Skip to main content

Full text of "The Nautilus"

See other formats


1e, NAUTILUS 


Volume 122, Number1 
March 28, 2008 
ISSN 0028-1344 


A quarterly devoted 
to malacology. 


EDITOR-IN-CHIEF 


Dr. José H. Leal 

The Bailey-Matthews Shell Museum 
3075 Sanibel-Captiva Road 
Sanibel, FL 33957 


BUSINESS MANAGER 


Mary Jo Bunnell 

The Bailey-Matthews Shell Museum 
3075 Sanibel-Captiva Road 
Sanibel, FL 33957 


EDITOR EMERITUS 


Dr. M. G. Harasewych 
Department of Invertebrate Zoology 
National Museum of 

Natural History 
Smithsonian Institution 
Washington, DC 20560 


CONSULTING EDITORS 


Dr. Riidiger Bieler 
Department of Invertebrates 
Field: Museum of 

Natural History 
Chicago, IL 60605 


Dr. Arthur E. Bogan 

North Carolina State Museum of 
Natural Sciences 

Raleigh, NC 27626 


Dr. Philippe Bouchet 

Laboratoire de Biologie des 
Invertébrés Marins et Malacologie 

Muséum National d'Histoire Naturelle 

55, rue Buffon 

Paris, 75005 France 


Dr. Robert H. Cowie 

Center for Conservation Research 
and Training 

University of Hawaii 

3050 Maile Way, Gilmore 409 

Honolulu, HI 96822 


Dr. Robert T. Dillon, Jr. 
Department of Biology 
College of Charleston 
Charleston, SC 29424 

Dr. Eileen H. Jokinen 
8234 E. North Shore Road 
Sault Ste. Marie, MI 49783 


Dr. Douglas S. Jones 

Florida Museum of Natural History 
University of Florida 

Gainesville, FL 32611-2035 


Dr. Harry G. Lee 
4132 Ortega Forest Drive 
Jacksonville, FL 32210 


Dr. Charles Lydeard 

Biodiversity and Systematics 
Department of Biological Sciences 
University of Alabama 

Tuscaloosa, AL 35487 


Bruce A. Marshall 
Museum of New Zealand 
Te Papa Tongarewa 
P.O. Box 467 
Wellington, NEW ZEALAND 


Dr. James H. McLean 
Department of Malacology 
Natural History Museum 
of Los Angeles County 
900 Exposition Boulevard 
Los Angeles, CA 90007 


Dr. Paula M. Mikkelsen 

Paleontological Research 
Institution 

1259 Trumansburg Road 

Ithaca, NY 14850 


Dr. Diarmaid O Foighil 

Museum of Zoology and Department 
of Biology 

University of Michigan 

Ann Arbor, MI 48109-1079 


Dr. Gustav Paulay 
Florida Museum of Natural History 
University of Florida 


Gainesville, FL 32611-2035 


Mr. Richard E. Petit 
P.O. Box 30 
North Myrtle Beach, SC 29582 


Dr. Gary Rosenberg 

Department of Mollusks 

The Academy of Natural Sciences 
1900 Benjamin Franklin Parkway 
Philadelphia, PA 19103 


Dr Angel Valdés 
Department of Malacology 
Natural History Museum 
of Los Angeles County 
900 Exposition Boulevard 
Los Angeles, CA 90007 


Dr. Geerat J. Vermeij 
Department of Geology 
University of California at Davis 
Davis, CA 95616 


Dr. G. Thomas Watters 
Aquatic Ecology Laboratory 
1314 Kinnear Road 
Columbus, OH 43212-1194 


SUBSCRIPTION INFORMATION 


The subscription rate per volume is 
US $47.00 for individuals, US $80.00 
for institutions. Postage outside the 
United States is an additional US 
$10.00 for regular mail and US 
$25.00 for air delivery. All orders 
should be accompanied by 
payment and sent to: THE 
NAUTILUS, P.O. Box 1580, 
Sanibel, FL 33957, USA, (239) 
395-2233. 


Change of address: Please inform 
the publisher of your new address at 
least 6 weeks in advance. All 
communications should inclide both 
old and new addresses (with zip 
codes) and state the effective date. 


THE NAUTILUS (ISSN 0028-1344) 
is published quarterly by The Bailey- 
Matthews Shell Museum, 3075 

Sanibel-Captiva Road, Sanibel, FL 


Periodicals postage paid at Sanibel, 
FL, and additional mailing offices. 


POSTMASTER: Send address 
changes to: THE NAUTILUS 
P.O. Box 1580 

Sanibel, FL 33957 


CONTENTS 


NAUTILUS 


Volume 122, Number 1 
March 28, 2008 
ISSN 0028-1544 


Philippe Bouchet 
Richard E. Petit 


Vittorio Garilli 


Diego G. Zelaya 


New species and new records of southwest Pacific Cancellariidae 
(Gastropoda) Fe oa. Qala ad deol Maio HG. gnerd Malan Gwe eR oe Re bE as | 


On some Neogene to Recent species related to Galeodina Monterosato, 1SS4, 
Galeodinopsis Sacco, 1895, and Massotia Bucquoy, Dautzenberg, and Dollfus, 
1884 (Caenogastropoda: Rissoidae) with the description of two new Alvania 
species from the Mediterranean Pleistocene 2... 0.0... 0000 ee 19 


Reallocation of Cyamiocardium crassilabrum Dell, 1964, into Perrierina 
Bernard, 1897 (Bivalvia: Cyamiidae) ......0.000000 0000000 eee eee 52 


Notice ................ 


MBLWHOI! Library 


JUN 1 @ “UAB 


Woops HOLE 
Massachusetts 02543 


THE NAUTILUS 122(1):1-1S, 2008 


Page | 


New species and new records of southwest Pacific 


Cancellariidae (Gastropoda) 


Philippe Bouchet 

Muséum National d’Histoire Naturelle 
55, rue Buffon 

75005 Paris, FRANCE 
pbouchet@mnhn.fr 


re.petit@att.net 


Richard E. Petit 
S06 St. Charles Road 
North Myrtle Beach, 


SC 29582 USA 


ABSTRACT 


Fifteen species of Cancellariidae referable to the genera Zead- 
mete, Admetula, Fusiaphera, Nipponaphera, and Trigonostoma 
are reported from depths between 200 and 700 m in New 

Caledonia and other island groups in the southwest Pacific. 
Twelve are new species: Zeadmete bathyomon new species, 
Zeadmete physomon new species, Zeadmete bilix new species, 
Admetula affluens new species, Admetula marshalli new spe- 
cies, Admetula bathynoma new species, Admetula lutea new 
species, Admetula emarginata new species, Nipponaphera argo 
new species, Nipponaphera agastor new species, Nippona- 
phera tuba new species, and Trigonostoma tryblium new spe- 
cies. All the Recent nominal species of Fusiaphera described 
from localities throughout the meee Pacific area are considered 
to be conspecific, the senior name being Fusiaphera mac- 
rospira (Adams and Reeve, 1850), now with ten synonyms. The 
ranges of Nipponaphera nodosivaricosa (Petuch, 1979) and 
Trigonostoma thysthlon Petit and Harasewych, 1987, are ex- 
tended to the South Pacific. 


INTRODUCTION 


The present paper is a continuation of our study of the 
deep-water cancellariid fauna of the Southwest Pacific, 
based on the material originating from recent expedi- 
tions in New Caledonia, Vanuatu, Fiji, Tonga, Wallis & 
Futuna, and the Solomon Islands. In a previous paper 
(Bouchet and Petit, 2002), we described the new genus 
Mirandaphera and nine new species in the genera Afri- 
cotriton, Merica, Sveltia, and Nipponaphera. We here 
deal with 15 species (12 new) in the genera Zeadmete, 
Admetula, Fusiaphera, and Trigonostoma, and add tur- 
ther species in Nipponaphera. Our review of the deep- 
water cancellariid fauna so far sampled in the southwest 
Pacific will be complete after a third paper (in prepara- 
tion) dealing with the genera Brocchinia, Microcancilla, 
and Gergovia. In addition, the cancellariid fauna of New 
Caledonia includes shallow-water to offshore species in 
the genera Scalptia (5 species) Tritonoharpa (several 
species), as well as the rediscovered endemic Merica 
semperiana, which we intend to deal with separately. 
Cancellariid radulae are not known to provide dis- 


criminating species-level characters, and we thus did not 
attempt to systematically examine them when we had 
live-taken specimens available. Much of the material re- 
ported in this series was collected in the 19S0—1990s and, 
at the time, fixed in formalin and then rinsed and dried. 
It is thus not adequate for nucleic-acid sequencing. More 
recent expeditions generate new material that is specifi- 
cally put aside for barcoding. Our treatment of the can- 
cellariid fauna is thus currently restricted to a descr iption 
of the shells, including the protoconch, but we may ex- 
pect that in the future it will be possible to test some of 
our species limits through molecular characters. 


MATERIALS AND TEXT CONVENTIONS 


In the lists of type and other material examined, indi- 
vidual lots in MNHN are unambiguously designated by 
the combination of cruise acronym (capitalized) and sta- 
tion number. DW (for Drague Warén) refers to dredge 
hauls, CP (for Chalut & Perche) to beam trawls; lv refers 
to live-taken specimens, dd to empty shells; spms to in- 
dividuals that cannot be unambiguously assigned to one 
of these two categories (esse -ntially commercially ob- 
tained specimens) Institutional acronyms are: AMNZ: 
Auckland Museum Auckland, New Zealand: BMNH: 
The Natural History Museum, London, Uk; DMNH: 
Delaware Museum of Natural History, Wilmington, 
Delaware, USA: MNHN : Museum National d’ Histoire 
Naturelle, Paris, France; NM: Natal Museum, Pieterma- 
ritzburg, South Africa, NMW: National Museum of 
Wales, Cardiff, UK: NSMT: National Science Museum, 
Tokyo, Japan; USNM: National Museum of Natural His- 
tory, Smithsonian Institution, Washington, DC, USA; 
WAM: Western Australian Museum, Perth, Australia. 


SYSTEMATICS 


Family Cancellariidae Forbes and Hanley, 1851 
Genus Zeadmete Finlay, 1926 


Type Species: Cancellaria trailli: Hutton, 1873, by 
original designation. Recent, New Zealand. 


Page 2 


THE NAUTILUS, Vol. 122, No. 1 


Discussion: The genus Zeadmete was Sas d_ by 
Finlay (1926: 429) who later (1930b: 242) considered it 
to be a subgenus of Oamaruia Finlay, 1924 from the 
Lower Miocene of New Zealand. Powell (1979: 224) also 
treated Zeadmete as a subgenus of Oamaruia. However, 
Garrard (1975: 44) and Wilson (1994: 173) restored 
Zeadmete as a full genus, as did Petit and Harasewych 
(2000: 151), who gave a brief discussion of its possible 
relationship to other taxa. Among other differences, the 
type species of Oamaruia, Admete suteri Marshall and 
Murdoch, 1920, has strong columellar folds whereas 
Zeadmete has weak, almost obsolete, folds. 

The genus Zeadmete, as interpreted here, occurs in 
the Miocene to Recent faunas of New Zealand and in the 
Recent faunas of South Africa, Fiji and New Caledonia 
(Z. kulanda Garrard, from Australia, is probably an Ip- 
hinopsis.) Expedition material in MNHN also contains 
an undescribed species from the Solomon Islands and 
another one from New Caledonia, both represented by 
specimens too imperfect to be named. All live offshore in 
300-600 m, with New Zealand records as shallow as 65 
meters. 


Zeadmete bathyomon new species 
(Figures 1-2) 


Description: Protoconch glassy, smooth, erect, of 
about 1.3 whorls, diameter 900 xm. Teleoconch of four 
whorls. Transition to teleoconch marked by a sharp 
growth line and a spiral cord forming shoulder angle, 
shortly followed by onset of weak sak sharp axial bs, 
forming small nodes at strengthening shoulder angle. Ad- 
ditional spiral cords appear before end of first teleoconch 


whorl. Shoulder angle prominent, bearing minute coro- 
nations formed by intersection of spiral cords and axial 
ribs. Sutural ramp strongly concave, bearing five to seven 
fine, low spiral cords. Suture attachment orthogonal. 
Shoulder raised, bearing three to five closely spac ed spi- 
ral cords. About 20+ wider, low spiral cords anterior to 
shoulder angle, extending down onto base. Fine, low, 
evenly space d axial ribs extending from shoulder angle to 
anterior end of shell; about 25 on last whorl; interspaces 
between axial ribs bearing closely packed fine growth 
lines. Low, narrow varices, formed. only by a slight raising 
of shell surface, at about 120° increments on final w horls, 
Oute r lip smooth, faintly sinuate adapically, curving back 

n prosocline direction abapically. Aperture elongate, 
ne aes elliptical, without lirations inside outer lip. Pa- 
rietal area ai very thin, narrow callus. Columella with 
three folds, the posterior two weak; anterior fold broad, 
situated on small siphonal fasciole, forming edge of in- 
ductural callus, which then forms edge of distinct sipho- 
nal canal. Exterior beige, rims of varices pale straw- 
yellow. 


Type Material: 


Wa. 


Holotype MNHIN 20498, 10.2 x 4.8 


Type Locality: South of New Caledonia, 22°17’ S, 
167°12’ E, 390 m [VAUBAN 1978-79: sta. 3]. 


Material Examined: Norfolk Ridge, south of New 
Caledonia.—VAUBAN 1978-79: sta. 3, 22°17’ S, 
167°12' E, 390 im, 1 lv (Figure 1).—BIOCAL: sta. 
DW77, 22°15’ S, 167°15' E, 440 m, 1 dd—BATHUS 2: 
sta. DW719, 99°48! S, =e E, 444-445 m, 1 lv.— 
SMIB 8: sta. DW166, 23°38’ S, 167°43" E, 433-450 m, 


Figures 1-5. Zeadinete. 1-2. Zeadmete bathyomon new species. 1. Holotype, height 10.2 mm; New Caledonia, Norfolk Ridge, 
22°17' S, 167°12’ E, 390 m [Vauban 1978-79 sta. 3]. 2. 23°3 167°43' FE, 433-450 m [SMIB S sta. DW166]. 3, 4. Zeadmete 
physomon new speci holotype height 15.1 mm, Fiji, 19°01’ S, 178°25' E, 500-516 m [BORDAU 1 sta. DWI14SS]. 5. Zeadmete 
bilix new specie holotype height 13.2 mm;, New Caledonia, Norfolk Ridge, 23°02’ S$, 168°16’ E, 335 m [SMIB 5 sta. DW98] 


P. Bouchet and R. E. Petit, 2008 


Page 3 


1 dd (Figure 2); Sta. DWI167, 23°38" S, 167°43' E, 43¢ 
452 m, | dd—NORFOL 7 l: sta. DW1666, 23°42’ S. 
167°44' E, 469-S60 m, 2 i —NORFOLK 2: 


DW 2024, 23°28’ S, 167°5 1’ E, 370-371 m, 1 dd. 


Etymology: From the Greek bathus, deep, combined 
with omos, shoulder, for the deeply concave sutural 
ramp. To be treated as a noun in apposition. 


Distribution: Known only from south of New Cale- 
donia, alive in 390-444 m. 


Remarks: — In Zeadmete finlayi Powell, 1940, the shoul- 
der is also raised but the sutural 1 ramp is not concave. In 
Z. bathyomon, the shoulder is even with, or raised above, 
the suture. Also, Z. finlayi has clathrate sculpture only on 
the adapical half of the last whorl, whereas the last whorl 
is entirely clathrate in Z. bathyomon, The aperture is 
two-thirds of the shell height in Z. finlayi, but in Z. 
bathyomon the aperture is just over half the shell height. 

The only other Zeadmete species we are aware of swith 
a sutural ramp that does not slope abapically is an unde- 
scribed species represented by two broken, worn speci- 
mens from a nearby station on the Norfolk Ridge 
[BERYX 11: sta. DW35, 23°33’ S, 167°16' E, 550-570 
m], which we leave undescribed because of the poor 
condition of the specimens. It differs from Z. bathyomon 
in having a shorter spire and a flatter sutural ramp that is 
devoid of spiral cords. It also has a larger protoconch 
with a diameter of 1100 jum. 


Zeadmete physomon new species 
(Figures 3+) 


Description: Protoconch smooth, glossy, of 1.5 
whorls, diameter 1250 wm. Transition to teleoconch 


marked by weak axial rib and faint spiral cords, both of 


which increase in strength rapidly. Teleoconch of four 
whorls, shell thin. Spiral cords flat, evenly spaced, with 
interspaces Ss lightly narrower than cords, about five on 
sutural ramp and 20+ anterior to shoulder angle. Axial 
ribs extending from suture to anterior end of te Teaeoiuoh, 
interspaces consider rably broader than ribs; about 20 ribs 
on last whorl. Sutural ramp slightly convex, sloping up to 
impressed suture. Aperture elongate- elliptical, only 
weakly angulate at shoulder, smooth within. Outer lip 
sinuous adapically but becoming prosocline anterior to 
periphery. Parietal wash fadisanct on holotype, distinct 
on one paratype. Columella almost vertical, bearing 
three folds, posterior fold on top of siphonal fasciole. 
Anterior folds callused, widely separated, anterior one 
almost obsolete, forming edge of short siphonal canal. 
Exterior cream with ill-defined pale brown band on pe- 
riphery of spire whorls; some ribs on last whorl have 
brownish tint. 

Type Material: Holotype MNHN 20519 (13.1 « 6.5 
mm) and 6 paratypes MNHN 20520-20521. 

Type Locality: Lau Ridge, Fiji, 19°01’ S, 175°25' E, 
500-516 m [BORDAU 1: sta. DW1488]. 
BORDAU |] 


Material Examined: — Fiji. sta. DW14S6, 


L9°O1' S, 178°26' EB, 395-540 m, 1 dd paratype MNHN 
20520.— Sta. DW1488, 19°O1' S, 178°25' EB, 500-516 m, 
6 dd, holotype MNTIN 20519, paratypes MNHN 20521 
(Figures 5-4), 


Etymology: From the Greek hae inflated, and omos, 
shoulder, Tor the appearance of the sutural ramp. To be 
treated as a noun in apposition. 


Distribution: 
listed above. 


Known only from Fiji at the two stations 


Remarks: Protoconch diameter in sigh ranges 
from 900 to 1300 jum. Zeadmete physomon differs from 
Z. bathyomon in having narrower cael cords with wider 
interspaces, fewer, more widely spaced axial ribs, and a 
slightly convex rather than strongly concave sutural 
ramp. Also, in Zeadmete physomon the axial ribs remain 
distinct below the periphery. 


Zeadmete bilix new species 
(Figure 5) 


Description: —Protoconch prominent, smooth, glassy, 

of 1.1 whorls, diameter $75 jum. Transition to teleoconch 
marked by onset of axial ribs and spiral cords. Teleo- 
conch high-spired, of five whorls. Spiral cords narrow, 
interspaces broader than cords. About three to four fine 
spiral cords on sutural ramp; eight cords anterior to 
shoulder angle on penultimate who. Last whorl with 
about 17 narrow spiral cords, one of which forms shoul- 

der angle and another, only slightly more prominent than 
those adjacent, is just posterior to periphery, giving te- 
leoconch a weakly biconic appearance. Axial Tbs A fine, 

spaced almost e qual to spacing of spiral cords, rendering 
teleoconch surface evenly reticulate; about 55 on last 
whorl. Sutural ramp narrow. Aperture narrowly elliptic. 

Outer lip orthocline adapically, becoming prosocline 
only at anterior end. Columella almost straight, bearing 
three folds: adapical one on siphonal fasciole, other two 
on a broad heavy callus; anterior one obsolete, forming 
edge of short siphonal canal. Periostracum pale brown, 
shell pale straw yellow. 


Type Material: Holotype (dd) MNHIN 20499 (13.2 x 


5.6 mm). 
Type Locality: Norfolk 1 AXidge, south of New Cale- 
donia, 23°02’ §, 168°16' E, 335 m[SMIB 5: sta. DW9S]. 


Material Examined: 
(Figure 5). 


Only known from the holotype 


Etymology: Latin adjective bilix, having a double 
thread, with reference to the two strong spiral cords 
around the shoulder angle. 


Distribution: 
type locality. 


Known only from New Caledonia at the 


Remarks: = Zeadmete bilix is placed in Zeadmete based 
on its columellar structure, which is identical to that of 
other Zeadmete species, but it may be immediately dis- 


Inge 
Page 4 


THE NAUTILUS, Vol. 122, No. 1 


tinguished from its congeners by its elongate shape and 
relatively short aperture. 

Zeadmete bilix superficially resembles the buccinid ge- 
nus Iredalula Finlay, 1926, but species of Iredalula have 
a recurved, notched siphonal canal and lack columellar 


folds. 


Genus Admetula Cossmann, 1889 


Type Species: Buccinwm evulsum Solander, 1766, by 
original designation. Eocene, British Isles. 


Discussion: Admetula is often placed in the synonymy 
of Bonellitia Jousseaume, 1887 ( (see discussions in Sacco, 
1894: 42: Cossmann, 1899: 33; Davoli, 1982: 62; Ver- 
hecken, 1986: 33; but not na a 2007: 286), but we 
regard Bonellitia as distinct, based on the angled whorls 
and muricated sculpture. We refer to Landau, Petit, 
and Marquet (2006) for further discussion. The genus 
Admetula is well represented in Paleogene and Neogene 
Tethyan faunas and in the Recent fauna along continen- 
tal margins at depths ranging from 75-700 m (Petit and 
Harasewych, 1991: 181). 

Many species of Admetula appear superficially very 
similar, ae cially based on published illustrations, yet 
can easily be distinguished when directly compi ared. 
Three discrete * “subgroups” can be recognized in the ge- 
nus. One consists of small attenuate species such as A. 
cornidei (Altimira, 1978), A. epula Petit and Harasewych, 
1991, and A. afra Petit and Harasewych, 2000. Larger, 
wide, rather thin-shelled species s such as A. bayeri Petit, 
1976, and A. emarginata (described herein) form a sec- 


ond group, with a third, intermediate group composed of 


thick shells with an anterior constriction of the last whorl 

such as A. vossi Petit, 1976, and A. deroyae (Petit, 1970). 
Early Tertiary species usually have well-formed varices at 

irre ‘cular intervals, a feature not found on Recent spe cles 


Admetula affluens new species 
(Figures 6-8) 


Description: Protoconch corroded on holotype, in 
other specimens consisting of 0.8 whorls, diameter 775 
wm, shiny, smooth apically, with six raised threads on 
ab: ipical part. Transition to teleoconch distinct, marked 
by onset of te : ‘oconch sculpture. Teleoconch spire high, 
spire angle 47 a ack consisting of 6.25 whorls, 
with ae of evenly spaced, narrow axial ribs crossed 
vy spiral cords forming small nodes at intersections. 
About 15 axial ribs on penultimate whorl, about 1S on 
ast whorl. About 12 primary spiral cords on last whorl, 
with spac ing ¢ qui al to that of axial ribs on shoulder and 
veriphery, more crowded on base; four to seven second- 
ary spire al cords in each inte rspace. Outer lip sharp 


yrosocline. \perture bea lirations. Parietal area with 


very thinly applied callus. Columella sloping, bearing two 
srominent folds extending to edge of inductural callus, 
with a third broad siphonal fold. Siphonal canal shallow, 
well defined. Exterior white with thick, pale olive-brown 


yeriostracum 


Type Material: Holotype MNHN 20500 (22 
mm) and 4 paratypes MNHN 20501-20502. 


7 ¥ 13:0 


Type Locality: SW of Malaita, Solomon Islands, 
09°46’ S, 160°53’ E, 611-636 m [SALOMON 1: sta. 
CP1S08]. 


Material Examined: Solomon Islands. SOLOMON 1: 
sta. CP1749, 09°21’ S, 159°56' E, 582-594 m, 1 dd.— 
Sta. CP1750, 09°16’ S, 159°55' E, 693-696 m, ie 
Sta. CP1751, 09°10’ S, 159°53’ E, 749-799 m, 2 lv (1 
paratype MNHN 20501).—Sta. CP1793, 09°13" S, 
160°0S' E, 505-510 m, 1 dd, 1 juv. dd (Figure 8).—Sta. 
CP1798, 09°21’ S, 160°29' E, 513-564 m, 2 lv, 1 dd.— 
Sta. CP1808, 09°46’ S, 160°53' E, 611-636 m, 1 lv (ho- 
lotype, Figures 6—7).—Sta. CP1859, 09°33’ S, 160°37’ E, 
283-305 m, | lv, 2 dd (3 paratypes MNHN 20502). Total 
of 13 specimens. (Largest specimen: 26.3 x 15.6 mm.) 


Distribution: Known only from the Solomon Islands, 
alive in 305-749 in. 


Etymology: From the Latin affluwens, an adjective 
meaning abundant or copious, with reference both to its 
large size and relative abundance in the Solomon Archi- 
pelago. 


Remarks: Admetula affluens superficially resembles 
the specimen figured as A. garrardi (Petit, 1974) by Ha- 
segawa (2000: 585, pl. 291, figure 26) but differs by hav- 
ing narrower axial ribs, a more constricted base and a 
more twisted columella. It lacks the lirations within the 
outer lip that are present in A. garrardi. It is our opinion 
that the specimens figured as A. garrardi by Hasegawa 
(2000) and Veriecken { 1997: oan figs. 11-13) are not 
that species. Verhecken (1986: 34-35, figs. 1-2) exam- 
ined, redescribed and poued the holotype of A. garrardi 
and explicitly deseribe d its multispiral protoconch 
whereas A. affluens has a paucispiral protoconch. 

Of the other species of Admetula in the tropical south- 
west Pacific, A. affluens is more similar to A. marshalli, 
but differs by its larger adult size, less solid shell, much 
weaker spiral cords and lack of apertural lirations. 


Admetula emarginata new species 
(Figures 9-11) 


Description: —Protoconch glossy, white, of 1.1 whorls, 
diameter L000 jzm, with five wile ly spi aced spiral cords. 
Transition to teleoconch indistinct, protoconch cords 
continuing as teleoconch cords. Teleoconch of five 
whorls, suture impresse sd, with sculpture of numerous 
prominent, well-defined axial ribs crossed by spiral cords 
forming small nodes at intersections. About 14 widely 
spaced axial ribs on penultimate whorl and about 12 on 
last whorl, final one enlarged into terminal varix. About 
eight primary spiral cords on last whorl, more closely 
spaced than axial ribs, with one or more secondary spirals 
in each interspace and about six prominent secondary 
spiral cords on spire whorls below rounded shoulder 
angle. Shell thin, axial ribs visible through last whorl. 


dO 


22.7 mm, Solomon Islands, 09°46S, 160 


Figures 6-17. Admietula. 6-8. Admetula affluens new species. 6-7. Holotype, height 22.7 
E. 611-636 m, [SALOMON 1 sta. CP1S80S]. 8. Protoconch, Solomon Islands, 09°13’ S, 160°OS’ FE, 505-510 m, [SALOMON I sta 


CP1793|. 9-11. Admetula emarginata new species. 9-10. Holotype, height 16 mm, Coral Sea, 20°03" S, 158°45' E, 315 m 
5’ S$, 158°30.5' E, 230 m [CORAIL 2 sta. D31]. 12-13. Admetula 


MUSORSTOM 5 sta. 335]. 11. Protoconch, Coral Sea, 19°33.5 
marshalli new species, holotype, height 14.7 mm, Fiji, 16°39’ S$, 179°57' W, 591-596 m [BORDAU 1 sta. CP1396], 14-15. Admetula 
21°19’ S, 175°01’ W, 225-233 m [BORDAU 2 sta. DW1521]. 15. 


lutea new species 14. holotype height 13.1 mm, Tonga, 2 
Protoconch Fiji, 18°09’ S, 178°39’ W, 290-300 m |[BORDAU | sta. DW1465]. 16-17. Admetula bathynoma new species 16. 
Holotype height 7.4 mm, New Caledonia, 22°52’ S, 167°23' E, 590-600 m [MUSORSTOM 4 sta. DW225]. 17. Protoconch, New 
22°52’ §. 167°16’ E, 530-541 m [BATHUS 2 sta. DW720]. All protoconchs at the same scale, scale bar 500 jzm 


( aledonia, 44 02 


Dawa cn 
Page 6 


THE NAUTILUS, Vol. 122, No. 1 


Outer lip prosocline, sinuous, with everted stromboid 
notch just anterior to periphery and another everted 
notch near base. Aperture with nine lirations, visible in- 
ternally at level of terminal varix, not extending to edge 
of lip or deeply into aperture. Parietal area with very 
thinly applied callus. Columella sloping, with two promi- 
nent folds extending to edge of inductural callus and a 
third broad siphonal fold. Siphonal canal shallow, well 
defined. Shell white, with weakly defined bands of pale 
brown at shoulder and base, also brown behind outer lip. 


Type Material: Holotype MNHN 20503 (16.0 x 10.9 
mim) and one paratype MNHN 20504. 


Type Locality: Coral Sea, 20°03" S$, 158°45" BE, 315 m 
|[MUSORSTOM 5: sta. 335]. 


Material Examined: Coral Sea. CHALCAL: sta. D31, 
19°33.5' S, 158°30.5' E, 230 m, 1 dd ( (paratype, Figure 
11)—MUSORSTOM 5: sta. 335, 20°03’ S, 158°45' E, 
315 m, 1 dd (holotype, Figure 9- 10).—E BISCO: sta. 
CP2571, ae S, 158°45' E, 298-309 m, 1 dd. 


Etymology: From the Latin emarginatus after the 
sinuous, everted outer lip. 


Distribution: Known only from the Coral Sea near the 
Chesterfield Islands, empty shells in 230-315 m, 


Remarks: Admetula emarginata differs from other de- 
scribed species of Admetula in having diffused colora- 
tion. It is much like A. bayeri Petit, 1976, from the west- 
ern Gulf of Mexico in having at the periphery a notch, 
which is not present in the similar species A. bathynoma. 
Admetula emarginata also differs from A. bathynoma in 
having less closely spaced spiral cords and axial ribs. The 
overall effect of the sculpture on A. emarginata is a strik- 
ing pattern of horizontal rectangles crossed by fine spi- 
rals. 


Admetula marshalli new species 
(Figures 12-13) 


Description: —Protoconch glossy, brown, of one whorl, 
diameter $25 jrm, with three wide ly spaced spire al cords. 
Transition to teleoconch marked by ‘a change in color and 
onset of axial ribs. Teleoconch of 5.5 whorls. Suture im- 
pressed. Spiral cords closely spaced, about 14 primary 
cords on last whorl with one secondary cord in most 
interspaces, forming small pointed be ads where they 
cross axial ribs. One spiral cord creates a modest angle 
between periphery and narrowly rounded shoulder: 
About 14 rounded axial ribs on last whorl, only final one 
enlarged into a varix. Fine, densely packed growth lines 
on entire teleoconch. Outer lip thin, ve ry indistinctly 
notched just anterior to periphery and with a slight ever- 
sion of the siphon: ul canal, Ape rture with nine Tie ations, 
visible interior to terminal varix but not extending deeply 
within aperture or to outer lip Parietal area with thin but 
distinct callus. Columella concave, bearing three distinct 
folds that extend out to edge of inductural callus. Poste- 
rior fold largest separated from central fold by a deep 


depression two anterioi folds form a shelf bordering 


short, broad siphonal canal. Shell white, periostracum 
yellow-brown. 


Type Material: Holotype MNHN 20505 (14.7 x 9.0 
mim) and 2 paratypes (all dd) MNHN 20506 (largest 
paratype: 17.0 « 9.7 mim). 

Type Locality: Fiji, Lau Ridge, 16°39’ S$, 179°57' W, 
591-596 m, [BORDAU I: sta. CP1396]. 


Material Examined: 
terial. 


Only known from the type ma- 


Distribution: Known only from the Lau Ridge in Fiji, 
empty shells in 591-596 m. 


Etymology: Named after Bruce Marshall (Museum of 
New Zealand, Wellington), in appreciation for the stan- 
dards of his descriptions and illustrations of the mollus- 
can fauna of New Zealand. 


Remarks: The New Zealand species A. superstes (Fin- 
lay, 1930b) is similar in shape and size, but differs in 
having a translucent white rather than brown protoconch 
with numerous fine, close spiral threads (versus 3 widely 
spaced cords), in having weaker axial costae on the te- 
leoconch, in lacking a terminal varix and internal lirations 
behind the apertural rim at maturity, and in that second- 
ary spiral sculpture is es rably weaker or entirely 
absent. Additionally, superstes has a much more 
strongly developed Lents wcum, with prominent spines 
at the. summits of the axial lamellae. Admetula superstes 
is endemic to northern New Zealand, living on muddy 
substrata in 79-550 m off the northeastern (northern- 
most record at 35°08" S) and southwestern North Island. 
A similar species (possibly a local variant) occurs off 
Three Kings Islands. There is no material of similar Soe 
cies from Norfolk Ridge south of Norfolk Island i 
NMNZ 


Admetula lutea new species 
(Figures 14-15) 


Description: —Protoconch glassy, of slightly more than 
one whorl, diameter 975 jum, worn but with traces of 
spiral cords visible on terminal portion. Transition to te- 
leoconch rather indistinct. Teleoconch of about 4.7 
whorls. Whorl profile regularly convex; suture im- 
pressed. Sculpture of prominent prosocline axial ribs 
crossed by weaker spiral cords; about nine to ten axial 
ribs on last whorl, eleven on penultimate whorl, some 
forming irregularly placed varices; four primary spiral 
cords on pe alan: ate whorl and about 12 on last whorl, 
with one to four secondary spirals in each interspace; 
spiral cords forming « slongate nodes where they cross 
axial ribs. Outer lip with very indistinct lirations, visible 
internally at level of te sininal varix, not exte nding either 
to edge of lip or deeply within aperture. Parie ied area 
with fhe callus. Inductura almost vertical. Columella 
bearing three folds: narrow anterior fold forming edge of 
short, recurved siphonal canal. Exterior ye sllow- bow: 
Periostracum thin, with low incremental lamellae and 
hairy projections on spiral cords. 


P. Bouchet and R. E. Petit, 2008 


Page 7 


Type Material: Holotype MNHN 20507 (13.1 « 7.8 
mm) and one paratype MNHN 20508. 


Tonga, 21°19’ S, 175°01’ W, 225-233 
2: sta. DW1521). 


Material Examined: — Fiji. BORDAU 1: sta. DW1465, 
18°09" S, 178°39' W, 290-300 m, 2 dd.—Tonga. BOR- 
DAU 2: sta. DW a 21°19’ S, nigec W, 225-233 m, 
1 dd ( (holotype, Figure 14).—Sta. CP1576, 19°42" S, 
174°18' W, 253-263 m, 1 dd ( (paratype, Figure 15). 


Type Locality: 
m |[BORDAU 


Etymology: From the Latin adjective luteus, meaning 
ve sJlow, in reference to the color of the shell. 


Distribution: Known only from Fiji and Tonga at lo- 
calities cited above. 


Remarks: This new species differs from its congeners 
by its strongly prosocline axial ribs forming irregularly 
place sd varices and its ae sculpture ‘with numerous 
secondary cords. It is also distinguished from other Ad- 
metula species by its yellow color. 

A specimen of le tula lutea new species in the Petit 
collection (No. 2872) is said to be from 50-200 m north 
of Taiwan, but is a dealer's locality designation that 
needs to be confirmed. ; 


Admetula bathynoma new species 
(Figures 16-17) 


Description: —Protoconch glassy, white, smooth, of one 
whorl, diameter 975 jum. Transition to teleoconch 
marked by onset of axial ribs and spiral cords. Teleo- 
conch of about four whorls. Last whorl beari ing about ten 
to 14 rounded axial ribs, crossed by prominent spiral 
cords, about five to six cords on spire whorls and about 
ten on last whorl, with fine secondary spirals in the in- 
terspaces. Final axial rib enlarged into a varix. Small 
nodes formed at intersections where spiral cords cross 
axial ribs. Sutural ramp narrow, almost flat; suture 
slightly impressed. Outer lip prosocline, rounded, 
smooth, without lirations within. Parietal area without 
wash or callus. Columella slightly inclined, bearing two 
prominent folds extending out to edge of faductinl cal- 
lus; a third descending fold f forming edge of short sipho- 
nal canal. No siphonal fasciole. Last whorl well rounded. 
Shell white. 


Type material: Holotype MNHN 20509 (7.4 x 5.0 
mm) and one paratype MNHN 20510. 


Type locality: Nortolk Ridge, south of New Cale- 
donia, 22°52’ S, 167°23' E, 590-600 m [MUSORSTOM 
4: sta. DW225]. 


Material examined: Norfolk Ridge. BIOCAL: sta. 
DW46, 22°53’ S, 167°17' E, 570-610 m, 1 dd 
paratype) —MU pier 4: sta. DW225, 22°52’ S, 
167°23' E, 590-600 m, 1 |v (holotype, Figure 
SMIB 5S: sta. ee ane 22°59'—23°00' S, 168°21'— 
168°23’ E, ee 558 m, 1 juv. dd—BATHUS 2: sta. 
DW720, 22°52’ S, 167°16' = 530-541 m, 1 juv. dd (Fig- 
ure 17). 


Distribution: 
491-610 mm. 


Known only from the Norfolk Ridge, in 


Etymology: From the Greek bathus, deep, and no- 
mos, place of living, treated as an adjective. 


Remarks: —Adimetula bathynoma resembles a juvenile 
A, marshalli but differs by being proportionally nar- 
rower, with less convex whorls and a larger protoconch 
(diameter 975-1000 jum versus $25 wm in A. marshalli). 
Admetula marshalli also differs in having pointed beads 
rather than small nodes formed at the intersections of the 
axial ribs and spiral cords. 

Kohn and Arua (1999: pl. 13, fig. 55) ilhistrated as C. 
icine Petit and Harasewych, 1986 an 8 mm high 
specimen of Admetula from the Early Pleistocene of Viti 
Levu, Fiji. Their specimen closely resembles the present 
species except that the Fiji fossil has lirations within the 
aperture. 


Genus Fusiaphera Habe, 1961 


Type Species: Cancellaria macrospira Adams and 
Reeve, 1850, fixed herein to comply with ICZN Article 
70.5. Recent, Japan. 


Discussion: Habe (1961a) proposed the genus Fusia- 
phera tor a species he identified as Cancellaria mac- 
rospira Adams and Reeve. Later in the same year, he 
(Habe 1961b) considered the species he had identified as 
C. macrospira to be a different species, which he then 
named Fusiaphera macrospiratoides Habe, 1961. The 
fact that the type species of Fusiaphera was based on a 
misidentified type species was noted by Verhecken 
(1986: 35), who stated that there is some confusion in the 
matter of the type species but did not make a definitive 
statement about the resolution of the problem. He did 
list, however, Fusiaphera macrospira (Adams and Reeve) 
as type species, as did Beu and Maxwell (1987: 55), who 
then stated “but misidentitie dP. Preally = Fusiaphera 
macrospiratoides Habe, 1961.” The Sonkision is Fe -mon- 
strated by the fact that Matsukuma, Okatani and Habe 
(1991: 179, pl. 111, fig. 2) figured the type specimen of 
F. macrospiratoides under the name F. macrospira. In 
the absence of a definitive statement by Verhecken and 
the queries used by Beu and Maxwell, it appears to be 
necessary to fix a type species to comply with Article 
70.3. Cancellaria mac rospira Adams i Reeve is here 
selected as the type species of Fusiaphera Habe, 1961. 
Fusiaphera is distinguished by a slender, non- 
umbilicate teleoconch with irregular varices and a thick 
outer lip with a well-defined aaa canal under the 
shoulder. The aperture is much like that of Scalptia but 
with a straighter columella. The genus Fusiaphera ranges 
from South Africa across the Indian Ocean, north to 
Japan, to Australia, New Caledonia and Fiji. It thus ap- 
pears to be restricted to the Indo-Pacific area with the 
earliest known occurrence in the Miocene of Australia. 
Darragh (1970: 168) referred the Australian Miocene 
species Cancellaria epidromiformis Tate, 1889 and C. 
exaltata Tate, 1889, to Fusiaphera, a placement accepted 


Page § THE NAUTILUS, Vol. 122, No. 1 


by Maxwell (1992: 166). However, three New Zealand Fusiaphera macrospira (Adams and Reeve, 1850) 

Eocene species, Uxia (?) marshalli Allan, 1926, Uxia (Figures 18—25) 

naroniformis Finlay, 1930, and Fusiaphera jenkinsi Max- 

well, 1992, place ed in Fusiaphera - Beu and Maxwell 

(1990) and ee ve 1992), are not considered by us to Cancellaria macrospira Adams and Reeve, 1850: 41, pl. 10, 
fig. 


be correctly placed. Also, Plesiotriton paytensis Olsson, 
1930, from the Eocene of Peru, was tentatively placed in 
Fusiaphera by Beu and Maxwell (1987: 55) but it is here 
excluded from the genus. It is possible that a new genus 


bo 


Cancellaria wilmeri G. B. Sowerby I, 1SS1: 637, pl. 56, fig. 
Cancellaria pallida E. A. Smith, 1899: 313, text- fig. a 

Cancellaria producta G. B. Sowerby TH, 1903: 220, pl. 4, fig. 5. 
3: 


Cancellaria | Trigonostoma) luscinia Melvill and oe n, 1903: 


will have to be erected to contain these Eocene taxa. 319, pl. 23, figs. 14-15. 

Many European Tertiary species of Unitas resemble Fu- Cancellaria exquisita Preston, 1905: 3, pl. 1, fig. 9. 

Sid} shera in form but the By lack a strongly delineated pos- Cancellaria tosaensis Habe, 1961a: Appendix 28, pl. 35, 
terior canal, fig. 21 


Figures 18-25. Fusiaphera macrospira (Adams and Reeve, 1850). 18. Height 20.2 mm Philippines, Balicasag I., said to be from 
130-230 m. 19. Height 30.1 mm, Japan, Mikawa. 20. inigeronpnat ees form, height 16.2 mm, Mikawa Issiki, Japan. 21. Holotype 
of Cancellaria wilmeri (BMNH 1881.5.20.30), height 11.8 mm. 22. tosaensis form, height 20.4 mm, Minabe Japan, 23. Height 15.0 

m, New Caledonia, 19°35’ S$, 163°25' E, 48 m [LAGON sta. 1192]. 24, 25. Height 11.0 mm, New Caledonia, 19°06’ S$, 163°10' 
| 1m | LAGON sta. 542 


P. Bouchet and R. E. Petit, 2008 


Page 9 


Cancellaria azumai Habe, 1961a: 72 
20. 
Cancellaria macrospiratoides Habe, 1961b: 433, pl. 2 
pl. 24, fig. 10. 
Fusiaphera dampierensis Garrard, 1975: 17, pl. 2, fig. 8. 
Fusiaphera eva Petit, 1980: 215, figs. 5, 6. 
Type Data: F. macrospira, C hina Sea, BMNH 
1969347, lectotype designated by Verhecken (1986: 36); 
illustrated by Higo, Callomon and Goto (2001: 99). 


. Appendix 28, pl. 35, fig. 


10; 


= 


C. wilmeri, Port Blair, Andaman Islands, holotype BMNH 
1881.5.20.30, herein Fig. 21. 

C. pallida, 25 fms, off Bonaparte Archipelago, NW Australia, 
holotype BMNH 1891.11.21.96. 

C. producta, 40 fms, off mouth of Umhloti River, Natal, syn- 
types BMNH _ 1903.7.27.76; SAM—A339 (Giles and Gos- 
liner, 1983: 28). 

C. (T.) luscinia, 40 fms, Arabian Sea, 18°58’ N, 71°45" E, 
syntypes BMNH 1903.12.15, 101-102 (2 syntypes); NMW 
1955.158.408. 

C. exquisita, Ceylon, holotype BMNH 1905.10.4.75. 

C. tosaensis, Kochi Prefecture, Shikoku, Japan, holotype 
NSMT-Mo 13287; illustrated by Higo, Callomon and Goto 
(2001: 99). 

C. macrospiratoides, Aiki Prefecture, Honshu, Japan, holotype 
NSMT-Mo 39781; illustrated by Higo, Callomon and Goto 
(2001: 99). 

C. azumai, Aiki Prefecture, Honshu, Japan, NSMT-Mo 13285a 
(illustrated by Higo, Callomon and Goto 2001: 99) is la- 
beled as type in NSMT but is not the figured specimen 
and is smaller than dimensions given for holotype. 

F. dampierensis, Delambre Island, Dampier Archipelago, 
northwestern Australia, holotype WAM 550-71. 

F. eva, west of central Bazaruto Island, southern Mozambique, 
holotype NM G4896. 


Description: Protoconch smooth, glassy, of two 

whorls with small initial nucleus, diameter 1000-1125 
wm, indicating planktotrophic larval development. Tran- 
sition to teleoconch marked by sharp axial rib followed by 
additional ribs and cords. Teleoconch slender, of about 6 
whorls. Axial ribs variable in number, 15 to 20 on last 
whorl of most specimens, extending adapically over 
slightly channeled sutural ramp to suture. Some ribs 
forming varices at irregular intervals, others projecting 
slightly above ramp. Spiral sculpture of fine cords, about 
15 on last whorl of most specimens, with weaker second- 
ary cords in most interspaces. Cords form small nodules 
where they intersect axial ribs. Aperture narrowly ovate. 
Outer lip thickened into a varix, with about 14-16 strong 
lirae that do not descend deeply into aperture. Inter el 
liration beneath sutural ramp forms edge of well-defined 
posterior canal. Parietal shield thin but well developed, 
with pustules on its outer edge. Columella with three 
folds. anterior one forming edge of short siphonal canal. 
Last whorl slightly constricte sd at base behind weak si- 
phonal fesciole. Exterior brown or white. Many brown 
specimens with a weak band of white just below periph- 
ery and white on the adapical ends of ribs and projections 
extending above the shoulder. 


Material Examined: Japan. Off Mikawa, 30 fms, 1 
spm (figured by Abbott and Dance, 1982)—Mikawa Is- 


siki, Aichi Pref., 50-70 m, 1 spm (Figure 20).—Off Min- 
abe, Wakayama Pref., 80-100 m, 2 spms (Figure 22)— 
Off Minabe, 100-200 m, 1 spm.—Off Mikawa, 50 fms, 1 
spm (Figure 19).—Off Tosa, 50 fins, 1 spm.— ‘Japan’, | 
spm.—Japan?”, 1 spm (gift from Habe with “azumai” in 
his hand).—Mikawa Bay, 50 m, 1 spm.—Off Mikawa, 40 
fms, | spm.—Enshu Kei, 2 spms. East China Sea. “180 
m”, | spm. Taiwan. Off Keelung, 50-110 m, 2 spms.— 
Off Keelung, 100— sae m, 5 spms. —Off SW Taiwan, | 
spm. —Off Kee slung, “deep water”, | spm. —Otf SW Tai- 
wan, 60 fms, | spm. Etahppine s. Off Aliguay Island, 
Mindanao, “240 m”, 2 spms.—Off Aliguay Island, 
“80-120 m”, 1 spm.—Off Balicasag Island, ‘ ‘240 m”, 1 
spm.—Off Balicasag, “130-230 m”, 1 spm (Figure 18) — 
MUSORSTOM 3: sta. DR140, 11°43’ N, 122°34' E, 93— 
99 m, 1 dd: Sta. CP 141, 11°45’ N, 122°45' E, 40-44 m, 
1 dd. Vietnam. no locality, “50 m”, 1 spm. Indonesia. 
Masalembo, Java, ca. 20 fms, 1 spm. (AIl above in Petit 
collection). Solomons. SOLOMON 1: sta. DW1760, 
8°47’ S, 160°O1' E, 172-179 m, 1 dd. Coral Sea. CHAL- 
CAL sta. D11, 20°31' S, 161°06’ E, 83 m, 1 dd. New 
Caledonia. LAGON: sta. 375, 22°32’ S, 167°08' E, 67-71 
m, 1 dd; Sta. 517, 19°09’ S, 163°35' E, 42 m, 2 dd: Sta. 
542, 10°06’ S, 163°10' E, 50 m, 3 lv (Figures 24-25); Sta. 

