name. Declaration 43 makes it clear that a nomen oblitum shall not replace a name that has been in current use for at least 50 years, and “current usage” has been defined as the usage of the name during the last 50 years by at least five different authors in 10 different publications. Search through malacological literature has revealed a published usage between 1934-1984 of the name Nassar- wus corpulentus by nine different authors in 12 different publications as follows: Nassarius corpulenta: BALES, 1938:45. Nassarius corpulentus: DEMOND, 1951:16, 1952:314, pl. 1, fig. 6; KEEN, 1958:409, fig. 571; MCLEAN, 1970:129; KEEN, 1971:606, fig. 1295; CERNOHORSKY, 1975:168, fig. 93, 1982:17-209; KAICHER, 1982: card 3148; ABBOTT & DANCE, 1982:179, fig. bottom row center. Nassa corpulenta: TURNER, 1956:44, pl. 5, fig. 3. Nassa corpulentus: OLIVEIRA et al., 1981:209. I am certain that more time and effort would have probably unearthed even a wider usage of Nassarius cor- pulentus, but this would have been purely of statistical interest. Unless a usage of the taxon Cancellaria nassifor- mis Lesson is found between 1934-1960, in which in- stance the name has been applied to a valid taxon (men- tion in synonymy, listing in an Index, or list of names does not qualify), the epithet Nassarius corpulentus (C. B. Adams) must be retained in nassariid nomenclature. LITERATURE CITED ApspoTT, R. T. & S. P. DANCE. 1982. Compendium of sea- shells. A color guide to more than 4,200 of the world’s marine shells. E. P. Dutton Inc.: New York. 411 pp. BaLes, B. R. 1938. Marine collecting on the west coast of Mexico. Nautilus 52(2):41-46. CERNOHORSKY, W. O. 1975. The taxonomy of some west American and Atlantic Nassariidae based on their type- specimens. Rec. Auckland Inst. Mus. 12:121-173, figs. 1- 93. CERNOHORSKY, W. O. 1982. Family Nassariidae Iredale, 1916. Suppl. 2:17-201-17-243. In: R. J. L. Wagner & R. T. Abbott, Standard catalogue of shells. 3rd ed. American Mal- acologists Inc.: Greenville, Delaware. DEMOND, J. 1951. Key to the Nassariidae of the west coast of North America. Nautilus 65(1):15-17. DEMOND, J. 1952. The Nassariidae of the west coast of North America between Cape San Lucas, Lower California, and Cape Flattery, Washington. Pacific Sci. 6(4):300-317, pls. ily INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLATURE. 1970. Declaration 43. Repeal of Article 23(b). Bull. Zool. Nomencl. 27(3/4):135-162. KaICcHER, S. D. 1982. Card catalogue of world-wide shells. Pack No. 31. Nassariidae. Part 1. S. D. Kaicher: St. Pe- tersburg, Florida. 105 cards. KEEN, A. M. 1958. Sea shells of tropical west America. Ma- rine mollusks from Lower California to Colombia. Stanford Univ. Press: Stanford, Calif. 626 pp. KEEN, A. M. 1971. Sea shells of tropical west America. Ma- rine mollusks from Baja California to Peru. 2nd ed. Stan- ford Univ. Press: Stanford, Calif. 1064 pp. McLgEan, J. H. 1970. New species of tropical eastern Pacific Gastropoda. Malacol. Review 2:115-130, figs. 1-41. OuiverraA, M. P. DE, G. DE J. R. REZENDE & G. A. DE CASTRO. 1981. Catalogo dos Moluscos da Universidade Federal de Juiz de Fora. Sinonimia de Familia, Género e Espécie. Minist. Educ. & Cultura: Juiz de Fora, Brasil. 520 pp. Petit, R. E. 1984. An earlier name of Nassarius corpulentus (C. B. Adams, 1852). Veliger 26(4):330. TURNER, R. D. 1956. The eastern Pacific marine mollusks described by C. B. Adams. Occas. Pap. Mollusks, Mus. Comp. Zool. Harvard 2(20):21-136, pls. 5-21. Observation of Predation on a Pleuronectid Fish by Navanax inermis (Opisthobranchia: Cephalaspidea) by Stephen A. Karl 1260 Oliver Ave., San Diego, California 92109, U.S.A. Navanax inermis (Cooper, 1862), which ranges from Monterey Bay, California, to the Gulf of California, Mexico, is an active and voracious predator reported to Notes, Information & News eat gastropods, annelids, arthropods, and fish (PAINE, 1963, table 1; BLAIR & SEapy, 1972). On 23 July 1981, a 40-mm long (2.0 g wet weight, displacing 0.9 mL seawater) Navanax inermis was ob- served off Naples Reef, Naples, California, with a 25-mm total length (19 mm standard) live spotted turbot (Pleu- ronichthys ritteri) in its pharynx. The animal was discov- ered and collected on a sand-rock interface in 13.7 m of water. When it was found, approximately one quarter (6.5 mm) of the turbot’s anterior region was inside the opisthobranch’s buccal mass. The turbot was still alive, as the tail was moving quickly from side to side. Although Navanax inermis has been previously reported to eat fish, these reports have been mainly of Porichthys myriaster (PAINE, 1963). These fish seasonally migrate inshore to mate and deposit their eggs on the underside of rocks. The male guards the eggs which remain attached to the rocks until the yolk sac is completely absorbed (BREDER & ROSEN, 1966). Because they are not able to swim the larvae are vulnerable to attack by N. inermis and this may account for PAINE’s (1963) observation of large numbers of 20-30-mm fish in the fecal remains from N. inermis. Whether or not N. inermis was actively pur- suing the turbot or if mucus secreted by the fish was in- volved in its detection and eventual capture is difficult to determine; the Navanax may have encountered the turbot only by chance. LITERATURE CITED Biair, G. M. & R. R. Seapy. 1972. Selective predation and prey location in the sea slug Navanax inermis. Veliger 15(2): 119-124. BREDER, C. M., JR. & D. E. ROSEN. 1966. Modes of repro- duction in fishes. Natural History Press: Garden City, New York. 598 pp. PAINE, R. T. 1963. Food recognition and predation on opis- thobranchs by Navanax inermis (Gastropoda: Opisthobran- chia). Veliger 6(1):1-9. International Commission on Zoological Nomenclature The following Opinions of potential interest to our read- ers have been published by the International Commission on Zoological Nomenclature in the Bulletin of Zoological Nomenclature, volume 42, part 3, on 30 September 1985: Opinion No. 1331 (p. 230). Sphaeriidae Jeffreys, 1862 (1820) (Mollusca, Bivalvia): placed on the Official list. Opinion No. 1350 (p. 283). Conus antiquus Lamarck, 1810 (Mollusca, Gastropoda): neotype suppressed. Erratum: Volume 28, Number 3 (2 January 1986) A typographical error has unfortunately escaped our attention in the article by Steven M. Chambers, “Two new bulimulid land snail species from Isla Santa Cruz, Page 461 Galapagos Islands,” which appeared in Volume 28, Number 3 (2 January 1986). In Table 2 the value given for the “No. whorls, other” of the holotype should be 5.25, not 4.25 as printed. California Malacozoological Society California Malacozoological Society, Inc., is a non-profit educational corporation (Articles of Incorporation No. 463389 were filed 6 January 1964 in the office of the Secretary of State). The Society publishes a scientific quarterly, The Veliger. Donations to the Society are used to pay a part of the production costs and thus to keep the subscription rate at a minimum. Donors may designate the Fund to which their contribution is to be credited: Operating Fund (available for current production); Sav- ings Fund (available only for specified purposes, such as publication of especially long and significant papers); or Endowment Fund (the income from which is available. The principal is irrevocably dedicated to scientific and educational purposes). Unassigned donations will be used according to greatest need. Contributions to the C.M.S., Inc., are deductible by donors as provided in section 170 of the Internal Revenue Code (for Federal income tax purposes). Bequests, lega- cies, gifts, and devises are deductible for Federal estate and gift tax purposes under sections 2055, 2106, and 2522 of the Code. The Treasurer of the C.M.S., Inc., will issue suitable receipts which may be used by donors to substan- tiate their tax deductions. Page Charges Although we would like to publish papers without charge, high costs of publication require that we ask authors to defray a portion of the cost of publishing their papers in The Veliger. We wish, however, to avoid possible financial handicap to younger contributors, or others without fi- nancial means, and to have charges fall most heavily on those who can best afford them. Therefore, the following voluntary charges have been adopted by the Executive Board of the California Malacozoological Society: $30 per printed page for authors with grant or institutional sup- port and $10 per page for authors who must pay from personal funds (2.5 manuscript pages produce about 1 printed page). In addition to page charges, authors of papers containing an extraordinary number of tables and figures should expect to be billed for these excess tables and figures at cost. It should be noted that even at the highest rate of $30 per page the Society is subsidizing well over half of the publication cost of a paper. However, authors for whom the regular page charges would present a financial handicap should so state in a letter accompa- nying the original manuscript. The letter will be consid- ered an application to the Society for a grant to cover necessary publication costs. Page 462 The Veliger, Vol. 28, No. 4 We emphasize that these are voluntary page charges and that they are unrelated to acceptance or rejection of manu- scripts for The Veliger. Acceptance is entirely on the basis of merit of the manuscript, and charges are to be paid after publication of the manuscript, if at all. Because these contributions are voluntary, they may be considered by authors as tax deductible donations to the Society. Such contributions are necessary, however, for the continued good financial health of the Society, and thus the contin- ued publication of The Veliger. Reprints While it was hoped at the “birth” of The Veliger that a modest number of reprints could be supplied to authors free of charge, this has not yet become possible. Reprints are supplied to authors at cost, and requests for reprints should be addressed directly to the authors concerned. The Society does not maintain stocks of reprints and also can- not undertake to forward requests for reprints to the au- thor(s) concerned. Patronage Groups Since the inception of The Veliger in 1958, many generous people, organizations, and institutions have given our journal substantial support in the form of monetary do- nations, either to The Veliger Endowment Fund, The Ve- liger Operating Fund, or to be used at our discretion. This help has been instrumental in maintaining the high qual- ity of the journal, especially in view of the rapidly rising costs of production. Ata recent Executive Board Meeting, we felt we should find a way to give much-deserved recognition to those past and future donors who so evidently have our best interests at heart. At the same time, we wish to broaden the basis of financial support for The Veliger, and thus to serve our purpose of fostering malacological research and publica- tion. Accordingly, it was decided to publicly honor our friends and donors. Henceforth, donors of $1000.00 or more will automatically become known as Patrons of The Veliger, donors of $500.00 or more will be known as Sponsors of The Veliger, and those giving $100.00 or more will become Benefactors of The Veliger. Lesser donations are also sincerely encouraged, and those donors will be known as Friends of The Veliger. As a partial expression of our gratitude, the names of donors in these different categories will be listed in a regular issue of the journal. Of course, we will honor the wishes of any donor who would like to remain anonymous. The Treasurer of the California Malacozoological Society will provide each member of the new patronage groups with a receipt that may be used for tax purposes. We thank all past and future donors for their truly helpful support and interest in the Society and The Velr- ger. Through that support, donors participate directly and importantly in producing a journal of high quality, one of which we all can be proud. Notes to Prospective Authors The increasing use of computers to prepare manuscript copy prompts the following notes. We request that the right margin of submitted papers be prepared “ragged,” that is, not justified. Although right-justified margins on printed copy sometimes look “‘neater,” the irregular spac- ing that results between words makes the reviewer’s, ed- itor’s, and printer’s tasks more difficult and subject to error. Similarly, the automatic hyphenation capability of many machines makes for additional editorial work and potential confusion; it is best not to hyphenate words at the end of a line. Above all, manuscripts should be printed with a printer that yields unambiguous, high-quality copy. With some printers, especially some of the dot-matrix kinds, copy is generally difficult to read and, specifically, the letters ‘‘a, p, g, and q” are difficult to distinguish, especially when underlined as for scientific names; again, errors may result. Other reminders are (1) that three copies of everything (figures, tables, and text) should be submitted to speed the review process, and (2) absolutely everything should be double-spaced, including tables, references, and figure leg- ends. Because The Veliger is an international journal, we oc- casionally receive inquiries as to whether papers in lan- guages other than English are acceptable. Our policy is that manuscripts