1129, 19°29’ S, 163°49' E, 40m, 3 lv, 2 dd: Sta. 1163, 
19°11’ S, 163°22’ E, 48m, 2 dd; Sta. 1168, 19°16" S 
163°09' E, 50 m, 1 lv; Sta. 1192, 19°35’ S, 163°25' E, 
48m, 1 lv, 1 dd (Figure 23)—MUSORSTOM 4: sta. 
DW151, 19°07’ S, 163°22" E, 200 m, 1 dd. Fiji. SUVA2: 
sta. DW44, Viti Levu, 17°5 1.7 S, 177°13" BE, 33 m, 1 dd. 


Dimensions: 19.3 « 8.4 mm (Coral Sea, CHALCAL 
sta. D11), 17.9 x 8.3 mm (New Caledonia, LAGON sta. 
1192), 14.8 « 7.3 (New Caledonia, LAGON sta. 1192), 
30.2 x 14.0 mm (Japan). 


Distribution: Natal, Mozambique, Arabian Sea, An- 
grias Bank, India, Ceylon, Andaman Islands, northwest 
Australia, Japan (from Izu Peninsula and Yamaguchi 
Pref. southwards), Taiwan, the Philippines, Vietnam, In- 
donesia, Solomon Islands, Queensland, Australia, New 
Caledonia and Fiji. Offshore from 20 to ca. 250 m. 


Remarks: Of the eleven nominal species that we in- 
clude under Fusiaphera macrospira, two have type lo- 
calities in the southwest Indian Ocean, one in the Ara- 
bian Sea, two in the Bay of Bengal, two off northwestern 
Australia, one off Borneo and fice off Japan. We have 
examined representative mi aterial from Japan, Taiwan, 
the Philippines and New Caledonia and fail to recognize 
more than one species. All specimens have in common a 
naticoid multispiral protoconch indicating planktotrophic 
development and we interpret the different names as 
individual rather than geographical variants, connected 
by intermediates. Of these, the nominal species FP’. mac- 
rospiratoides represents a form where the axial and spiral 
sculpture are of equal prominence, whereas F. tosaensis 
represents a form with much stronger axial sculpture. 
Hasegawa (2000: 585) was of the same opinion when he 


Page LO 


THE NAUTILUS, Vol. 122, No. 1 


stated that “[F. macrospiratoides| and other related spe- 
cies, such as F. azuwmai Habe, 1961 and F. tosaensis 
Habe, 1961 may be intraspecific forms of F. macrospira 
(Adams and Reeve, 1850).” We did not examine ex- 
tensive material from the Indian Ocean but published 
descriptions and illustrations suggest that the nominal 
species from this area also fall within the range of varia- 
tion of F. macrospira. Specimens from northwestern 
Australia described by Garrard (1975: 17-19) are, how- 
ever, distinctly smaller with average heights of 10.5 mm 
to 14 mm. 


Genus Nipponaphera Habe, 1961 


Type species: Nipponaphera habei Petit, 1972 by 
ICZN Opinion 1052; Recent, Japan. 


Discussion: Species of Nipponaphera have the angled 
outline of Trigonostoma but are less tabulate, have only 
a small umbilicus if one is present, and have a different 
columellar morphology. The genus has been utilized pri- 
marily to include species having an angled last whorl, a 
triangular aperture, and three columellar folds. Here we 
also place in Nipponaphera species with a rounded last 
whorl and rounded aperture, but which are united with 
those taxa traditionally included in the genus by the pe- 
culiar columellar morphology consisting of two anterior 
columellar folds situated on a slightly iaiked shelf, much 
like a widely bifurcate single fold. 


The genus Misteia Janssen, 1984, from the Miocene of 


The Netherlands, is similar in outline to the angled spe- 
cies of Nipponaphera, but has only two weak folds on the 
columella. 

Axelella Petit, 1988 (a replacement name for the pre- 
occupied Olssonella Petit, 1970) has been considered to 
be confined to the Americas (Petit, L970: $4; 1972: 104). 
What appears to be a neat division of genera between the 
Americas (Axelella) and the Indo-Pacific (Nipponaphera) 
is blurred by the enigmatic species Cancellaria agalma 
Melvill and Standen, 1901, from the Gulf of Oman, a 
species that appears to possess the characters of Axelella. 
Despite that species, which has not been studied in de- 
tail, it is our opinion that the similarities between Avelella 
and Nipponaphera are superficial and that they can be 
se ost ated by their columellar morphology. 

ecent species of Nipponaphera have previously been 
known from South Africa to the northwestern Indian 
Ocean, eastward to the Philippines and north to Japan. 
The range is now extended to New Caledonia. In addi- 
tion to the species treated by Bouchet and Petit (2002), 
we now include in Nipponaphera the following: N. semi- 
pellucida (Adams and Reeve, 1850) [described in Can- 
cellaria; previously placed in Cancellaria by Habe 
(1961b) and other Japanese authors}; teramachii 
Habe, 1961) [described in Tgenapners, placed 
Sealptia and Trigonostoma by various authors; placed in 
Nipponaphe ra by Habe (1961la pl. 36, fig. 4) on the plate 
caption although in the text it is ee in Trigonaphera|; 
N. nodosivaricosa (Petuch, 197 bY [described in Agatrix 
Olssonella), place din Nipponaphera by Bouchet and 


Petit (2002)|: N. quasilla (Petit, 1987), new combination 
[described in Cancellaria|, N. kastoroae (Verhecken, 
1997) new combination, and N. suduirauti (Verhecken, 
1999) new combination [the last two described in Axe- 
lella). 

The fossil record has not been completely searched for 
Nipponaphera but we have recognized Cancellaria 
chinenensis MacNeil, 1961 of the Olenaw a Pliocene and 
C. yonabaruensis MacNeil, 1961 of the Okinawa Mio- 
cene as belonging here. Also, Oyama, Hirose and Nish- 
imoto (1995) described the new species Nipponaphera 
taguchii from the Miocene of Japan and at the same time 
transferred Cancellaria sendoi Hatai, 1941, to the genus 
Nipponaphera. 


Nipponaphera nodosivaricosa (Petuch, 1979) 
(Figures 26-29) 


Agatrix (Olssonella) nodosivaricosa Petuch, 1979: 11, 
figs. 26, 27. 


Description: —Protoconch pale brown, of about one to 
1.2 whorls with fine spiral peek on final third. Tran- 
sition to teleoconch marked by prominent axial rib fol- 
lowed by onset of wide, rounded spiral cords and weak, 
poorly defined axial ribs. Teleoconch of 3.5 to 4.2 
rounded whorls. Spiral sculpture of 12 to 15 rounded 
primary spiral cords, with weaker secondary cords in 
each interspace; about six primary cords on penultimate 
whorl. Primary and secondary cords all bear extremely 
fine spiral threads. Axial sculpture of eight to 12 promi- 
nent, elevated ribs on last whorl, more numerous on 
earlier whorls: final one or two ribs becoming wider, 
forming varices. Spiral cords and interspaces erossed by 
fine, closely spaced growth lines, giving surface a linen- 
like appearance under low magnification. Suture im- 
pressed, sutural ramp convex. Aperture elongate, 
rounded. Last whorl slightly constricted behind siphonal 
fasciole. Outer lip only slightly prosocline, edge thin. In- 
terior of outer lip w ith ten to 14 strong lirations extend- 
ing deeply into aperture. Stromboid notch manifested by 
very slight indentation in outer lip. No parietal callus: 
some specimens with a thin wash on parietal area. In- 
ductural area covered with thin callus, which extends 
back over chink-like umbilicus. Columella with three 
folds, posterior one most prominent, almost perpendicu- 
lar to axis; two anterior folds sharply descending, situated 
on ends of u-shaped platform, anterior-most one forming 
edge of short but well-defined siphonal canal. Exterior 
cream to pale yellow-brown, many specimens with ir- 
regular markings; most specimens with two or three 
elite spiral cords at periphery of last whorl, with two or 
three dark brown cords above and below. 


Type Material: Holotype (11 x 9 mm, fide Petuch; 
12.9 x §.2 nm, fide Genie mn), DMNEH 126397. 

Type Locality: Off Balicasag Island, Philippines, from 
300 m depth. 

Material Examined: New Caledonia. BATHUS 1: 
sta. DW672, 20°48" S, 165°21' E, 347-366 m, 1 lv (Fig- 


Figures 26-38. Nipponaphera. 26-29. Nipponaphera nodosivaricosa (Petuch, 1979). 26. Height 16.9 mm, New Caledonia 
20°17’ S. 163°50' E, 500-600 m [BATHUS 4 sta. DWS9S8]. 27. Height 17.0 mm, New Caledonia, 20°48’ S, 165°21' E, 347-366 m 
BATHUS 1 sta. DW672]. 28. Protoconch, 21°45’ S, 166°37' E, 250 m [BATHUS | sta. CP713]. 29. Teleoconch microsculpture 
same specimen as 28, same scale. 30. Nipponaphera argo new species holotype, height $.6 mm; Coral Sea, 22°48’ S$, 159°24' E, 450 
m | MUSORSTOM. 5 sta. 300). 31-36. Nipponaphera agastor new species 31—32 holotype height 19.3 mm, Solomon Islands, 9°21’ 
S, 160°24' E, 357-359m [SOLOMON 1 sta. CP1S00]. 33. Height 12.4 mm, Fiji, 19°52’ $, 174°40' W, 383-393 m [BORDAU | sta 
CP156] 34. protoconch Vanuatu, 20°20’ S, 169°49' E, 400-440 m |MUSORSTOM 8 sta. CP963]. 35. Teleoconch microsculpture 
same specimen as 34 same scale. 36. Height 17.9 mm Philippines L1°O1’ N, 124°04' E, 214-246 m |MUSORSTOM 3 sta. CP145 
37-38. Nipponaphera tuba new species, holotype, he ight 20.7 mm, Vanuatu, 15°10’ S, 167°14" E, 394-421 m [MUSORSTOM §& 
sta. CP1LOST 


Page 12 


THE NAUTILUS, Vol. 122, No. 1 


ure 27), 1 dd.—Sta. CP713, a S, 166°37' E, 250 m, 

llv oo 98-29). BATHUS 2: sta. DW717, 22°44' S, 
err E, 350-393 m, 1 lv. BATHUS 4: sta. CPS97, 
20°16’ S. 163°52’ E, 305-350 m, 1 Iv.—Sta. DWS898, 
20°17' S, 163°50' E, 500-600 m, 1 dd (Figure. 26).— 
Sta. DW901, 19°03°S, 163°15' E, 297 m, 1 dd.—Sta. 
CP905, 19°02’ S, 163°16' E, 294-296 m, 1 lv.— Solomon 
Islands. SOLOMON 1: sta. CP1SO1, 9°25’ S, 160°26’ E, 
264-273m, | lv. (Dimensions of largest New Caledonia 
specimen: 18.5 x 11.6 mm.) 


Distribution: At this time Nipponaphera nodosivari- 
cosa is known only from New Caledonia, the Solomons 
and the Philippines (Springsteen and Leobrera, 1986; 
V erhecken, 1999). The Indonesian specimen figured as 
Axelella cf. nodosivaricosa by Verhecken (1997: 299, figs. 
5-7) was not attributed by him in his 1999 work to either 
N. nodosivaricosa or N. suduirauti (see below). In the 
New Caledonian dredgings, live specimens were taken 
from depths of 250-393 m and empty shells from as deep 
as 600 m. 


Remarks: Verhecken (1999) described the species 
Axelella suduirauti, here placed in Nipponaphera, distin- 
guished from N. nodosivaricosa based on protoconch 
characteristics. It was stated that N. suduirauti has a 
multispiral protoconch as opposed to the paucispiral pro- 
toconch of N. nodosivaricosa, Our specimen of N. no- 
dosivaricosa from the Solomon Islands has a protoconch 
that is difficult to attribute to one or the other of the two 
species and we believe that the separation between N. 
nodosivaricosa and N. suduirauti should be reevaluated, 
perhaps using molecular characters. We refer to Ver- 
hecken’s (1999) work for his discussion on the question. 


Nipponaphera argo new species 
(Figure 30) 


Description: —Protoconch smooth, of 1.1 whorls, diam- 
eter 1050 wm. Teleoconch of about 3.1 whorls, high- 


spired. Whorl profile angulated at shoulder. Sculpture of 


low, broad axial ribs and much finer spiral cords of rather 
even stre neth, exce pt for one on sutural ri amp and two at 
periphery of last whorl, which are more prominent than 
others: 14 axial ribs on penultimate whorl, seven on last 
whorl (specimen with severe growth scar and regrowth, 
distorting sculpture of last whorl); about 15 spiral cords 
on penultimate whorl and about 35 on last whorl, crossed 
by thin incremental riblets. Suture shallowly impressed. 
Last whorl slightly constricted behind siphonal fasciole. 
Outer lip thin, sharp, smooth within, lacking lirae. Inner 
lip with well-deve lope od parieti il shield, exte nding slightly 
over narrow umbilicus. Columella only slightly concave, 
with three folds: anterior two close toge the 1 much like 
one large bifurcate fold. Siphonal canal short, indistinct. 
Exterior uniformly very pale yellowish-white. 


Type material: Holotype MNHN 20511 (8.6 « 6.5 
mim) and one paratype MNHN 20512 


Type Locality: Argo Bank, Coral Sea, 22°48’ S, 
159°24' FE, 450 m [MUSORSTOM 5, sta. 300]. 
Material Examined: Coral Sea. MUSORSTOM 5, 
sta. 299, 22°48’ S, 159°24’ EF, eae: m, | dd 
(paratype ).—Sta. 300, 22°48’ S, 159°24’ E, 450 m, TH 
(holotype, Figure 30). 


Etymology: Named for the Argo Seamount, a promi- 
nent topographic feature of the Coral Sea, from which 
the specimens were collected; to be treated as a noun in 
apposition. 


Distribution: 
390-450 m. 


Coral Sea (Argo Seamount), dead in 


Remarks: Nipponaphera argo differs from N. goniata 
Bouchet and Petit, 2002 by its sculpture of oad low, 
non-lamellar axial ribs. Also, the spiral cords of N. argo 
are of more even strength, except for one on the choulder 
and two on the periphe ry, which a a slightly bian- 
gular aspect to shells of this species 


Nipponaphera agastor new species 
(Figures 31-36) 


Description: —Protoconch normally smooth, corroded 
on holotype, of 0.9 whorls, diameter SOO fxm. Proto- 
conch/teleoconch boundary indistinct due to corrosion, 
but distinctly marked by onset of teleoconch sculpture on 
specimens from Vanuatu. Teleoconch of five rounded 
whorls; spire angle 64°; suture deeply impressed. Axial 
sculpture of prominent, regularly spaced ribs, eleven on 
last whorl, eleven on pe anlbice ite whorl. Ribs rounded 
over steep, narrow sutural ramp. Spiral sculpture of 
evenly spaced, prominent spiral cords, eight on penulti- 
mate whorl, 13 on last whorl, with three to five secondary 
cords in each interspace; spiral cords rise over axial ribs, 
forming small nodules on primary cords at intersections. 
Numerous fine growth lines cross spiral cords, creating 
small imbrications. Last whorl slightly constricted behind 
siphonal fasciole. Outer lip prosocline. Inner margin of 
lip smooth apart from 16 prominent lirae « xtending into 
aperture; two indistinct lirae on parietal area. Columellar 
callus well developed, forming shield over chink-like um- 
bilicus. Columella with three almost equal folds; anterior 
one sloping sharply down at edge of small but distinct 
siphonal canal, which recurves abaxially. Exterior chalky 
white. 


Type Material: Holotype 20513 (19.3 x 12.3 mm) and 
4 paratypes MNHN 20514. 


Type Locality: Between ie idaleanal and Florida Is- 
land, Solomon Islands, 9°21’ S, 160°24' E, 357-359 m 
[SOLOMON I: sta. Cea. 


Material nara: Vanuatu. MUSORSTOM 5: sta. 
CP963, 20°20' S, 169°49' E, 400-440 m, 1 lv (Figures 
34-35) pen ‘BORD: AU 2: sta. CP 1561, 19°52’ S, 
174°40' W, 383-393 m, 1 dd (Figure 33).—Solomons. 
SOLOMON 1: sta. CP1L746, 09°23' S, 159°57' Ei, 302- 
396 m, 1 dd; Sta. CP1SO0, 9°21' S$, 160°24' BE, 357— 


P. Bouchet and R. E. Petit, 2008 


Page 13 


359m, 4 lv, 1 dd (holotype, Figures 31-32, and 
paratypes).—Philippines. MUSORSTOM 3: sta. CP145, 
L1°OL' N, 124°04" E, 214-246 m, 1 dd (Figure 36).— 
Punta Engano, tangle nets, approximately 60 fms, 1 spm. 


Etymology: From the Greek agastor, a noun in appo- 
sition, meaning near kinsman or brother, to highlight the 
similarity to N. nodosivaricosa. 


Distribution: Only known from the material exam- 
ined: Philippines, Solomons, Vanuatu and Tonga. Depth 
range in the southwest Pacific 360-400 m; in the Philip- 
pines possibly shallower. 


Remarks: Specimens from the Philippines s and the So- 
lomons are distinctly larger with adult sizes ranging from 
16.8 mm to 19.5 mm, whereas specimens from Vanuatu 
and Tonga are much smaller with adult sizes at 12.2 min 
and 12.38 mm respectively. 

Nipponaphera agastor is sympatric with N. nodosivari- 
cosa in the Philippines s and the Solomon Islands but dif- 
fers by its spiral sculpture with more numerous (3 to 5) 
secondary cords, not separated by an incised groove as in 
N. nodosivaricosa. It differs from N. tuba, w ath which it 
is sympatric in Vanuatu, by being more slender and lack- 
ing a stromboid notch. 


Nipponaphera tuba new species 
(Figures 37-35) 


Description: —Protoconch smooth, of 0.9 whorls, diam- 
eter SOO zm. Protoconch/teleoconch boundary distinctly 
marked by onset of teleoconch oo ae Teleoconch of 
five rounded whorls: spire angle 75°; suture deeply im- 
pressed. Axial sculpture of prominent, regularly spaced 
ribs, ten on last whorl, 14 on penultimate whorl: ribs 
rounded on steep, narrow sutural ramp. Spiral sculpture 
of evenly spaced cords of several strengths; primary cords 
evenly See by one secondary cord, resulting inter- 
spaces filled with two or three tertiary cords: spiral cords 
rising over axial ribs, forming small nodules on primary 
spiral cords at intersections. Numerous fine growth lines 
cross spiral cords, creating small imbrications. Last whorl 
slightly constricted behind siphonal fasciole. Outer lip 
prosocline. with distinct stromboid notch adapical to cen- 
ter of lip. Inner margin of lip smooth or slightly crenu- 
late, with 19 prominent lirae extending into aperture, 
four additional lirae beneath sutural ramp. Columellar 
callus well developed. forming shield over chink-like um- 
bilicus. Columella with three almost ae folds, anterior 
one sloping sharply down at edge of small but distinct 
siphonal canal, which recurves abaxially. Exterior yellow- 
brown with a band of white below periphe ry, bordered 
by indistinct bands of darker brown. 


Type Material: Holotype MNHN 20516 (20.7 « 15.0 
mm) and one paratype MNHN 20517 

Type Locality: Vanuatu, 15°10’ S, 167°14’ E, 394— 
421 m [MU SORSTOM 8: sta. CP1087]. 


Material Examined: Vanuatu. MUSORSTOM S: sta. 
CP1087, 15°10’ S, 167°14’ E, 394-421 m, 1 lv, 1 dd 


(holotype, Figures 37—38, and paratype).— Sta. [no data, 
mixed lot|, | a, 


Etymology: From the Latin tuba, a war trumpet, 
which this new species can, with some imagination, be 
reminiscent of; used as a noun in apposition. 


Distribution: Known only from Vanuatu. 


Remarks: Nipponaphera tuba differs from N. cy- 
phoma Bouchet and Petit, 2002 in being more rounded 
and robust. Also, the spiral sculpture never appears as 
incised lines as in N. cyphoma and N. nodosivaricosa. 
Nipponaphera tuba also has a thick outer lip that is not 
present in N. cyphoma. 


Genus Trigonostoma Blainville, 1827 


Trigona Perry, 1811: pl. 51. Type species: Trigona pellucida 
Perry, ISL], by monotypy. Not Trigona Jurine, 1807 (Hy- 
menoptera). 

Trigonostoma Blainville, 1827: 652. Type species: Delphinula 
trigonostoma Lamarck, 1822 (?= Buccinum scalare Gme- 
lin, 1791), by monotypy. Recent, Indo-Pacific 


Remarks: = Trigonostoma has a different taxonomic 
composition for various authors. We here use Trigono- 
stoma sensu lato to encompass the nominal genera Ven- 
trilia Jousseaume, 1887, Arizelostoma Iredale, 1936, 
Ovilia Jousseame, L887, and Extractrix Korobkov, 1955. 
We do not include Scalptia Jousseaume, 1887, Trigona- 
phera Iredale, 1936 and Cancellaphera Iredale, 1930, 
which are sometimes treated as subgenera of Trigono- 
stoma, 


Trigonostoma tryblium new species 
(Figures 39-44) 


Description: —Protoconch smooth, glassy, of one whorl, 
diameter 1050 jzm. Transition from protoconch to teleo- 
conch abrupt, denoted by onset of axial and spiral sculp- 
ture. Teleoconch of 2.5 whorls: spire depressed; umbili- 
cus broad. First teleoconch whorl with 16 axial ribs. Sec- 
ond whorl with about 18 axial ribs that have become 
somewhat obsolete. Spiral sculpture of broad, closely 
spaced cords with narrow interspaces; about seven to ten 
spiral cords on sutural ramp and 25 between shoulder 
angle and umbilical rim. Sutural ramp weakly concave, 
bat forming deeply channeled shoulder between shoul- 
der angle and impressed suture. Aperture narrowly 
ovate. Outer lip smooth. Inner lip forming parietal 
shield, partly covering deep, wide umbilicus. Columella 
with two descending folds. Anterior canal not con- 
stricted. Holotype exterior chalky white, with two broad, 
ill-defined brown bands, best seen through shell. 


Type Material: Holotype MNHN 20518 (7.0 « 6.5 


mm). 

Type Locality: North of Makira Island, Solomon Is- 
lands, 10°13’ S, 161°29' E, 381-383 m [SOLOMON 1: 
sta. CP1837]. 

Material Examined: Taiwan. TAIWAN 2000, sta. 
DW36, 21°54.8 N, 120°36.2 E, 305 m, Bashi Channel, | 


Page 14 


THE NAUTILUS, Vol. 122, No. 1 


Figures 39-47. Trigonostoma. 39-44. Trigonostoma tryblium new species. 39. Height $.3 mm, Taiwan, Bashi Channel, 21°54.8' 


N, 120°36.2' EF, 305 m [TAIWAN 2000 sta. DW36 


40-41. Holotype, height 7.0 mm, Solomon Islands, 10°13’ S, 161°29' E, 381-383 


m [SOLOMON 1 sta. CP1837]|. 42. Height 5.0 mm; New Caledonia, 23°03’ S, 166°5S' E, 397-400 m [BATHUS 2, sta. DW730] 
43-44. Protoconch and teleoconch microsculpture, same specimen as 42. 45-47. Trigonostoma thysthlon Petit and Harasewych, 


1987 


SMIB 1 sta. DW6 


dd (Figure 39). Solomons. SOLOMON 1: sta. CP1837 
10°13" S, 161°29" E, 381-383 m, 1 lv (holotype, Figures 
{0-41). New Caledonia. BIOCAL Sta. DW77, 22°15’ S 
167°15' FE, 440 m, 1 juv. lyv—BATHUS 2, sta. DW730 
'S, 166°58’" FE, 397-400 m, 1 juv. dd (Figures 42-44 


15. Height 16.9 mm, New Caledonia, 22°47’ S, 167°28' E 
{'S 


241-245 m [BATHUS 2 sta. CP728]. 46. Height 15.0 mm, Fiji, 
178°12.5' E, 200-215 m |MUSORSTOM 10: sta. DW1333]. 47. Protoconch, New Caledonia, 22°43’ S, 167°16' E, 300 


Etymology: From the Latin tryblium, meaning cup, a 
shape that the new species is somewhat reminiscent of; 
used as a noun in apposition 


Distribution: Known only from Taiwan, New Cale- 


P. Bouchet and R. E. Petit, 2008 


Page 15 


donia and the Solomons. Alive in 353-440 m; empty 
shells from 305 m. 


Remarks: Of the known Recent species of Trigono- 
stoma, Trigonostoma tryblium resembles only T. semid- 
isjuncta (Sowerby, 1849) in having primarily spiral sculp- 
ture. It differs from that specie s in bei ing more de- 
pressed, with a broader umbilicus. In T. se midusjunc ta 
there is no spiral sculpture on the sutural ramp. Also, the 
spiral sculpture of T. semidisjuncta is arranged in groups 
of cords separate od by wide furrows. 


Trigonostoma thysthlon Petit and Harasewych, 1987 
(Figures 45-47) 


Trigonostoma thysthlon Petit and Harasewych, 1987: 
8-13. 

Trigonostoma antiquata—Habe, 1961a: 435, pl. 24, fig. 14; pl. 
23, fig. S; 1961b: 73, pl. 36, fig. 8; Lan, 1979: 95, pl. 41, 
figs. 93, 93a; Abbott and Dance, 1982: 299 (second figure 


922 


in bottom row); Habe and Okutani, 1985: 233 (second 
figure in pottow row); Bosch, et al., 1995: 157, fig. 687. 
[not Cancellaria antiquata Hinds, 1843] 

Trigonostoma aie: pager ie and Takemura, 1963: 
Trigonaphera (2) plate, fig. 5. [not Cancellaria antiquata 
Hinds, 1843} 

Trigonostama (sic) thysthlon—Hasegawa, 2000: 581, pl. 290, 
fig. 11. 


79, figs. 5, 


~l 


Type Material: 
mm). 


Holotype, USNM 747301 (17.3 x 12. 


Type Locality: Off west coast of Wasir Island, West 
Wokam, Aru, Moluccas (5°30' S$, 134°12' E) in 56-73 m. 


Material Examined: New Caledonia. LAGON: sta. 
387, 22°39’ S, 167°07’ E, 225 m, 1 dd.—SMIB 1: sta. 
DW6, 22°43’ S, 167°16' E, 300 m, 1 dd (Figure 47).— 
BATHUS 2: sta. CP728, 22°47’ S, 167°28' E, 241-245 
m, | lv 
Bele, north New Caledonia, 1 spm. Fiji. MUSOR- 
STOM 10:sta. DW1333, 16°50.4' S, 178°12.5' E, 200— 
215 m, 1 dd (Figure 46). Solomons. SOLOMON 1: sta. 
DW 1850, 10°28’ S, 161°59' E, 139-261 m, 1 lv. 


Description (of a specimen from New Caledo- 
nia): Protoconch smooth, glassy, of 1.8 whorls, diam- 
eter 1050 tm. Transition to tele oconch marked by axial 
rib followed by both axial and spiral sculpture and flat- 
tening of the sutural ramp. Teleoconch of up to six tabu- 
late whorls. Sutural ramp flat, bordered by cord-like 
shoulder angle. Suture impressed. Axial ribs prominent, 
extending from suture across ramp, over shoulder angle, 
where they form recurved spines on many specimens, 
down and over siphonal fasciole, then inside umbilicus. 

Shoulder spines not formed on all ribs, but many ribs 
lacking spines form short. sloping buttress against pre- 
ceding whorl: about 12-15 ribs on last whorl, more nu- 
merous on earlier whorls: two thick, closely spaced ribs 
mark end of growth in adults: a few varix-like rib occurs 
earlier on some specimens. Fine, closely packed growth 
lines of varying number in axial interspaces slightly over- 
lap each other, producing a scabrous appearance. Spiral 


(Figure 45)—Dredged at “300-400 m”, off 


sculpture of very closely spaced cords, rising over ribs to 
form small nodes. Sutural r: amp with about 12-15 spiral 
cords of uniform prominence; about 15 primary spiral 
cords on last whorl between shoulder angle and siphonal 
fasciole; about ten fine secondary cords in each inter- 
space. Spiral cords of equal strength continue inside um- 
bilicus. Outer lip slightly prosocline. Outer lip thickened 
by varix, slightly serrate on margin, interior with eight to 
ten short irregular lirae, not extending quite to outer 

edge of lip and not extending deeply into aperture. Ap- 
erture triangular, with sme ull posterior notch under su- 
tural 1 ramp. Posterior portion of inner lip adpressed 
against siphonal fasciole and anterior quarter of penulti- 
mate whorl. Inner lip with two descending folds, poste- 
rior one slightly larger than anterior; third incipient fold 
present in some specimens. Umbilicus extends to proto- 
conch. Siphonal canal short. Exterior white with very 
faint orange-brown at shoulder. Specimens over 24 mm 
in height are known. 


Distribution: Gulf of Oman to the Moluccas, Japan, 
the Philippines, Solomon Islands, New Caledonia and 
Fiji. 

Remarks: The imbricate sculpture resulting from the 
overlap of growth lines, which is so holicenble on the 
New Cale doa specimens, is absent from T. antiqua- 
tum and is much less evident on T. thysthlon from other 
areas. However, we do not consider this difference to be 
of taxonomic significance. This sculpture isa promine nt 
feature of T, sculore (Gmelin, 1791), the type of the 
genus but that species has a more angular shape and its 
whorls are barely attached. 


ACKNOWLEDGMENTS 


We thank Bertrand Richer de Forges, the indefatigable 
as ee of the first author on many expeditions, for 
his skill and determination in exploring the deep-water 
benthos of the tropical Pacific. Ahmed Abdou did SEM 
illustrations of protoconchs and microsculpture; Del- 
phine Brabant and Philippe Maestrati did the digital 
photography and plate assembling. Bruce Marshall 
helped us interpret our material of Admetula from Fiji. 
Alan Beu helped us improve the manuscript by tracking 
internal inconsistencies. 


LITERATURE CITED 


Abbott, R. T. and S. P. Dance. 1982. Compendium of Seashells 
E. P. Dutton, Inc., New York, ix + 411 pp. [pre printing 
revised, 1983; 3"! printing, revised, 1956; see also Habe 
and Okutani, 1955}. 

Adams, A. and L. Reeve. 1848-50. Mollusca. In: A. Adams (ed.) 
The zoology c the voyage of H.M.S. Samar ngs under the 
command of Captain Sir Edward Belcher, ,F.RAS., 
F.G.S. during the years 1543-1546. Reeve, ae and 
Reeve, London. x + xv + 87 pp., pls. 1-24. [i-x (i-ii plate 
explanations to pls. 1-9; iii-x Introduction to Mollusca by 
Adams only), 1-24, pls. 1-8, 1 November 1848; 25-44, pls 


Page 16 


THE NAUTILUS, Vol. 122, No. 1 


10-17, 27 April 1850; i-xv, 45-87, pls. 18-24, 31 August 
1850. | 

Allan, R. S. 1926. Fossil Mollusca from the Waihao Green- 
sands. Transactions of the New Zealand Institute 56: 338— 
346, pls. 76-77. 

Altimira, C. 1978. Avance sobre algunos moluscos colectados 
por la expedicion “Atlor VH.” Resultados de las Expedi- 
ciones Cientificas del Buque Oceanografico “Cornide de 
Saavedra” 7: 169-171. 

Beu, A. G. and P. A. Maxwell. 1987. A revision of the fossil and 
living gastropods related to Plesiotriton Fischer, 1884 
(F Panty Cancellariidae, Subfamily Plesiotritoninae n. sub- 
fam.) with an Appendix: Genera ‘of Buccinidae Pisaniinae 
related to Colubraria Schumacher, 1817. New Zealand 
Geological Survey Paleontological Bulletin 54; 1-140, pls. 
1-30. 

Beu, A. G. and P. A. Maxwell. 1990, Cenozoic Mollusca of New 
Zealand. New Zealand Geological Survey Paleontological 
Bulletin 58: 1-518, pls. 1-57. 

Blainville, H. M. D. de. 1825-27. Manuel de malacologie et de 
conchyliologie. Paris. [1-647, 15825; 649-664, 109 pls., 
1827]. 

Bosch, D. T., S. P. Dance, R. G. Moolenbeek, and P. G. Oliver. 
1995. Seashells of Eastern Arabia. Motivate Publishing, 
Dubai, 296 pp. 

Bouchet, P. and Petit, R. E. 2002. New species of deep-water 
Cancellariidae (Gastropoda) from the southwestern Pa- 
cific. The Nautilus 116: 95-104. 

Brazier, J. 1877. List of marine shells, with descriptions of the 
new species collected during the Chevert Expedition. Pro- 
ceedings of the Linnean Society of New South Wales 1: 
169-181. 

Brocchi, G. B. 1814. Conchiologia fossile subapennina con os- 
servazioni geologiche sugli Apennini e sul suolo adiacente. 
ae “Reale, Milano. 2 vols. [1: 1-240, 2: 241-712, 
pls. 1-16.] 

Cemohorsky, W. O. 1972. Marine Shells of the Pacific, vol. 2 
Pacific Publications, Sydney, 411 pp., 68 pls. 

Cossmann, M. 1899. Essais de Paléoconchologie Comparée. 
Troisiéme livraison. Paris, 201 pp., § pls. 

Darragh, T. A. 1970. Catalogue of Australian Tertiary Mollusca 
(except Chitons). Memoirs of the National Museum of 
Victoria 31: 125-212. 

Davoli, F. 1982. Cancellariidae (Gastropoda). In: E. Montanaro 
Gallite lli (ed.), Studi monografici sulla malacologia mioce- 
nica iiodentse: Parte | — I Molluschi tortoniani di Mon- 
tegibbio. Paleontographia Italica 72: 5-73, pls. 1-7. 

Finlay, H. J. 1924. The molluscan fauna of Target Gully: Part 1. 
Transactions of the New Zealand Institute 55: 495-516. 

Finlay, H.J. 1926. A further commentary on New Zealand mol- 
luscan systematics. Transactions of the New Zealand In- 
stitute 57: 320-485, pls. 18-23. 

Finlay, H. J. 1930a. New shells from New Zealand Tertiary 
beds. Part 3. Transactions of the New Zealand Institute 61: 
19-84, pls. 1-6 


Finlay, H.J. 1930b. Additions to the Recent molluscan fauna of 


New Zealand. No. 3. Transactions of the New Zealand 
Institute 61: 222-247, pls. 42-45 

Garrard, T. A. 1975. A revision of Australian Cancellariidae 
(Gastropoda: Mollusca). Records of the Australian Mu- 
seum 30: 1-62 

Giles, E. and T. Gosliner. 1983. Primary type specimens of 
marine Mollusca (excluding Cephalopoda) in the South 


African Museum. Annals of the South African Museum 92: 
1-52. 

Gmelin, J. F. 1791. Caroli a Linné Systema Naturae per regna 
tria naturae. Editio decima tertia. Vol. 1, pt. 6 (Vermes): 
3021-3910, Lipsiae. 

Habe, : 1961a. Coloured illustrations of the shells of Japan 
(II). Hoikusha, Osaka. 152 pp., Appendix 42 pp., 66 pls. 
An English edition was issued in 1964, with a larger page 
size: Shells of the Western Pacific in Color, Voll Il. 
Hoikusha, Osaka. 233 pp., 66 pls. | 

Habe, T. 1961b. Description of four new cancellarid species, 
with a list of the Japanese species of the family Cancel- 
lariidae. Venus 21: 431-441, pls. 23-24. 

Habe, T. and T. Okutani. 1985. “Sekai kai-san kai-rui dai zukan’ 
[Compendium of world seashells]. Heibonsha Limited, 
Tokyo. 445 pp. [Japanese “edition” of Abbott and Dance, 
1982 but modified and must be treated as an original 
work]. } 

Hasegawa, kK. 2000. Cancellarioidea. pp. 580-585, pls. 289- 
291. In: T. Okutani (ed.), Marine Mollusks in Japan. Tokai 
University Press, Tokyo, 1173 pp. 

Hatai, K. 1941. Additional fossils from certain Miocene forma- 
tions. Bulletin of the Biogeographical Society of Japan 
11(13): 105-112, pl. 3. 

Higo, S., P. Callomon, and Y. Gotd. 1999. Catalogue and bib- 
liography of the marine shell-bearing Mollusca of Japan. 
Elle Scientific Publications, Osaka, 749 pp. 

Higo, S., P. Callomon and Y. Got6. 2001. Catalogue and bib- 
liography of the marine shell-bearing Mollusca of Japan. 
Type figures. Elle Scientific Publications, Osaka, 208 pp. 

Hinds, R. B. 1843. Description of ten new species of Cancel- 
laria, from the collection of Sir Edward Belcher. Proceed- 
ings of the Zoological Society of London 1543; 47-49. 

Hutton, F.W. 1873. Catalogue of the marine Mollusca of New 
Zealand, with diagnose: »s of the species. Colonial Museum 
and Geological Survey Department, Wellington, xx + 116 
pp.. 1 pl. 

Iredale, T. 1930. Queensland molluscan notes, no. 2. Memoirs 
of the Queensland Museum 10(1): 73-SS, es 9. 

Iredale, T. 1936. Australian molluscan notes: No. 2. Records of 
the Ainstralians Museum 19: 267-340, pls. 20-24. 

Janssen, A, W. 1954. An account of the Cancellariidae (Gas- 
tropoda) of Winterswijk-Miste (Miocene, Hemmoorian), 
The Netherlands. Scripta Geologica 68: 1-39, pls. 1-6. 

Jousseaume, F. P. 18S7a. Diagnoses de coquilles nouvelles de 
la famille des Cancellariidae (Mollusques gastéropodes). 
Le Naturaliste, Année 9, 2e Série [Année 1]: 163-165. 

Jousseaume, F. P. 1S87b. La famille des Cancellariidae (Mol- 
lusques gastéropodes). Le Naturaliste, Année 9, 2° Série 
[Année 1]: 155-157, 192-194, 213-214, 221-223. 

Kohn, A. J. and I. Arua. 1999. An Early Pleistocene molluscan 
assemblage from Fiji: gastropod faunal composition, pa- 
leoe scology and biogeography. Palaeogeography, Palaeocli- 
mz tology, Pali eoecology 146: 99-145. 

Korobkov, 1955. Spri avochnik i methodicheskoe rukovod- 
stvo es - ceca molljuskam. Briukhonogie. [Reference 
and methodological guide to Tertiz ary mollusks. Gas- 
tropoda]. G aes sKhizde it, Leningrad. 795 pp., 117 pls. 

Lamarck, J. B. P. A. 1S22. Histoire Naturelle des Animaux sans 
Vertebres. Paris: Vol. 7: 1-711. 

Lan, T.C. 1979. Rare halle of Taiwan in color. Privately pub- 
lished, Taipei. 144 pp., 63 pls. 

Landau, B., R. Petit and R. Marquet. 2006. The Early Pliocene 


P. Bouchet and R. E. Petit, 2008 


Page 17 


Gastropoda (Mollusca) of Estepona, Southern Spain. Part 
12. Cancellarioidea. Palaeontos 9: 61-101. 


MacNeil, F. S. 1961. Tertiary and Quaternary Gastropoda of 


Okinawa. United States Geological Survey Professional 
Paper 339: i-iv, 1-148, pls. 1- 2] [imprinted date of 1960 
but not published until 1961]. 

Marshall, P. and R. Murdoch. 1920. Some Tertiary Mollusca, 
with descriptions of new species. Transactions of the New 
Zealand Institute 52: 128-136, pls. 6-10. 

Matsukuma, A., T. Okutani, and T. Habe. 1991. World sea- 
shells of rarity and beauty. Revised and enlarged edition. 
Nz oe Science Museum, Tokyo, viii + 206 pp. 


Maxwell, P. A. 1992. Eocene Mollusca from the vicinity of 


Mebullps Bridge, Waihao River, South Canterbury, 
New Zealand: Paleoecology and systematics. New Zealand 
Geological Survey Paleontological Bulletin 65: 1-280. 

Melvill, J. C. and R. Standen. 1901. The Mollusca of the Per- 
sian Gulf, Gulf of Oman, and Arabian cee as evidenced 
mainly through the collections of Mr. F. W. Townsend, 
1893-1900: with descriptions of new sare Proceedings 
of the Zoological Society of London 1901: 327-460, pls. 
2)-24. 

Melvill, J. C. and R. Standen. 1903. Descriptions of sixty-eight 
new Gastropoda from the Persian Gulf, Gulf of Oman, and 


North Arabian Sea, dredged by Mr. F. W. Townsend, of 


the Indo-European Telegr: aph Service, 1901-1903. Annals 
and Magazine of Natural History, Series 7, 12: 289-324, 
pls. 20-23. 

Olsson, A. A. 1930. Contributions to the Tertiary paleontology 
of northern Peru: Part 3 
American Paleontology 17(62): 1-96, pls. 1-12. 

Oyama, kK. S. Hirose, and H. Nishimoto. 1995. On a new 
Miocene cancellariid, Nipponaphera taguchii n. sp. from 
the Katsuta Group, Okayama Prefecture, southwest Japan. 
Journal of the Hiba Society of Natural History 164: 19-25, 
pl. 2. 

Oyama, T. and Y. Takemura. 1963. The Molluscan Shells. Vol. 
6. [for details on this publication see Higo, Callomon, and 
Goto, 1999: 609]. 

Perry, G. 1811. Conchology, or the natural history of shells: 
Containing a new arrangement of the genera and species, 
illustrated by coloured engravings executed from the 
natural specimens and including the latest discoveries. 
William Miller, London. 4 pp. + 61 pls. with pl. explana- 
tions. 

Petit, R. E. 1967. Notes on Cancellariidae (Mollusca: Gas- 
tropoda). Tulane Studies in Geology 5; 217-219. 

Petit, R. E. 1970. Notes on Cancellariidae (Mollusca: Gas- 
tropoda) - II. Tulane Studies in Geology and Paleontology 
8: §3-88, pl. 1 

Petit, R. E. 1972. The cancellariid genus Nipponaphera Habe, 
1961. Venus 31: 103-105. 

Petit, R. E. 1974. Notes on Japanese Cancellariidae. Venus 33: 
109-115. 

Petit, R. E. 1976. Notes on Cancellariidae (Mollusca: Gas- 
tropoda) — HI. Tulane Studies in Geology and Paleontol- 
ogy 12: 33-43, pls. 1-2 

Petit. R. E. 1980. The Mozambique Cancellariidae (Mollusca: 
Gastropoda). Annals of the Natal Museum 24: 211-216. 

Petit. R. E. 1987. New names for two species of Cancellaria 
(Mollusca: Gastropoda). The Nautilus 101: 154 


. Eocene Mollusca. Bulletins of 


Petit, R. E. 1988. Axelella, new name for Olssonella Petit, 1970, 
a preoccupied taxon (Mollusca: Cancellariacea). The Nau- 
tilus 102: 130. 

Petit, R. E. and M. G. Harasewych. 1986. New Philippine 
Cancellariidae (Gastropoda: Cancellariacea), with notes 
on the fine structure and function of the nematoglossan 
radula. The Veliger 28: 436-443. 

Petit, R. E. and M. G. Harasewych. 1987. The Indo-West Pa- 
cific species of the genus Trigonostoma sensu stricto (Gas- 
tropoda: Cancellariidae). The Veliger 30: 75-81. 

Petit, R. E. and M. G. Harasewych. 1991. A new Admetula 
(Gastropoda: Cancellariidae) from South Africa. Proceed- 
ings of the Biological Society of Washington 104; 1S1— 
183. 


Petit, R. E. and M. G. Harasewych. 2000. Additions to the 
cancellariid (Mollusca: Neogastropoda) { fauna of South Af- 
rica. Proceedings of the Biological Society of Washington 
113: 145-154. : 


Petuch, E. J. 1979. Twelve new Indo-Pacific gastropods. Nem- 
ouria 23: 1-21. 

Powell, A. W. B. 1940. The marine Mollusca of the Aupourian 
Province, New Zealand. Transactions of the Royal Society 
of New Zealand 70(3): 205-248, pls. 28-33. 

Powell, A. W. B., 1979. New Zealand Mollusca. William Collins 
Publishers, Ltd, Auckland. i-xiv + 1-500 pp., 82 pls. 


Preston, H. B. 1905. Descriptions of new species of marine 
shells som Ceylon. The Journal of Malacology 12(1): 1-8, 
pls. 1-2 


Sacco, F. 1894. I molluschi dei terreni terziarii del Piemonte e 
della Liguria. Pt. 16 (Cancellariidae). C. Clausen, Torino. 
78 pp., 3 pls. [Reprinted 1987-89 by Atlante Malacologico, 
Roma. | 

Smith, E. A. 1899. Notes on some marine shells from North- 
West Australia, with descriptions of new species. Pro- 
ceedings of the Malacological Society of London 3(6): 


311-314. 

Solander, D. C. 1766. [descriptions of species]. In: Brander, G., 
Fossilia Hantoniensia collecta, et in Musaeo Britannico 
deposita, a Gustavo Brander. Londini, vi + 43 pp. 
9 pls. 

Sowerby, G. B., I. 1849. Descriptions of some new species of 
Cancellaria in the collection of Mr. H. Cuming. Pro- 
ceedings of the Zoological Society of London 1545: 136— 
138. 

Sowerby, G. B., If. 1881. Description of eight new species of 
shells. Proceedings of the Zoological Society of London 
1881: 635-638, pl. 56. 

Sowerby, G. B., HI, 1903. Mollusca of South Africa. Marine 
Inve stigations in South Africa 2: 213-232, pls. 3-5, 
Springsteen, F. J. and F. M. Leobrera. 1986. Shells of the 
Philippines. Cartel Seashell Museum, Manila. 377 pp. 
Tate, R. 1889. The gastropods of the older Tertiary of Australia. 
Part I. Transactions and Proceedings and Report 
of the Royal Society of South Australia 11: 116-174, pls. 

2-10. 

Verhecken, A. 1986. The Recent Cancellariidae of Indonesia 
(Neogastropoda, Cancellariacea), Gloria Maris 25(2): 29- 
66. 

Verhecken, A. 1997. Mollusca Gastropoda: Arafura Sea Can- 


Page 18 


THE NAUTILUS, Vol. 122, No. | 


cellariidae collected during the KARUBAR Cruise. In: A. 
Crosnier and P. Bouchet (eds), Résultats des Campagnes 
MUSORSTOM, Volume 16. Mémoires du Muséum Na- 
tional d'Histoire Naturelle 172: 295-323. 

Verhecken, A. 2007. Revision of the Cancellariidae (Molhusca, 
Neogastropoda, Cancellarioidea) of the eastern Atlantic 
(40°N—40°S) and the Mediterranean. Zoosystema 29: 281— 
364. 


Verhecken, A. 1999. A new species of the genus Axelella (Mol- 
lusca, Neogastropoda, Cancellariidae) from off Japan and 
the Philippines; with notes on related species. Bulletin de 
l'Institut Royal des Sciences Naturelles de Belgique, Bi- 
ologie 69: 23-30. 

Wilson, B. 1994, Australian marine shells. Prosobranch gas- 
tropods, Part 2 (Neogastropods). Odyssey, Kallaroo, 
370 pp. 


THE NAUTILUS 122(1):19-51, 2008 


Page 19 


On some Neogene to Recent species related to Galeodina 
Monterosato, 1884, Galeodinopsis Sacco, 1895, and Massotia 
Bucquoy, Dautzenberg, and Dollfus, 1854 (Caenogastropoda: 


> 


Rissoidae) with the description of two new Alvania species from 


the Mediterranean Pleistocene 


Vittorio Garilli 

APEMA Research and Educational Service 
Via Alla Falconara, 34 

1-90136 Palermo, ITALY 
vittoriogarilli@apema.eu. 


ABSTRACT 


Six species, related to the subgenera Galeodina, Galeodinopsis, 
and Massotia, are re-analyzed. Alvania francescoi new species 
(SE Sicily) and A. rosariae new species (SE Sicily and NW 
Peloponnesus) are described from Mediterranean Pleistocene. 
Galeodinopsis is regarded as the appropriate genus for Oli- 


gocene-Recent taxa having a quite conical shell close to that of 


some Alvania species and showing Manzonia-like combination 
of two microsculptural characters: the pitted surface on the 
spiral cords and the arrangement of the roughly prismatic ele- 
ments forming numerous and very fine spiral nes ids. Its type 
species, Rissoa tiberiana (previously known from Mediterra- 
nean Mio-Pliocene), lives along the tropical W African coasts, 
where it is known under the name A. fariai. As suggested by the 
oldest record of Galeodinopsis, the European Oligocene Rissoa 
duboisi, this genus very likely originated from a pre-Neogene 
Alvania group. The syntype of Rissoa prusi, a scarcely known 
species (Pleistocene of Rhodes), and material (also from type 
locality) of the almost unknown R. cingulata (from Sicily) and 
of its close relative, R. tenera (Mediterranean, Atlantic Moroc- 

co and Canary Islands), is shown. These three taxa and the type 
species of Galeodina and Massotia, are here tentative ‘ly con- 
sidered as belonging to Alvania sensu lato. With the exception 
of A. cingulata, all the discussed species have a multispiral 
protoconch. Generally, the protoconchs studied exhibit a sculp- 
tural pattem known in other rissoid taxa. Protoconch I of the 
type species of Massotia, A. lactea, is characterized by a coarser 
sculpture 


Additional Keywords 
Atlantic, Neogene-Recent, new species 


Aissoidae, taxonomy, Mediterranean-W 


INTRODUCTION 


The genus Alvania Risso, 1826, comprises one of the 
most diversified groups in the caenogastropod family 
Rissoidae, especially when considering the tropical east- 


ern Atlantic and the Mediterranean provinces. They in- 
habit a large variety of environments, from littoral to 
bathyal, and their geographical distribution is extensive, 

including the Mediterranean, Atlantic, Indo-Pacific, and 
the temperate Australian coasts (Ponder, 1985). Their 
stratigraphical distribution might extend back to the Late 
Cretaceous, but the first well-established records date to 
the early Tertiary (Ponder, 1985). With regard to the rich 
Mediterranean and E suropean Tertiary Alvania assem- 
blages, some of the most informative analyse ‘sare those 
of Sacco (1895), Seguenza (1903), Cossmann (1921), Lo- 
zouet (1998), Kowalke and Harzhauser (2004), and Chirli 
(2006). 

Many authorities, e.g. Monterosato (1554), Bucquoy et 
al. (1884), Weink matt | 1885), Kobelt (1888), Locard 
(1886), Nordsieck (1968, 1972), Jeffreys (1867, 1569), 
Pallary (1920), Wenz (1938), Warén (1975, 1974), Gofas 
and Warén (1982), Van Aartsen (1982a, 1982b) . Moolen- 
beek and Hoenselaar (1989, 1998), Van der Linden and 
Wagner (1989), Moolenbeek et al. (1991), Van der Lin- 
den (1993), Bouchet and Warén (1993), Peg wee 
Savelli et al. (1996), Palazzi (1997), Gofas (1999), Avila 
(2000), and Arduino and Arduino (2001), nee contrib- 
uted to the knowledge of the rich Recent Eastern Atlan- 
tic (especially the W Africa and the Macaronesian Prov- 
ince) and Mediterranean fauna. Ponder’s rissoid revision 

(1985), listing five Alvania subgenera, provided addi- 
tional perspective to the taxonomy. 

My attention is focused here on a relatively large, in- 
formal group of species of Alvania sensu lato character- 
ized by shells with wide and ovate aperture, lacking an 
internal denticulation of the outer lip, and often bearing 
varices on a well-developed, inflated body whorl. These 
species have been historically assigned to the subgenera 

Galeodina Monterosato, 1884, Gale ‘odinopsis Sacco, 
1895, and Massotia Bucquoy et al., 1554. This subgeneric 
settlement was partially rejected by Ponder (1985), who 


Page 20 


THE NAUTILUS, Vol. 122, No. 1 


modifying the sy’ stematic ar rangement of Monterosato 
(1884) and Bucquoy et al. (1884), stated that grouping of 
the numerous species of fee was very difficult at the 
subgenus level. The same author included Massotia ( (type 
species: Rissoa lactea Michaud, 1830) and Galeodinopsis 
(type species: R. tiberiana Coppi, 1876) in the Alvania 
sensu stricto group and mca considered Galeodina 
(type species Turbo carinatus Da Costa, 1778) as a valid 
subgenus on the basis of shell characters. Piani (1979) 
raised Galeodina to generic level and placed Rissoa cin- 
gulata Philippi, 1836, and R. tenera Philippi, 1844, 
therein. 

The principal aim of this work is to provide, for the 
first time, as complete as possible a taxonomic dataset 
based on shell features. Species such as Rissoa cingulata, 
loften misidentified as Alvania carinata (Da Costa, 
1778)], R. prusi Fischer, 1877, and R. tiberiana Coppi, 
1876, are very poorly known or, in the case of the last 
taxon, the generic attribution to Alvania appears incor- 
rect. I also describe two new species from the Pleisto- 
cene of the central Mediterranean area. 


MATERIALS AND METHODS 


Most of the examined material, all consisting of shells, is 
housed in the Muséum National d'Histoire Naturelle, 
Paris (Département Systématique et Evolution), in the 


private collections of Maurizio Forli (Prato, Italy) and of 


Stefano Palazzi (Modena, Italy). Remaining material is 
housed in the private collections of medshells.com (made 
available by Nino Adorni Sbrana, Grosseto, Italy), Ste- 
fano Rufini (Anguillara), in the Museum fiir Naturkunde 
(Humboldt University, Berlin), the Museo Geologico 
G.G. Gemmellaro and the Dipartimento di Geologia e 
Geodesia (both University of Palermo, Italy), in the Gou- 
landris Natural History Museum (Kifissia, Athens), the 
Dipartimento di Scienze della Terra of the University of 
Catania, Italy, the Zoologisch Museum of Amsterdam, 
the Muséum National d’Histoire Naturelle of Paris (Dé- 
partement Histoire de la Terre), and in my personal col- 
lection. Many other private collections were visited. 

In the parts dedicated to each species, a list of the 
material is provide ed with all information given in the 
original labels. An abbreviated list of essential synonymy 
and/or citations is provided. 

The fossil material (all from the Mediterranean area), 
from the deposits of Dattilo (NW Sicily), Cartiera 
Mulino (Vittoria, SE Sicily, type locality of Alvania 
francescoi new species and Alvania rosariae new spe- 
cies), and Kyllini (Elea, NW Peloponnesus, Greece), was 
obtained by washing bulk a on a serial sieves (0.5, 
1, and 2 mm square meshes) and sorted using a stereo- 
microscope. The same procedure was followed for the 
Aecent bulk sam] sles collected from Magnisi (Siracusa, 
SE Sicily, the ee locality of Rissoa cineulata Philippi, 
1836) and Mondello (Palermo, NW Sic ‘ily ) spec ifics ally to 


recover that species. Geological, stratigraphic and paleo- 
ecological information on the deposits of Cartiera 
Mulino, Dattilo and Kyllini are taken from Costa (1989), 
Garilli (1998; 2004), Garilli et al. (2005a; 2005b) and 
Garilli and Galletti (2007). The stratigraphic information 
on the Sicilian deposits of Birgi (Trapani) and Tommaso 

Natale (Palermo, originally attributed to the upper 
Pleistocene Thyrrenian by Ruggieri and Milone, 1973) 
are from Ruggieri and Unti (1988) and Hearty et al. 
(1986), respectively. For the remaining fossil ma- 
terial (mainly from Coll. Forli ex-coll. Palag7i), I followed 
the stratigraphic attributions reported by the col- 
lectors. 

For all the discussed species, at least five shells were 
studied by the scanning electron microscope (SEM) us- 
ing a Philips XL30 ESEM, except for the (sole) syntype 
of Rissoa prusi and R. cingulata. of which there were 
only three shells available. Specimens examined by SEM 
were cleaned in a Bransonic 5 ultrasonic machine using 
distilled water. Particular attention was given to proto- 
conchs and teleoconch microsculptures as potential 
sources of taxonomic characters at species level. The 
number of protoconch whorls were counted according to 
Verduin’s method (1977). 

Shells were measured using a stereo microscope pro- 
vided with a cross-line micrometer eyepiece. The posi- 
tion of any varix on the body whorl is indicated in degrees 
of the angles formed by the plane of the varix and of the 
outer lip. 

Geographic, bathymetric, and stratigraphic distribu- 
tion of each discussed species is based on the examined 
material and the literature. Published records were criti- 
cally evaluated on the basis of good illustrations or sat- 
isfactory de scriptions. This type of dataset allows for just 
a rough representation of the geographical distribution, 
e specially for the Adriatic Sea and the easternmost Medi- 
terranean basin, of which I found very little material 
from the collections studied. 

Abbreviations are used as following: DGUP: Diparti- 
mento di Geologia e Geodesia, Universita di Palermo, 
Italy; DSTC: Dipartimento di Scienze della Terra, Uni- 
versita di Catania, Italy; GNHM: Goulandris Natural 
History Museum, Kifissia, Athens, Greece; MGUP: Mu- 
seo Geologico G.G. Gemmellaro, Universita di Palermo, 
Italy; MNHN-DHT: Muséum National d'Histoire Na- 
turelle, Departement Histoire de la Terre, Paris, France; 
MNHN-DSE: Muséum National d'Histoire Naturelle, 
Departement Systématique et Evolution, Paris, France; 
MPOB: Dipartimento del Museo di Paleobiologia e 
del?Orto Botanico, Universita di Modena e Reggio 
Emilia, Modena, Italy; MSNCS: Museo Regionale di 
Storia Naturale e Mostra Permanente del Carretto Sicil- 
iano, Terrasini, Italy; ZMA: Zoologisch Museum Aimster- 
dam, Holland: ZMB: Museum fiir Naturkunde, Hum- 
boldt Universitit, Berlin, Germany; Coll.; collection; 
Coll. MF: Maurizio Forli collection, Prato, Italy; Coll. 
PAL: Stefano Palazzi collection, Modena, Italy; Sh(s) 
shell(s), used in the Material Examined sections only, 


V. Garilli, 2008 


Page 21 


SYSTEMATICS 


Family Rissoidae Gray, 1847 
Subfamily Rissoinae Gray, 1S47 
Genus Alvania Risso, 1826 


Type Species: = Alvania europea Risso, 1826 [synonym 
of A. cimex (Linnaeus, 1758)], subsequent designation by 
Nevill, 18S5 


Alvania carinata (Da Costa, 1778) 
(Figures 1-14) 


Turbo carinatus Da Costa, 1778: 102-103, pl. 8, fig. 10 
Rissoa trochlea Michaud, 1830: 16, fig. 4 

Galeodina cingulata (Philippi, 1S36).—Piani, 1979: 70-71, figs. 

2-3 

Description: Shell small, sturdy, conical and keeled to 
slender and turrited, reaching about 5 mm (rarely 6 mm) 
in height; 3.94.1 mm in width. Protoconch multispiral, 
conical, consisting of about 2.2 convex whorls. Proto- 
conch [ with 0.5 whorls, sculptured with six very fine 
spiral lirae and microscopic granules between them. Pro- 
toconch/teleoconch transition well-marked and sinuous. 
Protoconch H sculptured with sparse, spirally arranged 
microscopic Sade stronger in adapical direction, and 
one to two spiral ridge »s, one of them always very close to 
lower suture. In the irgest shells, te sleoconch formed by 
44.5 moderately convex whorls. Common morph (Fig- 
ures 2, 6-7) with teleoconch whorls mar kedly dominated 
by spiral sculpture, which consists of very strong cords 
(numbering 2-3, 3-6, 6-12, and 12-17 on first, second, 
third, and last teleoconch whorl, respectively). Second- 
ary, less conspicuous cords may occurr on last whorl. 
More marked spiral cords on adapical portion of whorls, 
at a certain distance from suture, give a characteristic 
keeled shape. Cords progressively less strong on basal 
area. Unkeeled morph (Figures 1, 4-5) characterized by 
a ser er shell shape, usually bears more spiral cords 
(1S—20) on last whorl. Axial sculpture always formed by 
numerous (32-50 on penultimate whorl), occasionally 
very narrow and lamella-like ribs, becoming obsolete to- 
ward base. Intersection of ribs with spiral cords gives an 
almost general clathrate pattern in unkeeled morph. In- 
tersection of spiral and axial sculptures nodular, usually 
forming squares (Figure 14), with i of last whorl 
where a rectangular pattern occurs (Figure 13). Micro- 
sculpture consists of very fine spiral threads (Figures 


13-14), covering all teleoconch surface, with exce ption of 


main spiral pattern. On early teleoconch whorls, spiral 
lirae often alternate with spiral alignments of micro- 
scopic pimples. Sutures slightly inclined. Last whorl well 
expande od, comprising 3/5 to 3/4 ( rarely more than 3/4, 
Figure 3) of shell height, often bearing one or two varices 
(mainly in keeled morph) with angles of 10°-340°. Ap- 
erture wide, ovate, slightly rounded to ; angled in the pos- 
terior part, comprising 3/5 to 3/4 of fase whorl height. 
Outer lip slightly prosocline, internally smooth, exter- 
nally markedly thickened by a strong rim very close to lip 
edge, and covered by spiral cords. Inner lip moderately 


arcuate and rather thickened, with a very narrow to dis- 
cretely expanded (Figure 3) callus delimiting a very small 
umbilical chink. 
Type Locality: Cornwall, southwestern England. 

Material Examined: Great Britain: Cornwall, Fal- 
mouth, 1 sh., coll. MF, 1974, E55A; Channel Islands, 
Herm, 11 shs, coll. MF, 09.1974, E54A. Atlantic France: 
Normandy, Carteret, 3 shs, coll. MF, Jul. 1973, E28A; 
lower Normandy, St. Pair, 7 shs, MNHN ain Denis, 
1945; Brittany, Finistere Anse de Bertheaume, 20-30 m, 
industrial dredging, 4 shs, MNHN coll. S. Gofas, 1978; 
Brittany, Cote-du-Nord Plomanach, fissures of infralit- 
toral rocks, 1 sh., MNHN coll. S. Gotas, 1973-78: Brit- 
tany, Finistere Roscoff, “les Cochons Noirs”, sand and 
conchiferous gravel, 20 m, 27 shs, MNHN coll. Gofas, 
Jul. 1994, Brittany, St. Lunaire, 20 shs, MNHN Coll. 
Fischer, 1898: Brittany, St. Servan, 10 shs, MNHN coll. 
Staadt, 1969: Brittany, St. Lunaire, 2 shs, MNHN coll. 
Ph. Dautzenberg ( figured i in Bucquoy - - 1854, pl. 35, 
figs. 1, 2); St. Lunaire, 5 shs, MSNCS, 7173 and 7174, 20 
Jun. 1970, on the beach at low tide; fan uy, Morlaix, 
Saint Michel en Gréve, 2 shs, coll. MF, 1976, E25A; 
Brittany, Saint Jacut, 3 shs, coll. MF, 06.1975, E12A; 
Brittany, Saint Jacut, 14 m, 6 shs, coll. MF, 04.1974, 
E12B: Britt: iny, Carnac, Quiberon, 4 shs, coll. MF, 1970, 
E16B; Brittany, Carnac, Quibe ron, Pointe de Couquel, 2 
shs, coll. MF, 1970, ELIA: Brittany, Saint Malo, 7 shs, 
coll. MF, 07.1973, EL3A; Brittany, St. Malo, Lizardrieux, 
1 sh., coll) MF, Aug, 1982, E69A. Atlantic Pyrenees, 
Aquitaine, St. Jean de Luz, Cote a infralittoral 
rocks, | sh., MNHN coll. S. Gofas, 19S0-S1; Atlantic 
ae Aquitaine, St. Jean de Luz, 73 shs, MNHN 
coll. H. Fischer, 1898: Aquitaine, Soulac, 1 sh., MNHN 
coll. A. Dolfus; Aquitaine, Hendaye, 2 shs, coll. MF, Jul. 
1976, E57A. Portugal: Algarve Sagres, Baie de Baleeira, 
(37°00.7' N, 08°55.0' W), tide zone, 1 sh., MNHN, Mis- 
sion Algarve, May 1988; Algarve Sagres, Ponta da Ba- 
leeira, (37°00.3' N, 08°55.5' W), 17-23 m, 5 shs, 
MNHN, Mission Algarve, May 1988; Algarve Sagres, 
Pontal dos Corvos, (37°01.3' N, 08°58.3' W), at the foot 
of falaise, 17-22 m, 5 shs, MNHN Mission Algarve, May 
1988; Albufeira (southern coast), Ponta de Castelo, 3-6 
m, 2 shs, coll. MF, OS Aug. 1985, E50B. Atlantic Moroc- 
co: asilah: mouth of Oued el Helou, conchiferous depos- 
its, beach, 6 shs, MNHN coll. S. Gofas, 1971-72. Strait of 
Gibraltar: Tanger, Grande Plage, conchiferous deposits, 
beach, 3 shs, MNHWN coll. S. Gofas, L970-S1; Cadiz, 
Getares, beach, 3 shs, coll. MF, ex coll. C. Bogi, legit 
Hanselaar, 2230 GET; Cadiz, Barbate, conchiferous de- 
posits, beach, 6 shs, MNHN coll. S. Gofas, 1976-S1; 
south Ceuta, Punta del Desnarigado, (35°53,.6' N, 
05°16.8' W), 16-20 m, 1 sh., MNHN coll. Bouchet, Go- 
fas and Lozouet, May 1996. Mediterranean Spain: Cas- 
tellon, Columbretes Islands, Espinosa Island, 5 m, 1 sh., 
coll. MF, 26 Jul. 1974, M273A; Malaga, 15 m, 7 shs, coll. 
MF, ex coll. Cesare Bogi, 2127IMA(V002G); Malaga, 
Algeciras, Torre del Almirante, 3-5 m, | sh., coll. MF, 28 


THE NAUTILUS, Vol. 122, No. 1 


V. Garilli, 2008 Page 23 


Figures 9-14. Alvania carinata (Da Costa, 1778), protoconch and details of sculpture. 9. Sicily, Palermo, Terrasini, “Magaggiare- 
Ciucca di Cinisi” beach, coll. tiene 146E) , protoconch. 10. Provence, Marseille, La Baule, small beach at 25 km west from Marseille, 
coll. PAL (212SBAU-VO008C), protoconch. 11-12. Shell from the same locality and collection, sculpture of protoconch T and II (11) 
and a detail of protoconch I (12). 13. Same shell as Figure 2, detail of teleoconch sculpture on the penultimate whorl, coll. PAL 
307B). 14. Detail of teleoconch sculpture on the last whorl, NW Sicily, Palermo, Terrasini, “Magaggiare-Ciucca di Cinisi” beach 
detritus of Miniacina, coll. PAL (146E). Scale bars: 100 wm in Figures 9-11 and 13-14; 50 zm in Figure 12. White arrows indicate 
the protoconch/teleoconch boundary, respectively. 


Sep. 1976, M207B; Malaga, Algeciras, 3 shs, coll. MF, 23 gloue, (43°10.6' N, 05°24.2' E), 33 m, 31 shs, MNHN 


Sep. 1976, M207A; Malaga, Fuengirola, 0.5-1 m, 1 sh., rec. H. Zibrowius Jun, 1996; Provence, Iles Embiez, con- 
coll. MF, 20 Aug. 1973, MG6SB; Malaga, Algeciras, Playa chiferous deposits, beach, 10 shs, MNFIN coll. S. Gofas, 
Getares, 6 m, 1 sh., coll. MF, Aug. 1983, M207E; 1968-70; Provence, Iles Embiez, Petit Rouveau, dredg- 
Malaga, Cabo Pino, detritus, 10 m, 2 shs, coll. § — ing of sandy conchiferous bottom, 3-5 m, 5 shs, MNHN 
(41.80g). Mediterranean Morocco: M’diq “(anc. coll. S. Gofas, 1968-70: Provence, St. Clair, infralittoral 
Rincon)” conchiferous deposits, beach, 1 sh., int rocks, (43°08.2' N, 6°23.2' E), 0-1 m, 1 sh., MNHN rec. 
coll. S. Gofas, 1971. Algeria: Alger, 1 sh., MNHWN coll. S. Gofas, Sep. 1992; Provence, St. Raphael, 2. shs, 
Locard. Mediterranean France: Languedoc, Carnon, MNHN coll. Locard: Provence, 9 shs, MNHN coll. Petit, 
conchiferous deposits, beach, 1 sh., MNHWN coll. S. Go- 1873; Provence, Sanary, | sh., MNHN coll. Locard:; Cor- 
fas, Aug. 1976: Languedoc, Roussilion, 3 shs, MNHN sica, Ajaccio, 3 shs, MNHN coll. Jousseaume, 1921; Cor- 
Coll. Dolfus, 1903; Languedoc, Roussilion, 1 sh., MNHN sica, Galeria, Punta Stollo, 6 m, 1 sh., coll) MF, 07 Aug 
coll. Ph. Dautzenberg (Moll. du Roussillon); Provence, 1984, M52D: Corsica, Pianottoli, Anse de Chevanu, 2 
Marseille, La Baule, small beach at 25 km west from shs, coll. MF, Jun. 1988, M77A. ee Djerba, Aghir 
Marseille, 3 shs, coll. MF, ex coll. C. Bogi, Oct. 1986, 5 m, Posidonia bed, 1 sh., coll. MF, S. Palazzi legit 06 
212SBAU (VOOSC); Provence, Marseille, Cape Cou- Aug. 1993, M79A; Djerba, Al Jazirah, 1- ~2.5m, | sh., coll 
ronne, 4 shs, coll. MF, ex coll. C. Bogi, Oct. 1987, MF, Aug. 1974, M9B. Italy: Friuli Venizia Giulia, Tri- 
2129COU (VOOSE): Provence, Marseille, Grand Con- este, beach, 7 shs, coll. MF, D. Di Massa legit 1976 


Figures 1-8. Alvania carinata (Da Costa, 1778), variation in shell shape and sculpture. 1. “Form” ecarinata Bucquoy et al., 1554 
shell from coll. Ph. Dautzenberg (Moll. du Roussillon), MNHN-DSE. 2. Typical, keeled morph, Liguria, Genova, Camogli, 42 m 
coll. PAL (307B). 3. Markedly keeled morph, corresponding to Rissoa trochlea Michaud, 1830, Strait of Gibraltar, Cadiz, Getares 
beach, coll. PAL (2230 GET). 4-5. Small, unkeeled morph, La Spezia, Monterosso, 15-30 m, coll. PAL (70A). 6. Typical morph with 
varice, Sardinia, Sassari, Capo Caccia, Cala della Calcina, 6 m, coll. PAL, (112A). 7. Profile view of the same shell as Figure 2. 8. 
Juvenile shell fitting well with the concept of Alvania cingulata (Philippi, 1836) sensu Piani (1979, figs. 2-3). Scale bars: 1 mm in 


Figures 1-7: 0.5 mm in Figure 8 


Page 24 


THE NAUTILUS, Vol. 122, No. 1 


307B; Liguria, Genova, Camogli, 42 m, 1 sh., coll. MF, 
06.1981, 271B; La Spezia, Monterosso, 15—30 m, detri- 
tus, 1 sh., coll. MF, A. Lugli legit Oct. 1978, TOA; La 
Spezia, Portovenere, 5-25 m, 3 shs, coll. MF, Oct. 1978, 
329A; La Spezia, Riomaggiore, 30 m, 1 sh., coll. MF, 
Aug, 1975, 312A; Liguria, La Spezia, Punta Mesco, 35 m, 
L sh., coll. MF, Jul. 1987, 176C; Tuscany, Livorno, San 
Vincenzo, Borraceia, 12 m, 1 sh., coll. MF, 12 Jul. 1987, 
190A; Tuscany, Livorno, Secche della Meloria, 6—17 m, 2 
shs, coll. MF, 1974, 34E; Tuscany, Livorno, Castiglion- 
cello, 4 shs, coll) MF, 1972, 21A; Tuscany, Livorno, 
Romito, off mouth of the torrent Chiona, 30-35 m, 2 shs, 
coll. MF, ex coll. C. Bogi, 36A; Tuscany, Livorno, Bagni 
Fiume, 20 shs, coll. MF, 1977, 34B; Tuscan Archipe slago, 
Island of Capraia, 100/400 m, 2 shs, coll. MF, ex coll. C. 
Bogi, 2126CAP(VO005G); Island of Capraia, Punta della 


Fica, 29 m, 1 sh., coll. MF, 15 Sep. 1985, 25R; Island of 


Capraia, Punta Civitata, 40 m, 1 sh., coll. MF, 19 Sep. 
1985, 28ST; Tuscan Archipelago, Gorgona Island, 35—40 
m, LI shs, coll. MF, Aug. 1978, 69A; Gorgona Island, 40 
m, 18 shs, coll. MF, legit C. Bogi, 194; Tuscany, Siena, 1 
sh., yellow sands from unknown layer, lower Pliocene, 
coll, MF ex coll. PAL, F22A; Tuscany, Grosseto, Punta 
Ala, Baia Verde, 5 m, 1 sh., coll. MF, G. Terzer legit OS 


Sep. 1974, 27A; Tuscany, Grosseto, Punta Ala, Punta 
Hidalgo, 2 shs, coll. MF, Jun. 1975, 27B; Tuscany, Gros- 


seto, Follonica, Cala Felice, 7 m, 1 sh., coll. MF, 25 Aug. 
1987, 55B: Tuscany, Grosseto, Island of Palmaiola, 2S m, 


2 shs, coll. MF, 14 Sep. 1986, 172A; Grosseto, Island of 


Elba, Capo Calamita, Scogli Corbelli, 46 m, 1 sh., coll. 
MF, Sep. 1972, 5E; Island of Elba, Scoglio Remaiolo, 35 
m, 1 sh., coll, MF, May 1980, 5k; Island of Elba, For- 
miche della Zanca, 20 m, 1 sh., coll. MF, Apr. 1954, 5P; 
Lazio, Rome, Civitavecchia, 2 shs, coll. MF, 1975, 137B; 
Sardinia, Sassari, Capo Caccia, Cala della Calcina, 6 m, 
conchiferous detritus at upper limit of Posidonia bed, 10 
shs, coll. MF, S. Palazzi legit 19 Aug. 1983, 112A; Sar- 
dinia, Sassari, bay at E of Faro di Capo Testa, 7-19 m, 1 
sh., coll. MF, 22 Aug. 1983, 125B; Nuoro, Capo Comino, 
Ruia Island, 0-2 m, | sh., coll. MF, 1974, 75A; Sardinia, 
Cagliari, Island of San Pietro, channel of San Pietro, 


2-4 m, Posidonia bed, 2 shs, coll. MF, G. Liuzzi legit 


10 Oct.1976, 264A; Sardinia, Cagliari, Island of 


Sant Antioco, Cala de Saboni, 1 sh., coll. MF, O7 Aug. 
1983, LI5B:; Sardinia, Cagliari, Island of Sant Antioco, 
Cala de Saboni, 14 m, 1 sh., coll. MF, 10 Aug. 1983, 
LI5E; Campania, Napoli, Island of Capri, Punta Vivara, 
6 m, 1 sh., coll. MF, 0S Sep. 1978, 254B; Campania, 
Napoli, Island of Procida, Marina Grande, 2.5-9 1m, 
2 shs, coll. MF, 1974, LO5A; Puglia, Bari, off Palese, 
12-13 m, 1 sh., coll. MF, 1] Sep. 1979, 77A; Puglia, 
Taranto, 2 shs, coll. MF, 1973, 84B; Puglia: Taranto, 
Campomarino, 2 shs., coll. MF, 12 Feb. 1977, 309A; 
Puglia, Taranto, Maruggio, 1 sh., coll. MF, 1977, 280A; 
Puglia, Brindisi, “Batteria Brin” beach, 14 shs, coll. MF, 
G. Oriolo legit 08.1970, 193D; Puglia, Brindisi, Punta 
Croce, 15 m, 19 shs, coll. MF, G. Oriolo legit Aug. 1974 
262A; Puglia, Brindisi, Torre Guaceto, 5 m, I sh., 


coll. MF, 1976, 42A; Brindisi, Lendinoso, 10-20 m, 1 sh., 
coll. MF, 1977, 279A; Puglia, Lecce, Gallipoli, Costa 
Brada, 2 shs, coll. MF, Apr. 1978, 91B; Lecce, Porto 
Cesareo, 2 shs, coll. MF, Oct. 1977, 46B; Lecce, Porto 
Cesareo, Torre Lapillo, 1 sh., coll. MF, 06 Jun. 1978, 
46C; Lecce, Marina di Ugento, 30 m, 1 sh., call MF, 
1977, 6SC: Lecce, Gallipoli, La Vecchia Torre, 2-6 m, 2 
shs, coll. MF, 1976, 91D; Lecce, San Cataldo, 1 sh., coll. 
MF, Apr. 1973, 260C; Lecce, Speechnulla, 1 sh., coll. 
MF, Oct. 1974, 281A; Calabria, Reggio Calabria, Pen- 
timele, from fisherman nets, 8 m, 1 sh., coll. MF ex coll. 
Sciano, VOG0A 194; Reggio Calabria, Laureana di Bor- 
rello, Pecoraio, 2 shs, (laver 2), lower Pleistocene, coll. 
MF ex coll. PAL, F24A; Sicily, Trapani, San Giuliano, 10 
shs, coll. MF, 25 Jul. 1978, 297A; Trapani, Scopello, Ton- 
nara, 12-20 m, 1 sh., coll. MF, 06 Apr. 1986, 333F; Sicily, 
Trapani, Egadi Islands, Favignana, Secea del Toro, 30 m, 
20 shs, coll. MF, A. Lugli legit 02 Jun. 1983, 272H; Egadi 
Islands, Favignana, 4 m, 17 shs, coll. MF; Egadi Islands, 
ae Cala Rotonda, 20-30 m, 3 shs, 04 Jun. 1983, 
coll. MF, 272G; Egadi Islands, Favignana, eteene Cor- 
rente, 30 m, 3 shs, coll. MF, O1 Jun. 1983, 272F; Egadi 
Islands, Favignana, Punta Sottile, 30-40 m, 4 shs, coll. 
MF, 30 May 1983, 272E; Egadi Islands, Marettimo, 
Punta Bassana, 41 m, 2 shs, coll. MF, 04 Jun. 1983, 140A; 
Egadi Islands, Levanzo, Faraglione, 7-12 m, 3 shs, coll. 
MF, 03 May 1979, 325B: Sicily, Pantelleria Island, Baia 
dei Fichi d’India, 33. m, 1 sh., coll. MF, Jul. 1983, 154M: 
Pantelleria Island, Punta Capace, 31 m, 1 Se coll. MF, 
Jul 1983, 154N; Sicily, Palermo, 1 sh., coll. MF ex coll. 

. Bogi, (VO5G6A) 194; Palermo, Bagheria, Aspra, 15.5 m, 
7 chs. coll. MF, Apr. 1973, 124C; Pz ilermo, Punta Raisi, 
Marina Longa, 1 sh., coll. MF, 28 Feb. 1979, 320A; Pal- 
ermo, Terrasini, “Magaggiare-Ciucea di Cinisi” beach, 
detritus of Miniacina, 14 shs, coll. MF, S. Palazzi legit 23 
Sep. 1977, 146E; Palermo, Island of Ustica, Scoglio del 
Medico, 25 m, 1 sh., coll. MF, Aug. 1980, 286k; Sicily, 
Messina, Milazzo, Capo Milazzo, Cala SantAntonio, 2 
shs, upper yellow sands, upper Pleistocene, coll. MF ex 
coll. PAL, F5A: Messina, Eolie Islands, Lipari, Secca del 
Bagno, 38-40 m, 3 shs, coll. MF, 05 Se p. 1979, 338A; 
Sicily, Cat: mia, Acitrezza, 3-25 m, 2 shs, coll. ME. Sep. 
1975, 22A: Sicily, Siracusa, Vendicari, bay, 2 shs, coll. 
MF, 13 Sep. 1977, 20A: Siracusa, Portopalo di C Capo 
Passero, 2-3 m, | sh., coll. MF, Jun. 1976, 63A; Siracusa, 
Peninsula of Magnisi, southern side, 1 sh., coll. V. Garilli, 
Jun, 2006; Sicily, Palermo, Tommaso Natale, 1 sh., late 
middle Pleistocene, MGUP 166/2/49; Sicily, Trapani, 
Birgi, 3 shs, upper Pleistocene, Tyrrhenian Stage, 
MGUP. 1765/36/14: Sicily, Trapani, 12 shs, upper Plei »18- 
tocene, Tyrrhenian Stage, MGUP. 358/3/42. Isle of 
Malta: Malte, 8 shs, MNHWN coll. Jousseaume, 1921. 
Croatia: Istria, Rovinij, between Rt. Muntrav, Hr. Mun- 
trav ancl Azino, S—L5 m, detritus from bottom, 11 shs, 
coll. MF, Palazzi legit 19 Jun. 1975S, MISSB; Istria, 
Savudrija, i. ach, 8 shs, coll. MF, S. Palazzi legit Oct. 
1975, M364A; Istria, Rovinij, 6-18 m, 19 shs, coll. MF, 
20 Jun. LOTS, MISSA; Istria, Vrsar, | sh., coll. MF, 03 


V. Garilli, 2008 


Page 25 


Apr. 1958, M7G6A; Istria, Umag, 1 sh., coll. MF, 1978, 
M1O9A; Istria, Umag, Taverna Lero, 0.3 m, 1 sh., 17 Jun. 
L978, coll. MF, MLO9B; Kaciack, dam of Cigale, 4m, 1 
sh., 29 Jul. 1983, coll. MF, MI4A. Greece (: Aegean Sea): 
NW Aegean, Island of Limnos, near Moudros, 3-5 m, 48 
shs, coll. MF, A. Lugli legit Aug. L991, MSOA; Sithonia, 
Ormos Panagias, 2-3 m, 1 sh., Aug. 1982, coll. MP, 
M31B; Sithonia, Nesis Dhiaporos, 33-34 m, 2 shs, coll. 
MF, 01 Aug. 1986, M31D. Unknown locality from Medi- 
terranean Sea: 120 shs, MNHN coll. Vayssiére. 


Habitat: The species is usually found at depths com- 
patible with the upper part of the shelf, the infralitoral 
stage of Pérés and Picard (1964). Rarely, I found material 
collected from the tidal zone. According to Gofas and 
Ponder (1991), Alvania carinata lives deeply buried un- 
der stones. It seems also to be linked (in the Mediterra- 
nean) to phanerogam beds. The finding of two shells 
from the Island of Capraia (Tuscan Archipelago), col- 
lected at a depth of 100-400 m, is much pre sbably due to 
lower shelf-slope transport. 


Distribution: Occurs probably throughout the Medi- 
terranean, but its presence in the easternmost coasts, 
from which I did not see any material, needs to be con- 


firmed. In the Atlantic it is recorded from the coasts of 


Great Britain south to Morocco. As fossil, it is rare in the 
Mediterranean Neogene where it is recorded from the 
lower Pliocene yellow sands near Siena. It becomes more 
frequent during the Mediterranean Quaternary, where I 
found it from the lower Pleistocene of Reggio Calabria, 
and the middle-upper Pleistocene of Sicily ( (Tommaso 


Natale, Capo Milazzo and Trapani). 
Remarks: A complete synonymy list was provided by 


Piani (1979) and Van Aartsen (1982). In various collec- 
tions, I found different lots of this species containing 
juvenile shells (not higher than 3 mm, see Figure §) 
identified as Alvania cingulata ( (Philippi, 1536), species 
hereafter re-described and discussed. This wrong deter- 
mination very likely follows the misidentification of Piani 
(1979, figs. 2-3) and Giannuzzi-Savelli et al. (1996, figure 
499). 

Typically, the shell of this species has a characteristic 
keeled shape due to the presence of well marked spiral 
cords that become very strong on the well-developed last 
whorl. The number of ee on adult whorls is slightly 
variable. 34 in the penultimate whorl. The not-ke meled 
morph, characterized by having a slender and turrited 
shape and more numerous cords (5-6 in the penultimate 
whorl), is the “form” ecarinata Bucquoy et al., 1854 (Fig- 
ures 1, 445), also known as minor-ecarinata Montero- 
sato, 1884 (probably corresponding to Rissoa lucullana 
var. cancellata Scacchi, 1836, as indicated by Piani, 
1979). It has been considered a Mediterranean subspe- 
cies of A. carinata by Van Aartsen (1982). I found no 
shells clearly belonging to this morph in Atlantic 
material studied (mainly from the MNHN-DSE collec- 
tions). However, in some Mediterranean areas (e.g. along 
the French coast in the MNHN-DSE eollertons), I 


found the typical morph together with unkeeled shells. 
Thus [am more inclined to consider it just a case of 
intraspecific variation, which appears well represented in 
Mediterranean populations. The keeled morph, charac- 
terized by having three spiral cords on subadult whorls 
and a very expi anded body whorl (more than 3/4 of the 
shell he ight, see Figure 3) fits well within the concept of 
Rissoa trochlea Michaud, 1830, which is certainly a syn- 
onym of A, carinata. 

Alvania carinata is the type species of Galeodina 
Monterosato, 1884, a genus created for cingulated, 
keeled, varicose rissoid shells with a wide aperture 
(Monterosato, 1SS4a, p. 163). This generic division can 
appear quite artificial, not being supported by appropri- 
ate and constant taxonomic features. Also, the consider- 
ation of Galeodina as a well-established subgenus of Al- 
vania, as indicated by Wenz (1938) and Van Aartsen 
(1982), appears doubtful. Ponder (1985), who synony- 
mized most of the available subgenera with Alvania 
sensu stricto, expresse d some uncertainty when consid- 
ering Galeodina as a valid taxon. He recognized the ig- 
norance of anatomical features as the main gap in solving 
this systematic question. In any event, the shell chaise 
ters alone do not provide enough support for Galeodina 
to stand as a subgenus. For example, the unkeeled 
morph, which usually lacks the main features of Galeo- 
dina, including the characteristic varices of the keeled 
morph, might be compared to the Alvinia Monterosato, 
1854, species group, with which it shares a slender, tur- 
reted shell shape. It is noteworthy that Cossmann (1921) 
cited Galeodina as a synonym of Alvinia. 


Alvania cingulata (Philippi, 1836) 
(Figures 15-26) 


Rissoa cingulata Philippi, 1S36: 152-153. 

Acinus cingulatus (Philippi)—Monterosato, 1$54b: 62. 

Cingula (Onoba) cingulata (Philippi, 1$36).—Nordsieck, 1968: 
46, pl. VIL, fig. 26.35 

Alvinia (Galeodina) cingulata (Philippi, 1836)—Nordsieck, 
1972: 181, pl. RVI, fig. 20. 


Description: Small, stately, conical-ovate shell reach- 
ing 3.5 mm in height, 2.2 mm in width. Protoconch pau- 
cispiral, consisting of little more than 1.5 convex ae very 
rounded whorls, sculptured by 5-6 very fine spiral lirae. 
Several prosocirte growth scars precede protoconch/ 
teleoconch transition (Figure 20). Teleoconch consists of 
about 4 cingulated convex a separated by slightly 

inclined sutures. Very early teleoconch whorls bear two 
main spiral cords crossed by numerous axial ridges. 

Sculpture of adult whorls consisting of spiral cords more 
conspicuous than axials, which are 5-6 and 10-11 on the 
penultimate and body whorls respectively. Axial sculp- 
ture formed by narrower (half the breadth of a spiral 
cord) ribs, numbering ae 40 on penultimate whorl, 
and which become obsolete toward shell base. Micro- 
sculpture consists of spiral rows of small tubercles that do 
not seem to appear on main sculpture (Figure 26). A few 
irregular scars cross these spiral rows. Last whorl rather 


THE NAUTILL 


V. Garilli, 2008 


Figures 24-29. 


Alvania cingulata (Philippi, 1536) from type locality (Peninsula of Magnisi, Siracusa, SE Sicily, coll. Garilli), and 


a slightly resembling Alvania sp., from the lower Pleistocene of Musala (Italy, Reggio Calabria, coll. MF ex coll. Pal, F95A), fitting 
well with the conce pt of A. cingulata stated by Monterosato (1SS4b). 24- 26. Alvania cingulata \pertural view of a subadult shell 


7 


5) and microsculpt ure (26). 27- 


24), sculpture on last whorl (5 


bars: 500 zm in figures 24, 27-28: 100 xm in figures 25, 29; 


protoce mnch/teleoconch b« vundary : 


inflated and well-developed, comprising little less than 


3/4 total height. Aperture wide and ovate, comprising 


about 4/9 and 2/3 of total height and last whorl respec- 
tively. Outer lip slightly prosocline, internally smooth, 
externally sli chily thickened near edge. Columellar side 
arcuate, with a thin callus forming a very narrow umbili- 
cal chink. Coloration consisting of quite large, reddish- 
brown bands on a whitish-cream background 


-29. Alvania sp. Apertural (27 
20 pm in Figure 26, The white arrow in Figure 29 indicates the 


), dorsal (28) views and protoconc h. Seale 


Type Locality: Peninsula of Magnisi, Siracusa, south- 


eastem Sicily. 
Type Material: 
coll. Philippi was de stroyed by Byne’s reaction (M 


One possible syntype, ZMB (2326) ex 


Glaubrecht, pers. comm., 2006) 


Material Examined: — Italy: NE Sicily, Messina, 2 shs 
ZMB ex coll. Monterosato, 80914 (original label by 


Figures 15-23. 


Alvania cingulata (Philippi, 1836). 15-17. Apertural (15), dorsal (16), profile (17) view of one shell from ZMB ex 
coll. Monterosato (80914), Messina, NE Sicily. 20, 23. Protoconch of the same shell. 18-19 and 21-22 


Another shell from the same 


lot apertural 18), dorsal (19 profil 21) view, and a detail of sculpture on the last whorl (22). Seale bars: 1 mm in figures 15-19 


and 21: 200 pm in Figure 22: 100 wm in Figure 20; 50 ym in Figure 23 


teleoconch boundary 


The white arrows indicate the approximate protocon h 


Page 28 


THE NAUTILUS, Vol. 122, No. 1 


Monterosato: Acinus cingulatus, 2 2. Messina, 1014); SE 
Sicily, Siracusa, Pemneuls of Magnisi, south side, detritus 
from the beach, 1 sh., coll Garilli, 6/2006 


Habitat: Unknown. It is noteworthy that sandy and 
Posidonia bottoms prevail in the south side of the Mag- 
nisi Peninsula (from where I collected the beached shell 
in Figure 24), while an almost paralic (pre-lagoon) envi- 
ronment (with Cerastoderma, Cerithium, and Gibbula 
spp., on a finely sandy bottom) and a rocky bottom with 
algae characterize the north and east sides. 


Distribution: — Its distribution appears to be limited to 
Sicily, with particular regard to the eastern coast, from 


Messina to Magnisi. It was recorded from Palermo by 


Monterosato (1872: 1875; 1878: 1884a; 1884b), who very 
likely followed the citations of Philippi (1544) and 
Weinkauff (1885). The latter author, who indicated Mon- 
dello (the beach near Palermo), also cited Ognina (Cata- 
nia) in eastern Sicily. To my knowledge, no f fossil record 
of this species exists. 


Remarks: This is a very rare, practically unknown spe- 
cies (see Van Aartsen, 1982a). Its identification has tra- 
ditionally been quite problematic, above all after the 
misidentification by Piani (1979, figs.2-3), who illus- 
trated a juvenile shell of Alvania carinata (Da Costa, 
1778) under the name Galeodina cingulata (Philippi, 
1836). The shell figured by Giannuzzi-Savelli et al. (1996, 
fig. 499), under the name Alvania (Alvania) cingulata 
(Philippi, 1836), is a subadult of A. carinata. 

My identification is based on three shells (one from 
the type locality, two from ZMB ex Monterosato coll. 
labelled as Acinus cingulatus), which agree well with the 
original description by Philippi (1836; see also 1544: 
128). A few diagnostic characters allow for its correct 
identification: the primary cingulated sculpture with 
raised cords, numbering 10-11 on the last whorl; the 
secondary axial pattern formed by very narrow ribs cross- 
ing the cords: the large ovate aperture: and the intern: ally 
eimaadh outer lip, slightly thickened near its edge. 

As Philippi ee ) noted, Alvania cinoulate may be 
compared with A. tenera, especially with the slender 
morph of the ae ‘Yr species (see Figure 78). Alvania te- 
nera has a smaller shell (usually not over 2.2 mm in 
height) with more raised cords (numbering 12-14 on the 
last whorl), and almost lamella-like ribs, which occur just 


on the interspaces between the cords (see discussion of 


A. tenera). Furthermore, A. cingulata has a paucispiral 
pre ytoconch. 

In addition, Nordsieck (1972) and Van Aartsen (1982) 
considered this species as belonging to Galeodina. How- 
ever, the similarity between Alvania cingulata and A. 
carinata (type species of Galeodina) is superficial. Sev- 
eral differences separate the two: A. cingulata has a less 
inflated last whorl bearing less numerous cords and lack- 
ing varices, its outer lip lacks a varicose thickness, and its 
spiral Sc at does not markedly dominate over axial as 
in the typic ral A. carinata. The latter species s also has a 
multispiral te conch. 


Monterosato (1584a and 1884b) included this species 
in his genus Acinus (synonym of Alvania according to 
Ponder, 1985), creating a link with the type species AL 
vania cimex (Linnaeus, 1758) (which is also the type 
species of Alvania, see Ponder, 1985: 36), mainly based 
on coloration. In reality, the similarity between the two 
species is limited to the color pattern of the shell and the 
nodular sculpture. Strong differences indicate that A. 
cingulata and A. cimex could be distant relatives, the 
latter having a more sturdy shell with a coarser sculpture 
(with subequal axial and spiral), and a proportionally 
smaller aperture provided with an internally denticulate 
outer lip. The concept of A. cingulata as expressed by 
Monterosato (1884b: p. 62) seems to be applicable to an 
interesting Alvania sp. (from the lower Pleistocene of 
southern Italy, Reggio Calabria, Musala, 19 shs, F95A, 
coll., Figures 27-29) rather than to the original descrip- 
tion by Philippi. The non- eo .. see the pau- 
cispiral protoconch in Figure 29) Alvania sp., illustrated 
here, showing only a mode srate aie ity with the teleo- 
conch sculpture of A. cingulata, has the characteristic 
general shape of A. cimex, with which it shares the ap- 
ertural features, including the denticulations on the inner 
part of the outer lip. Furthermore Monterosato himself 
determined the shells I have illustrated in Figures 15-23 
as A. cingulata. 