must be in English. In addition, authors whose first language is other than English should seek the assistance of a colleague who is fluent in English before submitting a manuscript. Subscription Rates and Membership Dues At its regular Annual Business Meeting on 25 September 1985, the Executive Board of the California Malacozoo- logical Society, Inc., set the subscription rates and mem- bership dues for Volume 29 of The Veliger. For affiliate members of the Society, the subscription rate for Volume 29 be US$25.00; this now includes postage to domestic addresses. For libraries and nonmembers the subscription rate will be US$50.00, also now with postage to domestic addresses included. An additional US$3.50 is required for all subscriptions sent to foreign addresses, including Can- ada and Mexico. Affiliate membership in the California Malacozoologi- cal Society is open to persons (no institutional member- ships) interested in any aspect of malacology. There is a one-time membership fee of US$2.00, after payment of Notes, Information & News which, membership is maintained in good standing by the timely renewal of the subscription. Send all business correspondence, including subscription orders, membership applications, payments for them, and changes of address to C.M.S., Inc., P.O. Box 9977, Berke- ley, CA 94709. Moving? If your address is changed it will be important to notify us of the new address at least six weeks before the effective date and not less than six weeks before our regular mailing dates. Send notification to C.M.S., Inc., P.O. Box 9977, Berkeley, CA 94709. Because of a number of drastic changes in the regula- tions affecting second class mailing, there is now a sizable charge to us on the returned copies as well as for our remailing to the new address. We are forced to ask our members and subscribers for reimbursement of these charges; further, because of increased costs in connection with the new mailing plate, we also must ask for reim- bursement of that expense. The following charges must be made: change of address and re-mailing of a returned issue— $2.75 minimum, but not more than actual cost to us. We must emphasize that these charges cover only our actual expenses and do not include compensation for the extra work involved in re-packing and re-mailing re- turned copies. Sale of C.M.S. Publications All back volumes still in print, both paper-covered and cloth-bound, are available only through “The Shell Cab- inet,” 12991 Bristow Road, Nokesville, VA 22123. The same applies to the supplements still in print, with certain exceptions (see below). Prices of available items may be obtained by applying to Mr. Morgan Breeden at the above address. Volumes 1 through 13, 24, 26, and 27 are out of print. Supplements still available are: part 1 and part 2, sup- plement to Volume 3, and supplements to Volumes 7, 11, 14, 15, and 16; these can be purchased from ‘““The Shell Cabinet” only. Copies of the supplement to Volume 17 (“Growth rates, depth preference and ecological succes- sion of some sessile marine invertebrates in Monterey Harbor” by E. C. Haderlie) may be obtained by applying to Dr. E. C. Haderlie, U.S. Naval Post-Graduate School, Monterey, CA 93940; the supplement to Volume 18 (“Chitons”’) is available from ‘““The Secretary,’ Hopkins Marine Station, Pacific Grove, CA 93950. Some out-of-print editions of the publications of C.M.S. are available as microfiche reproductions through Mr. Page 463 Steven J. Long. The microfiches are available as negative films (printed matter appearing white on black back- ground), 105 mm x 148 mm, and can be supplied im- mediately. The following is a list of items now ready: Volumes 1-6: $9.95 each. Volumes 7-12: $12.95 each. Supplement to Volume 6: $3.95; to Volume 18, $6.95. Send orders to Mr. Steven J. Long, Shells and Sea Life, 1701 Hyland, Bayside, CA 95524. A. Myra Keen (1905-1986) Malacologists everywhere have lost a friend and col- league, A. Myra Keen (23 May 1905-4 January 1986). She contributed frequently to the Veliger as author, re- viewer, and member of the Editorial Board, her kindness and care in reading manuscripts endearing her to Found- ing Editor Rudolf Stohler and to all whose papers passed her desk. A formal memorial is in preparation, but we acknowledge here our affection for the First Lady of Mal- acology and our appreciation of her as a teacher, author, nomenclatural expert, curator, and advisor. In one or more of these roles she affected the work of most malacologists during the last quarter of a century. Dr. Keen was associated from 1934 to 1970 with the Department of Geology, Stanford University. A psychol- ogy major (A.B., Colorado College; M.Sc., Stanford Uni- versity; Ph.D., University of California, Berkeley), she graduated during the depression when there were no jobs and directed her diagnostic skills and scientific back- ground to an interest in shells, eventually becoming an international authority in malacology and professor of pa- leontology. She taught advanced paleontology, biological oceanography, and curatorial methods, and concurrently wrote more than 75 scholarly papers and nine books. Thoroughly knowledgeable on taxonomic procedure, she chaired the nomenclatural committee of the Society of Sys- tematic Zoology and was directly involved in refining the rules by which all new animal taxa are named. As curator of the Stanford University research collections, Dr. Keen was proud of the fact that the Tertiary and Recent spec- imens were not only well arranged but also identified. She acquired shells from all over the world in return for iden- tifying duplicate lots, and maintained a type collection that exceeded 6,400 lots before it was transferred to the California Academy of Sciences. Visitors who climbed to her third-floor office recall the exhibits, as pleasing as they were informative, and the bookshelves lined with reprint boxes and journals. ‘There Dr. Keen painstakingly unscrambled “thorny problems,” as she called the more complicated nomenclatural puzzles, and administered the French exam required of many graduate students. She advised more than a dozen ad- vanced degree candidates in geology and biology (several now head the molluscan sections of some of the nation’s Page 464 most important institutions for malacological research). She had many foreign visitors, the most publicized being Emperor Hirohito of Japan, and carried out an immense correspondence, some of it scholarly and some leaving her chuckling over yet another request for “everything you know about shells.” Dr. Keen was able to envision projects of enormous scope and break them into increments she could accom- plish between lecture preparations, oral exams, classes, and caring for her invalid mother. Long before the word processor, she worked at an old manual typewriter, put- ting masses of information into a book that provoked a surge of research on eastern Pacific mollusks. Recognizing the gaps in our knowledge of west coast taxa described by early foreign workers, she obtained a prestigious John Simon Guggenheim Fellowship to visit European mu- seums and photograph the type specimens. Her greatest work, “Sea Shells of Tropical West America” (1971. 