In 1968, Nordsieck cited A. cingulata as Cingula 
(Onoba) cingulata. In reality, a vague similarity with 
some Cingula Fleming, 1828 or (more specifically) 
Onoba H. and A. Adams, 1852, nee s exists: e.g. Onoba 
(O.) carpenteri (Weinkauff) (see Ponder, 1985: fig. 
1 14a), bears a similar spiral sculpture. Anyway, the genus 
Onoba is characterized by usually elongate shell a a 
proportionally smaller and more rounded aperture. Its 
included species, usually lacking the axial sculpture, may 
have a much weaker axial pattern or this may consist of 
delicate ribs, which become obsolete towards the sutures 
(see Bouchet and Warén, 1993: figs. 1508-1509, 1514— 
1515). In addition, the protoconch sculpture apparent on 
some non- planktotrophic Onoba species resembles that 
of A. cingulata, in having few, very fine spiral widely 
spaced threads. This land of sculpture is also seen in 
Alvania (e.g. A. subsoluta (Aradas, 1S47), see Bouchet 
and Warén, 1993, fig. 1458). Cingula species appear to 
be less similar to A. cingulata in having a thick outer lip 
and lacking any axial sculptural pattern. 

A certain similarity exists with Alvania watsoni (Wat- 
son, 1873) from Madeira (see Ponder, 1985: fig. 102c—d), 
with which A. cingulata shares the general shell shape, 
characterized by an inflated last whorl provided with a 
large, ovate aperture. 


Alvania francescoi new species 
(Figures 30-43) 


Description: — Shell small, sturdy, conical-ovate, reach- 
ing 4.8 mm in height (4.3 mm in holotype) and 3.1 mm 
in width (2.8 min in holotype). Protoconch conical, mul- 


tispiral with partially immersed nucleus and convex 


V. Garilli, 2008 


age ¢ 
Page 29 


whorls. Protoconch I consisting of about 0.8 whorls 
sculptured by 5-6 fine spiral lirae interspersed spirally 
and irregul: uly with microscopic granules. Protoconch I 
consisting of about 1.2 whorls sculptured by pimples, 
which are irregular in size and arrangement. They are 
larger and less numerous in the adapical position; fused 
into very short prosocline tracts in central part. In the 
ab: :pical part of late protoconch, groups of pimples are 
irregularly fused into fine, discontinuous spiral ridges. 
Protoconch/teleoconch transition well-marked and sinu- 
ous. Teleoconch consisting of 3.84.6 (4.2 in holotype) 
rather convex whorls, with strong cancellate sculpture 
formed by intersection of equally Oe veloped spiral cords 
and axial ribs. Cords are narrow and well raised, num- 
bering 2, 6-7 (6 in holotype) and 13-15 (14 in holotype) 
in early, penultimate and body whorl, respectively. Two 
ads :pical cords are usually more pronounced in penulti- 
mate and last whorl. Cords are markedly stronger and 
more widely spaced on base. Secondary cords may com- 
mence on late body whorl, at a certain distance from 
outer lip. Axial seuptite consists of narrow raised ribs 
numbering 24—32 (26 in holotype) on the penultimate 
whorl. On base, ribs progressively become very narrow to 
obsolete toward columellar area. Intersection of spiral 
and axial sculpture, producing nodular small knobs, 
forms a rectangular pattern (with major side of rect- 
angles perpendicular to shell axis) except on last whorl, 
where a quadrangular to rhomboidal pattern occurs (Fig- 
ure 43). Microsculpture of very early teleoconch whorls 
consists of micropustules, sometimes fused, for ming spi- 
rally, discontinuous and irregular rows, mainly occurring 
between spiral cords (Figure 42); weak growth lines are 
present. Fine sculpture of the rest of teleoconch limited 
to weak growth lines Sa aie 43). Suture slightly in- 
clined. Last whorl well-developed, comprising about 
three fourths of shell height. It may bear single (paratype 


6) or double (paratype 4, Figure 37) varices, at angles of 


about 50° and 70° respectively. Aperture wide, ovate, 
and with a rounded profile in the posterior part, com- 
prising about two thirds to three fourths of last whorl 
height. Outer lip orthocline, internally smooth, externally 
markedly thickened by a strong, sometimes doubled 
(Figure 31), rim, occurring very close to its edge and 
covered by spiral cords. Inner lip moderately arcuate and 
rather thickened, with thin callus c lelimitating very nar- 
row umbilical chink. 


Type Locality: Lower Pleistocene of Cartiera Mulino 
36°56'57" N, 14°34'03" E), Vittoria, Ragusa, SE Sicily. 


The stratotype is the phanerogams-rich 3D1 layer of 


Costa (1989). This deposit crops out at about one hun- 
dred meters from the abandoned paper-mill known as 
Cartiera Mulino. Paleoenvironmental and stratigraphic 
information about this site are given by Costa (1989). 


Type Material: Holotype (4.3 x 2.5 mm), DGUP 
CMRG 005/488, coll. V. Garilli. Paratype 1 (4.1 = 2.7 
mim), DGUP CMRG 006/489, coll. V. Garilli; paratype 2 
4.8 x 3mm), ZMA Moll. 4.07.007 (ex CM GR 007/490, 
coll. Garilli); paratype 3 (4 x 2.7 mm), ZMB MB.Ga.2479 


(ex CMGR 008/49, coll. V. Garilli); paratype 4 (4.2 « 2.9 
mm), MNHN A25950 (ex CMGR 009/492, coll. 
Garilli); paratype 5 (4.6 x 3.1 mm), DGUP CMGR 010/ 
493, coll. V. Garilli; paratype 6 (4.5 x 3.1 mm), DGUP 
CM GR 011/494, coll. V. Garilli. All type material is from 
the type locality, 3D1 layer. DATE? 


Etymology: The species honors the name of both 
Francesco Garilli senior, my father and first mentor in 
my life, and of Francesco Garilli junior, my son. 


Habitat: All the shells came from the 3D1 layer of 
Costa (1989). The paleoenviromental reconstruction for 
this layer was characterized by Costa (1989) as compa- 
rable with the Mediterranean marine-marginal modern 
ecotone HP (Posidonia beds)-SVMC (sensu Pérés and 
Picard, 1964). It is interesting that the 3D1 layer is quite 
rich in Posidonia remains mainly consisting of leaves. 


Distribution: This species is known from type locality 
only, lower Pleistocene of Cartiera Mulino, Vittoria, Ra- 
gusa, SE Sicily. 


Remarks: This quite characteristic species shows a ga- 
leodiniform shape, having a quite inflated and well- 
developed (also with double varix) last whorl and a large 
aperture provided with an intermally smooth outer lip. 
However, its spiral sculpture does not markedly prevail 
over the axial, as in the typical Alvania carinata. Its mi- 
crosculpture, limited to the early teleoconch whorls, is 
similar to that observed in the analogous whorls of A. 
carinata, but in the latter this microsculpture is distrib- 
uted over the entire teleoconch. 

The strongly cancellated share pattern of A, 


francescoi new species, consisting of well-raised, sub- 


equal spiral and axial elements forming marked nodular 
intersections, resembles that shown io A. cimicoides 
(Forbes, 1844). The latter species, showing a more typi- 
cal Alvania shape, has a more slender shell, a less inflated 
and much less developed last whorl, and a proportionally 
smaller aperture with denticulations on the inner part of 
the outer lip. 

Alvania francescoi new species can also be compared 
with A. rosariae new species described later in this re- 
port. Remarks on their similarities and differences will 
be dealt with in the section dedicated to the latter spe- 
cies. 


Alvania lactea (Michaud, 1830) 
(Figures 44-49 and 53-61) 


Rissoa lactea Michaud, 1830: 9-10, figs. 11-12 


Rissoa (Massotia) lactea Michaud.—Bucquoy et al., 1884: 298, 
pl. 25, figs. 7-13 
Massotia Daje »rleini Monterosato ex-Schwartz ms, 1889: : 


Massotia lactea (Michaud) forma Dajerleini aoe 
1917: 12. 


Description: Shell sturdy, conical-ovate to cylindrical- 
ovate, occasionally almost ‘pupoid, reaching 5.2 mm in 


YAUTILUS, 


V. Garilli, 2008 Page 31 


Figures 38-43. 


Alvania francescoi new species, protoconch, teleoconch sculpture and microsculpture, lower Pleistocene of 
Cartiera Mulino, SE Sicily, Ragusa, Vittoria. 38. Protoconch, paratype 2. 39. Protoconch, transition between protoconch | and. IT, 
holotype. 40. protoconch 1, holotype. 41. Sculpture of early teleoconch whorls, note how the spiral microsculpture becomes lacking, 
paratype 2. 42. Microsculpture on the first whorl, paratype 2. 43. Detail of sculpture on the last whorl, note the numerous growth 
lines, paratype 4. Scale bars: 200 xm in figures 41 and 43; 100 jzm in figures 38-39: 50 jm in figures 40 and 42. Black and white 


arrows indicate the protoconch I/protoconch H and protoconch/teleoconch boundaries, respectively 


height and 3.3 mm in width. Protoconch conical, consist- 
ing of about two convex whorls. Nucleus partially im- 
mersed. Protoconch I consisting of about 0.8 whorls, 
sculptured by 5-7 spiral lirae and inte rspersed with nu- 
merous closely packed granules, which are spirally and 
irregularly arranged. Protoconch II consisting of about 
1.2 whorls sculptured by pimples irregular in size and 
arrangement. Pimples are fused into short inclined lines 
opisthocline and prosocline, Figures 53-54) or form one 


to three discontinuous spiral ridges in abapical aspect of 


whorl. Protoconch/teleoconch transition well-marked 
and sinuous. Large shells consist of 4.2-5 weakly to 
rather convex whorls. These may be sculptured by a 


finely cancellate pattern formed by the intersection of 


spiral cords and numerous delicate, narrow axial ribs, or 
by opisthocline, pronounced ribs. Ribs often start from 
early whorls and are crossed by finer spiral threads. In 
cancellate shells, cords and ribs are equal to subequal 
and numbering up to nine and 40 respectively on pen- 
ultimate whorl. In ribbed shells, cords and ribs are less 


numerous, numbering up to eight and 16 respectively on 
penultimate whorl. In early teleoconch whorls micro- 
sculpture consists of fine, irregular, and discontinuous 
spiral threads formed by groups of small granules, not 
covering main spiral sculpture: remaining teleoconch 
with a finely reticulate ultrastructure cove ring all primary 
sculpture, formed by intersection of very narrow and 
raised lamella-like axial ridges and very thin spiral 
threads. Suture slightly inclined and de ‘eply impressed. 
Body whorl well-exp: anded, comprising 4 of shell height. 
It may bear a varix, usually with angle of LS0°; rarely a 
double varix m: ty occur, especially in slender and cancel- 
late shells. Aperture wide, ovate to almost pyriforme, 
rounded on its posterior aspect, narrowed anteriorly, 
comprising two thirds of last whorl height. Outer lip or- 
thocline, internally smooth, externally thickened, mainly 
somewhat behind its edge, and covered by primary spiral 
sculpture. Sometime, in ribbed morphs, the outer lip 
descends vertically, so that last whorl appears ri ither 
cylindrical. Inner lip weakly arcuate, with a thin to rather 


Figures 30-37. 


profile (34) and dorsal (35) view; 36-37. Pages 
whorl. Scale bars: 1 mm 


Alvania francescoi new species, holotype a two paratypes, lower Pleistocene of C: 
Ragusa, Vittoria. 30-32. Holotype. shell in apertural (30), profile 


tiera Mulino, SE Sicily 
and dorsal (32) view. 33-35. Piece shell in apertural (33 


apertural (36) and dorsal (37) view, the last showing a double varix on the last 


THE NAUTILUS, Vol. 122, No. 1] 


V. Garilli, 2008 Page 33 


Figures 53-61. Alvania lactea (Michaud, 1830), protoconch and teleoconch sculpture and microsculpture. 53-54. Profile and 
dorsal views of protoconch from the same shell as Figure 44, Italy, Tuscany, Siena, Poggibonsi, Villa Pietratitta, podere Melograni, 


lower Pliocene. coll. MF ex coll. PAL (F55E). 52. Protoconch illustrating variation of the abapical spiral ridges, Italy, Tuscany, Siena 
Poggibonsi,. Villa Pietrafitta podere Sant Uliviere, lower Pliocene, coll. MF ex coll. PAL (F55C), 56-57. Sculpture of pre stoconch | 
56) and early teleoconch whorls (57) of the same shell as figures 44 and 53-54. 58-61. Early (58) and very early (59) teleoconch 
sculpture, and microsculpture of first (60) and last whorls (61), from the same shell as Figure 42, Monastir-Khenis, coll. Garilli. Scale 
bars: 200 jm in figures 57-58 and: 100 wm in figures 53-55, 58, 61; 50 wm in Figure 60; 20 wm in Figure 56. Black and white arrows 
indicate the protoconch I/protoconch II and protoconch/teleoconch boundaries, respectively 


thick narrow callus, leaving a very small umbilical chink. calities, all belonging to the French Mediterranean coast 
Very fresh shells show a ferruginous periostracum. Shell “Aode, Cette (Hérault) [now Séte], Callioure, Port- 
color white Vendre (Pyrénées orientales)” (With this citation I con- 
Type Locality: Michaud (1531) cited the following lo- sider the second edition, consulted in the MNHN-DSE 


Figures 44-52. Shells of Alvania lactea (Michaud, 1830) and Alvania prusi (Fischer, 1877). 44-49. Alvania lactea. 44. Italy 
Tuscany, Siena, Poggibonsi, Villa Pietrafitta, Podere Melograni, lower Pliocene, coll. MF ex coll. PAL (F55E), cancellated morph 
45. Tunisia. Monastir-Khenis, coll. Garilli, ribbed morh (var. dajerleini Monterosato). 46. Italy, Piemonte, Asti province, Pliocene 
MGUP coll. Doderlein (113B) shell close to the original description by Michaud (1830). 47. Italy, Piemonte, Alessandria, Tortona 
Miocene (Tortonian), MGUP coll. Doderlein (476B). 48. Profile view of the same shell as Figure 41. 49. Juvenile shell, Mediter 
ranean France, Provence, Marseille, La Baule, coll. PAL, (212SBAU-VO0SC). 50-52. Syntype of Rissoa prusi Fischer, 1877 
Quaternary of Rhodes, MNHN (DHT) coll. D’Orbigny (RO7495 apertural 50 profile 51) and dorsal (52) views. Scale bars 1 mm 


Page 34 


THE NAUTILUS, Vol. 122, No. 1 


library, of Michaud’s work “Descriptions de plusieurs . . 
published f or the first time in 1830. The date 1831 may 
be doubtful, 1832 being the most commonly cited date; 
I prefer to follow Palazzi ( 2003), who provide od helpful 
reasons to choose the former date.) 


Material Examined: 37 shs, coll. H. Fischer, with no 
locality. Atlantic France: Normandy, St Aubin Calvados, 
2 shs, MNHN; Normandy, St.Vaast, | sh., MNHN; Brit- 
tany, Finistére, Anse de Dionan, under stones covered 
with sand, 75 shs, MNHN legit. S. Gofas, 1973-78; Brit- 
tany, St. Lunaire, 40 shs, coll. MNHN coll. Dollfus, 1903; 
St. Lunaire, 15 shs, MNHN coll. Fischer; Brittany, 
Penthiérre, 4 shs, MNHN, P. Bouchet legit; Penthiérre 
(Morbihan), under stones covered with sand, low tide, 10 
shs, MNHWN coll. P. Bouchet; Penthiérre, 10 shs, 
MNHN, 27 Apr.1975; Aquitaine, Cote Basque, Hen- 
daye, conchiferous detritus, beach, 5 shs, MNHN coll. S 
sofas, 1981; Cdte Basque, Hendaye, infralittoral rocks, 3 
shs, MNHN coll. S. Gofas, 1980-S1; Céte Basque, Ond- 
arroa, infralittoral rocks, 1 sh., MNHWN coll. S. Gofas, 
1980-81; Cote Basque, St. Sebastian, infralittoral rocks, 
1 sh., MNHN coll. S. Gofas, 1980-81; Cdte Basque, St 
Jean de Luz, infralittoral rocks, 57 shs, MNHN coll. S. 
Gofas, 1980-81; St Jean de Luz, outside Cape Ste Barbe, 
tide zone, stones covered with sand, 4 shs, MNHN coll. 
S. Gofas, Dec. 1988; St Jean de Luz, outside cape Ste 
Barbe, tide zone, stones covered with sand, 49 shs, 
MNHN coll. S. Gofas, 1989; St Jean de Luz, 54 shs, 
MNHN coll. H. Fischer, 1898: be ee Guéthary, 7 
shs, MNHN coll. H. Fischer, ee Portugal: Algarve 
Sagres, Pontal dos Corvos, (37°01.3' N, 08°58.3' W), foot 
of falaise, 17-22 m, 1 sh., MNHN Micsone Algarve, O05. 
1988; Algarve Sagres, Baie de Baleeira, (37°00.7' N, 
08°55.0' W), tide zone, 1 sh., MNHN Mission Algarve, 
May 1988. Ailaniic Spain: Cantabria, Orifion prov. 
Santander, Punta de Sonabia, infralittoral rocks, 1 sh., 
MNHN coll. S. Gofas, May 1989; Asturias, Muros prov. 
Oviedo, playa de la Liana, infralittoral rocks, 2. shs, 
MNHN coll. J. Ortea-S. Gofas 08.89; Cadiz, Barbate, 
(36°10.9' N, 05°56.9' W), tide zone, infralittoral rocks, 4 
shs, MNHN réc. S. Gofas Apr. 1994; Cadiz, Chiclana, 
(36°22.5' N, 06°12.5' W), tide zone, infralittoral rocks 
and sands, 12 shs, MNHN réc. S. Gofas, Apr. 1994; 
Cadiz, Barbate, conchiferous detritus, beach, 3. shs, 
MNHN coll. S. Gofas, 1976-81; Mediterranean Spain: 
Malaga, Calahonda, conchiferous detritus, beach, 9 shs 
MNHN coll. S. Gofas, 1976-81; Malaga, Port de Mar- 
bella, conchiferous detritus, beach, 3 shs, MNHN coll. S. 
Gofas, 1978-81; Malaga, Benalmadena-Costa, conchifer- 
ous detritus, beach, 3 shs, MNHWN réc. S. Gofas, 1991— 
93. Atlantic Morocco: E] Jadida, (33°16' N, 08°29’ W), 
large beach, rocky platform, tide zone, 16 shs, MNHN 
réc. S. Gotas, 26 Sep. 1991; Fedala, Mannesmann beach, 
conchiferous detritus, beach, 20 shs, MNHWN coll. S. Go- 
fas, L970—72: Asilah, mouth of Oued el Helou, conchif- 
erous detritus, beach, 21 shs, MNHWN coll. S. Gofas, 
1971-72: Asilah, mouth of Oued el Helou, conchiferous 
detritus, beach, 30 shs MNHN coll. S. Gofas, 1972-80: 


Temara, (33°55' N, 07°00’ W), Sables @’Or beach, rocks 
and mud, 0-2 m, 24 shs, MNHN, MA4S, réc. S. Gofas, 
17 Sep.1991; Essaouira (formerly Mogador), (31°31' N, 
09°47' W), rocky platform, tide zone, 1 sh., MNHN, 
MAAS, réc. S. Gofas, 23 Sep.1991; Rabat, Lahlou, 
(34°02' N, 06°51’ W), conchiferous detritus, beach, 2 shs, 
MNHN réc. S. Gofas, 28 Sep. 1991; Essaouira (formerly 
Mogador), 4 shs, MNHN; Strait of Gibraltar, Morocco: 
Tanger, Grande Plage, conchiferous detritus, beach, 8 
shs, MNHWN coll. S. Gofas, 1970-81. Strait of Gibraltar, 
Spain: Ceuta Nord, Benzu, infralittoral rocks, 2 shs, 
MNHN coll. S. Gofas 1976-1981. Algeria: Oran, 6 shs, 
MNHN coll. Locard. Mediterranean France: Langue- 
doc, Roussilion, Banyuls sur Mer, near the beach de 
Paulilles, infralittoral rocks, 1 sh., MNHN coll. Bouchet 
and Gofas, Sep. 1980; Languedoc, Roussilion, 7 shs, 
MNHIN coll. Ph. Dautzenberg, (figured in Moll. Rouss. 
T. I pl. 35, figs.7-13); Languedoc, Roussilion, Séte, 6 shs, 
MNHN coll. Locard; Provence, Toulon, 8 shs, MNHN 
coll. Petit; Provence, Cannes, 21 shs, MNHN coll. Doll- 
fus, 1903; Provence, east coast, Iles Embiez, passe du 
Gaou, under stones covered with sand, 0-1 m, 5 shs, 
MNHN réc. S. Gofas, Aug. 1988; Provence, east coast, 
Iles Embiez, (43°04.3' N, 5°47.4" E), pase du Gaou, 
under stones covered with sand, 0-3 m, 2 shs, MNHN 
réc. S. Gofas, Aug. 1988; Provence, Iles Embiez, 
(43°04.3' N, 5°47.4' E) passe du Gaou, rocks, photophile 
algae under stones covered with sand, 0-3 m, 1 sh., 
MNHN réc. S. Gofas, Jun. 1995; Hes Embiez, cote Nord 
et Petit Rouveau, infralittoral rocks, 3 shs, MNHN coll. 
S. Gofas, 1968-70: Provence, Les Embiez, cote Nord et 
Petit Rouveau, (43°05' N, 5°47’ E), rocks, photophile 
algae, 0-1 m, 3 shs, MNHN réc. S. Gofas, Jun. 1995; Iles 
Embiez, conchiferous detritus, beach, 15 shs, MNHN 
coll. S. Gofas, 1968-70; Provence, St Clair, (43°08.2’ N, 
6°23.2' E), infralittoral rocks, 0-1 m, 2 shs, MNHN réc. 
S. Gofas, Sep. 1992; Provence, Porquerolles plage Notre 
Dame, (43°00.6' N, 6°13.8' E), rocks, 0-1 m, 1 sh., 
MNHN réc. S. Gofas, Sep. 1992; Provence, Le Dramont, 
conchiferous sand, beach, 3 shs, MNHN réc. J. Pelorce, 
1992: Provence, Marseille (Endoume, 43°16.9' N, 
05°21.0' W), littoral rocks, 1 sh., MNHWN réc. S. Gofas 
Apr. 1995; Provence, St. Raphael, 5 shs, sata 
Provence, Bandol, 16 shs, MNHN coll. Locard: 
Raphael, 2 shs, MNHN coll, Locard; Corsica, Ajaccio, VW 
shs, MNHN coll. Jousseaume; Corsica, Algajole, 35 shs, 
coll. MF, 2123. Tunisia: Monastir-Khenis, beach, 1 sh., 
coll. Garilli, legit Garilli and Galletti 4/2000; ae 
beach, 1 sh., coll. MF ex coll. C. Bogi, Jun. 1981, 
2163JER. Italy: Piemonte, Asti, Bi dichieri, “Grottino 
Monale”, 20 shs, yellow sands, middle-upper Pliocene, 
coll. MF ex coll. PAL, F1O4A:; Piemonte, Asti, 24 shs, 
Pliocene, MGUP coll. Doderlein, 113B; Piemonte, 
Alessandria, Tortona, | sh., Miocene, MGUP coll. 
Doderlein, 476B; Alessandria, Villalvernia, at the Cem- 
etery, 1 sh., Astian yellow sands, middle-upper Pliocene, 
coll. MF ex coll. PAL, FLISA: Emilia Romagna, Modena, 
Maranello, Fogliano, Gagliardella, Rio Grizzaga sands, 1 


V. Garilli, 2008 


icy) 
Ol 


Page « 


sh., middle Pliocene, coll. MF ex coll. PAL, F39A: Tus- 
cany, Siena, ne: ea Villa Pietratitta, Podere “La 
Vigna’, (SP 36, 4.9 km E side), 7 shs, lower Pliocene, 
coll. MF ex ot PAL, F55B; Siena, Poggibonsi, Villa 
Pietratitta, “Sbarra”, (SP 36, 5.2 km), 2 shs, sands, lower 
Pliocene, coll. MF ex coll. PAL, oo Siena, Castel- 

nuovo Berardenga, Terre Rosse, ( 104 km), 1 sh., 

sands, lower Pliocene, coll. oe ex call. PAL, F36A; 
Siena, Poggibonsi, Villa Pietrafitta, Podere Sant’ Uliviere, 
27 shs, lower Pliocene, coll. MF ex coll. PAL, F55C: 
Siena, Poggibonsi, Villa Pietratitta, Podere Melograni, 6 
shs, sands, lower Pliocene, coll. MF ex coll. PAL, F55E: 
Siena, Colle Val d’Elsa, Bibbiano, 2 shs, yellow sands, 
lower Pliocene, coll. MF ex coll. PAL, FLOSA; Tuseany, 
Livorno, Tuscan Archipelago, Island of Elba, Procchio, 
12 m, Posidonia bed, 2 shs, coll. MF ex coll. C. Bogi, 
2120PROC: Pai Terni, Ficulle, quarry near Chiani 
river, (SST1), 1 sh., Cidaris marly sands, lower Pliocene, 
coll. MF ex coll. PAL, F6SA; Sardinia, Nuoro, San Te- 
odoro, 6 shs coll. PAL; Puglia, Taranto, MSNCS 44744 
(ex 1505), 1974, BDA legit and det; Sicily, Messina, Mi- 
lazzo, Capo Milazzo, Cala S. Antonio, 12 shs, upper yel- 

low sands, upper Pleistocene, coll. MF ex coll. PAL, FSA: 
Sicily, Catania, Grammichele, C.da Catallarga, 2  shs, 
coarse sands, lower Pleistocene, coll. MF ex coll. PAL, 
F27A; Sicily, Palermo, 1 sh., MNHN coll. Petit: Sicily, 
Palermo, 2 shs, MNHN coll. Dollfus, 1903; Palermo, 
Tommaso Natale, 13 shs, upper Pleistocene, MGUP 167/ 
2/50; Palermo, Addaura, 1 sh., upper Pleistocene, 
MGUP 587/5/15; Sicily, Catania, 2 shs, MSNCS 44743 
(ex 1448), 4/1974, G. Gentile legit and det.; Sicily, Sira- 
cusa, Vendicari, 1 sh., coll. PAL, 2124. Adriatic Sea: 128 
shs, MNHN. Croatia: Zara, 8 shs, MNHN coll. Petit, 
1873. Greece: Evvia Island (Euboea), Loukissia, 4-5 m, 
1 sh., coll. PAL ex coll. Bogi Cesare, 2121LUC. Israel: 
Haifa, 9 m, 1 sh., coll. PAL ex coll. C. Bogi, 1994, 
2164HA. 


Habitat: As indicated by Jeffreys (1867), Gofas and 
Ponder (1991). and Bouchet (1978: 1992), this species 
typically lives buried under stones covered with sand at 
very shallow waters. I found very fresh shells collected 
from along all the upper part of the infralittoral stage. 


Distribution: All the Mediterranean, probably de- 
creasing is a in eastward direction. In the eastern 
Atlantic it lives from the British Islands (see also Jeffreys, 
1867: 1869) to Morocco. Alvania lactea has also been 
recorded from the Black Sea (Anistratenko and Star- 
obogatov, 1994). This species probably originated in the 
Mediterranean Neogene, from where it is recorded from 
the Miocene of north Italy. It becomes more common in 
the western and central Mediterranean Pliocene depos- 
its, where it is recorded from several localities of north, 
central. and south Italy (material herein studied: see also 
Sacco, 1895, and Chirli, 2006) and Spain (Estepona, Lan- 
dau et al., 2004). In the Atlantic Pleistocene, as reported 
by Landau et al. (2004), it is recorded from the post- 
glacial of Iceland. the North Sea Basin and the British 
Isles. As subfossil, it is recorded from Sweden (Huben- 


dick and Warén, 1969). In Mediterranean, it is common 
from the lower-upper Pleistocene of Sicily and is also 
recorded from the lower Pleistocene of Tusc: ny (Chirli, 
2006). 


Remarks: This is a very variable species especially 
with respect to sculpture and shell shape. The typical 
morph, as described and figured by Michaud (1831: fig. 
12), has a clathrate sculpture with the axial pattern domi- 
nating over the spiral one. The finely cancellated type, 

usually characterized by an elongate general shape and 
more convex whorls, corresponds to the forma minor- 
tenuisc ulpta Monterosato, 1917. Two varieties, semiacos- 
tata and fusulatovaricosa (the latter often bearing va- 
rices), were described by Sacco (1895) on similar mate- 
rial from the Pliocene of north Italy. This cancellate 
morph (Figures 44 and 47-48), ), which seems to be the 
only representation of the discussed species in the Mio- 
cene, is the most common in the Pliocene collections 
studied (see also Sacco, 1895: p. 28; Chirli, 2006: figs. 
9-11 and 15-16) and becomes less common in Quater- 
nary material. The markedly ribbed morph, descsbed by 
Sacco (1895) as var. laticostata (from the Pliocene of 
Italy), and better known as forma dajerleini Monterosato, 
1889, bears strong ribs, starting from the early teleo- 
conch whorls. According to Monterosato (1917) ), this re p- 
resents the Atlantic morph of Alvania lactea. 1 agree with 
Van Aatsen et al., considering it not exclusively an Atlan- 
tic morph, being present in some Mediterranean locali- 
ties (see Figure 45). It is noteworthy to remember that in 
a extensive lot (128 shells, MNHN-DSE) from the Adri- 
atic Sea, all the above mentioned morphs coexist. 

Especially when its typical morphs are considered, Al- 
vania lactea has a very characteristic ribbed, cylindrical- 
ovate shell and can not be confused with its congeners. 
Some problem might occur when considering the can- 
cellate morph, winch can be compared to ‘he recently 
described Pliocene species A. fredianii Della Bella and 
Scarponi, 2000 (see this article for the main differences 
between the two species). The same morph of A. lactea 
shows strong similarities with the new species described 
herein, A. rosariae (see discussion below for differential 
diagnosis). 

Alvania lactea is the type species of Massotia Bucquoy 
et al., 1884, which is considered a synonym of Alvania 
sensu stricto by Ponder (1985) on the basis of the num- 
ber of the metapodial tentacles, shown by this species, 
based on a description by Jeffreys (1867). In my opinion, 
a more exhaustive anatomical dataset could be provided. 
Nevertheless, even on the basis of its shell features and 
intraspecific variation, there is no reason to consider 
Massotia as a well-supported group. 

Alvania lactea has a very complicated teleoconch mi- 
ila ee which could ce srtainly be considered as a 
suite of distinctive characters, not observed in the 
closely-related taxa discussed in the present report. Early 
teleoconch whorls show a pattern similar to that ob- 
served in A. carinata, A. francescoi new species, and A, 
rosariae new species (detailed comparisons are provided 


Page 36 


THE NAUTILUS, Vol. 122, No. 1 


below), but the remainder of the shell is covered by very 
narrow, raised lamella-like axial ridges and very thin spi- 
ral threads. Also its protoconch I sculpture is rather dis- 
tinctive, showing a pattern similar to that observed in A. 
carinata, A. francescoi, A. rosariae, and A. tenera (as well 
as in many other rissoids), but having a coarser ornamen- 
tation. 


Alvania prusi (P. Fischer, 1877) 
(Figures 50-52) 


Rissoa (Alvania) Prusi P. Fischer, 1877: 80. 


Description: Shell small, Saaney conical, partially 

worn off, 4.7 mim in height, 3.2 mim in width. Protoconch 
multispiral, conical, consisting of about little more than 
two convex whorls. ‘Teleoeonel is formed by about four 
convex whorls, provided with a very delicate, sida ali 
sculpture almost giving a pitted shape to shell surface. 
This sculpture consists of moderately pronounced spiral 
cords crossed by apparently flat axial ribs. Spiral cords 
number 7-8 on penultimate whorl and 16 on body whorl, 
they are flatter and wider on shell base. Ribs, numbe ring 
about 40 on penultimate whorl, are lacking in basal part 
of last whorl. Sutures slightly inclined and deeply im- 
pressed. Last whorl well- expanded, rather angulated at 
the base and inflated, comprising % of shell height. 
Aperture wide, ovate to almost pyriforme, pointed adapi- 
cally, comprising 2 of the total height, 7% of last whorl 
height. Outer lip prosocline, internally smooth, externally 
with 1 an almost flat, wide thickening, apparently smoc th. 
Inner lip moderately arcuate, with a modest and narrow 
reflection on columellar area, leavi ing a very small and 
narrow umbilical chink (filled up with sediment). 


Type Locality: Isle of Rhodes 


Type Material: One syntype, RO7495 in MNHN- 
DHT, from type locality. 


Material Examined: Eastern Mediterranean Sea, 
Quaternary of Rhodes, type locality, MNHN-DHT 
(RO7495), 1 sh. 


Habitat: It is very difficult to characterize the paleo- 
ecological significance of this extinct species since no 
data about its paleoe nvironment are direc tly available or 
deducible. Furthermore, Fischer (1877) ) provide -d no 
precise indication about the site, stratotype, and its sedi- 
mentological and paleontological nature. Considering all 
the other molluscan species described from Rhodes by 
the same author, a shallow water depositional environ- 
ment (linked to the modern phanerogam infralittoral 
bottom) can be inferred. 


Distribution: The species is known only from the type 
locality, and its distribution appears to be limited to the 
Quaternary of Rhodes. 


Remarks: 


knowledge, the most recent treatment is that of Monte- 


This taxon is practically unknown. To my 


rosato (1917), who considered it as a separate species 
belonging to the Massotia group. 

Alvania prusi could be confused with juvenile shells of 
Alvania lactea (the form with cancellate sculpture, see 
Figure 49), which exhibit a very similar body whorl pro- 
file. Alvania prusi is quite close to A. fredianii Della 
Bella and Scarponi (2000) from the Pliocene of Tuscany. 
Both species share the general shell shape and the finely 
reticulated sculpture, tra the latter is of smaller size, has 
a paucispiral protoconch, deeper sutures, and a weakly 
denticulated outer lip (see Della Bella and Scarponi, 
2000: pls. 1 and 2). 


Alvania rosariae new species 
(Figures 62-79) 


Description: Shell small, sturdy, conical-ovate to tur- 
riform, moderately to markedly inflated, reaching about 
5 mm in height, 3.5 mm in width (holotype 4 mm in 
height and 2.7 mm in width). Protoconch multispiral, 
conical, consisting of about 2—2.1 convex whorls. Proto- 
conch I of about 0.8 whorls sculptured by 5-6 very fine 
spiral lirae irregularly interspersed with microscopic 
granules. Protocanch/teleaconch transition well marked 
aad sinuous adapically. Protoconch I sculptured by spi- 
rally arranged microscopic pimples (stronger in adapical 
portion) for ming one to two spiral threads, the lower very 
close to the suture, occurring on last whorl. Teleoconch 
formed by about 4.2 usually very convex w horls, sculp- 
tured by numerous axial ribs and slightly stronger spiral 
cords. The latter, rapidly increasing in count, number 
2-6, 6-10 ( se 11), and 26-34 (in specimens higher 
than 3.5 mm) on the first, penultimate, and last whorl of 
the teleoconch, respectively. Some secondary, less 
marked cords may occur on last whorl, usually close to 
the outer lip. T The adapical one, two, and three spiral 
cords on first, penultimate and body whorl, respectiv ely 
are more pronounced. In specimens higher than 3.5 mm, 
axial sculpture consists of 44-60 narrow ribs (52 in ho- 
lotype) on the penultimate whorl. Ribs become narrower 
and lamella-like to obsolete or lacking on shell base, par- 
ticularly close to columella. The same may occur on the 
terminal portion of the body whorl (Figure 66). The in- 
tersection of spiral cords w ith axial ae gives a charac- 
teristic cancellate and gently nodular shape, forming a 
rectangular (almost equilate ral in the central portion of 
the body whorl) pattern. The long axis of these rectangles 
is perpendicular to the shell axis on the central and 
abapical portion of whorls and parallel on upper part. 
Microsculpture consists of ve ry fine, sometime irregu- 
larly interspersed, spiral lirae ( Figure 7 77). These become 
obsolete on the pe ia ite and iboay a where nu- 
merous, very fine and narrow ribs ( (possible growth lines) 
occur. Lirae are continuous only on ade ipical portion of 
early teleoconch whorls and never cover primary spiral 
sculpture. Sutures slightly inclined and rather deeply 
Be ssed. Last whorl well-developed, comprising about 

; to % of the shell height, usually with a very convex 
iofile. It sometime bears one or two close varices which 


V. Garilli, 2008 


may form angles of 10° to 210°. Aperture HS ovate, 
comprising 2 to % of the total height; %% ¥4 of last 
whorl height. Outer lip orthocline (slightly eee in- 
ternally smooth, externally markedly thickened close to 
lip edge and covered by spiral cords. Inner lip moder- 
ately arcuated and rather thickened in the columellar 
area, where a very narrow umbilical chink occurs. 


Type Locality: — Lower Pleistocene of Cartiera Mulino 
(36°56'57" N, 14°34'03" E), Vittoria, Ragusa, southeast- 
ern Sicily. The stratotype is the 3D1 layer of Costa 
(1989). 


Type Material: Holotype (4.0 x 2.7 mm), DGUP 
CMRG 12/496, coll. Garilli; Paratype 1 (4.6 x 3, 2 mm), 
ZMA Moll. 4.07.014 (ex CMRG 13/497 coll. Garilli): 
paratype 2 (not-complete shell, 3 mm width), ZMA Moll. 
4.07.08 (ex CMRG 014/495 coll. Garilli); paratype 3 (3.6 
x 2.55 mm), DGUP CMRG 15/499, coll. V. Garilli; 
paratype 4 (4.5 x 3.3 mm), DGUP CMRG 16/500, coll. 
Garilli: paratype 5 (3.8 x 2.5 mm), MNHN A25951 (ex 
CMRG 17/501, coll. Garilli); paratype 6 (3.85 x 2.75 
mm), ZMB MB.Ga.2480 (ex CMRG 18/502, coll. 
Garilli); paratype 7 (3.6 x 2.4 mm), ZMB MB.Ga.2451 
(ex CMRG 19/503, coll. Garilli); paratype 5 (3.8 x 2.6 
mm), DGUP CMRG 20/504, coll. Garilli; paratype 9 (3.6 
x ¢ 6 mm), DGUP CMRG 21/505, coll. Garilli; paratype 
10 (3.95 x 2 2 6 mm), DGUP CMRG 22/506, coll. Garilli: 
ree 11 (4.2 «x 2.75 mm), DGUP CMRG 23/507, 
coll. Garilli; paratype 12 (3.7 x 2.55 mm), DGUP CMRG 
24/508, coll. Garilli; paratype 13 (3.9 x 2.5 mm), DGUP 
CMRG 25/509, coll. Garilli; paratype 14 (3.7 « 2.55 mm), 
DGUP CMRG 26/510, coll. Garilli; paratype 15 (3.5 x 
2.6 mm), DGUP CMRG 27/511, coll. Garilli; paratype 16 
(4 x 2.6 mm), DGUP CMRG 28/512, coll. Garilli; 
paratype 17 (3.9 x 2.7 mm), DGUP CMRG 29/513, coll. 
Garilli; paratype 15 (3.6 x 2.45 mm), GNHM ID 30.706 
(ex KIGR 3/514 coll. Garilli); paratype 19 (4 x 2.7 mm), 
GNHM ID 30.707 (ex KIGR 4/515 coll. Garilli); 
paratype 20 (4.5 x 2.7 mm), GNHM ID 30.708 (ex KIGR 
35/516 coll. Garilli); paratype 21 (4.7 x 3.4 mm), GNHM 
ID 30.709 (ex KIGR 6/517 coll. Garilli); paratype 22 (4.4 
x 29 mm), GNHM ID 30.710 (ex KIGR 7/518 coll. 
Garilli); paratype 23 (4 x 2.55 mm), GNHM ID 30.711 
(ex KIGR 8/519 coll. Garilli); paratype 24 (3.35 x 2.5 
mm), GNHM ID 30.712, (ex KIGR 9/520 coll. Garilli): 
paratype 25 (2.9 x 1.95 mm), GNHM ID 30.713 (ex 
ca 10/521 coll. Garilli); paratype 26 (not-complete 
shell ), GNHM ID 30.714 (ex KIGR 11/522 coll. Garilli): 
paratype 27 (4.15 x 2.65 mm), GNHM ID 30.715 (ex 
KIGR 12/523 coll. Garilli): ): paratype 28 (4.0 x 2.6 mm), 
GNHM ID 30.716 (ex KIGR 13/524 coll. Garilli); 

paratype 29 (not- complet e shell, 3.85 mm), GNHM ID 
30.717 (ex KIGR 14/525 coll. Garilli); paratype 30 (2.7 x 
1.9 mm), GNHM ID 30.718 (ex KIGR 15/526 coll. 
Garilli); paratype 31 (3 x 2.05 mm), GNHM ID 30.719 
ex KIGR 16/527 coll. Garilli); paratype 32 (3.5 « 2.5 
mm), GNHM ID 30.720 (ex KIGR 17/528 coll. Garilli): 
paratype 33 (3.55 x 2.5 mm), GNHM ID 30.721 (ex 


KIGR 18/529 coll. Garilli); paratype 34 (4.05 x 2.5 mm), 
GNHM ID _ 30.722 (ex KIGR 19/530 coll. Garilli); 
paratype 35 (4.95 « 3.25 mm), GNHM ID 30.723 (ex 
KIGR 20/531 coll. Garilli); paratype 36 (not ene: ), 
GNHM ID 30.724 (ex KIGR 21/532, coll. Garilli); 
paratype 37 (not measured), GNHM ID 30.725 (ex 
KIGR 22/533, coll. Garilli). Holotype and Paratypes 
4-17, from the lower Pleistocene of Cartiera Mulino, 
3D1 bed of Costa (1989), Vittoria, Ragusa, southeastern 
Sicily. Paratypes 1-3 from the same locality, 3D2 bed of 
Costa (1989). Paratypes 15-22 and 24-35 from the 
middle to upper Pleistocene of Kyllini, northwestern 
Pelopomnesus, N2 and H6 beds of Garilli et al. (2005a), 
respectively. Paratypes 36 and 37, same locality, from the 
lower to early middle Picistoesie P3 layer of Garilli 
(2005b) and from a late lower Pleistocene yellowish to 
reddish sandy layer about 50 m underlying the F14 bed 
of Garilli and Galletti (2007), respectiv ely, 

Another four, eee d paratypes are housed in 
DSTC (1 sh. from 3D1, 1 hi from 3D2 and : hae: from 
3C bed of Costa, 1989, all from the lot n° 18, as Galeo- 
dina carinata (Da Costa). One more fore pene 
paratype (ex CMRG 030/534 coll. Garilli), from the type 
locality, 3D1 layer, is in coll. MF (Prato), 


Etymology: 


saria. 


The species is dedicated to my wife Ro- 


Material Examined: The type material from the 
lower Pleistocene of Cartiera Mulino, Vittoria, Ragusa, 
SE Sicily, 18 shs, and from the late lower (1 sh., froma 
reddish to yellowish sandy bed about 50 m underlying 
the Fl4 bed ie Garilli and Galletti, 2007), lower to 
middle (2 shs, layer P3 of Garilli et al, 2005b). and 
middle to upper Pleistocene (13 shs, layer H6: 4 shs, 
layer N2) of Kyllini, Elea, NW Peloponnesus, Greece. 


Habitat: In the Do locality, the species was mainly 
recovered from the layers 3D1 and 3D2 which were 
linked to the ecotone SVMC-HP (sensu Pérés and Pi- 
card, 1964) by Costa (1989). In the Kyllini sites, this 
species was found in cerithids-trochids-rissoids assem- 
blage linked to the present biocenosis HP, characterized 
by the phanerogam Posidonia oceanica (Linnaeus) De- 
lile, 1813 (See Garilli et al. (2005a): Garilli et al. (2005b), 

and Garilli and Galletti (2007) for more detailed infor- 
mation about the paleoecological characteristics of the 
cited Kyllini strata.) 


Distribution: The species has a lower to middle-upper 
Pleistocene stratigraphic range, presently limited to SE 
Sicily and NW Peloponnesus, being recorded from the 
lower Pleistocene of Sicily (type locality) and from the 
lower to upper Pleistocene deposits of Kyllini, Greece 
(NW Peloponnesus). 


Remarks: Alvania rosariae represents a sort of inter- 
mediate form between A. carinata and A. lactea (cancel- 
late form). Compared with the former taxon, it shows 
some similarities in the type of intraspecific variation, 
showing very inflated to quite elongate shells (compare 


THE NAUTILUS, Vol. 122, No. | 


V. Garilli, 2008 


Page 39 


Figures 68-69 Res the “form” ecarinata of A. carinata, 
Figures l, 4, in usually bearing varices, and in the 
ch eee erie o the ¢ rarly teleoconch. In addition, their 
respective larval shells do not show relevant differences. 
Alvania rosariae new spe cies can be distinguished by 
absence of a keeled shape, having subequal and more 
numerous spiral and axial sculptural elements. The can- 
cellate form of A. lactea is comparable with A. rosariae 
new species, having a very similar sculpture and occa- 
sionally a similar shell shape (only in the very ovate 
morph, e.g. Figures 44, 47 and 65-66). However, A. ro- 
sariae new species differs by having more convex whorls 
and a very different microsculptural pattern, with only 
irregular, fine, often inte rrupted spiral threads covering 
the early teleoconch whorls. Furthermore, the proto- 
conch I of A. lactea bears a coarser sculpture, consisting 
of more numeorus and larger pimples and much more 
elevated spiral lirae. 

Alvania rosariae new species can also be compared 
with A. francescoi new species: both species have a ga- 
leodiniform shape, a similar microsculpture, and share a 
very similar sculptural pattern on the protoconch. Dif- 
ferences between the two species are mainly found in the 
teleoconch sculpture, which is strongly nodular and 
coarser in A. francescoi new species i a eee the 
latter has fewer spiral cords and axial fice. 

Alvania rosariae new species may resemble A. magis- 
tra Chirli, 2006, an interesting galeodiniform species 
from the Pliocene of N Italy and § Spain (Chirli, 2006: pl. 
. figs.13-16 and pl. 12, figs, 1- 8; Landau et al., 2004: pl. 

figs.3 B3a-e, as Alvania sp.), showing a quite similar 
ae cancellate sculpture, especially in the arrangement 
of the ad: ipical cords, and a quite wide, ovate aperture. 
However, the latter species shows an unusual outer lip 
profile, having a wide sinus at the upper aspect, its pro- 
toconch I shows a netted Manzonia-like sculpture, while 
protoconch II appears less sculptured than in A. rosariae 
new species 


Alvania tenera (Philippi, 1544) 
(Figures SO-S9) 


Rissoa tenera Philippi, 1544: 128-129, pl. 23, fig. 15. 
Galeodina tenera (Philippi, 1544) Pk ini, 1979: 71, 
fig. 4 


Description: Shell minute, conical, and subcarinate, 
to conical-ovate or elongate, moderately to markedly 
sturdy, reaching 2.2 mm in height and 1.4 mm in width. 
Protoconch multispiral, conical: with 2-2.3 convex 
whorls. Protoconch I consists of about 0.8 whorl, sculp- 
tured by six very thin spiral lirae and a few microscopic 
or anules between them. Protoconch I is sculptured by a 


few to abundant very small granules. These are more 
numerous on last half whorl, where they are spirally ar- 
ranged, forming very discontinuous and. irregular ridges. 
Fiotoconely te Tesconel transition clistinct, eh a slight 
(to very slight) sinuosity. Teleoconch consists of 3-3.5 
weakly to discretely convex whorls, quite variable in 
width. These are sculptured by well-raised, narrow, spi- 
ral, occasionally almost keel- like cords (in quite conical 
shells), which overide the axial sculpture. They number 
34, me 5, and 12- - on first, penultimate, and body 
whorl, respectively. Usually, weaker cords are present on 
shell base or close to upper suture. Axial sculpture 
formed by very narrow, prosocline, rows of short seg- 
ments, forming discontinous ribs, occurring between spi- 
ral cords and becoming very thin to lacking toward shell 
base. Ribs number 40-60 on last whorl. At the intersec- 
tions with axial elements, spiral cords generally appear 
very finely nodular (Figure 86). Last whorl well- 
expanded, with a rather rounded profile, sometimes in- 
flated, comprising about 73 to ‘4 (in subcarinate and 
conical shells) of total shell height. Aperture ovate, an- 
teriorly rounded, posteriorly angulated, comprising 
about 2 and % of total shell and last whorl height, re- 
spectively. Outer lip rounded, markedly prosocline, thin, 
internally smooth, externally with no thickening. Inner 
lip weakly arcuate, with a thin and narrow (Fi igure 82) to 
stronger and wider (Figure 80) callus, leaving a very nar- 
row senabilical chink. ¢ Solokilion usually consists of red- 
dish to brown spots on a cream-whitish or (rarely) 
brownish background, ~ 


Type Locality: Peninsula of Magnisi (originally indi- 
cated as “Peninsula Thapsum” from the old Greek name 
of Magnisi), Siracusa, southeastern Sicily. 


Material Examined: Atlantic: Canary Islands, Tene- 
rife, Pal-Mar, 6—S m, 1 sh., MNHN coll. P. Bouchet, 15 
Jul. 1980; Morocco: Asilah, mouth of Oued el Helou, 
conchiferous detritus, beach, 9 shs, MNHWN coll. S. Go- 
fas, 1971-72; El Jadida, (33°16" N, 08°29’ W), large 
beach, conchiferous detritus, beach, 5 shs, - NHN réc S. 
Gofas, 26 Sep. 1991; El Jadida, ( (33°16’ N, 08°29’ W), 

large beach, tide zone, 4 shs, MNHN réc S. Gofas, 26 
Sep.1991. Strait of Gibraltar: Spain, Cadiz, conchiferous 
detritus, beach, 2 shs, MNHWN coll. S. Gofas, 1976-81; 
Cadiz, Tarifa, beach, 4 shs, coll. PAL ex coll. C. Bogi, 
Jun. 1986, 2030TAR; Cadiz, Tarifa, Torre de la Pena, 
conchiferous detritus, beach, 3 shs, MNHWN coll. S. Go- 
fas, Aug. L981; Morocco, Tanger, Grande Plage, conchif- 
erous detritus, beach, | sh., MNHWN coll. S$. Gofas, L970— 
S81. Mediterranean: Spain, Andalusia, Punta della Mona, 
43 m, 5 shs, coll. PAL ex coll. C. Bogi, 2032PMO; Spain, 
Malaga, industrial dredging, 20-40 m, 1 sh., MNHN réc. 


Figures 62-69. 


Alvania rosariae new species, holotype and paratypes 16, 21, and 35, 62-63. Apertural (62), dorsal (63) and profile 


64) views of holotype, lower Pleistocene of Cartiera Mulino, bed 3D1 of Costa (1959), SE Sicily, Ragusa, Vittoria. 65-66, Paratype 
16, dorsal (65), illustrating varices on the last whorl, and apertural (66) view, same site and loc: lity. 67. Paratype 21, a very inflated 
last whorl and a strong varix opposite the outer lip, middle to upper Pleistocene of Kyllini, NW Pe loponnesus, N2 bed of ‘Garilli et 
al. (2005a). 68-69. Apertural (68) and dorsal (69) view of paratype 35, showing a quite slender shell, middle to upper Pleistocene 
of Kyllini, northwestern Peloponnesus, H6 bed of Garilli et al. (2005a). Scale bars 100 pm 


THE NAUTILUS, Vol. 122, No. 1 


V. Garilli, 2008 


Page 4] 


S. et C. Gofas, May 1991; Malaga, Calahonda, conchif- 
erous detritus, beach, 1 sh., MNHN coll.S. Gofas, 1976— 
Sl; Malaga, Benalmadena, conchiferous detritus, beach, 
6 shs, MNHN réc S. Gofas, 1991-93; Malaga, Mijas, 
detritus, LO m, 4 shs, ee Stefano Rutini; Mz ilaga, Cabo 
Pino, detritus, 10 m, 2 shs, coll. SR, (41.S80¢); Morocco, 
Cabo Negro, beach, 1 sh., MNHN coll. S. Gofas, det. W. 
Ponder, 1986; France, Provence, Marseille, Curry, 

beach, 3 m, 2 shs, coll. PAL ex coll. C. Bogi, Jun. 1986, 
2037 MAR; Provence, Marseille, La Baule, small beach at 
25 km west from Marseille, 9 shs, coll. PAL ex coll. C. 
Bogi, Oct. 1986, 2035BAU; Provence, Le Dramont, 
(43°24.7' N, 6°51.7 E), 22-30 m, 26 shs, MNHN réc. J. 
Pelorce, 1992; Provence, Les Embiez, cote Nord et Petit 
Rouveau, (43°05' N, 5°47’ E), rocks, algae, 0-1 m, 11 
shs, MNHN réc S. Gofas, Jun. 1995; Provence, 

Marseille, Cap Morgiou, “calque de la Triperie,” 
(43°12.2' N, 05°26.9' E), muddy sand , inside cavity, 22 
m, 4 shs, MNHN réc H. Zibrowius, Jun. 1996; Provence, 

Marseille, Grand Congloue, (43°10.6' N, 05°24.2' E), 33 
m, 50 shs, MNHN réc. H. Zibrowius, Jun. 1996; 
Provence, Les Embiez, passe du Gaou, rocks, photophile 
algae, (43°04.3' N 5°47.4' E), 0-3 m, 1 sh., MNHN réc 
S. Gofas, Jun. ae Provence, St. Clair, infralittoral 
rocks, (43°08.2’ N 6°23.2' E), 0-1 m, 1 sh., MNHN réc. 
S. Gofas, Sep. 1992: Tunisia, Sfax, 4 shs, MNHWN coll. 
Staadt, 1969; Italy, Liguria, Portofino, 1 sh., coll. PAL, 
2038: Italy, Tuscany, Livorno, Antignano, 0.5 m, brown 
algae on rocks, 1 sh., coll. PAL legit Bogi, Apr. 1999, 
2029: Livorno, under littoral rocks, 0.5-1.0 m, 5 shs, coll. 
PAL ex coll. C. Bogi, 2039LIV; Livorno, Meloria, 10-30 
m, 3 shs, coll. PAL ex coll. C. Bogi, 1995, 2031MEL; 
Livorno, Tuscan Archipelago, Island of Elba, Capoliveri, 
32 m, 2 shs, coll. PAL ex coll. C. Bogi, Aug. 1994, 
2036CAPOL; Tuscan Archipelago, Gemini Island, 
(southern side of Island of Elba), 11 m, 3 shs, coll. PAL, 
2034; Tuscan Archipelago, Isola del Giglio, Punta 
Fenaia, 32 m, Ish., coll. medshells.com ex coll. G. Rug- 
gieri; Tuscany, Grosseto, Argentario, 25 m, 1 sh., coll. 
medshells.com ex coll. G. Ruggieri, Jul. 1988; Italy, 
Lazio, Ostia, Tor Paterno, 33 m, 5 shs, coll. medshells. 
com ex coll. G. Ruggieri; Lazio, Roma, Santa Marinella, 
ex reti, 2 shs, coll. medshells.com ex coll. G. Ruggieri; 
Italy, Sardinia, Oristano, Santa Caterina di Pittinurri, 5 
m, 3 ahe. coll. PAL ex coll. C. Bogi, 30 Aug. 1986, 
2040SCP: Sardinia, Sassari, Island of Maddalena, beach, 
3 shs coll. PAL ex coll. C. Bogi, 2028IMA; Sardinia, 
Nuoro, Sant’Antioco, Cala Lunga, 20 shs, coll. med- 
shells.com ex coll. G. Ruggieri, 03.1989; Italy, Cam- 


pania, Peninsula of Sorrento, Punta Penna, 2 shs, coll. 
medshells.com ex coll. G. Ruggieri, 1955; Italy, Sicily, 
Palermo, 2 shs, MNHN coll. Locard; Sicily, 1 sh., 
MNHN coll. Petit, 1873; Palermo, 3 shs, ZMB ex coll. 
Monterosato, $1013 (originally labelled by Monterosato 
as Cingula tenera, 3, 1125, Palermo); Palermo, Arenella, 
2 shs, ZMB ex coll. Monterosato, $1014, (originally la- 
belled by Monterosato as “Cingula ie var... 5 2, 
Arenella, Palermo”); Sicily, Siracusa, Capo Passero, 16 
m, 3 shs, coll. medshells.com ex coll. G. Ruggieri, 05 Sep. 
1987. 


Habitat: This species is clearly limited to infralittoral 
depths. In the upper part of its distribution, it seems to 
live in very shallow waters, on algae. It likely lives also in 
the cavities occurring in infralittoral muddy sandy bot- 
toms. 


Distribution: In the western and central Mediterra- 
nean the species seems to be well distributed; Adriatic 
and eastern Mediterranean occurrences should be veri- 
fied. In the Atlantic it lives along the Moroccan coasts 
and in the Canary Islands. To my laow ledge, there is no 
fossil record of this species. 


Remarks: — This small species is characterized by having 
a variable shell shape and sculpture. The conical shells, 
bearing a strongly cingulated sculpture, which markedly 
prevails over the axial, are comparable with typical 
keeled morph of Alvania carinata. In addition, Cingula 
species provided with a strong spiral sculpture are com- 
parable to A. tenera. The ovate, slender shells of this last 
species, with a finely cingulated sculptural pattern, might 
vaguely resemble some species of Setia H. and A. Adanis. 
1854. Asa consequence, Piani (1979) ) and Van Aartsen 
(1982) included A. tenera in Galeodina, whereas Nords- 
ieck (1968; 1972) placed the same species in Setia and 
Cingula elas ely. The last two views should not be 
accepted. In fact, the species of Setia have a smooth or 
very slightly sculptured shell with more convex whorls 
than A. tenera, and Cingula species are characterized by 
shells usually lacking axial sculpture and having a very 
thick outer lip. 

The shells from eastern Atlantic (Canary Islands and 
Morocco) do not show meaningful differences from the 
Mediterranean ones studied. With regard to the resem- 
blance between A. tenera and A. carinata, it is manifest 
in the keeled, conical-inflated shell shape shown by their 
typical respective morphs, and in having a quite wide and 
developed aperture. However A. fenera never shows va- 


Figures 70-79. 


Alvania rosariae new species 70. Protoconch of the holotype, lower Pleistocene of Cartiera Mulino, bed 3D1 of 


Costa (1989). SE Sicily, Ragusa, Vittoria. 71. Protoconch of p paratype 16, showing variation of the abapical ridges on protoconch II, 
same locality and bed. 72. Dorsal view of protoconch of paratype 21, showing protoconch I sculpture and the abapical ridges on 
protoconch TI, middle to upper Pleistocene of Kyllini, NW Peloponnesus, N2 bed of Garilli et al. (2005a). 73-74. Holotype, detail 
of early protoconch (73) showing sculpture of protoconch [and protoconch I/protoconch H boundary, and dorsal view of protoconch 
74). 75. Sculpture of protoconch I, paratype 21. 76-79. Holotype, detail of teleoconch sculpture: early whorls (76), first to goes 
whorl (77), showing microsculpture, penultimate to last whorl (78), showing the microscopic incremental scars, and last whorl ( 
Scale bars: 100 2m in Figures 70-72, 74, 76 and 78-79: 50 jzm in Figure 77; 20 ym in Figure 75. Black and white arrows ened 
the protoconch I/protoconch IH and protoconch/teleoconch boundaries, respectively 


NNN eee 


THE NAUTILUS, Vol. 122, No. 1 


V. Garilli, 2008 


Page 43 


rices, bears a very delicate axial sculpture, its outer lip is 
always very thin, and usually has more colored shells. 


Genus Galeodinopsis Sacco, 1895 


Type Species: 
nal designation. 


Rissoa tiberiana Coppi, 1876 by origi- 


Galeodinopsis tiberiana (Coppi, 1876) 
(Figures S1-99) 


1862 Rissoa tuba Doderlein, 1862: 17 (nomen nudum) 
1876 Rissoa Tiberiana Coppi, 1876: 201-202. 

Manzonia fariai Rolain and Fermandes, 1990: 64-65, pl. 1, figs. 
4-6. 

Alvania fariai (Rolin and Fernandes, 1990).—Gofas, 1999: 8S— 
89, figs. 39-42. 

Alvania fariae (Rolain and Femandes, 1990)—Landau et al., 
2004: 41, pl. 7, figs. 3-4. 


Q 


Description: Shell conical, sturdy, reaching 3.75 mm 
in height and 2.75 mm in width. Protoconch multispiral, 
conical, with 2-2.2 convex whorls and a rather immersed 
nucleus. Protoconch I consists of about 0.7—0.8 whorls, 
with a netted sculpture, consisting of 7-8 very thin spiral 
lirae and numerous, irregular, short and very narrow axial 
segments occurring in interspaces between lirae. Proto- 
conch II is sculptured by very small, sparse granules, 
fused into 2-4 discontinuous ridges on central wad abapi- 
cal portions of latter part of fast. whorl. Groups of gran- 
ules form very short, prosocline segments on central 
area, mainly close to protoconchyteleoconch transition, 
which is marked by a quite sinous and thin lip. Teleo- 
conch formed by 3-3.5 convex whorls, sculptured by a 
primary pattern ‘of strong, slightly sinuous and opistho- 
cline ribs, numbering 12-14 on pemdinee whorl, be- 
coming very weak to lacking toward shell base. Ribs are 
covered by a secondary spiral sculpture, formed by flat 
narrower cords, numbering 4-5 on penultimate whorl, 
becoming more marked on shell base. Each cord bears a 
pitted microsculpture, consisting of microscopic subcir- 
cular pores forming S—10 spiral alignments (Figures 99). 
Between cords, numerous, closely spaced, very fine spi- 
ral ridges appear. They are for ined by rough ae ae 
tooth- like elements, extending perpendicularly rom 
shell surface. At their base, ridges have a lamella-like 
expansion which covers the interspaces occurring be- 
tween them. Last whorl inflated, well-expanded, on 
prising about % of total shell height, often bearing 1-2 
strong varices, most frequently just before the outer lip 
Aperture ovate, very rounded anteriorly, weakly angu- 
lated posteriorly, comprising about little less than ani 


3 of total shell and last whorl height, respectively. Outer 
lip sinuous, weakly opisthocline; internally smooth, with 
a thin rim on its edge; externally with a very marked 
varicose swelling, covered by spiral sculpture, and having 
a narrow ridge on its base, toward aperture, so that it 
appears double -rimmed. Inner lip weakly curved, with a 
very narrow columellar thicknening, forming an obsolete 
umbilical chink. Shell color white in Recent material. 


Type Locality: Coppi (1876) cited “La Tagliata”, an 
unknown name in toponymy (very likely referring toa 
recently deforested woodland) corresponding to the lo- 
cality Gagliarde lla (Maranello, Modena, Emilia Roma- 
ona, North Ite uly) (S. Palazzi personal comm., 2006). 


Type Material: A lot (PUM 13721), from Coppi coll, 
ee locality, with more than 100 possible syntypes (not 
seen), is housed in the MPOB, Modena. 


Material Examined: Senegal: Region de Dakar, 250 
m S.W. Cap Manuel, 12 shs, MNHWN coll. Marche- 
Marchad, dét. S. Gofas [The Nautilus 113: 8S—89, figs. 
40, 42]; Region de Dakar, S.W. Gorée large Cap Vert, 
250-150 m, ae MNHN coll. Marche-Marchad, det. S. 
Gofas [The Nautilus 113: SS—89, figs. 40, 42]; Region de 
Dakar, St. 56-1-10A Gorée 150-200 m, 1 sh., MNHN 
coll. Marche-Marchad, det. $. Gofas [The Nautilus 113: 
88-89, figs. 40, 42]; Angola: Luanda, Hha de Luanda, 
Cirealittoral, 120 m, 3 ae MNHN coll. S. Gofas, det. S. 
Gofas [The Nautilus 113(3): 88—S9, figs. 40, 42]; Luanda, 
Ilha de Luanda, circalittoral, 40-60 m, 1 sh., MNHN 
coll. S. Gofas ares det. S. Gofas [The Nautilus 113: 
88-59, figs. 40, 42]; Luanda, Au large de Mussulo 
(Mocéco), ” dredging 50-70 m, 2 shs, MNHN coll. S. Go- 
fas L9S1—1987, det S. Gofas (‘The Nautilus 113: 8S—S9, 
figs. 40, 42]; Au large de Mussulo, circalittoral, 90—LOO 
m, 2 shs, MNHN coll, S. Gofas, det. S. Gofas |The Nau- 
tilus 113: 88-89, figs. 40, 42]; Ambrizete, dredging, SO m, 
5 shs, MNHN coll. S. Gofas, det. S. Gofas [The N Nautilus 
113: SS—89, figs. 40, 42]; Ambrizete, (07°00' S, 12°20' E) 
sediment, 60 m, 3 shs, MNHWN coll. S. Gofas, 1983, det. 
S. Gofas [The N Nautilus 113: 88—S9, figs. 40, 42); Am- 
brizete, (06°57' S, 12°23’ E), sediment, 45 m, 1 sh., 
MNHN coll.S. Gofas, 1983, det. S. Gofas [The Nautilus 
113: SS—89, figs. 40, 42). Italy: Piemonte, Asti, Baldich- 
ieri, “Grottino Monale”, 2 shs, yellow sands, middle- 
upper Pliocene, coll. MF ex coll. PAL, FLO4A; Asti prov- 
ince, 9 shs, Pliocene, MGUP coll. Doderlein, 111A; Lig- 
uria, Savona, Rio Torsero, between Ceriale and Peagna, 
5 shs, clays, lower Pliocene, coll. MF ex coll. PAL, F5SA; 


Figures 0-89. 


Alvania tenera (Philippi, 1844). 80. Shell of a keeled and conical morph, Strait of Gibraltar, Spain, Cadiz, Tarifa, 


coll. PAL (2030TAR). 81-82. Conical-ovate, not-keeled morphs, illustrating the variable sculpture, France, Provence, Marseille 
Curry, coll. PAL (2037MAR). 83. Profile view of the same shell as Figure 80. 84. Profile view of an unkeeled morph, France 
Provence, Marseille, La Baule, small beach at 25 km west from Marseille, coll. PAL (2035BAU). 85-86. Sculpture of early teleoconch 
whorls (85) and detail of sculpture (86), Italy, Tuscany, Livorno, Antignano, coll. PAL (2029). 87-88. Protoconch (87) and detail of 


protoconch I ($8) of the same shell. 89. Protoconch of the same shell as Figure 84, showing sculpture variation. Scale bars: 0.5 mm 
in Figures 80-84: 200 jzm in Figure $5; 100 jum in Figures 86-87 and 89: 20 pm in Figure 88. Black and white arrows indicate the 
protoconch I/protoconch II and protoconch/teleoconch boundaries, respectively. 


THE NAUTILUS, Vol. 122, No. 1 


V. Garilli, 2008 


Page 45 


Emilia Romagna, Modena province, 35 shs, Pliocene, 
coll. Doderlein, 111B; Emilia Romagna, Modena prov- 
ince, 1 sh., Miocene, MGUP coll. Doderlein, 474: Emilia 
Romagna, Piacenza, Lugagnano Val d’Arda, 2 shs, “calan- 
chi di valle” (marls), middle-upper Pliocene, coll. MF ex 
coll, PAL, FI3A; Piacenza, Castell’Arquato, Monte Pa- 
dova, 1 sh., blue clays, middle Pliocene, coll. MF ex coll. 
PAL, FIGA; Emilia Romagna, Parma, San Nicomede, 
Stirone river, 50 shs and fragments, lower clays middle- 
upper Pliocene, coll. MF ex coll. PAL, FI4B: Emilia 
Romagna, Modena, Marano on the Panaro, Panaro river, 

1 sh., clays stormy layers, middle-upper Pliocene, coll. 
Me ex coll. PAL, FSOA: Emilia Romagna, Modena, Ma- 
ranello, Fogliano, Gagliardella (type locality), Rio Griz- 
zaga, 60 shs, sands, middle Pliocene, coll. MF ex coll. 
PAL, FS9A; Tuscany, Siena, San Donato, Ciuciano, 
Prison, | sh., clays and sands, lower Pliocene, coll. MF ex 
coll. PAL, F112A; Siena, Castiglioncello del Trinoro, 
Poggio Rotondo, 3 shs, marls, lower Pliocene, coll. MF ex 
coll. PAL, F54A; Lazio, Rome, Magliano Sabina, Cla- 
docora yellow sands, 23 shs, lower Pliocene, coll. MF ex 
coll. PAL, FI5A: Sicily, Palermo, Altavilla Milicia, rigth 
side of Milicia river, 12 shs, sands, lower-middle 
Pliocene, coll. MF ex coll. PAL, F2A; Palermo, Partitico, 
Trappeto, Lido Ciammarito to Nocella river mouth, 11 
shs, clays, lower Pliocene, coll. MF ex coll. PAL, F72A. 


Habitat: In the Atlantic Ocean, the species has a lower 
shelf-upper slope distribution (see Gofas, 1999), is indi- 
cated by the fossil Mediterranean occurrences. A shal- 


lower and more restricted distribution, limited to shelf 


paleoenvironments, with sandy to muddy bottoms. 


Distribution: The species lives in the eastern Atlantic, 
from Senegal to northern Angola (see also Gofas, 1999). 
It was also collected from the coasts of Mauritania (S. 
Palazzi, pers. comm., 2006). The species has a Mediter- 
ranean Miocene to Pliocene paleoditribution, being re- 
corded from the Miocene of northern Apennines 
(Modena), the Pliocene of northern (Piemonte, Liguria, 
Toscana, Emilia Romagna), central (Lazio) and insular 
(northwestern Sicily) Italy, south Spain (Estepona) Lan- 
dau et al. (2004, as Alvania fariae), and Algeria (Coss- 
mann, 1921). In Atlantic, it is recorded fom the Portu- 
a middle Pliocene of Mondego Basin (Landau et al., 
2004). The citation of Wenz (1938: 616), according to 
Ww hich the species lived in the Oligocene (up to Fieesae 
of Europe, North Africa and North America), should be 
verified. The species was not found in the very rich mol- 


luscan assemblages from the Oligo-Miocene of south- 
western France (Lozouet, 1998; 1999). 


Remarks: —Gofas (1999) and Landau et al. (2004), dis- 
cussing and re-describing this species, originally de- 
scribed as Manzonia fariai by Rolan and Fermandes 
(1990) from West Africa, did not recognize its identity as 
Rissoa tiberiana Coppi, 1876, the latter being a common 
species from the Mediterranean Neogene, particularly 
from the Pliocene. This species, origin ally not illustrated 
by Coppi (1576), was figured by Sacco (1895: figs. 67, 
a-bis and 68, a—b), who de signated it as the type species 
of the subgenus Gale odinopsis Sacco, 1895. More re- 
cently, this species was illustrated by Cossmann (1921: 
pl. 1, figs. 55-56) and Wenz (1938: fig. 1715). Compari- 
son between fossil material of R. tiberiana (Figures 90— 
94, also from topotype material; see also Landau et al., 
(2004: pl. 7, figs. 3-4), to that of Rolain and Fernandes 
oun. aL , figs. 4-6), and of Gofas (1999: figs. 39-42) 
strongly éonficus the above mentioned synonymy. The 
rather conical, ribbed shell with an inflated last whorl, 
the frequent presence of varices on the last whorl, and 
the double-rimmed outer lip are the most characteristic 
features of this species, which shows a modest variability 
in the number and strength of ribs and in the spire el- 
evation (see Figures 90-92). 

Gofas (1999) moved this species from Manzonia Bru- 
sina, 1870 [type species Manzonia crassa (Kanmacher, 
1798), see Figures 104-107] to Alvania based on the lack 
of the characteristic punctate spiral sculpture of the 
former taxon. This does not appear appropriate. In fact, 
in reasonably well-preserved shells, the primary spiral 
cords clearly bears a microsculpture consisting of regu- 
lar, spirally arranged pits, quite like M. crassa (compare 
Figure 99 with Figure 106). This spiral pitted micro- 
sculpture, sonsilared a typical Manzonia character by 
Moolenbeek and Faber (1987), was indicated by 
Bouchet and Warén (1993) as not restricted to this genus 
(occurring in Alvania, Gofasia Bouchet and Warén, 
1993, and with a rough similarity, in Rissoininae species, 
see Gofas, 1999, figs. 79-80, 85, and 89]). The same 
authors interpreted it as a symplesiomorphy retained in 
Manzonia and in other rissoid genera. However , the par- 
ticular structure of the secondary, very fing spiral 
threads, formed by roughly prismatic elements growing 
perpendicularly to the Shell surface, is a chowacker share d 
by the Manzonia species, never observed in Alvania, and 
retained only in the recently described genus Gofasia 
(see Bouchet and Warén, 1993, fig. 1557). The combi- 


Figures 90-99. Galeodinopsis tiberiana (Coppi, 1876). 90-93. Shells from the type locality, showing variability and varices, middle 


Pliocene of Italy, Emilia Romagna, Modena, Maranello, Fogliano, 


Gagliardella, Rio Grizzaga sands, coll. MF ex coll. PAL (F394). 


94. Profile view of a shell from the middle- ‘upper Pliocene of Italy, Emilia Romagna, Parma, San Nicomede, Stirone River, coll. MF 
ex coll. PAL (F14B). 95-96. Protoconchs from the same locality, note variation of the abapical sculpture just behind the transiction 
to teleoconch. 97. Detail of protoconch I, showing the nette d microsculpture and the partially immersed nucleus, same shell as 
Figure 95. 98. Detail of teleoconch sculpture from the shell as Figure 90. 99. Detail of tele oconch microsculpture from the shell as 
Figures 95-96: note the pitted pattern on the spiral cord and the structure of the fine spiral ridges. Scale bars: | mm in Figures 90-94; 
100 pm in Figures 95-96; 50 jm in Figure 98; 20 pm in Figures 97, 99. Black and white arrows indicate the protoconch I/protoconch 


I and protoconch/teleoconch boundaries, respectively. 


er ee a ee 
AUTILUS, Vol. 122, No. 1 


V. Garilli, 2008 


Page 47 


nation of these characters (pitted sculpture and structure 
of fine spiral threads) represents a quite singular and 
original feature, which should be regarded as limited to 
Manzonia-related species.: e.g. M. darwini Moolenbeek 
and Faber, 1987 (pl. 1, fig. 18), M. crispa CW atson, 1873) 
(see Moolenbeek and Faber, 1987, pl. 3, fig. 54), M. 
boogi lanzarottii Moolenbeek and Faber, 1987 | pl 2, fig. 
39), M. spreta (Watson, 1873) (see Moolenbeek and 

Faber, 1987, pl. 3, fig. 57), M. vigoensis (Roldan, 1983) 
(see Bouchet and Warén, 1993, p. 656, fig. 1499) and 
several others. The double rimmed outer lip and the 
netted sculpture of protoconch I are also characters 
shared by Manzonia species. However, it must be con- 
sidered that these two features, considered by Ponder 

(1985: 46) as typical of Manzonia sensu stricto, should be 
a aled with suspicion, being as they are shared by 
well- recognized Alvania species. In regard to the proto- 
conch, species such as A. testae (Aradas and Maggiore, 
1844), A. setlandica (Montagu, 1815) (see Bouchet and 
Warén, 1993, figs. 1386-1387 and 1502; Landau et al., 
2004, pl. 9, figs. 1b-Id), the Pliocene Alvania magistra 
Chirli, *006 (pl 11, fig. 16 and pl. 12, figs. 1-3), have the 
same sculptural pattern on protoconch I. A. tomentosa 

Pallary, 1920), which has a paucispiral protoconch, also 
hie this sculpture (see Bouchet and Warén, 1993, 
fig. 1388). 

Rissoa tiberiana could be regarded as one (probably 
the sole) of the few survivors of a group of species close 
to Manzonia sensu lato, which very likely originated in 
the upper Paleogene. The European Oligocene Rissoa 
duboisi Nyst, 1843 (Figures 100-103) certainly belongs 
to this group. Both these species share several intersting 
characters: a quite conical Alvania-like shell shape, often 
provided with varices on the last whorl, the arrangement 
of the basal cords (not so strong as in Manzonia sensu 
stricto, where keels occur on the shell base), the above 
mentioned combination of the microsculptural pattern, 
and the kind of axial sculpture (with slightly sinuous, less 
pronounced ribs than those shown by Manzonia), and a 
double, weakly opisthocline outer lip. All this leads me to 
revalue Galeodinopsis as the useful generic placement 
for such Manzonia-related species. 

In overall appearance, G. duboisi (Nyst, 143) strongly 
resambles G. tiberiana, from which it differs princips ally 
in having less numerous cords and ribs and less convex 
whorls (see also Ponder, 1985, fig. 100c). Another com- 
parable species is the Recent Macaronesian M. spreta 
(Watson, 1873), which has a similar shell shape but dif- 
fers from G. tiberiana in having a more delicate axial 
sculpture (the spiral cords being large and almost flat) a 


more rounded and smaller aperture, and a paucispiral 
protoconch (see Moolenbeek and Faber, 1987, figs. 
47a—b and 55-57). With some significant reservations, it 
could be regarded as belonging to Galeodinopsis. The 
species M. foraminata (Lozouet, 1998), originally de- 
scribed as Alvania (from the upper Oligocene : south- 
western France, see Lozouet, 1998, fig, 9f-h), M. mou- 
linsi (dW Orbigny, 1852) (see Lozouet, 1998S, fig. a from 
the French upper Oligocene), M. scalaris (Dubois, 1831) 
(Kowalke and Harzhauser, 2004, fig. Sd, from the middle 
Miocene, Badenian, of Austria, Hungary, Poland and Ro- 
mania and the Miocene of Russia) and the Recent Man- 
zonia crispa (Watson, 1875) (of which I studied 3 shells 
from Madeira, ZMA Moll. 101.0, ex coll. R.G. Moolen- 
beek; see also Ponder, 1985, fig. LOOA and Moolenbeek 
and Faber, 1987, text-fig. 46, acl pl. 3, fig. 52-54) show 
less affinities, having a more slender chell with more 
curved ribs (protruding over the suture in M. scalaris), 

which become stronger on the base of the shell. All these 
four species appear more related to Manzonia than to 
Galeodinopsis. 

Sacco (1895) and Cossmann (1921) indicated Rissoa 
multicostata Speyer 1864 (pl. 41, figs. 3-5, from the Oli- 
gocene of Germany) as a possible Galeodinopsis species. 

I did not see any shells of this Alvinia sensu lato-like 
species, which more closely resembles the group of Al- 
vania zetlandica (Montagu, 1815) and A. weinkauffi 
(Weinkauff, 1S68 ex Sdhwartz ms.). The original illustra- 
tions show a turreted shell with a more finely cancellate 
(not-ribbed) sculpture, bearing almost orthocline axial 
ribs, characters which militate against placement in Ga- 
leodinopsis. 

The material of Rissoa tuba Doderlein, 1862, housed 
in the MGUP Doderlein’s collection (from the Miocene 
and Pliocene of North Italy), belongs to this species. 
Anyway, Doderlein (1562: 17) just listed this taxon with- 
out providing a description or a valid reference, so that R. 
tuba must he considered a nomen nudum. 


CONCLUDING REMARKS 


As indicated by Ponder (1985), the systematic grouping 
of Alvania species at the subgeneric level is quite diffi- 
cult and putative groups usually fold into synonymy with 
Alvania sensu stricto. This viewpoint appears be appli- 
cable to the species studied in this report (except for 
Rissoa tiberiana Coppi, 1876). The subgeneric division 
into Galeodina and Massotia lacks any valid and convinc- 
ing basis in shell features (especially on consideration of 


Figures 100-107. 


Galeodinopsis duboisi (Nyst, 1843) (100-103), from the upper Oligocene of Hessen (Germany, SE of Kassel, 


Hiessizoh: Lichten, Glimmerode, coll. MF ex coll. Pal, FEI5A) and Manzonia crassa (Kanmacher, 1798) (104-107), type species of 
Manzonia Brusina, 1870, from the middle-upper Pleistocene of Kyllini (NW Peloponnesus, Greece, N2 bed of Garilli et al., 2005a). 

100-102. Apertural (100 and 102) and profile (101) views. 103- 104. Microsculptures, showing the typical pitted surface on the flat 
cords and the microstructure of the narrow spiral threads in G. duboisi (103, same shell as Figure 101) and M. crassa (104, same shell 
as Figure 105). 105-106. Apertural (105) and profile (106) view of shell. 107. Detail of protoconch I, showing the netted sculpture 
Scale bars: 1 mm in Figures 97-98; 20 jm in Figure 100: 10 zm in Figure 99. The black arrow indicates the protoconch I/protoconch 


Il boundary. 


Page 45 


THE NAUTILUS, Vol. 122, No. 1 


the often extensive intraspecific variation). In effect, a 
link between the markedly keeled and inflated shells 
(e.g. A. carinata), and the more typical Alvania-shape 
taxa could be hypothesized. It could be articulated and 
summarized in the transition A. carinata-A. lactea (via A. 
francescoi new species-A. rosariae new species) to more 
typical Alvania secies. A parallel trend could be con- 
structed: A. tenera and A. cingulata, showing a progres- 
sive weakening of the spiral sculpture, the proportional 
reduction in the expansion of the body whorl and the 
formation of the outer lip thickening. Furthermore, 
other interspecific trends, involving various nominal 
groups of Alvania, could be constructed when consider- 
ing the widely variable shells of most of the species dis- 
cussed here. Nowstanding this, I cannot deny that all 
these hypothetical links appears tenuous and that these 
taxa, sharing a quite inflated and well-expanded body 
whorls, a wide, internally smooth aperture, seem some- 
what distant from Alvania sensu stricto This point of view 
could lead to regard them as belonging to an inclusive 
taxonomic group (preferably at subgeneric level). Galeo- 
dina could serve as the appropriate taxonomic unit. Nev- 
ertheless, since I believe that appropriate anatomical 
studies (of which there is no exhaustive data-set) should 
be used to solve this question, I have preferred a more 
open-ended option, placing the discussed species here, 
from carinata to tenera, in Alvania sensu lato 

The re-analysis of the Neogene-Recent R. tiberiana, 

type species of Galeodinopsis, has lead me to reevaluate 
the latter taxon as the appropriate genus for grouping 
species characterized by a particular telec conch micro- 
sculpture (consisting of ve ry fine spiral threads, formed 
by roughly prismatic ele ments, extending perpendicu- 
larly respect from the shell surface, plus the pitted sur- 
face on the primary spiral cords, as seen in Manzonia) 
and by having a conical Alvania-like shell shape. The 
oldest representative of this genus is the European upper 
Oligocene R. duboisi. The upper Oligocene M. forami- 
nata and M. moulinsi are probably more linked to Man- 
zonia sensu lato than to Galeodinopsis while the Recent 
Macaronesian M. spreta could be doubtfully regarded as 
belonging to Galeodinopsis. 

Because of its affinity with Alvania, onsite 
probably derived from some Oligocene group of that 
ee and could be regarded as a transitional link to 
Manzonia. It is interesting that, as indicated by Lozouet 
(1998), the genus Alvania had a surprising radiation in 
the upper Oligocene, creating a strong diversification. 
Furthermore, the Manzonia sensu stricto spe cies, char- 
acterized by the quite slender shell with strongly opis- 
thocline ribs and ve ry marked spiral cords on the shell 
base, seem to be well-established from the European 
Neogene, where they are represented by the following 
species: M crassa, M. falunica, M. pontilevie nsis, and M, 
scalaris. 

All the discussed species have a multispiral proto- 
conch, indicating a planktotrophic larval development, 


with the exception of A. cingulata, which is the most 


geographically restricted species, limited to Sicilian wa- 
ters. The reason for its very limited geographical distri- 
bution is not certainly solely attributable to its non- 
planktotrophic larval development. Our knowledge 
about this species is not encouraging: I just know that: 


@ There is no known fossil record . . . 
Mediterranean acquisition? 

@ From a large amount of bulk samples (about 50 liters 
from Magnisi and 20 liters from Mondello), I picked 
only a single eroded shell! . . . Is it still a living species? 


is it a very recent 


Among species with planktotrophic development dis- 
cussed in this report, three show an eastern Atlantic- 
Mediterranean distribution: A. carinata, A. lactea, and 
A. tenera. Among these, the first two species have a Neo- 
gene to Recent distribution, being known from the 
Pliocene and the Miocene of Mediterranean basin, re- 
spectively, and show a quite extensive east Atlantic dis- 
tribution, being commonly recorded from the English 
Channel to Moreceo. Regarding the third species, living 
in Mediterranean and along the Atlantic Morocco aad 
the Canary Islands coasts, I “did not find any Mediterra- 
nean or Atlantic fossil records. The extinct species A. 


francescoi, A. prusi, and A. rosariae are limited to the 


Mediterranean Pleistocene, while Galeodinopsis tiberi- 
ana has a Miocene-Pliocene Mediterranean distribution 
and lives along the W African coasts from Mauritania to 
northern Angola. 

In general, the protoconchs of the species reported 
here do not indicate any distinguishing taxonomical char- 
acters at the supri aspecific, ane, in most cases, at the 
species level, being characterized by sculptural patterns 
shown by several rissoid species. Among the plank- 
totrophic species, the commonest sculptural pattern on 
protoconch I (observed in A. carinata, A. francescoi, A. 
lactea, A. rosariae, and A. tenera), consisting of fine spiral 
lirae and scarce to abundant pimples between them, is 
shown by A. cancellata (Da Costa, 1778) (see Giannuzzi- 
Savelli et al., 1996, fig. 408b), A. beani (Hanley in 
Thorpe, 1544) (see Giannuzzi-Savelli et al., 1996, fig. 
412), A. cimex (Linneus, 1758) (see Ponder, 1985, 
figs. S6C-B), A. cimicoides (Forbes, 1844) (see Bouchet 
and Warén, 1993, fig. 1385 and Giannuzzi-Savelli et al., 
1996, fig. 408d), 4. dingde nsis Uae n, 1967) (see Pon- 
der, 1985, figs. cas G), A. ge ryonia (Nardo, 1847) (see 
Giannuzzi-Savelli et al., 1996, fig. 3950), A. hispidula, 
Monterosato, 1554 (see Gotas, 1999, fig. 26), A. punctura 
(Montagu, 1803) (see Giannuzzi-Savelli et al., 1996, fig. 
436), A. stephanensis Lozouet, 1998 (fig. 9D), Crisilla 
semistriata (Bouchet and Warén, 1993, fig. 1535), Rissoa 
parva (Da Costa, 1778) (see Ponder, 1985, figs. 79B), 
and several other species, The sculptural pattern on pro- 
toconch I (also observed in G. tiberiana), consisting of 
1-3 more or less discontinuous, abapical to central, spiral 
threads (also formed by short, 
ments close to the beginning of the teleoconch) and tu- 
bereles on the remaining surface, is also shown by most 


fused, prosocline seg- 


of the above mentioned species and several others. Al- 


V. Garilli, 2008 


Page 49 


vania lactea could represent a slight exception, having a 
quite characteristic, easily distinguish: thle protoconch | 

with a coarser sculpture than that shown by the other 
studied species. A. tenera shows a similar protoconch I 
sculptural pattern, with few to abundant very small gran- 
ules spirally arranged, which may form very discontinu- 
ous and irregular ridge »s. Likewise this kind of sculpture 
is shared by other psseils (e.g. A. tarsodes (Watson, 
1886) (see Bouchet and Warén, 1993, fig. 1450) and 
Crisilla semistriata). The sculpture of the paucispiral 


protoconch of A. cingulata, which is almost a replica of 


protoconch I of the first group discussed above, repre- 
sents a quite common pattern shown by several non- 
planktotrophic rissoids: e.g. A. argillensis Lozouet, 1998, 
A. macandrewi (iensont 1868), Lironoba multilirata 
(T. Woods, 1878), Onoba gianninii (Nordsieck, 1974) 
and, with a moderate similarity, by A. swbsoluta (Aradas, 
1847), Onoba semicostata ( Montagu, 1803), and A. viro- 
dunensis Lozouet, 1998 (see Ponder, 1985, fis 89A, 109E 
and 126C; Bouchet and Warén, 1993, figs. 1458, 1525; 
Lozouet, 1998. figs. 1OF and 10k). ; 

There is similarity between the West African rissoid 
assemblages, including the Macaronesian province, and 
the European ones, with particular regard to those from 
the Mediterranean Neogene. This similarity is perhaps 
more marked than it has been indicated by Gofas (1999). 
The case of G. tiberiana is a further (see Monegatti and 
Raffi, 2001, and Garilli and Galletti, 2007) interesting 
case of a molluscan species that lived in the Mediterra- 
nean Neogene and today occurrs along the West African 
coasts. In this view, it is noteworthy to remark that, as a 
whole, most of the species here included in Galeodinop- 
sis lived in the European Oligo-Pliocene while its living 
representings occurs along the West Africa and Macaro- 
nesian Provinces. In addition: A. tenera, living in the 
Mediterranean, Atlantic Morocco, and in the Canary Is- 
lands (Tenerife), should be regarded as a new record, 
further supporting the Geciisse d similarity. The rissoid 
Rissoina dOrbigny. 1840, species from the Mediterra- 
nean Plio- Pleistocene (see the good illustrations of 
Greco, 1974, figs. 11, 13, 15, 17 ad Chirli, 2006, pl. 23 
figs. 7-12). usually cited as R. decussata Moses 
1803), is very likely the same taxon as living along the W 
African coasts, S40 Tomé and (¢ Cape Vena Islands [see 
Gofas, 1999: 97, figs. 69-73, and treated as R. punc- 
tostriata (Talavera, 1975)]. 

The rather common presence of varices on the last 
whorl (a rare character in rissoids) of A. carinata, A. 
francescoi new species, A. lactea, A. rosariae new spe- 
cies, and G. tiberiana should not be regarded as a salient 
taxonomic character at the supraspecific level, being 
present in quite unrelated species (e.g. A. carinata and 
G. tiberiana). 


ACKNOWLEDGMENTS 


This article would not have been possible without the 
generous support of friends, private collectors, and re- 


searchers. | particularly thank Maurizio Forli (Prato), 

Stefano Palazzi (Modena), Nino Adorna Sbrana (Gros- 
seto), Stefano pasa (Anguill: wa) for lending a large 
amount of precious and indispensable material. Many 
other people kindly made their collections available. 

R.G. Moolenbeek (ZMA, Amsterdam) donated three 
shells of Manzonia crispa from ZMA Moll. collection. 
Rossana Messina (Palermo) provided a special support, 
measuring most of the investigated shells and assisted me 
during the bibliographical research in the DGUP library. 
Luca Calle (Monreale), a companion for all the field- 
works in Kyllini and Cartiera Mulino, also help me in 
providing SEM images. My special thanks to M. 
Glaubrecht (ZMB, Berlin), who very kindly lent precious 
material (with particular regard to the two Alvania cin- 
culata shells from Monterosato collection), to D. Merle, 
J. M. Pacaud (MNHN-DHT, Paris), V. Héros, and P. 
Lozouet (MNHN-DSE, Paris), who Meer helpful 
and kind assistance during my visits in MNHN. A. Rosso 
and I. Di Geronimo (DSTC, Catania) kindly allowed the 
visit to the malacological collections of the DSTC. Kathie 
Way (British Museum of Natural History, London), H. J. 
Niederhéfer (Staatliches Museum fur Naturkunde, 
Stuttgart) and P. Serventi (MPOB, Modena) provided 
information on type material. Thanks to M. Fiore, P. 
[acopelli, V. P. Li Vigni (MSNCS, Terrasini), who al- 
lowed the visit to the conchological collections of the 
MSNCS. C. D’Arpa (MGUP, Palermo) lent some fossil 
material housed in the MGUP. Even when the library 
was moving, G. Barranca (DGUP, Palermo) very kindly 
allowed me the access to some papers. Riccardo Gian- 
nuzzi-Savelli (Palermo) and R. G. Moolenbeek helped 
me in some bibliographic researches. R. G. Moolenbeek 
and S. Palazzi also provided very useful and constructive 
comments on an early manuscript. 

I am grateful to D. Merle, R. G. Moolenbeek, H. 
Scholz (ZMB) and E. Theodorou-Vardala (GNHM, Ki- 
fissia), who ne provided the catalog numbers for 
type material of A. francescoi new species and A. rosariae 
new species. I am also grateful to H. G. Lee (Jackson- 
ville, Florida) and P. Lozouet (MNHN, Paris) who re- 
viewed the manuscript and provided valuable comments. 
H. G. Lee also kindly helped improve my shaky English. 
This work also benefited from critical re -adings by S. Go- 
fas (Departamento de Biologia Animal, C Gierded de 
Malaga) and José H. Leal (The Bailey-Matthews Shell 
Museum, Sanibel, Florida). 


LITERATURE CITED 


Aartsen, J. J. Van. 1982a. Synoptic Tables of Medit. & Europ 
Conchology. Gen. Alvania (Subg. Alvinia & Galeodina). 
La C ‘onchiglia 14 (162-163): 8-9 

Aartsen, J. J. Van.1982b. Synoptic Tables of Medit. & Europ 
Conchology. Genus Alvania (Sbg. Arsenia & Alvaniella) 
La © onchigli: 1 14 (164-165): 4-6. 

Aartsen, J. J. Van, H. P. M. G. Menkhorst, and E. Gittenberger 
1984. The marine Mollusca of the Bay of Algeciras, Spain, 
with general notes on Mitrella, Marginellidae and Tur- 
ridae. Basteria Supplement 2: 1-135 


Page 50 


THE NAUTILUS, Vol. 122, No. 1 


Anistratenko, V. V. and Ya. I. Starobogatoy. 1994. Molluscs of 


the Superfamily Rissooidea (Gastropoda Pectinibranchia 
Rissoiformes). La C onchiglia Anno XXVI, 271: 41-48. 

Arduino, G. and M. Arduino, 2001. Catalogo illustrato. Ris- 
soidae del Mediterraneo, Naturama (ed.), Palermo. 50 pp. 
Also available at http:/Avww.turin.net/shell. 

Avila, S. P. 2000. The shallow-water Rissoidae (Mollusca, Gas- 
tropoda) of the Azores and some aspects of their ecology. 
Iberus 1S: 51-76. 

Bouchet, P. 1978. Coquillages des cétes atlantiques et de la 
Manche. Les editions du Pacifique, Papeete, 144 pp. 
Bouchet, P. 1992. Terre océane. Imprimerie N Nationale, 1SS pp. 
Bouchet, P. and A. Warén. 1993. Revision of the Northeast 
Atlantic bathyal and abyssal sae i a Bollettino 

Malacologico, Supplement 3 3: 579-840. 