2nd ed. Stanford University Press: Stanford, Calif. 1064 pp.), is the standard reference for eastern Pacific mollusks. In private life, Dr. Keen was a member of the Religious Society of Friends, a quiet pacifist, and an ardent feminist. Shy, but sure of her convictions, she firmly opposed smok- ing and frequently wrote letters in support of wildlife and conservation. She enjoyed classical music, especially Brahms, poetry referring to the sea, and keeping in touch with students and close friends. Her thorough, meticulous research and thoughtful ways will continue to inspire all who share her interests and build on her work. Judith Terry Smith Donald Putnam Abbott (1920-1986) Stanford University Professor Emeritus Professor Don- ald P. Abbott died at his home in Honolulu, Hawaii, on 18 January 1986. He was 65 years old. Dr. Abbott is survived by his wife, University of Hawaii Botany Pro- fessor Isabella A. Abbott and daughter Ann K. Abbott of Honolulu. Don Abbott will be greatly missed by not only his family but also his numerous colleagues, former grad- uate students, and the hundreds of other students who had the opportunity to take the courses he taught between 1950 and 1982 on the Stanford campus and at the Hop- kins Marine Station in Pacific Grove, California. Donald Abbott was born in Chicago, Illinois, on 14 October 1920, and there received his primary and second- ary education. He enrolled as a freshman at the Univer- sity of Hawaii in 1937, at the age of 16, and was awarded a bachelor’s degree in Zoology in 1941. His master’s de- gree program at the University of Hawaii was interrupted by the attack on Pearl Harbor, and he turned to instruct- ing in Zoology at the University of Hawaii until enlisting in the U.S. Army in 1943. Donald Abbott married his undergraduate classmate, Isabella Aiona, in 1943, and they remained in Hawaii throughout the war years. In The Veliger, Vol. 28, No. 4 1946, after his discharge from the Army, Abbott enrolled in the graduate program in Zoology at the University of California at Berkeley, receiving a master’s degree in 1948 and a Ph.D. in 1950. Stanford University hired Dr. Abbott to its faculty at the Hopkins Marine Station in 1950, and he remained there until his retirement in 1982. His summer inverte- brates course at Hopkins became one of the best known in the world and was an introduction to marine animals that launched many students on their lives’ work. A con- summate teacher, Don Abbott provided a model that a generation of graduate students still seeks to emulate in style and substance. Abbott was a major force in the es- tablishment of a Hopkins Marine Station “spring course” that eventually attracted national attention for its success in involving undergraduate students and faculty in joint, original research in marine biology. Donald Abbott was major advisor to 26 doctoral and 10 master’s degree stu- dents, and an ex officio mentor to many more. In recog- nition of his superior teaching accomplishments, Abbott received a number of prestigious teaching awards from Stanford University. Professor Abbott’s major research interests were in the biology and taxonomy of tunicates and in animal phylog- eny. He was the author of numerous research papers and co-author of two books: “Coral island: Portrait of an atoll” (1958, with Marston Bates) and “Intertidal Invertebrates of California” (1980, with R. H. Morris & E. C. Had- erlie). A book of Dr. Abbott’s drawings of invertebrate animals is currently being prepared for publication in the careful editorial hands of his former student Galen H. Hilgard. Another of Abbott’s former students, Prof. A. T. Newberry of U.C. Santa Cruz is completing a summary of the Hawaiian tunicate fauna on which Abbott had been working at the time of his death. The international community of invertebrate zoologists has lost one of its finest colleagues, best teachers, and most creative thinkers. Don Abbott will be sorely missed. Michael G. Hadfield American Malacological Union, Meeting The 52nd annual meeting of the American Malaco- logical Union will be held in historic Monterey, Califor- nia, from 1-6 July 1986, at the new Sheraton Hotel. The new Sheraton is adjacent to Fisherman’s Wharf, within a few blocks of Steinbeck’s Cannery Row, and is surround- ed by many historic sites. An international symposium on opisthobranch mol- lusks in honor of Dr. Eveline Marcus is being planned and organized by Dr. Terrence Gosliner and Dr. Michael Ghiselin. A second international symposium on molluscan morphological analysis is being planned by Drs. Carole Hickman and David Lindberg. There will be contributed papers and a poster display. A special cephalopod session honoring Dr. S. S. Berry is being organized by Dr. Roger Notes, Information & News Hanlon (you can watch the boats fish for squid from the hotel). A workshop on photography is planned as well as the traditional auction. Tentative field trips will include tours of the California Granite Canyon Shellfish Culture Lab- oratory, the Moss Landing Marine Laboratories, the rich molluscan intertidal areas of the Monterey Peninsula, and some special fossil-rich cliffs. A dredging trip on the re- search vessel “Cayuse” is also planned. The meeting will be highlighted by a special afternoon affair at the all new Page 465 Monterey Bay Aquarium, with its spectacular kelp tank and associated displays, and a banquet featuring Mac- Arthur Fellow Dr. Michael Ghiselin as the speaker. The Western Society of Malacologists will meet jointly with the AMU for this meeting, ensuring that this will be an outstanding meeting. For further information contact: Dr. James Nybakken, AMU President, Moss Landing Marine Laboratories, Box 223, Moss Landing, CA 95039. EDITOR’S NOTE: The Tables of Contents and Au- thor Index for papers that appeared in Volume 28 will be printed in the first issue of Volume 29. The Veliger 28(4):466-468 (April 1, 1986) THE VELIGER © CMS, Inc., 1986 BOOKS, PERIODICALS & PAMPHLETS A Guide to the Common Molluscs of South-western Australian Estuaries by Frep E. WELLS, with photography by Clayton W. Bryce. 1984. Western Australian Museum: Francis Street, Perth 6000, Australia. 112 pp.; 41 pls. Paperback, $3.50 Australian, plus postage. This compact book, intended as an easy identification guide for professional scientists, students, and interested amateurs, illustrates 82 of the common and widespread macromollusks occurring in estuaries of southwestern Australia, from Esperance to the Moore River (75 km north of Perth). Included are 50 gastropods, 31 bivalves, and 1 chiton. Although all are to be found in the estuaries covered, the species’ salinity tolerances and distributions vary considerably in specifics: some are characteristic of purely freshwater habitats, whereas others show distinctly marine affinities. For each species, a black-and-white half- tone plate is provided, along with other basic information: family, common name, scientific name, a description of the shell or animal, habitat and geographical range, and notes. In addition to these 82 descriptions, the book con- tains a glossary, many useful references, and an appendix listing species known to occur in estuaries of the region but not presented in detail. D. W. Phillips Archaeogastropod Biology and the Systematics of the Genus Tricolia (Trochacea: Tricoliidae) in the Indo-West-Pacific by ROBERT ROBERTSON. 1985. Monographs of Marine Mollusca, Number 3. American Malacologists, Inc.: P.O. Box 2255, Melbourne, FL 32902. 104 pp.; 96 pls. Sta- pled, $13.50. In Number 3 of “Monographs of Marine Mollusca,” Robertson has provided a useful biological and systematic treatment of the genus 777colia in the Indo-West Pacific, in which nine species are accepted and treated. The title’s prominent reference to coverage of “archaeogastropod bi- ology” is perhaps a bit misleading, as only a few pages deal with the biology of species other than T7icolia and these with only selective aspects. Also, the rather muddy printing of the many half-tone illustrations has not done justice to the author’s undoubtedly fine original photo- graphs. However, the work as a whole is a scholarly monograph of the Indo-West Pacific representatives of a world-wide, temperate and tropical genus, placed in a context of relevant work on other archaeogastropods; as such it should be of interest to general students of mollusks as well as specialists on archaeogastropod systematics. D. W. Phillips Larval Forms and Other Zoological Verses by WALTER GARSTANG. Reprinted 1985. University of Chicago Press: 5801 S. Ellis Ave., Chicago, IL 60637. 98 pp. Paperback, $5.95. Garstang, a Professor of Zoology at Leeds and a Fellow of Lincoln College at Oxford University, first published a collection of his verses about ontogeny and phylogeny in 1951. These delightful, witty verses express, in a whimsical way, Garstang’s fascination with larval forms but also capture the more serious essence of several sci- entific debates on evolutionary biology, past and present. Certainly one of the more famous verses, “The Ballad of the Veliger, or How the Gastropod got its Twist,” is a marvelous presentation of Garstang’s notion of the adap- tive advantage of gastropod torsion. In addition to the original 26 verses, this 1985 edition also includes Gar- stang’s Presidential Address to the British Association for the Advancement of Science (entitled “The Origin and Evolution of Larval Forms”) and a new Foreword, by Michael LaBarbera, that provides explanatory notes for many of the verses and relates Garstang’s work to the evolutionary thinking of his time. This book is sure to be enjoyed by students, teachers, and professional scientists alike, indeed by anyone who shares Garstang’s fascination with the lives of larvae and the relationships between on- togeny and phylogeny. D. W. Phillips The Printer’s Catch An Artist’s Guide to Pacific Coast Edible Marine Animals by CHRISTOPHER M. DEWEES. 1984. Sea Challengers: 4 Sommerset Rise, Monterey, CA 93940. 112 pp.; 63 color pls. Hardcover, $26.95. Although the pages of the “Books, Periodicals & Pam- phlets” section are usually reserved for publications hav- ing a formally scientific and distinctly molluscan focus, occasionally there appears a book that presents the beauty of nature in such a way that our readers may wish to know about it. Such a book is “The Printer’s Catch.” Constructed around 63 color plates of original fish rub- bings (gyotaku) made by naturalist, fisheries biologist, and artist Christopher Dewees, the book combines art and Books, Periodicals & Pamphlets nature in an unusual and pleasing way. Each print, beau- tifully reproduced (as seems to be standard for Sea Chal- lengers publications) is accompanied by the common and scientific names of the subject and by pertinent life history, fisheries, and consumer information. Also provided are methods and materials for gyotaku, which some see as a unique form of scientific illustration and documentation as well as art. Although most of the prints are of marine fish, covering the majority of Pacific coast fish families, eight species of mollusks are included in the prints. Enjoy. D. W. Phillips The Distributions of the Native Land Mollusks of the Eastern United States by LesLig HuBRICcHT. 1985. Fieldiana, Zoology, New Se- ries, No. 24. 191 pp., 523 distribution maps. $23.00. Terrestrial malacologists have waited a long time for the publication of Mr. Leslie Hubricht’s distribution maps of eastern United States land mollusks. This book contains maps and habitat notes for the 523 species and subspecies of native land mollusks of the eastern United States (east of the western boundaries of North and South Dakota, Nebraska, Kansas, Oklahoma, and Texas east of the Pe- cos River) recognized by Hubricht. The maps summarize data from 55 years of collecting by Mr. Hubricht (about 43,000 lots), his identifications of material for other work- ers, 20 years of examining material in major eastern United States museums, and data from PILsBRy’s (1939-1948) monograph. An important feature of the book in addition to the distribution maps is the inclusion of nomenclatural and systematic changes since publication of Pilsbry’s monograph. Because this book is not intended as an iden- tification guide, it contains no descriptions of taxa. The five sections of the book are (1) a short introduc- tion, (2) an annotated systematic list, (3) a list of refer- ences from which distribution records were taken, (4) the distribution maps, and (5) an index to taxa. The index is alphabetical for all levels of taxa from subclass to subspecies, and includes synonyms of taxa from genus to subspecies. For example, Glyphyalinia rhoadsi austrina, formerly known as Retinella rhoadsi austrina, can be found in the index under six permutations of the tri- nomials. The