Bucquoy, E., P. Dautzenberg, ae é Dollfus. 1884. Les Mol- 
lusques marins du Roussilon. Vol. I, parte 2: 197-356. 
Bailliére & fils, Paris. 

Chirli, C. 2006. Malacofauna aaa toscana. be 5. Cae- 
nogastropoda Cox, 1859. C. Chirli (ed), 
Firenze, 144 pp. 

Costa, B. 1989. La malacofauna pleistocenica della Cartiera 
Mulino (Vittoria, Ragusa) In: I. Di Geronimo ( (ed). Atti 3° 
simposio Ecologia e Paleoecologia delle comunita benton- 
iche, Catania-Taormina, 12-16 Ottobre 1985; 477-500, 

Coppi, F. 1876. Frammenti di Paleontologia Modenese. Bol- 
lettino del Regio Comitato Geologico d'Italia. Roma. Vol. 
VII, bollettino 1 e 2: 190-209. 

Cossmann, M. 1921. Essais de Paléoconchologie Comparée. 
Douzieme Livraison, Paris, 345 pp. 

Della Bella, G. and D. Scarponi. 2000. Una nuova specie di 
Alvania (Rissoidae) nel Pliocene del Mediterraneo. Gior- 
nale di Geologia, serie. 3°, vol. 62, supplemento: 63-68. 

Doderlein, P. 1862. Cenni geologici intorno la giacitura dei 
terreni miocenici o superiori dell'Ttalia centrale. Estratto 
dagli Atti del X Congresso degli Scienziati Italiani, Siena, 
settembre 1862, 25 pp. ; 


Tavernelle, 


Fischer, P. 1S77. Diagnoses Molluscorum in stratis fossiliferis 
sem 


insulae Rhodi jacentium. Jounal de Conchyliologie, 3 


série, Tome XVII, vol. XXV: 7S—S1. 


Garilli, V., M. Crisci, and R. Messina. 2005a. A new species of 
Gibbula (Gastropoda: Trochidae) from the Pleistocene of 


Killini (north-western Peloponnesus, Greece). Bollettino 
della Societa Paleontologica Italiana 44: 47-53. 

Garilli, V., B. Dell’ Angelo, and E. Vardala-Theodorou. 2005b. 
Polyplacophora from the Pleistocene of Kyllini (NW 
Peloponnese, Greece). Bollettino della Socie ta Paleonto- 
logica Italiana 44: 127-144. 

Garilli, V. and L. Galletti. 2007. Kyllinia parentalis (Neogas- 
tropoda: Turridae), new genus and species from the Plio- 


Pleistocene of the central Mediterranean area. Journal of 


Paleontology $1: 746-753 

Giannuzzi-Savelli, R., F. Pusateri, A. Palmeri and C. Ebreo. 
1996. Atlas of the Mediterranean sea shells. Vol. 2 (Cae- 
nogastropoda part 1: Discopoda-Heteropoda). La Con- 
chiglia, Roma, 258 pp. 

Gofas, S. and A. Warén, 1952. Taxonomie de quelques especes 
du genre Alvania (Mollusca, Gastropoda) des cotes Iberisque 
et Marocaines. Bollettino Malacologico 18: 1-16. 

Gofas, S. and W.F. Ponder. 1991. The habitat and relationships 
of Plagyostila asturiana (Gastropoda, Rissoidae). Bulletin 
du Muséum National d'Histoire Naturelle. Section A, zo- 

13: 39-47. 

Gofas, S. 1999. The West African Rissoidae (Gastropoda: Ris- 


ologie biologie et écologie animales, 


Jeffreys, J. € 


sooidea) and their ees to some European species. 
The Nautilus 113; 78-101. 

Greco, A. 1974. La eoulaebriana pliocenica di contrada Cerausi 
presso Serradifalco (Caltanissetta). Geologica Romana IX: 
975-314. 

Hearty, P. J., G. H. Miller, C. E. Stearns, and B. J. Szabo. 1986. 
Aminostratigraphy of Gu: fe rnary shorelines in the Medi- 
terranean basin. Geological Socie ty of American Bulletin 
97: S5O—-S5S. 

Hubendick, B. 7 A. Warén. 1969. Smasniickor vid svenska 
viistkusten. Sliktet Alvania. Sirtryck ur Goéteborgs 
eau Museums Astryck 1969: 55-61. 

Jeffreys, J. G. 1867. British ¢ ‘onchology, or an account of the 
Mollusca which now inhabit the British isles and the sur- 
rounding seas. Vol. IV. John van Voorst, London, 467 pp. 

1869. British Conchology, or an account of the 
Molluses which now inhabit the British isles and the sur- 
rounding seas. Vol. V. John van Voorst, London, 258 pp. 

Kobelt, W. 1888. Prodromus Faunae Molluscorum Testace- 
orum Maria Europaea Inhabitantium. Bauer u. Raspe, 
Nurnberg, iv + 550 pp. 

Kowalke, T. and M. Harzhauser. 2004. Early ontogeny and 
palaecology of the Mid-Miocene rissoid gastropods of the 
central Parate thys. Acta palaeontologica Polonica 49 (1): 
111-134. 

Landau, B., R. Marquet, and M. Grigs. 2004. The early 
Pliocene eaeopons (Mollusca) of Este »pona, southern 
Spain. Part 2: Orthogastropoda, Neotaenioglossa. Paleon- 
tos 4: 1-108. 

Linden, J. Van der, and W. M. Wagner. 1989. Alvania multi- 
quadrata spec. nov. from the Canary Islands (Gastropoda, 
Prosobranchia: Rissoidae), Basteria 53(1—3): 35-37, 

Linden, J. Van der. 1993. Alvania obsoleta spec. nov. from the 
Azores (Gastropoda, Prosobranchia: Rissoidae). Basteria 

57: 79-82. 

Locard, A. 1856, Catalogue Général des Mollusques vivants de 
France. Mollusques marins. Genéve et Bale, Henry 
Georg, Lyon and Paris, x + 77S pp- 

Lozouet, P. 1998. Nouvelles especes de gastéropodes (Mol- 
lusea: Gastre ypc rda) de IC Jligocene et du Miocene inférieur 
d Aquitaine (sud-onest de la France). Cossmanniana 5: 
61-102. 

Lozouet, P. 1999. Nouvelles espeéces de gastéropodes (Mol- 
lusca: Gastropoda) de POligocéene et du Nia ne intérieur 
d’ Aquitaine (sud-ouest de la France). Partie 2. Cossman- 
niana 6; 16S. 

Michaud, A. L. G. 1831. Descripitions de plusieurs nouvelles 
especes de coquilles du Genre Rissoa (Fréminville), 2.° 
Edition, Imprimerie de F.G, Levrault, Strasbourg, 19 pp., 
1 ph. 

Monegatti, P. and S. Raffi. 2001. Taxonomic diversity and 
stratigraphic distribution of Mediterranean Pliocene bi- 
valves. Palaeogeography, Palaeoclimatology, Palaeoecol- 
ogy 165; 171-193. 

Monterosato, T. 1872. Notizie intorno alle conchiglie mediter- 
ranee, Ufficio Tipogratico di Michele Amenta, Palermo, 
61 pp. 

Monterosato, T. 1875, Nuova rivista delle conchiglie mediter- 
ranee. Atti dell Accademia Palermitana di Scienze Lettere 
ed Arti, If, 5. 50 pp: 

Monterosato, T. IS78. Enumerazione e sinonimia delle con- 
chiglie mediterranee. Giornale di Scienze Naturali ed 
Economiche, Palermo 13: 61-115. 


V. Garilli, 2008 


Page 51 
fo) 


Monterosato, T. 1SS4a. Conchiglie littorali mediterranee. II 
Naturalista Siciliano 3(6): 159-163. 

Monterosato, T. 1SS4b. Nomenclatura generica e specifica di 
alcune conchiglie mediterranee. Stabilimento Tipogratico 
Virzi, Palermo, 152 pp. 

Monterosato, T. 1S89. Coquilles marines marocaines. Journal 
de Conchyliologie 37: 20-40. 

Monterosato, T. 1917. Molluschi viventi e quaternari lingo le 
coste della Tripolitana. Bollettino Societa Geologica Ital- 
iana, serie III, 4: 1-28. 

Moolenbeek, R. G. and M. J. Faber.1987, The Macaronesian 
species of the genus Manzonia (Gastropoda: Rissoidae), 
Parts 1-3. De Kreukel 23:1- 16, 23-31, 166-179. 

Moolenbeek, R.G. and H.J. Hoenselaar.1989. The genus Alva- 
nia on the Canary Islands and Madeira (Molluscs, € yas- 
tropoda), part 1.Bulletin, Zodlogisch Museum, Univer- 
siteit van Amsterdam 11: 215-228. 

Moolenbeek, R.G., H. J. Hoenselaar, and M. Oliverio. 1991. 
The rissoid species described by J.-J. Oberling. Bollettino 
Malacologico 27: 107-120. 

Moolenbeek, R. G. and H. J. Hoenselaar. 1998. The genus 
Alvania on the Canary Islands and Madeira (Mollusca, 
Gastropoda), part 2. Bulletin, ZoGlogisch Museum, Uni- 
versiteit van Amsterdam 16: 53-62. 

Nordsieck, F. 1968. Die europiischen Meeres- Gehiius- 
eschnecken (Prosobranchia). Gustav Fischer Verlag, Stutt- 
gart, i-vili+273 pp. 

Nordsieck, F, 1972. Die europiiischen Meeres- Gehtius- 
eschnecken (Opisthobranchia mit Pyramidellidae, Rissoa- 
cea). Gustav Fischer Verlag, Stuttgart, i-xiii+327pp. 

Pallary, P. 1920. Exploration scientifique du Maroe organisée 
par la Société de Géographie de Paris et continuée par la 
Société des Sciences Naturelles du Maroc. Deuxiéme fas- 
cicule. Malacologie (1912). Larose, Rabat et Paris, 108 pp. 

Palazzi, S. 1997. Un problema ancora... spinoso. I Rissoidi del 
Mediterraneo. La Conchiglia. Anno XXIX, n° 285: 35-40. 

Palazzi, S. 2003. Loose notes, 2. Biosophia Archive 368. Avail- 
able at http:/Avww.dodoline.it/biosophia/docs/mollusca/ 
mollusca.htm. 

Pérés. J. M. and J. Picard. 1964. Nouveau manuel de bionomie 
benthique de la mer Méditerranée. Recueil des Travaux 
de la Station Marine d’Endoume, Faculté des Sciences de 
Marseille 31: 1-137. 


Philippi, R.A. 1836. Enumeratio molluscorum Siciliae cum 
viventium tum in tellure terziaria fossilium, quae in itinere 
suo observavit. Vol. I: Schropp, Berolini, Berlino, xiv + 267 
pp: 

Philippi, R.A. 1S44. Enumeratio Molluscorum Siciliae cum 
viventium tum in tellure tertiaria fossilium, quae in itinere 
suo observavit. Vol. 2, Eduard Anton, Halis Saxonum 
(Halle), iv + 303 pp. 

Piani, P. 1979. Rissoacea mediterranei. Digesta I. Le specie 
mediterranee del genere Galeodina Monterosato, 1884. 
Bollettino Malacologico 15: 67-73. 

Rolan, EB. and F. Fernandes. 1990. Tres nuevas especies del 
género Manzonia (Mollusca, Gastropoda) para la costa 
Occidental de Africa. Publicagdes Ocasionais da Sociedade 
Portuguesa de Malacologia 15: 63-68. 

Ruggieri, G. and G. Milone. 1973. La Macrofauna del Tirre- 
niano di Tormmaso Natale. Bollettino della Societa Pale- 
ontologica Italiana 12: 217-222. 

Ruggieri, G. and M. Unti. 1988. Una malacofauna del Tirre- 
niano (Pleistocene Superiore ) ) di Birgi Nuovo ( (Trapani). ). 1] 
Naturalista Siciliano XII, serie quarta: 19-32. 

Sacco F., 1895. I molluschi dei terreni terziarii del Piemonte e 
della Liguria. Parte XVHI, Carlo Clausen, Torino, 51 pp. 

Seguenza, L. 1903. Rissoidi neogenici della provincia di Mes- 
sina. Palaenthographia Itelicg Vol. IX: 35-60. 

Speyer, O. 1864. Die Tertiirfauna von Séllingen bei Jerxheim 
im Herzogthum Braunschweig. Paleontographica 9: 247- 

338. 

Verduin, A. L977. On a remarkable dimorphism of the apices in 
many groups of sympatric, closely related marine gastro- 
pod species. Basteria 41: 91-95. 

Warén, A. 1973. Revision of the Rissoidae from the Norwegian 
North Atlantic expedition 1576-78. Sarsia 53: 1-14. 
Warén, A. 1974. Revision of the Arctic-Atlantic Rissoidae (Gas- 

tropoda, Prosobranchia). Zoologica Scripta 3: 121-135, 

Weinkauftf, H.C. 1888. Die Gattungen Rissoina und Rissoa. 
Verlag von Bauer & Raspe, Nurnberg, 205 pp. 

Wenz, W. 1935. Gastropoda, Teil 1: Allgemeiner Teil und 
Prosobranchia. In: O.H. Schindewolf. Handbuch der 
Paliiozoologie, Band 6. Gebr. Borntraeger, Berlin, vi + 
1639 pp. 


THE NAUTILUS 122(1):52-55, 2008 


Page 52 


Reallocation of Cyamiocardium crassilabrum Dell, 1964, into 
Perrierina Bernard, 1897 (Bivalvia: Cyamiidae) 


Diego G. Zelaya 

Departamento Biodiversidad y Biologia 
Experimental 

Facultad de Ciencias Exactas y Naturales, 
U.B.A. 

and 

Division Zoologia Invertebrados 

Museo de La Plata, Paseo del Bosque s/n, 
1900 

La Plata, ARGENTINA 
dzelaya@tenym.unlp.edu.ar 


ABSTRACT 


Re-examination of the holotype of Cyamiocardium crassila- 
brum Dell, 1964, allowed confirmation that the species was 
wrongly allocated to the genus Cyamiocardium. This material, 
pe with additional specimens from the westemm Malvinas 

Falklands) Islands allowed for a re-description of that species. 
The species is also properly illustrated through scanning elec- 
tron microscopy, and its update sd generic place ment discussed. 
The presence of a “taxodont” hinge plate is the main morpho- 
logical character supporting the re-allocation of this species in 
the genus Perrierina Bernard, 1897. Information on the brood- 
ing condition of the species is provided. 


Additional Keywords: Perrierina, Cyamiidae, Magellanic Region 


INTRODUCTION 


Study of the mollusks collected by the R/V WILLIAM 
Scoresby in the Magellan Region and adjacent Antarctic 
waters resulted in he discovery of several species new to 
this region, among them, a small bivalve described by 
Dell (1964) as Cyamiocardium crassilabrum. The only 
subsequent record of this species was reported by Cas- 
tellanos (1980) from the western Malvinas (Falkland) Is- 
lands. 

The genus Cyamiocardium was introduced by Soot- 
Ryen (1951) to allocate C yamium dentic ulatum BE. A. 
Smith, 1907 (the type species by original designation), a 
species widely distributed throughout the Antarctic Re- 
gion (Lamy, 1910, 1911; Powell, 1958; Dell, 1990). Other 
Antarctic species assigned to this genus by Soot-Ryen 
(1951) were Cyamiocardium rotundatum (Thiele, 1912), 
C. dahli Soot-Ryen, 1957, and Cyamiocardium crassila- 
brum Dell, 1964, from the Magellan Region. In the con- 
text of a systematic revision of these species (currently in 


progress), the type material and additional specimens of 
“Cyamiocardium” crassilabrum were examined. As a re- 
sult, I have concluded in that the placement of this spe- 
cies in Cyamiocardium is incorrect. In the present paper 
the generic placement of this species is revised, and the 
taxon is re-described and properly illustrated by first 
time. 


MATERIALS AND METHODS 


This study is based on dried preserved specimens col- 
lected during the R/V SHINKAT Maru expedition (1978— 

79). Voucher specimens have been deposited in the col- 
lections of the Museo de La Plata (MLP). Photographs of 
the holotype of Cyamiocardium crassilabrum were re- 
ceived from The Natural History Museum (NHM), Lon- 
don. For comparative purposes, specimens of Cyamio- 
cardium denticulatum, C. rotundatum, and C. dahli were 
also examined. 

The specimens reported by Castellanos (1980) as Cya- 
miocardium crassilabrum could not be located either at 
the MLP or the Museo Argentino de Ciencias Naturales 
“Bernardino Rivadavia” (MACN). The material here 
studied, originally at the Instituto Nacional de Investiga- 
ciones y Des: urrollo Pe ssquero (INIDEP), comes from the 
same SHINKAL MARU sampling station than the material 
previously reported by Castellanos (1980). 

Shell measurements were made under a stereoscopic 
microscope, according to the following criteria: L: maxi- 
mum antero-posterior distance; H: maximum dorso- 
ventral clistance, perpe ndicular to L: W: maximum dis- 
tance across valves, perpendicular to H. Shell morphol- 
ogy Was studie fal unde ‘y scanning ¢ le sctron mic roscope; for 
this, Philips XL30 TMP and JEOL JSN-6360 LV scan- 


ning Cc le ke microscope “Ss Were USe ad. 


D. Zelaya, 2008 


SYSTEMATICS 


Cyamiidae G. O. Sars, 1S7S 
Perrierina Bernard, 1897 


Perrierina crassilabrum (Dell, 1964) new combination 
(Figures 1-12) 

Cyamioc — crassilabrum Dell, 1964: 204, fig. 
numbers 1, 2, and pl. 6, figs. 1, 2; Castellanos, 1980: 13 


Type Locality: 50°17’ S, 60°06" W, station 211, R/V 
WILLIAM SCORESBY, 161-174 m 


Material Examined: Photographs of the holotype 
(NHM 1962863); 7 dried specimens and | valve, 51° 29’ 
S, 61° 50’ W, Malvinas (Falkland) Islands, 192 m (MLP 
12606). 


Literature Records: 50°17’ S, 60°06’ W, station 210, 
R/V WILLIAM Scoresby, 161 m; 50°35’ S, 57°20’ W, 
station 229, R/V WILLIAM Scoresby, 210-271 m (Dell, 
1964); 51°29" S, 61°50’ W, 192 m (Castellanos, 1980). 


Distribution: Only known from the Atlantic sector of 


the Magellan Region, in the vicinity of the Malvinas 


(Falkland) Islands, 161-271 m. 


Description: Shell solid, small (maximum L: 6.3 mm), 
outline nearly circular (H/L = 0.99 + 0.01, n = 6), sli ghtly 
inequilateral, not inflated (W/H = 0.65 + 0.01, n = 5), 
whitish, glossy. Anterior margin short, curved, a 
ous with ventral margin, Ww hich is ev enly arcuate; poste- 
rior end rounded, slightly expanded (Figures 1, 2). Juve- 
nile shell slightly ovate in outline, elongated antero- 
posteriorly (Figure 3 3). Beaks full, chbeennil: directed 
slightl y anteriorly. Baiwaoneh ovate, smooth, about 590 
wm in diameter (Figure 5). Shell surface sculptured with 
rounded, strongly marked and regularly spaced radial 
cords, 50-60 in larger specimens (Figure 6); cords sepa- 
rated by interspaces wider than ribs. Regularly sepa- 
rated, microscopic commarginal threads also present. 
Radial sculpture also evident on inner margins, where 
they produce prominent crenulations (Figures 7-9). 
Hinge plate narrow, somewhat broader anterior to beaks, 
where the cardinal teeth are inserted. Right valve with 
large, hooked cardinal 3: 3a high, solid, triangular, en- 
larged at base, which is Say bifid: 3b. delicate, 
narrow, nearly straight, one-third of size of 3a (Figures 7, 
10). Left valve with prominent triangular cardinal 2, thin 
and styliform cardinal 4b behind resilifer, and : slender 
but solid, nearly straight anterior tooth (referred to as 
“cardinal 4a” by Berard (1597) and as “anterior lateral 
IP” by Lamy 1917)) (Figures 8, 9, 11). Both valves with 
two series of tubercles anterior and posterior to beaks, 
producing the appearance of a taxodont hinge plate (Fig- 
ures 10, 11); each series composed of four tubercles; 
posterior series stronger than anterior in juveniles ( Fig- 
ure 12). In both series, tubercles diminish in size and 
degree of development from beaks to anterior and pos- 
terior ends, where they are followed by marginal crenu- 
lations. Internal ligament somewhat solid, located in 


short, oblique resilifer posterior to cardinal teeth. Exter- 
nal ligament short. 


Biological Observations: One of the specimens ex- 
amined contained numerous embryos between the as- 
cending and descending lamellae of inner demibranchs. 
E mbryos were in different stages of development, rang- 
ing from incipient (unshelled) to well-developed speci- 
mens, the latter numbering 32, and reaching 700 pm in 
diameter (Figure 4). These observations are consistent 
with that by Dell (1964), who reported 44 “developing 
young” contained in one of the specimens he studied. 


Remarks: = Perrierina was proposed by Bernard (1897) 
for P. taxodonta (type species by monotypy; illustrated by 
Bernard, 1897; fig. 3) a species he described from fle 
Stewart (New Zealand). The description of the genus was 
merged with that of the type species, in which Bernard 
(1897) described the presence of several “lamellae” an- 
terior and posterior to the cardinal teeth, resembling a 
taxodont hinge. This character, infrequent among Cy a- 
mioidea, is also present in Legrandina Tate aed May, 
1901 (type species: L. beriordr Tate and May, 1901, by 
original designation), an Australian genus regarded by 
Ponder (1971) and Powell (1979) as a subgenus of Per- 
rierina. 

Perrierina and Cyamiocardium have as common char- 
acters the number, morphology, and arrangement of 
hinge teeth, ie., the presence of a hooked anterior car- 
dinal 3 in the right valve and two prominent cardinal 
teeth (cardinals 2 and 4) in the left valve. The specimens 
studied here generally agree with these characteristics, 
but, additionally, the hinge examined showed several an- 
terior and posterior dubeniles which, as a group produce 
the appearance of a taxodont hinge, a diagnostic charac- 
ter for Perrierina that is absent in Cyamiocardium. Fur- 
thermore, the studied specimens lack the posterior 

“pseudo-lateral tooth” described by Smith (1907) for the 
type species of Cyamiocardium. Another character that 
is useful in separating the genera Cyamiocardium and 
Perrierina is the presence in the former of mantle margin 
papillae (Soot-Ryen, 1951). According to Ponder (1971), 
mantle margin papillae are absent in Perrierina. Untor- 
tunately, the poor prese rvation of the specimens on 
which this paper is based made it impossible to confirm 
the state of this character. Nevertheless, the differences 
in hinge morphology call for the reallocation of “Cya- 
miocardium” crassicostatum into Perrierina. 

Perrierina crassilabrum shows some characteristics in- 
termediate between the subgenera Perrierina and Leg- 
randina. The presence of prominent beaks is a charac- 
teristic shared with Perricrina, whereas a hinge plate 
broader at the base of the cardinals is characteristic of 
Legrandina. Furthermore, there are some other features 
of P. crassilabrum not previously known for other species 
of Perrierina or Legrandina, such as the relatively large 
adult size, the nearly circular shell outline of adults, and 
the strong radial ornamentation. 


THE NAUTILUS, Vol. 122, No. 1 


Page 54 


Perrierina crassilabrum. 1,9. Holotype (NHM 1962863). 1. External view of left valve. 9. Internal view of left valve. 


Figures I-12. 
2-8, 10-12. Specimens from 51° 29'S, 61°50! W, SEM micrographs. 2. External view of an adult specimen, right valve. 3. External 


iew of a juvenile. 4. External views of larvae removed from an adult. 5. Dorsal view, showing protoconch and hinge plate 6. Detail 
of shell microsculpture. 7. Inner view of right valve. 8. Internal view of left valve, 10, 11. Detail of adult hinge plates. 10. Right valve 
anterior tooth, 12. Detail of hinge plate of a juvenile 


ri 


sb = cardinal teeth, L = ligament. 11. Left valve. 2, 4b = cardinal teeth, at 
Seale bars: 1-3, 7-9 = 1 mm; 4 = 100 wm: 5 = 200 jam: 6 = 50 wm; LO-12 = 500 pom 


of 2.3 mm length 


D. Zelaya, 2008 


Page 55 


The familial placement of Perrierina is somewhat con- 
fusing; the genus was a sssively placed within the 
Mactridae ( Bernard, 1897), Le apeont ie (Dall, 1899), and 
Crassatellidae (Suter, nae amy, L917). Later, Mar- 
wick (1928) regarded the “taxodont” lamellae of the 


hinge as a character sufficient to warrant proposition of 


the family Perrierinidae, but this family was subse- 
quently regarded as a synonym of C yamiidae ( (e.g. Cha- 
van oe Thiele, (1934), Fle sming (1948), Ponder 
(1971), and Powell (1979), The Cyamiidae is a family 
eee well-diversified in the Magellanic Region, 
where a total of 12 species belonging to Cyamium, Cya- 
miocardium, Gaimardia, and Kidderia have been re- 
ported and recognized as valid by Zelaya (2005). The 
present paper represents the first record in M: agellanic 
waters of a species of Perrierina, a genus thus far known 
only from Australia, Tasmania, and New Zealand. 


ACKNOWLEDGMENTS 


The author is grateful to A. Roux (INIDEP) who kindly 
made available the material collected during the R/V 
SHINKAI MARU cruise; K. Way and A. MacLellan (NHM) 
kindly sent photographs of the holotype of Cyamiocar- 
dium crassilabrum: C. Digiani (MLP) helped with older 
literature: and C. Ituarte provided valuable criticism and 
suggestions on an early version of the manuscript. 

The author is member of the National Research Coun- 
cil for Science and Technology (CONICET), Argentina. 
This study was partially supporte ed by a grant froth the 
Western Society of Malacologists, made possible by the 
Santa Barbara Malacological Society, the Southwestern 
Malacological Society, the San Diego Shell Club, and the 
Northern California Malacological Club, and by grant 
PICT 282 from “Agencia Nacional de Promocién Cientf- 
fica y Tecnolégica,” Argentina. 


LITERATURE CITED 


Bernard, F. 1897. Sur quelques coquilles des Lamellibranches 
de Il fle Stewart. Bulletin du Muséum d’ Histoire Na- 
turelle 7: 309-314. 

Castellanos, Z. A. de. 1980. Micromoluscos poco conocidos del 
sur argentino-chileno, Neotropica 25 (74): 133-140. 
Chavan, A. 1969. Superfamily Cyamiacea. Pp. 537-543. In: 

Leslie Reginald Cox et al., Part. N [Bivalvia], part. 6, vols. 
1 and 2: xxxvii + 952 pp. Treatise on invertebrate paleon- 
Sieg Lawrence, Kansas. Geological Society of America 

& University of Kansas. 
Coan, E. V., P. V. Scott y F. R. Bernard. 2000. Bivalve seashells 


of Western North America. Santa Barbara Museum of 


Natural History. Monographs number 2, Studies in Biodi- 
versity n°2, 764 pp. 


Dall, W. H. 1899. Synopsis of the Recent and Tertiary Lep- 
tonacea of North America and the West Indies. Proceed- 
ings of the United States National Museum 21: 873-897, 
pls. S7-SS. 

Dell, R. kK. 1964. Antarctic and sub-Antarctic Mollusca: Am- 
phineura, Scaphopoda and Bivalvia. Discovery Reports 33: 
eee pls. 2-7, 

Dell, R. 1490. Antarctic Mollusca with special reference to 
the aan of the Ross Sea. Bulletin of the Royal Society of 
New Zealand 27: 1-311. 

Egorova, E. N. 1982. Biological results of the Soviet Antarctic 
Expeditions, 7. Explorations of the fauna of the seas 26 
(34); 1-143. 

Fleming, C. A. 1948. New species and genera of marine Mol- 
lases a from the Southland a Transactions of the Royal 
Society of New Zealand 77 (1): 72-92. 

Lamy, E. 1910. Mission dans ee dirigée par M.Le Dr 
Charcot (1905-1910), collections recueillies par M. le Dr. 
Jacques Liouville. Bulletin du Muséum National 
d Histoire Naturelle 7: 388-394. 

Lamy, E. 1911. Gastéropodes prosobranches, scaphopodes et 
pélécypodes. Deuxieéme Expédition Antarctique Francaise 
(1908-1910) commmandée par le Dr. J. Charcot. Sciences 
Naturelles: Documents Scientifiques. 32 pp., pl. 1 

Lamy, E. 1917. Révision des Crassatellidae vivants du Muséum 
dHistoire Naturelle de Paris. Journal de Conchyliologie 62 
(4): 197-270 + pl. 6. 

Marwick, J. 1928. The Tertiary Mollusca, Catham Islands. 
Transactions and Proceedings of the New Zealand Insti- 
tute 58 (4): 432-506. 

Ponder, W. F. 1971. Some New Zealand and Subantarctic bi- 
valves of the Cyamiacea and Leptonacea with desc riptions 
of new taxa. Records of the Dominion Museum 7 (13): 
119-141. 

Powell, A. W. B. 1951. Antarctic and Subantarctic Mollusca: 
Pelecypoda and Gastropoda collected by the ships of the 
Discovery Committee during the years 1926-1937. Dis- 
covery Reports 26: 49-196, pls. 5 5-10. 

Powell, A. W. B. 1958. Mollusca from the Victoria-Ross quad- 
rants of Antarctica. B.A.N.Z.A.R. Expedition B, 6: 165— 
215. 

Powell, A. W. B. 1979. New Zealand Mollusca. Marine, land 
and freshwater shells. Auckland, William Collins Publish- 
ers, 500 pp. 

Smith, E. A. 1907. Mollusca. 5. Lamellibranchiata. National 
Antarctic Expedition 1901-1904. Natural History 2, Zool- 
ogy (Vertebrata, Mollusca, Crustacea): 1-6. 

Soot-Ryen, T. 1951. Antarctic Pelecypods. Scientific Results of 
the Norwegian Antarctic Expeditions 1927-1925, 32: 1-46 
+ 1 pl. 

Suter, H. 1913. Manual of the New Zealand Mollusca. John 
MacKay, Government Printer, Wellington. 1120 pp. 
Thiele, J. 1934. Handbuch der sy stematischen Weichtierkunde. 
Gustav Fischer Verlag, Jena. Bd 3, Dritter Teil, pp.1193- 

1528. 

Zelaya, D. G. 2005. The bivalves from the Scotia Are islands: 
species richness and faunistic affinities. Scientia Marina 69 
(suppl. 2): 113-122. 


THE NAUTILUS 122(1):56, 2008 Page 56 


Notice 


THE 2008 R. T. ABBOTT VISITING CURATORSHIP 


The Bailey-Matthews Shell Museum is pleased to invite applications for the 2008 R. T. Abbott Visiting Curatorship. 


The Curatorship, established originally in accordance with the wishes of the late Dr. R. Tucker Abbott, Founding Director of the 
Shell Museum, is awarded annually to enable malacologists to visit the museum for a period of one week. Abbott Fellows are 
expected, by performing collection-based research, to assist with the curation of portions of the Museum’s collection and to provide 
one evening talk for the general public. The Museum collection consists of marine, freshwater, and terrestrial specimens. A large 
percentage of our holdings have been catalogued through a computerized database management system; part of the catalogue is 
already available for searches online at: www.shellmuseum.org/collection.html. A substantial portion of the time will be available for 
research in the collection, but field work in southwest Florida can be arranged. The R. T. Abbott Visiting Curatorship is accompanied 
by a stipend of $1,500. 


Interested malacologists are invited to send a copy of their curriculum vitae together with a letter detailing their areas of taxonomic 
expertise and research objectives, and to provide a tentative subject for their talk. Send materials to: 


Dr. José H. Leal, Director 

The Bailey-Matthews Shell Museum 
P.O. Box 1550 

Sanibel, FL 33957 


jleal@shellmuseum.org 


Applications for the 2008 Visiting Curatorship should be sent electronically to the above e-mail address no later than May 15, 2008, 
or postmarked by that date if sent by regular mail. The award will be announced by mid-June 2008. Questions about the Visiting 


Curatorship should be sent to the e-mail address above, or by phone at: 


(239) 395-2233; fax (239) 395-6706 


Sponsored in part by the State of Florida, Department 
of State, Division of Cultural Affairs, the Florida Arts 
Council and the National Endowment for the Arts 


v 


ENDOWMENT 
FOR THE ARTS 


Ibi 


Oi 


| 


| 


‘il 


5 WHSE 05828 


INSTRUCTIONS TO AUTHORS 


THE NAUTILUS publishes articles on all aspects of the 
biology, Balbentatndy: and systematics of mollusks. 
Manuscripts describing original, unpublished research 
and review articles will be considered. Brief articles, not 
exceeding 1000 words, will be published as notes and do 
not require an abstract. Notices of interest to the mala- 
cological community will appear in a notices section. 


Manuscripts: Each original manuscript and accompa- 
nying illustrations should be submitted to the editor pref- 
erably via e-mail or as hardcopy in triplicate. 

Text must conform to the dimensions of 82 x 11-inch 
paper, double-spaced, and single-column throughout (in- 
cluding literature cited, tables, and figure captions). Au- 
thors should follow the general recommendations of Sci- 
entific Style and Format—The CSE Manual for Authors, 
Editors, and Publishers, available from the Council of 
Science Editors at www.councilscienceeditors.org. The 
first mention of a scientific name in the text should be 
accompanied by the taxonomic authority, including year. 
Latinized names and other words to be printed in italics 
must be underlined; leave other formatting indications to 
the editor. Metric, not English, units are to be used. Geo- 
chronologic modifiers should be capitalized only when 
units are formally recognized: for instance, use Late Cre- 
taceous but early Miocene. Likewise, only modifiers of 
formally recognized chronostratigraphic units are capi- 
talized: use Lower Jurassic but upper Oligocene. 

The sequence of sections aadld be title page, ab- 
stract, introduction, materials and methods, results, dis- 
cussion, acknowledgments, literature cited, tables, figure 
captions, figures. The title page should include the title, 
author’s name(s) and address(es). If corresponding au- 
thor is not the senior author, please indicate. The ab- 
stract should summarize in 250 words or less the scope, 
main results, and conclusions of the article. Abstracts 
should be followed by a list of additional key words. All 
references cited in the text must appear in the Literature 
Cited section and vice-versa. Please follow a recent issue 
of THE NAUTILUS for bibliographic style, noting that 
journal titles must be unabbreviated. Information on 
plates and figures should be cited only if not included 
within the pagination of cited work. Tables must be num- 
bered and each placed on a separate page. If in doubt, 
please follow a recent issue of the journal for sequence of 
sections and other style requirements. 


Illustrations: [llustrations are rendered either at full- 
page width (maximum width 17 cm) or column width 
(maximum width 8.2 cm). Please take these dimensions 
into consideration when preparing illustrations. Page- 
width illustrations ideally eeald span the entire width of 
printed page (17 cm). “Tall” page-width illustrations 
should be avoided, square or “landscape” formats work 
better. Please design plates accordingly, such that there 
will be enough space Heft at the bottom of printed page 
for plate caption. (Digital technology has made this task 
much easier.) 


All line drawings must be in black, clearly detailed, 
and completely labeled. Abbreviation definitions must be 
included in the caption. Line drawings must be high 
resolution files at least 600 dpi (dots per inch) resolution 
at actual size. Standard digital formats for line drawings 
include .tif, .bmp, .psd, .eps, and .pdf. 

Photographs may be submitted in black-and-white or 
color, preferably in RGB mode if in color. Standard digi- 
tal formats for photographs include .tif, -psd, .jpg, or .pdf. 
Photographs must be high resolution files at least 300 dpi 
resolution at actual size. 

If more than one figure is included in an illustration, 
all figures are to be consecutively numbered (Figures 1, 
2135010) NOT Pigures ‘lAy. VB 1G). 22) NOR Plate 1, 
Figure 1, .. .). In illustrations with more than one figure, 
make sure that blank areas between figures is kept to a 
minimum, thereby allowing for more area for each indi- 
vidual figure. 

Compressed files (e.g., jpg) may be used to facilitate 
transmission of files during original submission, but may 
not be acceptable at final submission (see below). 


Voucher Specimens: Deposition of the holotype in a 
recognized institutional, public collection is a require- 
ment for publication of articles in which new species- 
level taxa are described. Deposition of paratypes in in- 
stitutional collections is strongly encouraged, as is the 
deposition of representative voucher specimens for all 
other types of research work. 


The Editorial Process: Upon receipt, all manuscripts are 
assigned a number and acknowledged. The editor reserves 
the right to return manuscripts that are substandard or 
not appropriate in scope for THE NAUTILUS. Manu- 
scripts deemed appropriate for the journal will be sent 
for critical review to at least two reviewers. The review- 
ers’ recommendations will serve as basis for rejection or 
continuation of the editorial process. Reviewed manu- 
scripts will be sent back to authors for consideration of 
the reviewers’ comments. The revised version of the 
manuscript may at this point be considered accepted for 
publication by the journal. i 


Final Submission: Authors of accepted manuscripts are 
required to submit a final version via e-mail to the editor 
at jleal@shellmuseum.org. Please do not send low-resolu- 
tion or compressed illustration files at this stage. Send any 
files larger than 20 Mb on a CD or DVD to the editor. 


Proofs: After typesetting, proofs will be sent to the au- 
thor. Author should read proofs carefully and send cor- 
rections to the editor within 48 hours. Changes other than 
typesetting errors will be charged to the author at cost. 


Offprints: An order form for offprints will accompany 
the proofs. Offprints will be ordered through the Lae 
Authors with institutional, grant, or other research sup- 
port will be asked to pay for page charges at the rate of 
$60 per page. 


@ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper) 


THE NAUTILUS 


Volume 122, Number 2 
June 25, 2008 
ISSN 0028-1544 


A quarterly devoted 
to malacology. 


EDITOR-IN-CHIEF 

Dr. José H. Leal 

The Bailey-Matthews Shell Museum 
3075 Sanibel-Captiva Road 
Sanibel, FL 33957 


BUSINESS MANAGER 


Mary Jo Bunnell 

The Bailey-Matthews Shell Museum 
3075 Sanibel-Captiva Road 
Sanibel, FL 33957 


EDITOR EMERITUS 


Dr. M. G. Harasewych 
Department of Invertebrate Zoology 
National Museum of 

Natural History 
Smithsonian Institution 
Washington, DC 20560 


CONSULTING EDITORS 
Dr. Riidiger Bieler 
‘ Departnient of Invertebrates 
Field Museum of 
Natural History 
Chicago, IL 60605 


Dr. Arthur E. Bogan 

North Carolina State Museum of 
Natural Sciences 

Raleigh, NC 27626 


Dr. Philippe Bouchet 

Laboratoire de Biologie des 
Invertébrés Marins et Malacologie 

Muséum National d'Histoire Naturelle 

55, rue Buffon 

Paris, 75005 France 


Dr. Robert H. Cowie 

Center for Conservation Research 
and Training 

University of Hawaii 

3050 Maile Way, Gilmore 409 

Honolulu, HI 96822 


Dr. Robert T. Dillon, Jr. 
Department of Biology 
College of Charleston 
Charleston, SC 29424 

Dr. Eileen H. Jokinen 
$234 E. North Shore Road 
Sault Ste. Marie, MI 49783 


Dr. Douglas S. Jones 

Florida Museum of Natural History 
University of Florida 

Gainesville, FL 32611-2035 


Dr. Harry G. Lee 
4132 Ortega Forest Drive 
Jacksonville, FL 32210 


Dr. Charles Lydeard 

Biodiversity and Systematics 
Department of Biological Sciences 
University of Alabama 

Tuscaloosa, AL 35487 


Bruce A. Marshall 
Museum of New Zealand 
Te Papa Tongarewa 
P.O. Box 467 
Wellington, NEW ZEALAND 


Dr. James H. McLean 
Department of Malacology 
Natural History Museum 
of Los Angeles County 
900 Exposition Boulevard 
Los Angeles, CA 90007 


Dr. Paula M. Mikkelsen 

Paleontological Research 
Institution 

1259 Trumansburg Road 

Ithaca, NY 14850 


Dr. Diarmaid O Foighil 

Museum of Zoology and Department 
of Biology 

University of Michigan 

Ann Arbor, MI 48109-1079 


Dr. Gustav Paulay 

Florida Museum of Natural History 
University of Florida 

Gainesville, FL 32611-2035 


Mr. Richard E. Petit 
P.O. Box 30 
North Myrtle Beach, SC 29582 


Dr. Gary Rosenberg 

Department of Mollusks 

The Academy of Natural Sciences 
1900 Benjamin Franklin Parkway 
Philadelphia, PA 19103 


Dr. Angel Valdés 
Department of Malacology 
Natural History Museum 
of Los Angeles County 
900 Exposition Boulevard 
Los Angeles, CA 90007 


Dr. Geerat J. Vermeij 
Department of Geology 
University of California at Davis 
Davis, CA 95616 


Dr. G. Thomas Watters 
Aquatic Ecology Laboratory 
1314 Kinnear Road 
Columbus, OH 43212-1194 


SUBSCRIPTION INFORMATION 


The subscription rate per volume is 
US $47.00 for individuals, US $80.00 
for institutions. Postage outside the 
United States is an additional US 
$10.00 for regular mail and US 
$25.00 for air delivery. All orders 
should be accompanied by 
payment and sent to: THE 
NAUTILUS, P.O. Box 1580, 
Sanibel, FL 33957, USA, (239) 


395-2233. 


Change of address: Please inform 

the publisher of your new address at 
: Z 

least 6 weeks in advance. All 

communications should include both 

old and new addresses (with zip 

codes) and state the effective date. 


THE NAUTILUS (UISSN 0028-1344) 
is published quarterly by The Bailey- 
Matthews Shell Museum, 3075 
Sanibel-Captiva Road, Sanibel, FL 
33975. 


Periodicals postage paid at Sanibel, 
FL, and additional mailing offices. 


POSTMASTER: Send address 
changes to: THE NAUTILUS 
P.O. Box 1580 

Sanibel, FL 33957 


NAUTILUS 


Volume 122, Number 2 
June 25, 2008 
ISSN 0028-1544 


ay te 


CONTENTS 


Luiz Ricardo L. Simone Revision of the genus Spinosipella (Bivalvia: Verticordiidae ), with 
Carlo M. Cunha descriptions of two new species from Brazil .. 2... 0.000.002 57 
Jonathan Hendricks Late Eocene Conus (Neogastropoda: Conidae) from Florida, USA... 2... 79 


Roger W. Portell 


Janine O. Arruda Synonymization of Neohyalimax Simroth, 1596, and Omalonyx dOrbigny, 
José W. Thomé 1837, with a re Seine en of Omalonyx brasiliensis (Simroth, 1896) 
(Gastropoda: Succineidae) .. 2... 26. eee 94 
Roland Houart Rehabilitation of Ergalatax martensi (Schepman, 1892) (Gastropoda: 
Muricidae), senior synonym of Ergalatax obscura Houart, 1996, and description 
of Ergalatax junionae, new name for Morula martensi Dall, 1923 ........ 99 
Guido Pastorino Two new deep-sea muricids (Gastropoda) from Argentina... .......... 107 


Fabrizio Scarabino 


MBLWHOI Library 
JUN 2 & 9Ang 


WOODS HOLE 
Massachusetts 02543 


THE NAUTILUS 122(2):57-78, 2008 


Page 57 


Revision of the genus Spinosipella (Bivalvia: Verticordiidae), 
with descriptions of two new species from Brazil 


Luiz Ricardo L. Simone 

Carlo M. Cunha! 

Museu de Zoologia da Universidade de 
Sao Paulo 

Caixa Postal 42494 

04299-970 Sao Paulo, BRAZIL 
Irsimone@usp.br 
‘carlomagenta@gmail.com 


ABSTRACT 


A revision of the deep-water verticordiid genus Spinosipella is 
provided, based on conchological and anatomical characters. 
The genus is considered distinct from Verticordia (of which it 
was considered a subgenus) based on the strong ribs, prickly 
surface, reduction of lunula, relative large size, weakly spiral 
valve shape, and other ees The following species are 
considered in the genus: (1) Spinosipella agnes new species, 
ranging from Florida, U Ne to Rio de Janeiro, Brazil, and also 
including the Porcupine Abyssal Plain in the North Atlantic; (2) 
S. tinga new species, occurring from Rio de Janeiro to ie 
Grande do Sul, Brazil: S. ac a (Philippi, 1544), 
Pliocene fossil from ee es Italy; (4) S. deshayesiana (Fis. 
cher, 1862), from south and central ene. (S. ericia Hed- 
ley, 1911, the type species of the genus, was oe astes to bea 
new synonym of S. deshayesiana); and (5) S. costeminens 
Poutiers, 1981), from the tropical w es a The five species 
differ mainly in conchological details of the number and size of 
ribs, of the prickly sculpture, shape of the shell, of the hinge 
and the degree of convexity. Anatomical description is also 
provided for the two Pacific species, which differ among them- 
selves mainly by the size of the pair of renal folds. From the 
standpoint of anatomical characters, the more significant are: 
the wide lithodesma; the elongation of the auricles, crossing the 
roof of pallial cavity; a tall digital fold in posterior region of 
supraseptal chamber; the low but wide palps: the muscular, 
gizzard-like stomach; the complete separation of both constitu- 
ents of the hermaphroditic gonad (a ventro-posterior testicle 
and a centro-dorsal ovary), and a complete fusion of the visceral 
ganglia. 


Additional Keywords: Mollusca, Anomalodesmata, Septibran- 
chia 


INTRODUCTION 


The Verticordiidae is a family of Ag epee bivalves 
comprised of carnivorous and mostly d deep-water species. 
They are typically small (less than 10 mm) but some 
species reach 30-40 mm. They are mostly radially sculp- 
tured and usually have nacreous inner surface. 


The genus Spinosipella Iredale, 1930 (type Verticordia 
ericia Hedley, 1911, by original designation) is usually 
considered a subgenus of Vertieordia Sowerby, 1844 
(e.g., Thiele, 1934; Moore, 1969; Abbott and Dance, 
1983). The genus encompasses species with shell having 
prickly outer surface, lunule very reduced, thick walls. 
and generally larger size (up to 30 mm). In addition to 
the type species, S. ericia, three other species are cur- 
rently included in this genus, S. acuticostata (Philippi, 
1844), from Atlantic and Mediterranean (middle Tertiary 
to Recent); S. deshayesiana (P. Fischer, 1862a) and S. 
costeminens (Poutiers, 1981), from Indo-Pacific. Some 
authors have considered S. deshayesiana as an Indo- 
Pacific occurrence of S. acuticostata (e.g., Nobre, 1936; 
Crozier, 1966; Rosenberg, 2005). 

Examination of worldwide samples, with an emphasis 
on the Western Atlantic, showed that two species actually 
exist in the Atlantic. Both are separate from the fossil S. 
acuticostata. In addition, it was possible to reorganize the 
Indo-Pacific species, mainly because of the abundant 
material deposited at the Muséum national d'Histoire 
naturelle, Paris (MNHN), which results from several ex- 
peditions. A revision of the taxonomy and a necessary 
re-definition of taxa are provided in this paper, as part of 
a larger project revising Western Atlantic mollusk tax- 
onomy, based on morphology. 