distribution maps make up the bulk of the book. They are drawn on base maps showing the counties of the United States. The maps are of good quality and are easily readable. Nowhere are states or counties on the maps identified. A separate map is used for each taxon. County records are shown with different symbols used to indicate living individuals, Pleistocene fossil occurrences, and oc- currences known only from river drift. This is the first time that many of the distribution records have been pub- lished. This book can be compared to British mapping of non- Page 467 marine Mollusca by vice-counties which began in 1876 (KERNEY, 1982). More recently, the Conchological Soci- ety of Great Britain and Ireland has mapped detailed distributions of British species of terrestrial molluscs on a grid system of 10 x 10 km squares (KERNEY, 1976). Hu- bricht’s distribution maps are based on counties (mean 1746 km? per county, my calculation) and are comparable to the British mapping by vice-counties. Finer scale map- ping and more intensive surveys of United States land Mollusca will undoubtedly follow. Not only should distribution maps show where individ- uals of a species occur, maps should also show enough detail to reveal possible correlations between the geo- graphic range of a species and various environmental fac- tors such as geology, vegetation, and climate, and maps should be able to show temporal changes in the distribu- tion of a species (KERNEY, 1967). Hubricht’s distribution maps provide sufficient detail for environmental compar- ison, although no maps of environmental factors are in- cluded. Unfortunately, Hubricht’s maps give only a par- tial indication of temporal changes. Map symbols showing Pleistocene occurrences indicate reductions but not expan- sions in range, and range changes in historical times are not shown. Furthermore, as the title indicates, introduced species are not included in this book, although their dis- tributions would be interesting from a perspective of his- torical distribution change. In the systematic list Hubricht gives, for each taxon, a list of references (exclusive of titles) published since Pils- bry’s monograph that affect the systematics or nomencla- ture of the taxon, and brief notes on habitat. Remarks or notes on variation are included for about one-tenth of the taxa. Systematists may be frustrated that Hubricht intro- duced in this book a systematic revision of taxa for which some of the changes in rank or status have not been jus- tified in print. Hubricht promises that justification for these changes will be published elsewhere. The new clas- sification is similar to previous arrangements of other au- thors, but reflects Hubricht’s opinions on the status of some taxa. For example, Hubricht uses the suborders Au- lacopoda and Holopodopes, but he did not use the long recognized suborder Holopoda. One might expect Hubricht’s list of references affecting the status of taxa, or naming new taxa, to replace the paper by MILLER et al. (1984) because Hubricht covers all native United States terrestrial Mollusca in contrast to the work of Miller et al. which updated only volume 1 (Helicacea and Polygyracea) of PILSBRY (1939-1948). However, each work has a few unique references for the taxa they have in common, so the two works complement each other. The distribution maps, updated systematic list, and ref- erences make this a necessary book for terrestrial mala- cologists, and the maps are of value to biogeographers. Ecologists will find it useful in seeking correlations of molluscan distributions with environmental parameters. Page 468 Quaternary paleontologists will find the fossil and Recent distribution records useful for documenting temporal changes in species ranges. The references are valuable to people tracing the nomenclatural history of a species. En- vironmental consultants will find this a helpful book as well. Hubricht points out that this is a working document. It is a starting point for future, more intensive mapping of faunal distributions. At the same time it is a useful and significant step toward documenting and understanding the distributions of United States native Mollusca. Literature Cited KERNEY, M. P. 1967. Distribution mapping of land and fresh- water Mollusca in the British Isles: a brief history and fu- ture prospects. J. Conch. 26:152-160. KERNEY, M. P. 1976. Atlas of the non-marine Mollusca of the British Isles. Institute of Terrestrial Ecology, Cambridge. 199 pp. KERNEY, M. P. 1982. Vice-comital census of the non-marine Mollusca of the British Isles (8th edition). J. Conch. 31(1): 63-71. MILLER, W. B., R. L. REEDER, N. BABRAKZAI & H. L. Fair- BANKS. 1984. List of new and revised Recent taxa in the North American terrestrial Mollusca (north of Mexico) published since 19 March 1948, part 1. Tryonia 11:1-14. Pitsspry, H. A. 1939-1948. Land Mollusca of North America (north of Mexico). Acad. Natur. Sci. Phila., Monograph No. 3, 2 volumes in 4 parts: 2007 pp. Timothy A. Pearce Genus Clypeomorus Jousseaume (Cerithiidae: Prosobranchia) by RICHARD S. Housrick. 1985. Smithsonian Contri- butions to Zoology, no. 403. 131 pp., 62 figs., 33 tbls. A recurring problem arising in literature pertaining to the ecology, physiology, or natural history of mollusks is doubt about the actual identity of species studied. This problem is more acute with certain groups of mollusks than with others and blame cannot rest entirely with the nonsystematist(s) whose results are in question. It is not their fault that the systematics of abundant and ecologi- cally important or otherwise interesting species are poorly understood. A case in point is the cerithiid genus Clypeo- morus, species of which commonly occur in high density populations on tropical intertidal shores where they con- stitute an ecologically important group of microphagous herbivores. The tortuous nomenclatural history of many of the species, and the high degree of interspecific simi- larity and intraspecific variability, have led to uncertainty of identification in the literature on the ecology and re- production of members of the genus. These problems in nomenclature and variation, and the identities of the species studied in the nonsystematic literature if that can be de- termined, have been admirably sorted out by Houbrick. The Veliger, Vol. 28, No. 4 Richard Houbrick has brought many forms of evidence to bear on the systematics of this very confusing group. In addition to nomenclatural history and detailed study of shells, radulae, and soft parts, he includes data on bio- geography, the fossil record, habitat, ecology, and repro- duction, in assessing the status of 15 Recent and