MATERIALS AND METHODS 


A detailed list of the material examined follows each 
species description. Specimens generally belong to mu- 
seum collections. Most material consists of shells exam- 
ined under a stereomicroscope. Some few Pacific 
samples have preserved soft parts in 70% ETOH. They 
were dissected by standard techniques, under stereo- 
microscope, with specimen immerse in alcohol. All dis- 
secting steps were digitally photographed; all drawings 
were made with the aid of a camera lucida. In the case of 
the material examined of Spinosipella deshayesiana and 


Page 58 


THE NAUTILUS, Vol. 122, No. 2 


S. costeminens, as the quantity of examined lots is very 
large, mainly thorugh courtesy of staff at MNHN (Paris), 
the list only contains the country and the quantity of 
specimens. The full list of examined lots of these speci- 
mens is being published elsewhere, in a complementary 
paper (Simone and Cunha, in press). 

Abbreviations used in figures are: am, anterior adduc- 
tor muscle; an, anus; au, ausicle: bs, byssus; by, byssal 
gland or furrow; ce, cerebral commissure; ce, ‘cerebral 

ganglion; ej, connective tissue; em, circular muscle layer; 

co, cerebro-visceral connective: er, crustacean inside 
stomach: ev, ctenidial (efferent) vein; dd, ducts to diges- 
tive diverticulae; dg, digestive diverticula; es, esophagus; 
fa, foot aperture of mantle; fm, posterior foot retractor 
muscle; fr, anterior foot retractor muscle; ft, foot: ga, 
genital ape rture; ge, gastric epithelium; gi, gill; he, 
Renienock ic, infra- septal chamber; in, intestine: ki, kid- 
ney; Im, lateral muscle; lo, longitudinal muscle layer; It, 
lithodesma: mb, mantle border; mf, fused mantle edge; 
mg, radial mantle gland; mo, moueh: mp, mantle ten- 
tacle: ms, mantle muscles of incurrent siphon; mt, 
mantle; mu, muscular tissue; ne, nephropore; nv, nerve; 
oy, ovary; pa, posterior adductor muscle; pe, pericar- 
dium; pg, pedal ganglia; pi, papilla of excurrent chamber 
roof; pm, pallial muscles; pp, palp; rs, renal fold; rt 
rectum; se, excurrent siphon; sh, shell; si, incurrent si- 
phon; sm, septum muscle; sp, septum, ss, style sac; st, 
stomach; su, supra-septal chamber; sy, crystalline style; 
ts, testis; um, shell umbo; ve, ventricle: vg, visceral gan- 
glia; vm, visceral mass. 

Abbreviations of institutions: AMS, Australian Mu- 
seum at Sydney, Australia; EGC, Emilio Garcia collec- 
tion, FMNH, Florida Museum of Natural History, 
Florida, USA; HGLC, Harry G. Lee collection; 
INVEMAR-MHNMC, project of Museo de Historia 
Natural Marina de Colombia; MHNMC, Museo de His- 
toria Natural Marina de Colombia (Programa de Biodi- 
versidad y Ecosistemas Marinos); MNHN, Muséum na- 
tional d'Histoire naturelle, Paris, France: MZSP, Museu 
de Zoologia da Universidade de Sao Paulo, Brazil; 
RLPC, Rafael La Perma collection (Universita di Bali, 
Italy). 

Specimens from other verticordiid species were also 
examined for comparative purposes. This material in- 
cludes: 


Haliris fischeriana Dall, 1881; 


UNITED STATES OF AMERICA. Florida, Fowey Light, 130 
m depth, MZSP 19934, 2 valves (R.V. Eouis sta. 184). 
BRAZIL. Rio de Janeiro; 22°34’ S, 40°29" W, 213 m 
depth, MZSP 18751, | valve (on Laminarias, W. Besnard 
col, est. IX). Rio Grande do Sul: 30°42’ S, 49°03’ W, 
182-156 m depth, MZSP 18750, 5 valves (21 Aug. 1972); 
32°55! §, 50°34’ W, 99 m depth, UFRG 1688, | shell and 
10 valves (sta. 45, 6839 dredge, 04/iv/1998) 


Euciroa sp. 


BRAZIL. Rio Grande do Sul, off Tramandar, 30°42’ S, 49°03’ 
W, 182-186 m, MZSP 18750, 5 valves (R/V W. Besnard, 


GEDIP sta. 1556, 21 Aug. 1972), MADAGASCAR. 600 m 
depth, EGC 23588, 1 shell. MOZAMBIQUE. off Moron- 
dava, Channel Madagascar, 600-800 m depth, MZSP 
61516, 3 shells (Trawled by local fisherman, May 2002). 


Euciroa elegantissima (Dall, 1881). 


UNITED STATES OF AMERICA. Florida, 27°16’ N, 
$4°58.99' W, 457 m depth, EGC 13005, 6 valves (dredged, 
R/V PELICAN), 24°09’ N, 82°31’ W, about 64.3 km off 
Southwest of Ke y West, 549 m depth, EGC 23688, 1 shell 
(R/V OREGON II col, cruise #45, sta. 13362); Monroe Co. 
Straits of Florida, 549 m depth, FMNH 209892, 1 shell 
(Frank Lyman col.), 24°15.1' N, 82°11.71' W, 525 m 
depth, FMNH 164794, 1 valve (G.H. Burgess, et. al. 
GHB-90-5, 23 Apr. 1990). COLOMBIA. Santa Marta, 
Cerro de Punta Betin, A.A. 1016, (MHNMC INVEMAR), 
MHNMC 2782, 3 valves, MHNMC 2781, | valve. 


SYSTEMATICS 
Genus Spinosipella Iredale, 1930 


Iphigenia Costa, 1850: 398 (type species by original designation 
ippagis acuticostatus Philippi. 1844) ( (pre -occupied) 
(non Schumacher, 1817). 

Spinosipella Iredale, 1930: 388 (type species by original desig- 
nation Verticordia ericia Hedley. 1911); Poutier and Ber- 
nard, 1995: 142. 

Verticordia (Spinosipella): Thiele, 1934: 1428; Moore, 1969: 
$55. 

Diagnosis: Shell relatively large, obese, with spiral 

valves. Surface prickly including on radial ribs. Radial 
ribs tall, extending beyond shell margin. Lunula very re- 
duced. 


es Sage SHELL: From small to medium size (up 
to 30 mm). Width/length ratio usually about 1. Color 
opaque-whitish. Sculptured by strong and tall radial, 
weakly curved ribs, triangular in section, bulging weakly 
beyond shell edge, alternating in both valves. Surface 
spiny, coasted by uniform sized, very small bulbs, 
covering almost entire outer surface. Lunula very re- 
duced. Umho projected, weakly spiral. Right valve with 
single tall, pointed and broad cardinal tooth. Left valve 
with low, broad tooth ( (posterior to tooth of right valve), 
and plane cardinal concavity as socket of tooth of right 
valve. Ligament just anterior to anterior hinge tooth, in- 
serted at some distance from median line, in approxi- 
mately middle way between hinge medial edge and um- 
bonal cavity (Figures 10, 12). I thode sma aide, curve, 
occupying about 0.25 of ‘hinge length, possessing a pair of 
lateral ligamental articulations (Figures 93, 96-105), 


List of Included Taxa: S$. acuticostata (Philippi, 
1844): S. agnes new species; S. costeminens (Poutiers, 
1981); S. deshayesiana (P. Fischer, 1862a) [=S. ericia 
(Henley, L911); S. tinga new species. 


Spinosipella agnes new spec ies 
(Figures 1-18, 27-29, 31, 55) 


Verticordia acuticostata.—Nobre, 1936; 303-304; 1938; 769— 
770; Abbott, L974: 563; Abbott and Dance, 1983: 375 


L. R. L. Simone and C. M. Cunha, 2008 Page 59 


Figures 1-14. Spinosipella agnes new species. Shells. 1-10. Holotype (length 23.2 mm) shell. 1. Left valve, outer view. 2. Right 


valve. 3. Right valve, inner view; 4. Left valve, inner view. 5. Dorsal view. 6. Posterior view. 7. Anterior view. 8. Detail of shell surface 
in SEM, middle region of right valve. 9. Detail of inter-umbonal region, dorsal view. 10. Hinge, ventral-inner view. 11-14. Paratypes 
11. HGLC. from Florida, left valve, outer view; 11 mm. 12. Same, ventral view, valves opened for showing whole view of hinge 
13-14. EGC 17419. from Colombia, outer view of right and left valves; 15 mm 


Page 60 THE NAUTILUS, Vol. 122, No. 2 


Figures 15-29. Spinosipella new species. Shells. 15-18. 8. agnes paratype AMNEL 162803, Florida, specimen with remains of soft 
l4.5 mm. 19-26. S. tinga type specimens. 19-21. Holotype, left valve, outer, inner and dorsal views; 16.9 mm, 22-24. 


Paratype MORG 18085, right valve, dorsal, outer and inner views; LO.1 mm, 25. Holotype detail of hinge, left valve. 26. Paratype 
MORG 18085, right valve, detail of hinge. 27-29. S. agnes 
idinal tooth. 28-29. Paratype INV-MOL 2943 


ng mayor difference 


27. Holotype, left valve, detail of hinge: arrow indicating well-developed 
from Colombia, a specimen ol equivalent size of main types of S. tinga 
compare with Figures 19-20): inner and outer views; 17.0 mm note developed posterio1 cardinal toot 
Inge wWYrow fewe! tall I ancl more spaced ribs anc more projected ribs at edges 


L. R. L. Simone and C. M. Cunha, 2008 


Page 61 


(fig.); van Aartsen, 1992: 45; Poppe and Goto, 1993: 139; 
?McLean and Geiger, 1998: 27, 109 (fig.); Salas, 1996: 46; 
Rosenberg, 2005 (part) (non Philippi, 1$44). 

Verticordia deshayesiana.—Rosenberg, 2005 (part) (in syn- 
onymy) (non Fischer, 1S62a). 


Type Mater ial: HOLOTYPE, MZSP 36917; BRAZIL. 
Rio de Janeiro, off Cabo Frio, 23°41’ S, 41°03’ W, 750- 
SOO m depth (o.t.). PARATYPES. UNITED STATES 
OF AMERICA. Florida; Off Cape Canaveral, 903 m 
depth, USNM 64039, 1 right valve; SE of Sand Key, 
AMNH 248458, 4 shells (Jan. 1970, |. M. Bijur Collec- 
tion), AMNH 245459, 2 shells ee 1970, J. M. Bijur 
Collection); Monroe County, S.E. Sand Key, 270 m 
depth, HGLC, 2 shells, FMNH 154594, 1 specimen 


(dredged, Jerry Phelps col., Jun. 1970); 120.6 kin east of 


Daytona, 29°17’ N, 79°27" W, 878 m depth, USNM 
§10590, 1 shell and 1 left valve (R/V. OREGON, sta. 6690, 
9 May 1967); Marquesas Key, 24°15’ N, 82°13’ W, 278— 
419 m depth, 1 left, USN M 810889, 1 right valve (R/V. 
Blake, A. Agassiz 1877-1878). COLOMBIA. off Carta- 
gena, 10°28’ N, 75°42’ W, 280 m, MHNMC 2203, 1 
valve (E-47), 10°31’ N, 75°37’ W, 309 m, MHNMC 
2775. 1 valve (E-141), Palomino, Dibulla, 11°29’ N, 
73°27' W, 476 m, MHNMC 3104, 4 valves, (E-21), Gua- 
jira, Bahia Honda, 12°31’ N, 72°8’ W, 452 m, MHNMC 
2943, 1 shell (E-12), Guajira Peninsula, 12°30’ N, 72°08’ 
W, 470 m depth, EGC 17419, 1 shell; Cabo de la Vela, 
12°19’ N, 72°42’ W, 464 m, MHNMC 3087, 2 valves 
(E-19), Islas del Rosario, °10" N N, 76°01’ W, 510 m, 
MHNMC 2208, 2 valves (E-78). BRAZIL. Rio Grande 
do Norte: 206 m depth, ne 84627, 1 shell (Sta. D-22. 
10 Nov. 2001); Pernambuco; 690 m depth, MZSP 84628, 
1 shell (Sta. D-11). 


Diagnosis: Shell with 15-17 radial ribs; smooth pre- 
umbonal region wide (about 0.25 of shell length); prickly 
sculpture chaotically organized. Width/] ength ratio in 
each valve approximately 0.57. Posterior eienmal tooth of 
left valve hinge well developed; main cardinal tooth of 
left valve relatively low and cylindrical; main cardinal 
tooth of right valve tall (about 0.2 of valve width) and 
pointed. 


Description: SHELL: Up to 22 mm, equivalve, inflated, 
each valve symmetrically and weakly spiral (1 whorl) 
(Figures 7, 9). Color white. Degree of convexity (width/ 
length) in each valve approximately 0.57. Outer surface 
spiny, opaque forming an irregular mosaic (Figures 8, 9). 
Umbones located in middle region of dorsal surface, spi- 
ral, high. divergent, separated from each other at about 
Vs of shell width ( Figures 1-4, 13-18, 55). Sculptured by 
strong, uniform, arched, radial ribs, from 15 to 17 in each 
valve. Posterior edge about twice as wide as anterior 
edge. Between umbo and anterior edge a concavity bear- 
ing transversal ribs, es wider than ribs of remaining 
region (Figures 5, 7, Pre-umbonal region smooth, 
narrow, 0.2 of shell le set (Figures 5, 7, 9). Anterior, 
ventral and posterior edges forming zigzag (Figures 3, 4): 
tips of this zigzag coinciding with tips of each a ly 
encasing in concavity of opposite valve (Figures 1, 2, 11, 


13-16). Inner surface iridescent, whitish, glossy; includ- 
ing hinge (Figures 3, 4, 10, 12, 17, 18). Hinge with a large 
eandinal tooth in right valve, stubby, tall (about 20% of 
valve width), broadly pointed, we vakly curved forwards 
(Figures 3, 10, 12, 18), circular in section: correspondent 
socket in left valve shallow, restrict to dorsal surface; this 
socket flanked by small tooth in each side, anterior 
smaller and lower than posterior (Figures 4, 10, 12, 17, 
27-28 |arrow]). Ligament just anterior to anterior hinge 
tooth, inserted at some distance from median line (Fig- 
ures LO, 12), approximateh sly midw: ay between hinge me- 
dial edge and umbonal cavity. Sears of adductor muscles 

shallow ( (Figures 3, 4, 17, 18, 28): anterior scar elliptical 
(longer “igiee vere), located close to anterior edge, 

area about Vis of inner surface of valve; posterior scar 
circular, about 0.33 larger than anterior scar, located 
close to posterior shell edge. Pallial line continuous, lo- 
cated at wide distance from shell ec dge, about 0.33 of 
distance between ventral and umbonal height. 


Measurements cone height, width, in mm): 


Holotype: 20.1 by 23.2 by 22.2; EGC 17419: 15.6 by 15.5 
by 15.5. 


Geographic Distribution: Florida, USA, to Rio de 


Janeiro, Brazil. 
Habitat: 


Material Examined: Types. BARBADOS. USNM 
63200, 3 valves (Blake Coll., sta. 100). CUBA. Havana; 
Gulf of Mexico, 419 m depth, USNM 63201, 3 right, 4 
left valves (Blake Coll., sta. 5). PORTUGAL. Porcupine 
Bank; USNM 63204, 2 right valves (Jeffreys Coll., Por- 
cupine Exp. 1870). UNITED STATES OF AMERICA. 
Florida; Gulf of Mexico, off Cape San Blas, 309 m depth, 
USNM 323871, 1 left, 1 right valve (sta. 2400); 120.6 km 
east of Daytona, 29°17’ N, 79°27’ W, 878 m depth, 
USNM 810590, 1 shell and 1 left valve. BRAZIL. Es- 
pirito Santo (R/V MartIon-DUFRESNE MD55, May 1987); 
off Conceigao da Barra, 18°59’ S$, 37°50’ W, 637 m 
depth, MNHN, 10 valves (sta. CB76); off Pontal da 
Regéncia, 19°34’ S, 38°55’ W, 340-360 m depth, 
MNHN 1 valve (sta. CB92). 


Muddy bottoms, 270-900 m. 


Etymology: The specific epithet refers from the 
Greek agnes, meaning pure, an allusion to the whitish 
color of the shell. 


Remarks: The above listed examined material that was 
not designed as types are normally lots with eroded 
specimens, or sometimes they have aberrant characters. 
This is the case of the MNHN material collected off 
north coast of Espirito Santo, Brazil. They actually are 
free valves that resemble the Pacific species Spinisopella 
costeminens, in having a weakly larger radial thread be- 
tween middle and posterior thirds. ond in lacking ante- 
rior tooth in hinge. As they can represent another spe- 
cies, they are not designec Tas types; on the other hand, 
the material is not sufficiently well-preserved for further 
analysis. Because of they can only represent an extreme 


Page 62 


THE NAUTILUS, Vol. 122, No. 2 


of variation of the S. agnes, they are listed as additional 
examined material of this species. 


Spinosipella tinga new species 
(Figures 19-26, 30, 32) 


Verticordia acuticostata: Marini, 1974: 242, figs. 5, 6 (non Phil- 
ippi, 1544). 

Verticordia (Haliris) acuticostata: Rios, 1975: 262, pl. 85., fig. 
1261; 1985: 282, pl. 99, fig. 1391; 1994: 304, ‘i 104, figs. 
1489 (non Philippi, 1844). 

Verticordia deshayesiana: Marini, 1974: 242 (in synonymy); 
Rios, 1975: 262; 1985: 282; 1994: 304 (in synonymy); 
Rosenberg, 2005 (part) (Gin synonymy) (non Fischer, 
1862a). 


Diagnosis: Shell with 17-18 radial ribs; smooth pre- 
umbonal region very narrow (less that 1% of shell length). 
W idth/length ratio in each valve approximately 0.47. Pos- 
terior cardinal tooth of left valve hinge absent; main car- 
dinal tooth of left valve low and cy adie: al; main cardinal 
tooth of right valve lower (about 10% of valve width) and 
rounded. 


Description: SiHeLL: Up to 11 mm, equivalve, in- 
luted, both valves weakly spiral (1 whorl) (Figures 21, 

). Color white. Degree of convexity width length) in 
i valve approximately 0.47. Umbones located in 
middle region of dorsal surface (Figures 19, 20, 23, 24); 
umbones weakly spiraled, somewhat high, divergent, 
separated from each other. Sculpture of strong, uniform, 
arched, radial ribs, 17-18 in each — Outer surface 
opaque, covered by a mosaic of small, blunt, loosely 
aligned spines par allel to radial ribs ( Figure 3 2). Anterior 
edge almost same size as posterior edge. A concavity 
bearing transversal ribs of same width as remaining ribs 
between umbo and posterior edge (Figures 21, 22, 30). 


Anterior, ventral, and posterior edges rounded, tips of 


ribs prominent (Figures 19, 20, 23, 24), fitting with con- 
cavity in opposite valve. Inner surface iridescent, whitish, 
glossy, including hinge. Hinge with a somewhat large 
cardinal tooth in right valve (Figures 20, 25); cardinal 
tooth stubby, tip rounded, flat in cross-section, tooth 
length about 10% of valve width: correspondent socket in 
left valve shallow, restricted to dorsal surface; this socket 
flanked by small, low, posterior tooth (no anterior tooth) 
(Figures 24,26). L igament just anterior to anterior hinge 
tooth, inserted at some distance from midline, approxi- 
mately midway be stween hinge medial edge and umbonal 
cavity. Scars of adductor muscles shallow ( (Figures 20, 
24); anterior scar elliptical (longer dorso-ventrally), lo- 
cated close to anterior edge, area about Vis of inner sur- 
face of valve; posterior scar circular, about / larger than 
anterior scar, located close to poste rior shell edge. Pallial 
line with a very weak pallial sinus, located at wide dis- 
tance from shell e dge, about “% of distance between ven- 
tral and umbonal height. 


Measurements (respectively length, height, width, 
in mm): THolotype: 16.9 x 15.3 x 9.6 (single valve); 
MZSP 18752: Paratype #1, $8.6 x §.6 x 4.1 (1 valve): 


Paratype #2, 11.9 x 11.4 x 5.4 (1 valve); MZSP 18753: 9.5 
x 9.6 x 4.8 (1 valve). 


Type Material: Holotype, MZSP 19345, 1 valve, from 
type locality (R/V W. BEsNarD, GEPID Est. 458, 9 Dec. 
1968. Paratypes, Rio de Janeiro, Cabo de Sao Tomé, 
31°08’ S, 49°31’ W, 182-253 m,.1 valve, MZSP 18752 
(R/V W. BESNARD, GEDIP st. 1858, 6 Aug. 1972): 22°34’ 
S, 40°29’ W, 213 m, 1 valve, MZSP 18753: (R/V W. 
BESNARD, st. IX, 1] Feb.1969), 100 m, 2 valves, MORG 
18085 (R/V ALMIRANTE SALDANHA, Mar. 1972), off 
Solidao, 240 m, 2 valves, MORG 31888 (R/V ATLANTICO 
SuL, Exp. Coltro, 14 Oct.1993). 


Type Locality: BRAZIL. Rio Grande do Sul, off AL 
bardao, 33°29’ S, 50°44’ W, 200 m, muddy bottom. 
Geographic Distribution: 
ro to Rio Grande do Sul. 


Brazil, from Rio de Janei- 


Etymology: The specific epithet refers to the color 
white of the shell, from the Tupy language: tinga. 


Spinosipella acuticostata (Philippi, 1844) 
(Figures 33-40) 


Hippagus acuticostatus Philippi, 1S44: 42 (pl. 14, fig. 19) [fossil 
in Lamati valley, Calabria, Italy]. 
Verticordia acuticostata: Micali and Villari, 1991: 353. 


Spinosipella acuticostata: Poutiers and Bernard, 1995: 143, 
150; 
Diagnosis: Shell with 12-13 radial ribs; pre-umbonal 


region narrow, smooth; posterior cardinal tooth of left 
valve hinge shallow or absent; main cardinal tooth of left 
valve longer and flat (Figures 35, 40); main cardinal tooth 
of right valve shallower (Figures 34, 37) (about 10% of 
valve width). 


Description: SHELL: Up to 24 mm; width/length ratio 
approximately 1 (Figures 36-35) to 1.5 (Figures : 39, 40). 
Degree of convexity (width/length) in each valve approxi- 
mately 0.55. Outer surface spiny, opaque, spines forming 
radially aligned mosaic parallel to ribs (Figures 36, 38, 

39). Sculpture of strong, uniform, arched, radial ribs, 
12 13 in each valve. Posterior edge about twice as broad 
as anterior edge. A concavity bearing transversal ribs 
weakly broader than ribs of remaining region between 
umbo and anterior edge (Figures 35, 39); pre-umbonal 
region narrow, smoooth (Figure 33), about 10% of shell 
length. Anterior, ventral, and posterior edges forming 
zigzag (Figures 35, 37, 40). Hinge with a large cardinal 
tooth in right valve, stubby, tall (about 10% of valve 
width), broadly pointed, weakly curved anteriorly (Fig- 
ures 34, 37), circular in section: corre spondent socket in 
left valve shallow, restricted to dorsal surface; this socket 
sometimes flanked by small tooth in each side, anterior 
absent or very weak (Figures 35, 40). 


Measurements (respectively length, height, width, 
inmm): KRLPC #1: 11.4 x 14.5 x 6. 6 (valve); #2: 10.0 x 
9.8 x 4.7 (valve). 


Geographic Distribution: Mediterranean. Pliocene 
fossil from south Italy (Calabria and Sicily). 


L. R. L. Simone and C. M. Cunha, 2008 Page 63 


Figures 30-47. Spinosipella species. Shells. 30. S. tinga, Holotype, left valve, dorsal view. 31. S. agnes, Paratype INV-MOL 2943 
from Colombia, a specimen of equivalent size of Holotype of S. tinga for showing major differences (compare with Figure 30); dorsal 
view: 17.0 mm: note higher convexity, fewer, taller and more spaced ribs. 32. S. tinga, left valve, SEM of Holotype, showing prickly 
S( ulpture 3340. S. acuticostata. 33-38. Pliocene fossil from Messina, Italy, USNM 63202. 33 Specimen 2 dorsal-slightly anterio1 
view. 34. Specimen 3 right valve, detail of hinge. 35-36. Specimen |] left valve, inner and outer views; 13.5 mm. 37-38. Specimen 
2. right valve, inner and outer views, hinge broken; 12.2 mm. 39-40. RLPC, from Rometta, Italy, left valve specimen with long 
shape; 19.0 mm. 41-47. S. deshayesiana. 41-43. Paratype 1 of S. ericia AMS 032068, left valve, inner, outer and dorsal views; 3.5 
mim. 44. Paratype 2. right valve, inner view; 2.6 mm. 45 Type specimen of S japonica ANSP 49639, right valve: 5.2 mm. 46-47. 
ANSP 292986 (from India), right valve, outer and inner views; 10.5 mm 


Page 64 


THE NAUTILUS, Vol. 122, No. 2 


Paleohabitat: 
bathyal environments. 


Material Examined: ITALY. Sicily, Messina, 38°11’ 

15°34' E, Seguenza, USNM 63202, 2 left, 2 right 
valves. Middle Pliocene outcrops at Rometta, 4 valves, 
RLPC. 


Spinosipella deshayesiana (Fischer, 1862) 
(Figures 41-54, 66, 67, 72-82, 93-102) 


Verticordia Deshayesiana Fischer, 1862a: 35-36 
11) [China Sea]. 

Verticordia japonica A. Adams, 1862: 224. 

Verticordia ericia Hedley, 1911: 96; Prezant, 1998: 421 (fig. 
9.16A) 

Spinosipella deshayesiana.—Poutiers and Bernard, 1995: 110— 
112, 143, 159, 161 (figs. 7-9). 

Spinisopella ericia.—Poutiers and Bernard, 1995: 143, 159. 

Verticordia acuticostata—McLean and Geiger, 1998: 109 (non 
Philippi, 1544). 


(pl. 5, fig. 10- 


Diagnosis: Shell with 16-19 radial ribs uniformly dis- 
tributed, closely packed; pre-umbonal region narrow, 
smooth. Each rib bearing well- developed crests with 
small, prickly granules. Posterior cardinal tooth of left 
valve hinge abecak: main cardinal tooth of left valve low 
and flat, “with insertion of anterior valve edge approxi- 
mately in middle region of this tooth; main car Aint il tooth 
of right valve high (about 10% of valve width) and 
pointed. 


Description: SHELL: Up to 1S mm. Color white. De- 
gree of convexity (width/length) in each valve approxi- 
mately 0.57. Outer surface spiny, spines organized some- 
what radially, parallel to ribs; each rib with well- 
developed crests with small, prickly granules (Figures 42 
45, 46, 48-54). Sculpture of strong, “uniform, arched, ra- 
dial ribs, 16-19 in each valve (Figures 42, 45, 46, 50), 
somewhat closely packed. Posterior edge about twice 
broader than anterior edge. A concavity bearing trans- 
versal ribs similar to ribs of remaining region present 
between umbo and anterior edge (Figures 43, 53); pre- 
umbonal region narrow, smooth, about 10% of shell 
length (Figures 43, 53). Anterior, ventral, and posterior 
edges forming zigzag (Figures 42, 44, 47, 50, 51, 59, 67, 
100), with tips projecte ri longer, and narrower. Hinge 
with a large cardinal tooth in right valve, stubby, tall 
(about 10% of valve width), broadly pointed, somewhat 
flat (Figures 44, 47, 51, 59, 67); correspondent socket in 
left valve shallow, restrict to dorsal surface; this socket 
flanked by small posterior tooth, with insertion of ante- 
rior valve edge approximately in middle region of this 
tooth (Figures 41, 50), anterior tooth absent (Figures 41, 
50) 

LITHODESMA (Figures 93, 96-99): Saddle-shaped, 
hemi-cylindrical. Dorsal surface concave (Figures 97— 
98), flanking ventral surface of hinge, along Vs of hinge 
length; located just posterior to teeth. Left and right 
edges straight, turned upwards and medially, connected 
with valves by dark-brown ligament inside umbonal cav- 
ity closer to hinge inner edge (Figures LOO-102). Outer 


Middle and upper Pliocene beds of 


surface convex, covering dorsal-middle, inter-umbonal 
region of visceral mass (Figure 75). Anterior and poste- 
rior edges concave; anterior edge slightly deeper and 
with tenuous slope. Both edges covered by opaque, yel- 
lowish periostracum (Figures 96, 101, 102), Lithodesma 
thickness equivalent to that of shell. 

Major MUSCLES (FIGURES 72-75, 77, 78, 80, 91, 92): 
Both adductor muscles similar in size and position (Fig- 
ures 72-75), near valve edges; insertion size equivalent to 
1/20 of valves inner surface each; approximately two 
times taller than wide; outer length about half of inner 
length, with insertion in valves gr ndatly oblique (Figures 
75, 91, 92). Anterior adductor muscle with anterior re- 
on about 3 times narrower than posterior region, di- 
vided transversally (dorsoventral) in two similar halves 
(quick and slow components). Posterior adductor muscle 
similar to, but inverted arrangement in comparison to 
anterior adductor muscle; components different, how- 
ever, one of them horseshoe-shaped, occupying ventral 
and posterior sides (Figure 77); another component fill- 
ing internal region of muscle, only exposed in posterior 
and dorsal siden ( Figure 77). Pair of anterior foot retrac- 
tor muscles long aud narrow (Figure 80); originating just 
dorsal to anterior adductor muscle in area equivalent to 
1/10 of adductor (Figures 75, 80); running ventrally and 
posteriorly; spreading after insertion in anterior and lat- 
eral regions of foot base. Pair of posterior foot retractor 
muscles similar to anterior pair, but about half narrower 
(Figures 75, 78, 80); originating just dorsal to posterior 
adductor muscle in area equivalent to 1/20 of that ad- 
ductor; running ventral and anteriorly; inserting in pos- 
terior and lateral regions of foot base. Pair of palp 
muscles, septal and pallial muscles described below. Pair 
of foot protractor muscles absent. 

Foor AND Byssus (FIGURES 73, 74, 80): Foot conical, 
pointed; estimated volume equivalent to 1/6 of that of 
chamber of valves; base located in middle region of ven- 
tral surface of visceral sac. Byssal furrow shallow and very 
narrow, length about half of that of foot, offset ventrally 
and distally, lying along posterior surface and midline, 
ending at short distance from foot apex. Byssus found in 
a single specimen, brown, with single filament, narrow: 
proximal end attached to distal region of byssal furrow. 

MANTLE (FIGURES 72-74): Dorsal fusion of mantle 
lobes about 3 of their edges, along entire hinge length 
and about “% of valves height toward ventral, in both 
sides. Edges of mantle lobes with two folds. Inner fold 
fused between two lobes along entire posterior half (ex- 
cept for siphonal apertures) (Figure 74). Both lobes free 
from each other along anterior half, up to dorsal level of 
anterior adductor muscle; in this region both folds are of 
similar size, with height equivalent to 25 of valves height. 
Mantle edges thick, muscular, insertion relatively thick i in 
pallial line (Figure 72, pm). Pallial muscles originating in 
pallial line in location about 13 from ventro-dorsal dis- 
tance; no clear pallial sinus. Incurrent siphon as aperture 
of a septum formed by fusion of inner mantle edge folds; 
aperture about 5 of posterior fused region of mantle, 


L. R. L. Simone and C. M. Cunha, 2008 Page 65 


gures 48-63. Spinosipella species. Shells. 48-54, 56-59. S. deshayesiana, adult specimens. 48-53. MNHN (Sta. CP1475, Fiji 

5 mm. 48. Left valve, outer view. 49. Right valve, outer view. 50. Left valve, inner view. 51. Right valve, inner view. 52. Posterior 
view. 53. Anterior view. 54. Dorsal view, HGLC, from Philippines; 11.8 mm. 55. S. agnes paratype, BMNH, 18.2 mm. 56-59. 
Syntypes of S. deshayesiana MNHN. 56. Outer view, specimen 1, left valve. 57. Outer view, specimen 2 (possibly figured by Fische1 
1$62a), right valve. 58. Inner view, specimen 1. 59. Inner view, specimen 2; length = 8 mm. 60-63. S. costeminens Holotype MNHN 
60. Outer view, left valve. 61. Outer view, right valve. 62. Inner view, left valve. 63. Inner view, right valve; 17 mm 


a 
Page 66 


THE NAUTILUS, Vol. 122, No. 2 


Figures 64-71. Spinosipella species. Shells. 64-65. S. costeminens Holotype MNHN. 64. Right valve, anterior view. 65. Left 
valve, anterior view; 17.0 mm. 66-67. S. ericia Holotype AMS, right valve. 66. Outer view. 67. Inner view; 5.8 mm. 68-71. S 
costeminens lacking projections on ribs, MNHN (Sta. CP 992, Vanuatu), right valve. 68. Outer-right view. 69. Anterior view. 70. 
Posterior view. 71. Inner view; 29.0 mm. 


longer dorso-ventrally (Figure 76); walls thick, muscular; 
outer surface flanked by 9 tentacles surrounding siphonal 
aperture; all tentacles of similar size, turned inwards, 
somewhat conical, tip blunt and rounded, length equiva- 
ent to that of siphonal aperture; single unpaired tentacle 
ocated ventrally; five secondary smaller tentacles located 
externally, midway between siphonal aperture and 
mantle edge, of similar size, about Ys of size of major 
entacles, well separated from each other, one of them 
ocated in ventral region of siphonal aperture, other four 
ocated laterally, in ventral half of siphon (Figure 76). 
Incurrent siphon a small pore located in small elevation, 
approximately midway between excurrent siphon and 


nge: a pair of small tentacles similar to secondary ten- 


acles of incurrent siphons, located laterally, in dorsal 
region of siphon base (Figures 74, 76, 77). Radial mantle 


gland present along mantle edges outer fold (Figures 73, 
92, mg), occupying about half of outer fold volume, situ- 


ed closer to inner surface of this fold. 
79 
19 


PALLIAL CAVITY (FIGURES 75, 94, 95): Occupying 


about 70% of volume of valves. Transversal, horizontal 
septum located approximately midway in animal, i.e., su- 


pra- and infra-septal chambers of equivalent length (Fig- 
ure 72). Paired palps low, wide, bilobed folds (Figures 


73, 8] 


( 


)5, pp) that occupy anterior third of dorsal sur- 


face of infra-septal chamber, permanently open as a fun- 
nel. Pair of palp muscles (Figures 72, 74, 79, lm) located 
laterally; originating in anterior region of umbonal cavity, 
in a distance from origin of anterior foot retractor equiva- 
lent to ¥3 of anterior adductor muscle height; located in 
same horizontal level of origin of anterior foot retractor; 
size equivalent to V4 of that of anterior foot retractor; 
running ventrally attached to mantle for a distance 
equivalent to Y of valve height: spreading after insertion 
in lateral region between inner and outer hemipalps. 
Palp muscles also connect anterior end of septum. Sep- 
tum with two constituents: external one produced by a 
fold of mantle (about 73 of septum area); internal pro- 
duced by gill (Figures 73, 95). External septum element 
thick, muscular; posterior muscles originating as a pair, 
just dorsal to posterior adductor muscle (Figure 77, ms): 
running ventrally immersed in mantle, at some distance 
from each other (equivalent to half of their width) and 
from midline, gradually becoming wider and thicker, in 
anterior surrounding posterior surface of posterior ad- 
ductor muscle and lateral edges of excurrent siphon; 
some secondary muscular bundles originating from cen- 
troposterior region of posterior adductor muscle uniting 
with main, vertical bundles (Figure 77); muscles spread- 
ing within septum in region between incurrent and ex- 


L. R. L. Simone and C. M. Cunha, 2008 Page 67 


gi au 


Figures 72-75. Spinosipella deshayesiana. Anatomy. 72. Whole specimen just extracted from shell, right view. 73. Same, right 


mantle lobe in its infra-septal region removed, right-slightly ventral view, left shell valve also shown. 74. Same, right mantle lobe 
almost completely removed, right portion of septum also removed. 75. whole specimen, dorsal-slightly right view, most of mantle and 


dorsal integument artificially shown as transparent, lithodesma (It) shown in its in situ topology. Scale bars = 2 mm 


Page 65 THE NAUTILUS, Vol. 122, No. 2 


pa ; . 
vg pl ne ga he ft PP 


Figures 76-79. — Spinosipella deshayesiana. Anatomy. 76. Detail of region of siphons, posterior-slightly right view. 77. Peri-anal 


chamber, right view, adjacent region of right mantle lobe sectioned and deflected to show inner surface and muscles, inferior region 
or right mantle lobe removed along median line. 78. Reno-pericardial region and adjacent structures, right view, right wall of 
pericardium removed. 79. Whole right view, showing topology of genital system, reno-pericardial structures, palps, main ganglia and 


muscles, most structures artificially shown as transparent. Scale bars = 2 mm 


L. R. L. Simone and C. M. Cunha, 2008 


Pase 66 
Page 69 


80 


Figures $0-S2. 


Spinosipella deshayesiana. Anatomy. 80. Whole right view, emphasizing digestive structures, main musculature 


and main nervous ganglia; topology of some adjacent structures also shown, everything else represented by transparency. 81. Same, 
anterior region of digestive structures opened longitudinally, some objects inside stomach preserved, topology of some adjacent 
structures also shown, 82. Visceral ganglia (left), ventral view, and pedal ganglia (right), postero-dorsal view. Scale bars = 1 mm. 


current siphons. Outer component of septal muscles in- 
serted in shell just ventral to posterior adductor muscle 
(Figure 72, sm), in area equivalent to Yio of that of ad- 
ductor muscle insertion. Internal element of septum 
constituted by gills. Gill with both demibranchs narrow, 
of similar size, flattened, in same plane of remaining 
septum: both gills surrounding posterior and lateral re- 
gions of foot base (Figures 73, 75, 95). Gill attached to 
remaining septum via tissue; gill attachment to foot by 
cilia. Connection between gill filaments of 6-7 longitu- 
dinal, equidistant bridges of similar width of filaments. 
Papilla situated in posterior region of roof of supraseptal 
chamber (Figures 78-80, 94, pi), positioned just ventral 


to visceral ganglia, internally solid: length about Yo of 


posterior adductor muscle length and about % of it in 
width: tip broadly pointed, normally turned to anterior. 

VISCERAL MASS (FIGURES 72, 75, 79): Strongly bilobed, 
as internal mould of well-separated umbos (Figures 72, 
75). Most dorsal structures, just inside valve apexes, 
formed by sponge-like connective tissue. Pair of ovaries 
cream in color, occupying central and dorsal regions sur- 
rounding stomach and digestive diverticula, reaching 
dorsal areas up to dorsal sponge-like connective when 
fully developed. Testes brown, consistence harder, lo- 
cated ventrally and laterally, totally separated from ova- 
ries; anterior region irregularly digitiform (Figures 75, 


79, ts). Digestive diverticula situated compressed be- 
tween stomach and gonads, color greenish-beige; occu- 
pying about “% of visceral volume. Stomach and intestine 
lying in central region, occupying about “4 of visceral 
volume (Figures 80). Reno-pericardial structures located 
just anterior to posterior adductor muscle and posterior 
foot retractor muscles, with volume approximately “% of 
visceral volume (Figures 75, 79). 

CIRCULATORY AND EXCRETORY SYSTEMS (FIGURES 75, 
78): Pericardium located at short distance anterior to 
posterior adductor muscle; with about half of reno- 
pericardial volume, and with a pair of expansions toward 
anterior, surrounding roof of pallial cavity where lies pair 
of auricles. Auricles connecting to anterior end of gills, in 
short isolated ctenidial vein (Figure 75, cv); abruptly 
curving towards posterior and dorsal; after this curve, 
auricles increasing gradually, surrounding obliquely pe- 
riphery of visceral mass in roof of pallial cavity (Figures 
72, 75), walls thin, translucent: close to midline auricles 
abruptly narrowing and connecting to ventricle (Figures 
75, 78); posterior region relatively lobed. Ventricle lo- 
cated in center of pericardium, surrounding intestine: 
relatively narrow. Kidney mostly solid, color dark purple- 
almost black; most of renal gland located just anterior to 
posterior adductor muscle, ventral to pericardium (Fig- 
ures 75, ki); a pair of folds originating from this region, 


Page 70 


THE NAUTILUS, Vol. 122, No. 2 


running long roof of pallial, supraseptal cavity, just ven- 
tral and external to auricles (Figures 74, rf), this pair of 
folds with about ¥% of supraseptal chamber height, run- 
ning posteriorly in middle region of roof of this chamber, 
gradually approaching visceral mass towards anterior, 
fusing to visceral mass after running about “% of chamber 
length (Figure 74). Pair of nephropores as small slits 
located in posterior region of supraseptal chamber, cov- 
ered by posterior end of renal fold, just dorsal to of 
posterior retractor muscles of foot ( Figure 74, ne) 

DIGESTIVE SYSTEM (FIGURES 80, 81): Palps partially 
described above (pallial cavity), widely fused as pair of 
folds along midline (Figures 95, pp). Mouth central ( Fig- 
ure 95, mo), with sphincter relatively well developed. 
Esophagus with about 5 of visceral mass length, not 
attached to anterior adductor muscle, width about ¥ of 
that of anterior adductor muscles; wall relatively thick, 
muscular; inner surface with about 20 longitudinal, nar- 
row, low folds as continuation from those of palps (Fig- 
ure 81). Stomach main chamber with about “4 of visceral 
mass volume, elliptical, anteroposteriorly longer; walls 
thick, muscular (Figure 91, st). Gastric inner surface 
smooth; two pairs of ducts to digestive diverticula 
present, each one located in ventro-lateral region just 
posterior to esophageal insertion. Stomach normally con- 
taining 3-4 isopod crustaceans (Figure $1, cr) ). Style sac 
with about / of gastric main chamber volume, Ipeater lin 
middle of gastric ventral wall, somewhat elliptical (longer 
dorso-ventrally); ory stalline style occupying entire style 

sac (Figure Sl, inner surface of style sac smooth, 
lacking any fold ete it from intestine; gastric 
shield lacking. Intestine a single sigmoid loop with about 
half of style sac width. Inner surface simple, smooth. 
Intestinal portion crossing through pericardium in some- 
what anteroposterior direction. Rectum attached to dor- 
sal and posterior surface of posterior adductor muscle, 
with about 7% of remaining intestinal width. Anus simple, 
sessile, located in ventral third of posterior surface of 
posterior adductor muscle (Figure 77). 

GENITAL SYSTEM (Partially described above under VIs- 
CERAL Mass): Pair of testes and ovaries conv erging to a 
single common, short duct, of about “is of visceral mass 
length. Genital pores small slits located at short distance 
from ne phropores ( (Figures 74, 75, 94, ga). 

CENTRAL NERVOUS SYSTEM (FIGURES 80, 82): Cerebral 
ganglia somewhat triangular, each ganglion with volume 
equivalent to Vis of that of anterior adductor muscle; 
anterior end narrow, possessing thick pair of nerves run- 
ning to pallial region dorsal to palps; pair of ventral 
nerves also thick, originated in middle region of ganglia, 
running ventrally to palps; Posterior a originating 
(Figures SO, 106); cere bral 
commissure length about 3 of posterior surface of an- 


cere bro- visce ral connective 


terior adductor muscle. Pair of cerebro-visceral commis- 
sures relatively thick, running though visceral mass be- 
tween stomach and testes. Pair of pedal ganglia located 
in ventral third anterior pair of pedal retractor 
muscles, touching these muscles, both totally fused with 


each other along midline, almost forming a sphere, vol- 
ume of both equivalent to that of each cerehial ganglion; 
pedal nerves and cerebropedal connectives originating 
subterminally in posterior surface of ganglia. Pai of vis- 
ceral ganglia located anterior to ential surface of poste- 
rior adductor muscle; both also totally fused with each 
other along median line, being somewhat squared in ven- 
tral view; size equivalent to that of pedal pair of ganglia; 
cerebrovisceral connectives and siphonal nerves located 
in vertices. 


Measurements (respectively length, height, width, 
inmm): HGLC: 11.5 by 12.2 by 12.0; MNHN (Sta. 
DW11): 15.7 by 17.7 by 9.2 (valve); MNHN (Sta. 
CPS889): 19.7 by 17.0 by 9.1 (valve). 


Geographic Distribution: South and Central Indo- 


Pacific in 146-805 m depth. 


Material Examined:  Paratypes of S. ericia: AUSTRA- 
LIA; South Cape Wiles, 174-153 m, 35°39’ S, 136°40" E, 
AMS 032068, 1 left, 1 right valves (Zoological Results of 
the F.1.S. ENDEAvouR, 28 Aug. 1909). 


Other Material Examined: Holotype of S. japonica: 
JAPAN. ANSP 49639, 1 shell. MNHN. SW PACIFIC. 
Loyaute Islands, 16 lots [122 v]. TONGA IS. 12 lots [59 
v|. GUAM. Marianas Islands, 3 lots [15 v]. AUSTRALIA. 
South Cape Wiles, 1 lot [6 v]. NEW CALEDONIA. 
South, 3 lots [7 specimens]; Banc Esponge, 2 lots [3 
specimens]: Chee jield Plateau, 1 specimen. PHILIP- 
PINES. Aliguri Is. 2 lots [1 ete and 3 v|; Bohol 
Sea, Off Balicasag island 1 lot [1 v]. FHL. 1 specimen. 
MYANMAR (BURMA). 1 lot [5 v] ae: North Chan- 
nel, 1 lot [4 v]; N.W. of Tavoy L., 1 lot [11 v]. ANDA- 
MANS SEA. 1 lot [1 v]. THAILAND. Phuket I, 1 lot [11 
v]; Andaman Sea, 1 lot [1 v| (Details in Simone and 
Cunha, in press.) 


Spinosipella costeminens (Poutiers, 1981) 
(Figures 60-65, 65-71, S3—92, 103-108) 


bl ay (Spinosipella) ) costeminens Poutiers, 1981: 351 (pl. 
. figs 144, text fig 5) 

Soacoe lla costeminens _Poutiers and Bernard, 1995: 110, 
143, 158 (figs. 1-2). 


Diagnosis: Shell with 16-17 tall radial ribs, those 
more posterior to middle surface very taller, normally 
possessing blade-like projections along tip; 3-4 more 
posterior abruptly lower, preceded by a very tall, carina- 
like rib. 