fossil Clypeomorus species. His conclusions are based upon ex- tensive field research, study of thousands of specimens from major museum collections worldwide, and exami- nation of relevant type material. These data are summa- rized and presented in parallel fashion under subject headings, in tables, and with clearly reproduced photo- graphs for each species treated. This consistent arrange- ment makes it particularly easy to compare data among species. The extensive lists of material examined would have been less obtrusive if placed in an appendix, but this hardly detracts from the usefulness of the work. Houbrick also calls attention to gaps in our knowledge of Clypeo- morus where future research opportunities lie. Of special interest to evolutionary biologists are possible examples of sibling species and candidates for incipient speciation. A refreshing trend in several recent molluscan mono- graphs including this one is the inference of species phy- logeny by cladistic methods. In such attempts it often be- comes clear that even the most basic information is missing for some species and unresolved polychotomies may result; but even so, hypotheses of relationships are rigorously formulated and presented in a manner condusive to fur- ther testing. Unfortunately, typographical errors (other- wise rare in this work) necessitate caution in the inter- pretation of the cladogram supplied. For example, Clypeomorus adunca is scored as beaded in Table 5 (char- acters 1 and 7) although the description and illustrations indicate it has smooth sculpture. Similarly, C. subbrevicula is scored as unbeaded at character 1, but is beaded ac- cording to the description and illustrations. On the clado- gram (Figure 1) C. subbrevicula is indicated as having state 2 of character 1 (beaded sculpture) although states O (present) and 1 (absent) are the only possibilities pre- sented in Table 5. Seven character changes are indicated for C. adunca, but there are actually 9 if the scores for characters 1 and 7 are corrected. None of these apparent errors change the topology of the resulting cladogram, however, and the thoroughness and clarity used in pre- senting characters and methods make the errors easy to spot. The publication of this monograph should make future nonsystematic work on Clypeomorus more useful by facil- itating the proper identification of this interesting group. Systematic monographs have extra value when they treat organisms that are difficult to identify and are used in nonsystematic work. Good systematics, as exemplified in this monograph, not only inform, but also provide direc- tion for nonsystematic studies. Michael G. Kellogg Information for Contributors Manuscripts Manuscripts must be typed on white paper, 812” by 11”, and double-spaced throughout (including references, figure legends, footnotes, and tables). If computer generated copy is to be submitted, margins should be ragged right (7.e., not justified). To facilitate the review process, manuscripts, including figures, should be submitted in triplicate. The first mention in the text of the scientific name of a species should be accompanied by the taxonomic authority, including the year, if possible. Underline scientific names and other words to be printed in italics. Metric and Celsius units are to be used. The sequence of manuscript components should be as follows in most cases: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, lit- erature cited, figure legends, figures, footnotes, and tables. The title page should be on a separate sheet and should include the title, author’s name, and address. The abstract should describe in the briefest possible way (normally less than 200 words) the scope, main results, and conclusions of the paper. Literature cited References in the text should be given by the name of the author(s) followed by the date of publication: for one author (Smith, 1951), for two authors (Smith & Jones, 1952), and for more than two (Smith e¢ al., 1953). The “literature cited” section must include all (but not additional) references quoted in the text. References should be listed in alphabetical order and typed on sheets separate from the text. Each citation must be complete and in the following form: a) Periodicals Cate, J. M. 1962. On the identifications of five Pacific Mitra. Veliger 4:132-134. b) Books Yonge, C. M. & T. E. Thompson. 1976. Living marine molluscs. Collins: London. 288 pp. c) Composite works Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135. In: R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal invertebrates of California. Stanford Univ. Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Figures and plates Figures must be carefully prepared and should be submitted ready for publication. Each should have a short legend, listed on a sheet following the tables. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photographs for half-tone plates must be of good quality. They should be trimmed off squarely, arranged into plates, and mounted on suitable drawing board. Where necessary, a scale should be put on the actual figure. Preferably, photographs should be in the desired final size. It is the author’s responsibility that lettering is legible after final reduction (if any) and that lettering size is appropriate to the figure. Charges will be made for necessary alterations. Processing of manuscripts Upon receipt each manuscript is critically evaluated by at least two referees. Based on these evaluations the editor decides on acceptance or rejection. Acceptable manuscripts are returned to the author for consideration of comments and criticisms, and a finalized manuscript is sent to press. The author will receive from the printer two sets of proofs, which should be corrected carefully for printing errors. At this stage, stylistic changes are no longer appropriate, and changes other than the correction of printing errors will be charged to the author at cost. One set of corrected proofs should be returned to the editor. An order form for the purchase of reprints will accompany proofs. If reprints are desired, they are to be ordered directly from the printer. Send manuscripts, proofs, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. CONTENTS — Continued Reproductive cycle in the freshwater mussel Diplodon chilensis chilensis (Mol- lusca: Bivalvia). SANTIAGO, PEREDO/AND ESPERANZA\ PARADA G)- (tase a eke ly Ris tenia eae Egg capsule and young of the gastropod Beringius (Neoberingius) friele: (Dall) (Neptuneidae). RICHARD Ay MAGCINTOSH (0 CUE SIN Monee) <0). Wes 013th ae Cea The systematic position of Royella sinon (Bayle) (Prosobranchia: Cerithiidae). RICHARD'S:; HOUBRICK)) (enh) a ERRNO ait Casula Coke de ea New Philippine Cancellariidae (Gastropoda: Cancellariacea), with notes on the fine structure and function of the nematoglossan radula. RICHARD E. PETIT; AND) M.(G: FIARASEWYCHS <9 00). ee A new species of Helminthoglypta (Gastropoda: Pulmonata: Helminthoglypti- dae) from the Cuyamaca Mountains of southern California. RICHARD: L. (REEDERG 13 4:00. Sa Gaetan ie Cle aoe. beer A new species of Ischnochiton (Mollusca: Polyplacophora) from the tropical eastern Pacific. ANTONIO. J. FERREIRA WA CoN UE Ue Ne oi Sie Sede oh Three temperate-water species of South African gastropods recorded for the first time in southwestern Australia. FRED EE. WELLS AND RICHARD N-KIEBURN |. 0.2.23) 3.) 9 02 Indomya, a new subgenus of Pholadomya from the Middle Jurassic of Kachchh, Western India (Bivalvia: Pholadomyidae). AKG JATIIUY s \ieceiiil (Oh etalon VEU amebiasis) ea! LOS UN Ree aa rr NOTES, INFORMATION & NEWS Retention of Nassarius corpulentus (C. B. Adams, 1852) in west American nassarlid nomenclature. WiALTER Ok CERNOHORSK Yipes se ee ti a Ec Observation of predation on a pleuronectid fish by Navanax inermis (Opistho- branchia: Cephalaspidea). STEPHEN AU KARL jlo" coated pekaianrs cables eles yoni ey ey ne SNe Be RNG eae J ~ rey) 5 Wy = “ary & \ i] 2 oe = Ormrsy m ~~ D ) My Wn CALS 2 Yig Au Non Sy SY Muy: m 2 hin Li = wo ee = = 7) = ,LIBRARIES SMITHSONIAN : NOMATILSNIUNVINOSH MINS) Gal uMea ig MUNBIR ARTES. INSTITE SVE < z S = ye 2 Zz SN WS oH Zz = > = See = = \\ > Wo & = S) ae a) iw or x DS Wei BRR” FB D D S BY 2 g 2 YS : SY = E 2 E 2 E 2 - a = See = > = > = ee = nD ~ A (Tp) a z Ww) Pad (2) vs z (Tp) Ne AILS HT NMINOSH TNS Voi o ise in —!NSTITUTION NOILNLILSNI NVINOSHLINS SI1UVE re a ays it 2 ud 2 ta # tu | be = WW . (oe _- x. A Yi, — oe aS fad x H NWS < =f xt 4, Wy a4 x 4 x : Se Z a “gy 2 E 5 = Nae 0 = eines fo) a je O = fe) = | e Zz ie} Zz = ca — Zz SA a LIBRARIES SMITHSONIAN _INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYUGIT LIBRAR {ES SMITHSONIAN INSTIT i z . oe S Ce = Las = : Si -_ oO O O <= 2) 7 — : = 2 : 5 E 3 = 50 = kz > =) D> =) =) ke > = = : = Ee ae Ee 3 g = a = & a a - i a ia) ¥ Ww m se YW) sf D z a a ms z oO z a | NOILALILSNI NVINOSHLINS S3!IYVUGIT LIBRARIES INSTITUTION NOILALILSNI NVINOSHLINS S3IYY n z w - ra wy ee) = ¥ Ww = 2) a | © , = = Wy, a = < = wy, = = | _ GL Z me = Y y, = : Gy: 2 Yo 3 : : 2 Wy: : 8 MY hi ? 8 Yh 2 NB 2 @ bi be? a Yn 2, 25) = oY 2 = 2 Os 2G ; a 2 a oe a 5 a 2 | LIBRARIES SMITHSONIAN MUU RSI SNS EO MIMO Uy LIBRARIES SMITHSONIAN INSTITU > on = Ss = n Ss | - Ww 2 us o Ww 2 WwW by F : z wa = : “ = Yu % : < io < 2 .< a < U7 fe a r oc € oc = a = o Y. fos c ) a S) = S = S a 3S 2 a 2 a, 2 - z a 2 | NOILNLILSNI NVINOSHLINS S3IYVYUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS S3luvil z f= z UE 3 G = & nu = | > 2 es gv 2) ef S) a Ss = = = a ee a = ay - : 2 - = F = . = E u m 2 aa z He z x Z ~ 122) » = — : = = | (LIBRARIES SMITHSONIAN INSTITUTION NOILALIESNIT TUWUGIILLIBRARIES _INSTIT | c Sie = < WS = < = se Se = < NS r WX = > sr ‘S =; Seo = Sj > Wi EOL = S = = 2 ly 5. : 2 OV E We 2 BS g z SG yg 2 : ENN | = SK 2 = 2 = 2 Cy i 2 =z WY = gos = > = > = ss S 7) 2 a soe wo 2 7%) Byte no jNOLLNLILSNI NVINOSHLIWS S3IYVYSIT_ LIBRARIES SMITHSONIAN _INSTITUTION NOILNLILSNI NVINOSHLIWS Satu | — (dp) = . J z SN os Wl 4 uw 2 ON ra x n We on pd Bet au i OD. a wn” Las c = NS ce = He Uh fig = a = a 2 2 z =¢i7 3 < Z on | c SQN o = oc We = a a 2 NY: 5 = Uy = q a ] J 7 Sey 2 ay iz a = Seed TI BRAR TES, SMITHSONIAN, INSTITUTION NOLLNLILSNI NVINOSHLIWS (S31 ¥ Yai T 7 LIBRARIESS SMITDSONIANINeiI | J — se) wo s oo = ow ; - Gy ® E > We 5 : : 3 » re ye Vd. > = > \ SS E > res “, ) = UY fl. “20 = e) RN WW = 70 — feo) z Oo 2 ee B z o NOILALILSNI_NVINOSHLINS SAIUVYGIT_ LIBRARIES SMITHSONIAN INSTITUTION NOILLILSNI NVINOSHLINS S3IUV El ? a ; ; | ft = = , = = < z < Sy, | Li Wy : Ne w (op) ss Giff | PE ie Z E * 2 Fe 2 = 2 “7 A a = = = SS > = Si s >" | ” Pe, i?) ne iia ” z (7p) io ) LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI_ NVINOSHLIWS LIBRARIES SMITHSONIAN INSTITUM e Z a a 5 z wn Z ani ) uJ a at 7) ee 7} cd no Ke ( zs “ = bes = te = a VA = 5 | . ap = WW], = WX = Re, 3 Bae SINS = GE 2 Xk %S x re) Zz \N 2 = xO) ry) 2 Ss ox ZON n> S ¢ YP m ON. DC. Lv Cami m Ee “nis m m 'N! D — = a) ale OILNLILSNI NVINOSHIINS Salyvegl_LIBRAR | ade e anes INSTT OTON mo Pe luv Sie ee = ze Z a es : = a Atty y z = = : \ A LL Af 4 S = “YY y ay 3 s GY DB o o a ee B b Yt ZY. r 0 GG x O 2 ies Ba o UG “Wy E 2 GO ‘ = = Zz Yi E 2 ip : = > SE = aye Ss > GQ = > ; Fe 7) BN 4 () nae n Zz IBRARI ES. SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLIWS $3 byuvudit LIBRARI ES SMITHSONIAN _INSTIT| cm. el a zal fod = fad : = as = re ioe] —_ oO — wo = es IN ss a) = | 59) as Be = 2 WANE z F i> F > = = WL" 5 = a? = o = o ae = a z a z B lai pp oMITHSONIAN INSTITUTION NOLLNLILSNI Ee INOSHTING > 3 IuYVvVyudit ao! B RARI PMU CTT UNS = =i =) ee = = 8 = As = Ss 2 = Y fy = re Ne : 2 \y i GY; : é - \ , 2 wn ‘~ AN . Ww YH Ze wn B) (2) nD) 7p) WS WE nN Es GE: g 3 = . = Se Wr ee eS =. = = = OX Ss = = \ > = > = = = oS = w cae Zz w 2 w 72) 10 IENIILSNI_NVINOSHLINS Sa 1yVvug Tale BRARI ES SMITHSONIAN INSTITUTION pNOLLNLILSNI_NVINOSHLINS S34 ldV ay - a us Ai us G a 4 | ea, ac 4 NX = a = a ee s » < (= s VY ¥ (as < , (= (Ss ? feed A ANY & 4 fa = a& 4 a oy 2 5 ca a = 5 = Zz Say Jj Zz a; 2 3 z J LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3iy¥vudit SHIA I So eine al Male iul Zz z - z a = ee) = ow = o = mo!) =n ce =) D 4 > a = a Ke: > = > KE > ky = és vA = 0 = 0 = os; 7 a 7) a = w Z # Z a ie as Z a OILNLILSNI p JIUVESIILLIBRARIES | SMITHSONIAN INSTITUTION ee yoalav = S : z Se = z el, & e < ; Zz = ire as Zz = z Sy 4 9 z 2 z 2 a 2 ad = 2 = 2 2 : = 2 SMITHSONIAN _INSTITUTION NOILNLILSNI MC AS) tuvud 1 OM ITHSONIAN _ INSTITU ee 2 RG z = pa ee B 2 a RN ben & se £ 4 “ae. ¢vyZ YZ aa, a et NN SS fae = [eae = % GO ad < Sallie OS i? =A <% a SS SSSy > ‘\N A \ 2 2 hK~ * > (f= WW E ee) PN WW aa Po] re a PE mo) NYAS = 2 ANS = a Bir = b = WY 2 ae 2 3 Z A Z a ane IBRARI Bee BSOMIANT INS TINUTION NOVNLISNI MIs UAT 8 1UVvyuald ae B RARI HENAN Us ‘ z aot = : < = i") = < =| S = Ye fy Zz =| z A A z 7 fe) t tttty. o =e ) AN: = 5 ao a 7) Vy Wp m) 7) a VN ¥ 77) S = 9 YF fy = S F N\A 2 = 5 = ia s > = SX = = 2 Z 77) 2 7) 7) \ NVINOSHLINS _LIBRARI ES SMITHSONIAN INSTITUTION Cee cee 7 2 us es “Y, eo n ~ — : : : 2 G4 : + Nyy = a a Yee fp 7 A a cal Ss. < c c < Vy GE Ge c RK ce =a] ee =a o Puede 2 Gianna a sl SN SS on | i | | 00842 693 N INSTITUTIO} | ii 3 9088 | =—— —__ ————— == A ae nag OS Se ese eae Batam soar evs e oe ee Sea S Hn Th eee ea Cinremeat v UP WASP UTE NIaR 46 Wurueu RPE Uiy “WTA ES Wwe 9 weet rare PRS ty cicart