Description: SHELL: Up to 30 mm. Color white. De- 
gree of convexity (width/length) in each valve approxi- 
mately 0.50. Outer surface prickly, with somewhat cha- 
otic organization (Figures 60, 61, 65-70). Sculptured by 
strong, uniform, arched, radial ribs, from 16 to 17 in each 
valve (Figures 60, 61); ribs increasing from region ante- 
rior to umbo to region between middle and posterior 
thirds, last ribs in this region taller and more separated 
from each other, last one on a weak carina (Figure 70); 
larger ribs normally possessing blade-like, projection 


L. R. L. Simone and C. M. Cunha, 2008 Page 71 


A AW} 
% \ \ Wo 


SP gi 


mb ft pm Im 


Figures 83-86. Spinosipella costeminens. Anatomy. 83. Whole specimen with right valve extracted, right view. 84. Specimen 
extracted from shell, posterior view, showing siphonal area. 85. Whole right view, some portions of right mantle lobe extracted, 
particularly regions ventral to septum, and ventral and dorsal to renal fold (rf) to expose inner surface; cerebral ganglion (ce) seen 
by transparency. 86. Same, ventral-slightly right view. Scale bar = 5 mm. 


along tip; posterior third as a slope, having 3-4 ribs simi- umbonal region narrow, smooth about 10% of shell 
lar to those of anterior region; blade like projection ab- length (Figures 64, 65, 69). Anterior, ventral, and poste- 
sent in some specimens (Figures 68-71). Posterior edge rior edges forming zigzag (Figures 62, 63, 71, 103), with 
about twice broader than anterior edge. Between umbos tips longer and narrower projected in those middle and 
and anterior edge a concavity bearing transversal ribs larger ribs. Hinge with a large cardinal tooth in right 


similar to ribs of remaining region (Figures 64, 65): pre- valve, stubby, tall (about 10% of valve width), broadly 


Page 72 THE NAUTILUS, Vol. 122, No. 2 


in 


Figures 87-90. Spinosipella costeminens. Anatomy. 87. Whole right view, mainly showing digestive tubes and main ganglia, 
topology of some structures also shown. Scale bar = 5 mm. 88. Scheme of layers of tissue in indicated region of stomach. Scale bar 
= 0.5 mm. 89. Fore- and midgut opened longitudinally for exposing inner surface (same scale of Figure 87). 90. Foot, ventral-slightly 
posterior view, sectioned transversally in two levels to show inner layer of tissues. Scale bar = 1 mm. 


pointed, somewhat flat (Figures 63, 71, 103); correspon- tionally shorter and wider (Figures 104-105). Length 
dent socket in left valve shallow, restrict to dorsal sur- about Ys to % of hinge length, and about 1.5 times aden 
face; this socket flanked by small posterior tooth, with and long. 
insertion of anterior valve edge approximately in middle Matin MUSCLE SYSTEM (FIGURES 83-87): Characters 
region of this tooth (Figure 62), anterior tooth absent similar to those in preceding species. Anterior adductor 
(Figure 62). muscle about 20% dorso-ventrally longer (Figures 83, 85). 
Additional details for this species see Poutiers (1981), Foor AND Byssus (FIGURES 85, 84, 90, 107): ): Shape 
Poutiers and Bernard (1995). and disposition similar to those in S. deshayesiana. Byssal 
LITHODESMA (FicuRES 103-105): Characters similar gland relatively deep, running immersed in ventral re- 
to those in preceding species, differing in being propor- gion of pedal musculature at about half of byssal furrow 


Figures 91-108. —Spinosipella species. Anatomy. 91. S. costeminens, middle horizontal, longitudinal section through visceral mass 
at same level as pericardium (MNHN sta. CP767, Mallory, 5 zm), Scale bar = 2 mm. 92. Same, detail of posterior region of mantle 
border. Scale bar = 1 mm. 93-102. S. deshayesiana. 93. Detail of hinge region of left valve with lithodesma (It) still attached, right 
view. Scale bar = 2 mim. 94. Detail of posterior region of supraseptal chamber, right view, right mantle lobe removed (MNHN sta. 
Oy 1499). Scale bar = Imm. 95. Infraseptal chamber roof, ventral view, right mantle lobe removed (MNHN sta. CP767). Scale bar 
2 mm. 96-99. Lithodesma (MNHN sta. DW739). Scale bar = 1 min. 96. Ventral view. 97. Dorsal view. 98. Posterior-slightly dorsal 
view. 99. Posterior view. 100. Same specimen, empty shell, ventral view, valves slightly open, lithodesma still in situ. Scale bar = 2 
mm. 101. Same, detail of hinge and lithodesma. 102. Same, ventral-slightly anterior view. 103-108. S. costeminens. 103. Shell, 
ventral view, valves open, lithodesma still attached to left valve (MNHN sta. CP1460). Scale bar = 2 mm. 104-105. Lithodesma, same 
ia other specimen), dorsal and ventral views respectively. Scale bar = 1 mm. 106. Detail of anterior region, right view, integument 
removed, mainly showing right cerebral ganglion (ce) (same lot). Scale bar = 1 mm. 107. Infraseptal chamber roof, ventral view, right 
mantle lobe removed (MNHN CP767). Scale bar = 1 mm. 108. Detail of posterior (siphonal) region, posterior view (MNHN 
CP1460). Scale bar = 2 mm 


Page 74 


THE NAUTILUS, Vol. 122, No. 2 


length towards dorsal (Figure 90, by). Thick muscular 
layer surrounding a nucleus of conective tissue (Figure 
90, cj). : 

MANTLE (Ficures 84—S6, 92, 108): Characters similar 

to those in preceding species, with fo llowing distinctive 

characters. Pair of secondary tentacles positioned be- 
tween incurrent and excurrent siphons (Figures $4, 108); 
remaining tentacles similar in size and position. Ventral 
pair of feninele »s of incurrent siphon generally symmetri- 
cal. Zigzag formed by mantle edge having second: wy 
folds positioned in more distal tips, possibly elated to 
taller radial shell ribs (Figure 108). Radial mantle gland 
(Figure 92) similar to S. de shayesiana. 

PALLIAL Cavity (FiGuRES 85-86, 107): Characters 
similar to those in preceding species, except for wider 
platform between posterior region of gills as part of sep- 
tum (Figure 107). 

VISCERAL Mass (FIGURES 85—S7): Characters similar 
to those in preceding species, differing mainly by wider 
region separating pair of renal folds in supr: aseptal cham- 
ber (Figure $5). 

CIRCULATORY AND EXCRETORY SYSTEMS (FIGURES 55, 
91): Pericardium and heart with characters similar to 
those in S. deshayesiana (Figure 91). Kidneys of similar 
features, differing mainly by enlargement of pair of renal 
folds (Figures 85— 86, rf), taller and wider, almost divid- 
ing supraseptal chamber in two—internal and external 
halves. Height of renal fold about S0% of that of su- 
praseptal chamber height. In addition to an enlargement, 
both renal folds still have posterior end in more anterior 
position and wider separation between folds and visceral 
mass (Figure 85). 

DIGESTIVE SYSTEM (FIGURES 87-89): Characters simi- 
lar to those in preceding species. Esophagus with about 
3 of visceral mass length, running horizontally, perpen- 
dicular to posterior surface of anterior adductor muscle 
(Figure 87, es). Stomach main chamber with longer re- 
gion as a blind-sac projected posteriorly. Gastric wall 
constinited by external layer er of weak connective tissue 


Figure 109. 


Geographic distribution of Spinosipella spp 


+ 8. aceulicostata #5. tinga 


S. deshavesiana ¥% S. agnes | 


(Figure 88, cj), two thick muscular layers of similar size, 
Wi ith outer layer of longitudinal muscle and inner layer of 
circular muscle (Figure 88, lo and cm). Imner surface of 
stomach (Figure 89) with posterior ead of esophageal 
folds clearly more evident that together form a flat fold. 
Another ventral fold surrounding apertures to digestive 
diverticula. Gastric style narrower (about Ys of gastric 
width); internally a pair of tall folds separating intestinal 
from sty le sac components (Figure 8S, ss, in). 

GENITAL SysTEM: Characters similar to those in pre- 
ceding species. Separated masculine and feminine com- 
ponents of gonad shown through histological sections in 
Figures 91(ts, ov). 

CENTRAL NERVOUS SYSTEM (FIGURES 87, LOG): Three 
ganglia with similar localization and size to those of pre- 
ceding species. 


Measurements (respectively length, height, width 
in mm): MNHN (Sta. 1361): 22.0 by 98.1 by 12:5 
(valve); MNHIN (Sta. CC996): 20.0 by 24.3 by 14.3 
(valve); MNHN (Sta. CP992): 19.6 by 23.3 by 12.6 


(valve). 
Geographic Distribution: Tropical West Pacific. 


Depth Range: 750-925 m. 


Material Examined: Holotype; Additional material 
(MNHN): SW PACIFIC. 4 lots [32 v, 11 specimens]; 
Wallis Is., 6 lots [15 v]; Bane Combe, 5 Lots [28 v]; 
Fortuna Is., 5 lots [1S v|; Bane Waterwitch, 2 lots [3 v]; 
Bane Tuscarora, 29 lots [63 v]; South Vanuatu - Monts 
Gemini, 4 lots [4 v, 1 specimen]; TONGA IS. 8 lots [52 
v|; Eua Is. 6 lots [12 v]; Seamount, 6 lots [29 vi South of 
Nomuka group, | lot [25 v|; Ha’apai Group, 2 lots [4 v]; 
N Ha’apai group, 3 lots { 6 v|; NW Tongatapu, 3 lots [16 
v|; SW Tongatapu, 5 lots [22 v]; Tongatapu, 6 lots [8 v]; 
S. Nomuka group, 2 lots [6 v]; Vava’ group, 1 lot [2 v]; 
NEW CALEDONIA. 5 lots [5 v, 5 specimens]; Lord 
Howe, | lot [1 v]; Bane Nova, 2 lot [8 v, 1 specimen]; 


North New Caledonia, 10 lots [tota 20 v]; South New 


Indie 


Ocean 


@ S. costeminens 


L. R. L. Simone and C. M. Cunha, 2008 


Page 75 


Caledonia, 13 lots [46 v, | specimen]; off Norfolk, 1S lots 
[9S v]; Banc Esponge, 11 lots [144 v|; Bane Kaimon- 
Maru, 9 lots [3S v|; Bane Antigonia, 1 lot [1 v]; Banc 
Jumeau-West, 4 lots [17 v]; Bane Introuvable, 7 lots [16 
v]; Bane Stylaster, 1 lot [1 v] ; Volcans Hunter and Mat- 
thew, 2 lots [2 v]; S.E. New Caledonia, 2 lots {2 v]; East 
New Caledonia, 6 lots [30 v] Banc Capel, 1 lot [lota 12 v}; 
Banc Kelso, 1 lot [6 v]; I. Loyaute, 22 lots [44 v]. FIJI. 
South of Viti Levu, 42 lots [328 v]; Southeast of Viti 
Levu, 17 lots [57 v|; Bohol/Sulu Seas, 2 lots [5 v|; Bohol 
Sea - Balicasag Island, 3 lots [5 v]; Bordau, 1 specimen; 
TAIWAN. Bashi channel, 2 lots [3 v]; South China Sea, 
1 lot {2 v]; East Taiwan, 2 lots [5 v]. (Details in Simone 
and Cunha, in press.) 


DISCUSSION 


THE GENUS SPINOSIPELLA WITHIN THE VERTICORDHDAE. 
Despite their larger size, the prickly outer surface of the 
shell, and the reduction of the hinule, which differenti- 
ates Spinosipella from the remaining verticordiids, this 
taxon has traditionally been considere d a subgenus of the 
genus Verticordia. This set of characters is sufficient in 
my opinion to allocate Spinosipella as a separate genus. 
This view was previously defended by the author of the 
genus (Iredale, 1930) and by Poutiers and Bernard 
(1995). Other distinctive characters are the spiral um- 
bones (Figures 5, 7, 21, 22, 33, 54, 53), the tall, some- 
what uniform radial sculpture, triangular in section; and 
the obesity of the valves. The spiral umbones and the 
obesity of Spinosipella are quite similar to those in the 
fossil genus Pecchiolia Savi and Meneghini in Murchison, 
1850 [type-species (by monotypy): Pecchiolia argentea 
Savi and Meneghini in Murchison, 1850 (= Chama ari- 
etina Brocchi, 1814) middle Tertiary, Europe] ( (Keen, 
1969: 857), from which Spinosipella differs in having 
well-developed ribs and zigzag edges. 
The full genus status of Spinosipella is based on the 
differences with the typical Verticordia sensu stricto 
[tvpe species (by monotypy) Verticordia cardiiformis 
Sowerby, 1844], such as the higher size and obesity of the 
valves: the additional devel opment of the prickly surface 
(which also covers the radial ribs, whereas in Verticordia, 
when a prickly surface is present, it does not cover the 


Table i. 


Spinosipella 


Character acuticostata Spinosipella agnes 


radial ribs), the absence of lunule; the spiral fashion of 
both valves; and the similarity among the radial ribs (rep- 
resentatives of Verticordia usually have an unusually 
larger rib or space between ribs). The same set of char- 
acters also differentiates Spinosipella from Trigonulina 
dOrbigny, 1842 [type species (by monotypy) T. ornata 
d'Orbigny, 1842] in the sense of Jung (1996: 46-47). 

Repre sentatives of Spinosipella also resemble those of 
the genera Haliris Dall, 1856, and Euciroa Dall, 1881, by 
their larger size, convexity, and prickly shell surface. Spi- 
nosipella differs from those two genera, however, in the 
Rigier degree of convexity, reflected in more obese 
shells in its species; in the much more developed and 
taller radial ribs; higher degree of spiralization of the 
valves; and in the expansion of the ribs beyond the shell 
margin. 

Further analysis on the verticordiid systematics and 
phylogeny can be found in the literature (e. g., Pelseneer, 
1888; Calvan: Plawén and Haszprunar, 1982: Bieler and 
Mikkelsen, 1992). 


COMPARISON BETWEEN THE 
SPINOSIPELLA SPECIES 


The differentiation between the five species of Spino- 
sipella is summarized in the respective diagnoses and in 
Table 1. The degree of differentiation in the samples of 
each species examined allows for specific separations. 
The number of radial ribs is the most notable feature; 
despite certain a small amount of intraspecific variation, 
the number of radial ribs is somewhat constant in each 
species, at least in specimens of larger size. The fossil S. 
acuticostata is the species with fewest ribs, 12-13 ( Fig- 
ures 36, 38, 39), while S. deshayesiana has the largest 
number of ribs, 16-19 (Figures 46, 48, 49, 54, 53). The 
other species possess an intermediary number of ribs. 
The species of Spinosipe la usually have radial ribs of 
relatively uniform size; the single exception is S. costem- 
inens, which has ribs clearly increasing posteriorly; in the 
posterior shell slope, however, the ane abruptly reduce 
in size, although in some specimens, particularly in the 
young ones, dhs character is not so clear, i.e., the ribs are 
saimewiat uniform-sized. The shell inflation is well de- 
veloped in most Spinosipella species, but this is clearer in 


Comparison of characters between the five studied species of Spinosipella. 


Spinosipella Spinosipella Spinosipella 
tinga deshayesiana costeminens 


Distribution Mediterranean Tropical W. Atlantic; 


Caribbean: to SE Brazil 


Shell Inflated Strongly Highly 
Sculptured between — Radial Disorganized 
radial ribs 
Prickly ribs outer Rough Rough 
surface 
Number of Ribs 12-13 15-17 


Size (mm 20.0 20.2 


South and Central 
Indo-Pacitic 

Strongly 

Radial 


S-SE Brazil Tropical West Pacific 
Weakly 


; Highly 
Radial 


Disorganized 
Weakly prickly — Strongly prickly Rough 

17-18 18-19 16-17 
10.4 11.5 20.0 


Page 76 


THE NAUTILUS, Vol. 122, No. 2 


the larger specimens; while the young specimens are 
éonsiderebly flatter (Figures 41-45). The prickly outer 
shell surface is an outstanding character of the Spino- 
sipella species; however, this character is conservative 
among the five species; the single exception is the rela- 
tively chaotic arrangement in S. agnes (Figure 8) and S. 
costeminens, while in the remaining species a radial ar- 
rangement is apparent (parallel to the radial ribs) (Figure 
32). The Pacific species S. deshayesiana has much larger, 
crispy prickles along the tip of the ribs (Figures 42, 45, 
46, 48, 49). This is lacking in the remaining species, 
except in some very young specimens (e.g.. USNM 
810889, S. agnes, 6 mm), where the prickles, aware: 
are not fully dev eloped. The prickly surface is strongly 
damaged in eroded specimens (Figure 55), becoming 
‘dies completely smooth. Spinosipella de shayesiana, 
perhaps because of this character, has the distal tips of 
the zigzag edges of the shell even longer and more pro- 
jected (Figures 41, 44, 47, 50, 51, 59, 67). The series of 
radial ribs is interrupted in the region between the um- 
bos, where a triangular smooth area appears. This area is 
particularly large in S. agnes (Figures 7, 9), but is prac- 
tically absent in S. tinga (Figures 2 1, 22): it is narrow in 
the remaining three species. The size of the specimens 
appears to be another distinctive feature, as S. tinga is 
small (around LO mm), whereas the remaining species 
are larger (20-30 mim). The hinge does not vary much 
between the Spinosipella species; however, some par- 
ticularities exist. The posterior tooth of the left valve is 
well developed in S. agnes [Figures 4, 10, 12,2 27, 28 
(arrow)], very low in S. acuticostata (Figures 35, 39), ‘ae 
practically absent in remaining species iiieues 20, 25, 
50). The tall and pointed car dinal tooth of the right valve 
is more developed in S. agnes, in such it is also sharply 
pointed and curved (Figures 3, 10, 12). In the remaining 
species this tooth is weak ly Beare and more founded 
(Figures 26, 34, 47, 51). 

The geographic and stratigraphic distribution are 
somewhat mutually exclusive for most of the species 
(Fig. 72); Spinosipe lla acuticostata is the only Mediter- 
ranean species, S. agnes occurs from Florida to Rio de 
Janeiro, S. tinga is found from Rio de Janeiro to Rio 
Grande do Sul, along the Brazilian coast. The fine- 
resolution distribution of the Indo-Pacific species is still 
unclear, but $. deshayesiana and S. costeminens, appear 
to be sympatric. Spinosipella acuticostata is a fossil spe- 
cies, occurring in Pliocene strata, while the remaining 
species are found in the Recent. Apparently no Recent 
Spinosipella occur in the Mediterranean. 

\ll samples of Spinosipella from the Atlantic and 
Mediterranean have previously been acce is ‘ot as belong- 
ing to the single species S. acuticostata (e. Abbott. 
1974: Abbott and Dance, 1983: Rios, 1994). See: 
ane ieee »s of the conchologici il, geographic, and. strati- 
or phic differences, show that the separation into three 
spe c1es is warrante od, As the shi ape che mge Ss conside rably 
during ontogeny, a spec imen of S agnes at same size as 
the holotype of S. tinga was chosen to show the differ- 


ences between those species. Figures 25-31 illustrate 
these differences. Spinosipella agnes has fewer, taller, 
and more widely spaced ribs than S. tinga (Figures 19, 
29). The shape of the shell edge is much more uniform 
in S. tinga than in S. agnes, in that the tips of the ribs are 
more expanded, extending longer beyond the shell mar- 
gin (Figures 20, 24, 28). The posterior cardinal tooth in 
the hinge of the left valve is present in S. agnes, in ani- 
mals larger than 5-6 mm, while this toath is never 
present in S. tinga (Figures 20, 25, 27-28, arrow). The 
degree of i rags is higher in S. agnes and in S. tinga 
(Figures 30, 31); S. agnes has a degree of convexity 
(width/lensth) ) in each ie of about 0.5 57, while S. tinga 
it is 0.47. 

The comparison of the previously valid species Spino- 
sipella ericia, including paratypes (Figures 41-44), and S. 
deshayesiana, does not reveal any distinction between 
them. Normally, specimens of sinaller size were identi- 
fied as S. ericia, and the large ones as S. deshayesiana. 
But examination of shell features along a growth series 
show a complete gradient linking the two taxa. The same 
lack of distinction is found in the literature for both spe- 
cies, including the original descriptions. For these rea- 
sons, despite the fact that S. ericia is the type species of 
the genus, the older name S. de shou should be 
ise Furthermore, a type specimen of S. japonica was 
also examined (Figure 45), confirming the synonymy of 
this species with S. de shayesiana. 

The distinction between the Pacific species Spino- 
sipella deshayesiana and S. costeminens is not always 
easy. With the large quantity of specimens kindly pro- 
vided by the MNHN (Paris), it was possible to analyze 
the degree of variation of both species. Spinosipella 
costeminens mostly has samples with shell possessing the 
outstandingly large, carina-like spiral ridge between the 
middle ea posterior thirds of the shell, "but sometimes 
this ridge is not so different from the others, and the 
animal become more rounded, similar to S. deshayesi- 
ana. The distinction is based mainly on the presence of at 
least a weak carina in the region between middle and 
posterior thirds, and also by the more robust ridges of S. 
costeminens specimens ( (Figures 60, 61), while those of S. 
deshayesiana lack any clear radial carina and the ridges 
are more delicate, uniform and apparently close fren 
each other (Figures 46, 48). 

The lot USNM 63200 includes 3 valves (2 left and 1 
right), collected in Barbados, the known geographic dis- 
sibutiod of Spinosipella agnes. However the right valve 
has the characters of S. de shaye sianda, pictend of hese of 
S. agnes. In addition, is looks different in the state of 
conservation, color and associated sediment, from the 
other 2 valves of the same sample. 


DISCUSSION ON ANATOMY 


More in-depth anatomical descriptions and. discussions 
on verticordiids are provided by Allen and Turner 


L. R. L. Simone and C. M. Cunha, 2008 


Page 77 


(1974), who studied 19 species of several genera. How- 
ever, no information on the anatomy of the genus Spi- 
nosipella is found in the literature. Although anatomical 
information is available here only for two of the five 
species of the genus (of course one of them is a Pliocene 
fossil), some systematic inferences can be made based on 
the scenario given in the literature the Verticordiidae and 
related families (Allen and Turner, 1974, and others, e.g., 
Fisher, 1860, 1862b: Pelseneer, 1888; Nakazima, 1967; 
Allen and Morgan, 1981). Besides the conchological 
characters discussed above, some anatomical features are 
possibly restricted to Spinosipella, such as: the wide lith- 
odesma (Figures 93, 96—LO4, It); the simplified siphonal 
tentacles (Figure 10S), which normally have secondary 
papillae; the papill: von the root of the excurrent chamber 
(Figures 7S—SO, 85, 94: pi); the absence of incurrent 
valve in infraseptal chamber. However, wide lithodesma 
have been reported for Policordia lisbetae Knudsen, 
1970 (fig. 90), which has very different shell and pallial 
tentacular characters. The study on the incurrent sipho- 
nal structures is o f particule uw import: imce in septibranchs, 
as the modified incurrent siphon is the main structure 
used in prey capture (Morton, 1987). 

On the other hand, some features appear to be char- 
acteristic of Verticordiidae, such as: elongation of lateral 
region of kidneys; the muscular stomach (see also Pur- 
chon, 1956, 1963): the separation between testis and 
ovary. By the proximity of the a daca from anterior 
adductor muscle, by the lack of incurrent valve, and by 
the simplified buc cal structures, e. ., lack of buccal cavity 
and tongue, it is possible to suggest that Spinosipella is a 
basal taxon inside Verticordiidae. Unfortunately, no 
member of the genus was analyzed in the recent com- 
parative studies on anomalodesmatans (Harper et al, 
2006). 


ACKNOWLEDGMENTS 


The authors are grateful to the researchers who loaned 
the material for this study: Winston Ponder and Ian Loch 
(AMS) for types of Spinosipella ericia; Néstor E. Ardila 
(MHNMC) and Emilio Garcia (EGC) for S. agnes (Co- 
lombia) and S. deshayesiana (Philippines); Harry G. Lee 
for S. agnes (Florida) (MHNMC) for a large lot of S. 
agnes (Colombia): and especially to Philippe Bouchet 
and Philippe Maestrati, MNHN, for the loan of a huge 
quantity of lots coming from several places of the ere | 
mostly from the Indo-Pacific. For thorough comments 
and additional information about S. acuticostata we 
thank Rafael La Perna. For material of Haliris fisheriana 
we thank to Daniel Mansur Pimpao (PPG-BAN, 
UFRGS). For help with SEM procedures, we thank Lara 
Guimaraes (MZSP). For Rachel Collin, Smithsonian In- 
stitution at Panama, we thank for the oe in the text and 
language. We thank also both referees and the Editor for 
the thoughtful correction on the manuscript. T his project 
is supported by FAPESP (Fundagao de Amparo a Pes- 
quisa do Estado de Sao Paulo), procs. no. 03/05860-6, 


04/02333-8, and a “Treinamento Técnico 3” grant, under 
the supervision of Antonia Cecilia Z. Amaral and Luiz 
R.L. Simone. 


LITERATURE CITED 


Abbott, R. T, 1974. American Seashells, second edition. Van 
Nostrand Reinhold Company. New York, 663 pp., 24 pls. 

Abbott, R.T. and S. P. Dance. 1983. Compendium of Sea- 
shells. E.P. Dutton, Inc. New York, 411 pp. 

Allen, J. A. and R. E. Morgan, 1981. The functional morphol- 
ogy of Atlantic deep water species of the families Cuspi- 
dariidae and Poromyidae (Bivalvia): an analysis of the evo- 
lution of the se ptibr: inch condition. Philosophical T rans- 
actions of the Royal Society of London (B) 294 (107: 
413-546. 

Allen, J. A. and J. F. Tumer, 1974. On the functional anatomy 
of the family Verticordiidae (Bivalvia) with descriptions of 
new species from the abyssal Atlantic. Philosophical 
Transactions of the Royal Society of London (B) 268(S94): 
401-536. : , 

Bieler, R. and P. M. Mikkelsen. 1992. Preliminary phylogenetic 
analysis of the bivalve family Galeommatidae. American 
Malacological Bulletin 9: 157-164. 

Costa, O. G. 1850. Paleontologia del Regno di Napoli. Atti della 
Accademia Pontaniana 5; 398. 

Crozier, M. A. 1966, New species and records of Mollusca from 
off Three Kings Islands, New Zealand. Transactions of the 
Royal Society of New Zealand, Zoology 8(5): 39-49. 

Fischer, P. H. 1860. Note sur les genres Hippa igus et Verticor- 
dia. Journal de ¢ onchyliologie 8 [(4) 4] (3): 295-300. 

Fisher, P. H. 1862a. Description d'une espece ae de Ver- 
ticordia. Journal de Conchyliologie 10: 35-36, pl. 5 

Fischer, P. H. 1862b. Sur l'anatomie des Hinnites. Journal de 
Conchyliologie 10 [(3)2] (3): 205-217, pl. 11. 

Harper, E. M, H. Dreyer, and G. Steiner, 2006. Reconstructing 
the Anomalodesmata (Mollusca: Bivalvia): morphology 
and molecules. Zoological Journal of the Linnean Society 
148; 395-420. 

Hedley, C. 1911. Reports on the Mollusca obtained by the 
F.LS. “Endeavour”, chiefly off Cape Wiles, South Austra- 
lia. Part 1. Zoological Results of the Fishing Experiments 
Carried out by the F.LS. “Endeavour”, 1909-10, 1: 90— 
114, pls. 17-20, 

Iredale, T. 1930. More notes on the marine Mollusca of New 
South Wales. Records of the Australian Museum of Syd- 


ney 17 (2): 384407. 
Jung, P. 1996. tia ne Pi aleontology of the northern Domini- 
can Republic. The families Cuspidariidae and Verti- 


cordiidae alles Bivalvia). Bulletins of American Pale- 
ontology 110 (351): 35-75. 

Keen, A. M 1969. Superfamily Poromyacea Dall, 1886, In: 
Moore, R. C. (ed.) Treatise on Invertebrate Paleontology. 
Part N, vol. 2, Mollusca 6 Bivalvia. The Geological Society 
of America and University of Kansas. Lawrence, pp. 491- 
952. 

Knudsen, J. 1970. The systematics and biology of abyssal and 
hadal Bivalvia. Galathea Report 11: 1-241, 20 pls. 

Marini, A. C. 1974. O Género Verticordia Wood, 1844 (Bi- 
valvia, Verticordiidae) na plataforma continental brasileira. 
Papéis Avulsos de Zoologia, Sao Paulo 28 (13): 241-244 

McLean J. H. and D. Geiger. 1995. IT Species inventory. In 
Beck, T., T. Metzger, and A, Freiwald. BIAS Biodicersity 


Page 78 


THE NAUTILUS, Vol. 122, No. 2 


inventorial atlas of macrobentic seamount animals. 
Friedrick-Alexander University. Nuremberg, 127 pp. 

Merlano, J. M.D. and M. P. Hegedus. 1994. Moluscos Del 
Caribe Colombiano: Un catélogo ilustrado. Fundacién 
Natura, 291 pp., 74 pls. 

Micali, P and A. Villari. 1991. Le specie malacologiche di Salice 
(Messina) Institute da Giuseppe Seguenza. Atti Acca- 
demia Petoritana dei Pericolati, Classe I 67 (suppl. 1) 
345-363. 

Morton, B.S. 1987. Siphon structure and prey capture as a 
guide to affinities in the abyssal septibranch Anomalodes- 
mata (Bivalvia). Sarsia 72: 49-69. 


Nakazima, M. 1967. Some observations on the soft parts of 


Halicardia nipponensis Okutami. Venus 25: 147-158, pls. 
6-9. 

Nobre, A. 1936. Moluscos marinhos de Portugal, Segundo vol- 
ume. Companhia Editora do Minho. Barcelos, 378 pp.. 
pls. S1-S6. 

Nobre, A. 1938, Fauna malacolégica de Portugal I. Moluscos 
marinhos e das aguas sé alobras. Porto, 806 pp.. 57 pls 
Pelseneer, P. 1888. Report on the anatomy of the deep- sea 

Mollusca collected by H. M.S. Challenger in the years 


1873-76. Report on the Scientific Results of the Voyage of 


H. Mi a Challenger during the years 1873-76, .. . Zool- 
ogy 27(2)[74]: 42 pp., 4 pls. 

Philippi. . A. 1844. Enumeratio Molluscorum Siciliae cum 
viventium tum in tellure tertiaria fossilium, quae in itinere 
suo observavit. Verlag Eduardi Anton, Halle. 2. iv + 303 
pp.. pls. 13-28. 

Poppe, G. T. and Y. Goto. 1993. European seashells, volume 2. 
Verlag Christa Hemmen, Wiesbaden, 221 pp. 

Poutiers, |. M. 1981. Resultats des campagnes Musorstom. | — 
Philippines (18-25 mars 1976). Mollusques: Bivalves. Mé- 
moires ORSTOM 91: 325-356. 

Poutiers, J. M. and F. R. Bemard. 1995. Carnivorous bivalve 
mollusks (Anomalodesmata) from the tropical western Pa- 
cific Ocean, with a proposed classification and a catalogue 


of Recent species. Résultats des Campagnes Musorstom, 
vol 14. Mémoires du Muséum national d'Histoire na- 
turelle 167: 107-187. 

Prezant, R. S. 1998. Superfamily Verticordioidea, pp. 420-422. 
IN Beesley, P. L.; Ross, G. J. B. and Wells, A. {eds.]. Mol- 
lusea: the southern synthesis. Fauna of Australia. Vol. 5. 
CSIRO Publishing, Melboume. Part A, xvi + 563 pp. 

Purchon R. D. 1956. The stomach in the Protobranchia and 
Septibranchia (Lamellibranchia). Proceedings of the Zoo- 
logical Society of London 127: 511-525. 

Purchon, R. D. 1963. Phy logenetic classification of the Bivalvia, 
with special reference to the Septibranchia. Proceedings 
of the Malacological Society of London 35(2-3): 71- 80. 

Rios, E. C. 1975. Brazilian marine mollusks iconography. 
Fundacao Cidade do Rio Grande. Rio Grande, 331 pp., 91 
pls. 

Rios, E. C. 1985. Seashells of Brazil. Fundagao Cidade do Rio 
Grande. Rio Grande, 328 pp., 102 pls. 

Rios, E. C. 1994. Seashells of Brazil, second edition, Fundacao 
Universidade do Rio Grande. Rio Grande, 368 pp., 113 
pls. 

Rosenberg, G. 2005. Malacolog version 4.0.1. Website http:// 
data.acnatisci.org/wasp. Philadelphia (accessed in Feb/ 
2006). 

Salas, C. 1996. Marine bivalves from off the southern Iberian 
Penninsula collected by the Balgim and Fauna I expedi- 
tions. Haliotis 25: 33-100. 

Salvini-Plawén, L.v. and G. Haszprunar. 1982. On the affinities 
of Septibranchia (Bivalvia). The Veliger 25; 83-85. 

Thiele, ]. 1934. Handbuch der sy stematischen Weichtierkunde. 
Vol. 3. pp. 779-1022. Gustav Fischer Verlag. Jena (now 
Stuttgart). Translation by Bieler, R. and P. Mikkelsen. 
1998. Handbook of Systematic Malacology. Parts 3-4; pp. 
1193-1690. Smithsonian Institution Libraries, W ashing- 
ton, DC. 

van Aartsen, J. J. 1992. European marine Mollusca: notes on 
less well-known species. La Conchiglia 23 (264): 45-48. 


THE NAUTILUS 122(2):79-93, 2008 


Page 79 


Late Eocene Conus (Neogastropoda: Conidae) 


from Florida, USA 


Jonathan R. Hendricks 
Department of Geology 
University of Kansas 

Lawrence, KS 66045-7613 USA 


jrhendri@ku.edu Gainesville, F 


Roger W. Portell 

Division of Invertebrate Paleontology 
Florida Museum of Natural History 
University of Florida 

L 32611-7800 USA 
portell@fnmh.ufl.edu 


ABSTRACT 


The neogastropod genus Conus is likely the most diverse ma- 
rine animal genus, but has an Eocene to Pleistocene fossil 
record that remains poorly understood. We discuss the fossil 
record of Eocene Conus from Florida and recognize three 
species: Conus sauridens Conrad, 1833, and two new species, 
Conus palmerae and Conus alleni. We also re-describe C. sau- 
ridens, identify its likely synonyms, and add new information 
about its geographic range. The new species C. palmerae is 
restricted to the upper Eocene Ocala Limestone of Florida, 
while C. alleni new species occurs in the upper Eocene Ocala 
Limestone of Florida and the upper Eocene Moodys Branch 
Formation of Louisiana. 


Additional Keywords: Mollusca, Fossil, 
Limestone 


Gastropoda, Ocala 


INTRODUCTION 


With over 1500 described fossil and extant species, Co- 
nus Linnaeus, 1758 (cone snails) may be the most diverse 
marine animal genus (Réckel et al., 1995). Molecular 
sequence data have offered valuable insights into the 
relationships of extant Conus species (e.g., Duda and 
Kohn, 2005), but the deep evolutionary history of Conus 
remains unclear: although about 1000 fossil s species have 
been described ( Rocke! et al., 1995), many of these are 
likely synonymous (e.g., see Hendricks, in press). Many 
fossil Conus species are based only upon type specimens 
and/or were described without reference to comparable 
material from nearby, contemporaneous fossil faunas. 
Unraveling the early evolutionary history of Conus will 
only be possible by careful examination ‘and interpreta- 
tion of its rich and well- preserved fossil record. 

Some of the oldest Sabeuinl: ited Conus fossils are 
from the early Eocene (Ypresian) of England and 
France. and possibly Pakistan (Kohn, 1990). Occurrence 
records of middle Eocene Conus in North America show 
that the genus had achieved a wide geographic distribu- 
tion soon after its first appearance in the fossil record. 
For example, Table 1 (which is derived in large part from 


Kohn and Anderson, 2008) lists names that have been 
applied to Eocene Conus hon the southeastern United 
States. Our purpose is not to revise the systematics and 
describe the fossil record of all of these taxa, but rather 
to focus on those species occurring in Florida, which 
have thus far received only eae attention (Rich- 
ards and Palmer, 1953). This lack of attention is like ‘ly 
related to the preservational nature of these Floridian 
specimens, almost all of which are preserved as internal 
and/or external molds rather than shell material. While 
we focus on Conus fossils from the upper Eocene Ocala 
Limestone of Florida, we also consider likely synonyms 
of Conus sauridens Conrad, 1833, a widespread and tem- 
porally persistent taxon that will require additional future 
attention. 


GEOLOGICAL BACKGROUND 


Formally named by W. H. Dall (in Dall and Harris, 
1892), the Ocala limestone represented all limestones 
exposed in central Florida at that time. A late Eocene 
age was not determined for the Ocala limestone, how- 
ever, until Cooke (1915) correlated the unit with deposits 
in Mississippi and Alabama. Based on lithology and micro- 
fauna, Applin and Applin (1944) divided the Ocala Lime- 
stone into an upper member and lower member. Vernon 
(1951) redefined the Ocala Limestone by restricting it to 
Applin and Applin’s (1944) upper member and placed 
their lover member in the Moodys Branch Formation. 
Further, Vernon (1951) subdivided the Moodys Branch 
Formation (as pertaining to Florida) into the basal Inglis 
Member and overlying Williston Member. Puri (1953) 
renamed Vernon's (1951) Ocala Limestone (restricted) 
as the Crystal River Formation. Later, Puri (1957) el- 
evated the Ocala Limestone to group status (Ocala 
Group)—consisting, from oldest to youngest, of the In- 
glis, Williston, and Crystal River formations—and elimi- 
mated the Moodys Branch Formation de ssignation for 
Florida Eocene de ‘posits. Base d on macro- and micro- 
fossils, Toulmin (1977: 117) correlated the Inglis For- 
mation with the lower Moodys Branch Formation in 


Page SO 


THE NAUTILUS, Vol. 122, No. 2 


Table 1. Names applied to Eocene Conus from the U.S. Coastal Plain, including names of some younger species that are here 
considered synonyms of Eocene taxa. Type specimen abbreviations: H, holotype; L, lectotype; and F, figured. 


Taxon 


Type 


Type locality 


Present disposition 
according to this study 


sauridens Conrad, 1833 
claibornensis I. Lea, 1833 
gyratus Morton, 1834 
parcus H. C. Lea, 1841 
mutilatus Tuomey, 1852 
tortilis Conrad, 1855 
alveatus Conrad, 1865 


agaaaaan 


C. subsauridens Conrad, 1865 
C. pulcherrimus Heilprin, 1879 


C. jacksonensis Meyer, 1885 


C. deperditus var. subdiadema 
de Gregorio, 1890 
C. improvidus de Gregorio, 1890 


C. (Conospirus) granopsis de 
Gregorio, 1890 

C. smithvillensis Harris, 1895 

C. cormacki Harbison, 1944 


C. (Leptoconus) santander 
Gardner, 1945 

C. (Leptoconus) haighti 
Gardner, 1945 

“Conus sp. A” Palmer in 
Richards and Palmer, 1953 

“Conus sp. B” Palmer in 
Richards and Palmer, 1953 


lant 


). eracens Hoerle, 1976 

C. alveatus spiralis Dockery in 
MacNeil and Dockery, 1954 

C. (Lithoconus) smithvillensis 
var. Dockery, 1980 

C. (Lithoconus) nocens Garvie, 
1996 

C. (Lithoconus) smithvillensis 

var. Dockery, 1980 in 

Campbell (1995) 


ANSP 14854 (L) 

Lost 

ANSP 211 (H) 

ANSP 13161 (L) 

None 

ANSP 13196 (H) 

ANSP. 13446 (L); See MacNeil 
and Dockery (1954, PI. 35, 
Fig. 26) 

ANSP 53812 (H) 

AMNH-FI 10175 (H?) 


Unknown, but specimen 
purported to be holotype 
figured by Harris and 
Palmer (1947, pl. 62, fig. 17) 

PRI 26436 (H) 


Lost (see Palmer and Brann, 
1966) 

Lost (see Palmer and Brann, 
1966) 

BEG 34656 (H) 

ANSP 16415 (AH) 


USNM 495181 (H) 

USNM 495182 (H) 

UF 108683 

UF 108858 (designated here as 
holotype of C. palmerae, 
new species; formerly FL 
Geol. Survey I-7634) 

USNM 220109 (H) 

USNM 376678 (H) 

MGS 590 (F) 

UT-TMM 962TX22 (H) 


UNC 15445 (F) 


Claiborne, Alabama 

Claiborne, Alabama 

South Carolina, locality unknown 
Claiborne, Alabama 

Wilmington, North Carolina 
Jackson, Mississippi 

Vicksburg, Mississippi 


Probably Claiborne, Alabama 
Claiborne, Alabama 


Jackson, Mississippi 


Claiborne, Alabama 
Claiborne, Alabama 
Claiborne, Alabama 
Smithville, Texas 
Santee Cooper Canal, 
South Carolina 
Moseleys Ferry, Texas 
Arroyo Veleno, Texas 


Gulf Hammock, Florida 


Gulf Hammock, Florida 


Calhoun County, Florida 
Smith County, Mississippi 


Near Newton, Mississippi 
Bastrop County, Texas 


Near Cross, South Carolina 


C. sauridens 
C. sauridens? 
Nomen dubium 
C. sauridens 
Nomen dubium 
C. sauridens 
C. sauridens 


C. sauridens 

Eosurcula pulcherrimus 
(a turrid) 

C. sauridens 


C. sauridens 
Nomen dubium 
Nomen dubium 


C. smithvillensis 
Nomen dubium 


C. sauridens? 
C. haighti 
Unidentifiable 


C. palmerae, new species 


. sauridens? 


ag 


_ sauridens 
C. smithvillensis? 
C. sauridens? 


C. smithvillensis? 


Alabama, Mississippi, and Louisiana. He also correlated 
the Williston Formation with the upper Moodys Branch 
Formation in Alabama, Mississippi, and Louisiana. For 
detailed correlations, see the Correlation of Stratigraphic 
Units of North America—Gulf Coast Region (1988). 
Based on lithology, Scott (1991)—in order to follow 
the North American Stratigraphic Code (North Ameri- 
can Commission on Stratigraphic Nomenclature, 
1983)—reduced the Ocala Group to formational rank 
and returned to the terminology used by Applin and 
\pplin (1944). Therefore, current designations used by 


the Florida Geological Survey and United States Geo- 
logical Survey are: lower member of the Ocala Lime- 
stone (formerly Inglis Formation) and upper member of 
the Ocala Limestone (formerly Williston and Crystal 
River formations). See Figure 1 for a brief history of 
stratigraphic divisions of the Ocala Limestone in Florida. 

In outcrop and shallow subsurface, the Ocala Lime- 
stone occurs in northwestern peninsular Florida and a 
small area of the Florida panhandle adjacent to Georgia 
and Alabama. Lithologically, the Ocala Limestone is a 
relatively pure carbonate. Non-carbonate minerals 


J. R. Hendricks and R. W. Portell, 2008 


Page $] 


SERIES | STAGE |APP!P & Appling Vernon, 1954 


1944 Puri, 1957 


Scott, 1991 


Upper Member 
Ocala Limestone 
(restricted) 
Crystal River 
Formation 


Upper Member 


Williston Member 
Ocala Group 
Williston Formation 


Z 
=< 
Z 
Oo 
WY 
< 
UO 
< 
= 


Ocala Limestone 
Ocala Limestone 


UPPER EOCENE 


Lower Member 
Moodys Branch Formation 


Inglis Member 
Inglis Formation 
Lower Member 


Figure 1. History of stratigraphic divisions of the Ocala 
Limestone in Florida (modified from Oyen and Portell, 2001). 


(quartz, chert, and clay minerals) represent less than 5% 
of the rock volume in the lower member and less than 
10% in the upper member (Oyen, 1995). The lower 
member (formerly Inglis Formation) is primarily a clean 
packstone and grainstone that represents a higher en- 
ergy, subtidal environment while the upper tener 
(formerly Crystal River Formation) has a lithology of 
muddy pac -kstone and wackestone interpreted to signify a 
lower energy (below wave base), deeper subtidal deposit 
(Fenk, 1979). 

The Ocala Limestone contains diverse and abundant, 
shallow water marine invertebrate fossils consisting pri- 
marily of foraminifera, mollusks, and echinoids. Taxa 
with calcitic shells are preserved as body fossils, while 
taxa that had aragonitic shells typically occur as internal 
and external molds, although on rare occasions they can 
be found as omen of calcite or silica. Most of 
the calcitic shelled mollusks (e.g., oysters and scallops) 
and those found as pseudomorphs have been well docu- 
mented (Harris, 1951; Richards and Palmer, 1953). Few 
of the many moldic Ocala Limestone mollusks, however, 
have received the critical attention that they need (Por- 
tell and Vokes, 1997) 


MATERIALS AND METHODS 


Most of the specimens examined in this study are from 
the Florida Museum of Natural History (FLMNH) Di- 


vision of Invertebrate Paleontology at the University of 


Florida (UF) and locality details for individual UF speci- 
men lots listed below can be accessed online via the 
FLMNH Invertebrate Paleontology database at http:// 
www.flmnh.ufl.edu/invertpaleo/search.asp. Besides UF 


specimens, some type specimens from the Academy of 


Natural Sciences, Philadelphia (ANSP), the Texas Natu- 


ral Science Center at the Univ ersity of Texas ( (specimens 
carry the acronym TMM for Texas Memorial Museum), 
and the Texas Bureau of Economic Geology (BEG; 
specimens now at the Texas Natural Science Center) 
were also examined. Other specimens referred to in the 
text are from the Geological Survey of Alabama (GSA), 

the American Museum of Natural History (AMNH), the 
Mississippi Geological Survey (MGS), the Paleontologi- 
cal Research Institution (PRI), the University of North 
Carolina (UNC), and the United States National Mu- 
seum of Natural History (USNM), 

Morphospecies were recognized from museum speci- 
mens (primarily those at the FLMNH) using the criteria 
discussed by Smith (1930), Réckel et al. (1995), and 
Hendricks (in press) and morphological terminology and 
measurements collected follow those authors. Two terms 
are introduced here that relate to characters states of the 
subsutural flexure, which is “the backward curving or 
bending of the shell aperture below the suture of whorl 
contact” and “{bJeing a feature of the apertural margin it 
is strikingly manifest in the growth lines [on the sutural 

ramp] of well- preserved cones” (Smith, 1930: 284). 
These two terms are: symmetrically curved subsutural 
flexure and diagonal subsutural flexure. A symmetrically 
curved subsutural flexure (also see the more generalized 
“curved type” described by Smith, 1930) has a maximum 
point of curvature centered between the margins of the 
sutural ramp. A diagonal subsutural flexure forms a 
straight or eee straight | ine that crosses diagonally 
(abaxioventrally) across the sutural ramp. Also see 
Muniz-Solis (1999) for illustrations of different subsu- 
tural flexure morphologies. 

Measurements were collected using digital and dial 
calipers and include: shell length (SL), maximum diam- 
eter (MD), aperture height (AH), height of maxi- 
mum diameter (HMD), and spire angle (SA). From 
these measurements, the following morphometric ratios 

(developed by Réckel et al., 1995) were computed 
to characterize shell sh: ape: relative diameter (RD; 
RD = MD / AH); position of maximum diameter (PMD; 
PMD = HMD / AH); and relative spire height (RSH; 
RSH = [SL - AH]/SL). Recognized Floridian morpho- 
species were compared with previously described species 
of Eocene Conus from the southeastern United States. 

Although the simple internal molds of Conzs fossils 
offer little taxonomically relevant information, room tem- 
perature vulcanizing (R.T.V.) silicone rubber casts made 
from external molds are often very useful for recognizing 
species. The external molds discussed herein were first 
gently cleaned to remove loose debris and then impreg- 
nated with polyvinyl butyral (Butvar-76), if needed, to 
consolidate the limestone. Once dried, clay dams were 
constructed around the outside of the limestone blocks 
containing the external molds. Then, de-aerated R.T.V. 
silicone rubber was slowly poured into and above the 
molds and again de-aerated in a vacuum chamber. Later, 
the cured rubber casts were carefully lifted from the 
molds. For more information regarding this technique 


Page 82 


THE NAUTILUS, Vol. 122, No. 2 


see Chaney (1989) or visit http://paleo.cc/casting/ 
silsum.htin. 


SYSTEMATICS 


Our examination of Eocene Conus fossils from Florida 
resulted in the recognition of one distinctive morpho- 
species that was first described by Conrad (1833) as 
C. sauridens, and two new species (C. palmerae and 
C. alleni) which are described below. 


Family Conidae Fleming, 1822 
Genus Conus Linnaeus, 1758 


Conus sauridens Conrad, 1833 
(Figures 2-10, 15-22; Table 2) 


Conus sauridens Conrad, 1833: p. 33; Conrad, 1835: pl. 15, fig. 
7. For comprehensive synonymy listings, see Palmer 
(1937), Harris and Palmer (1947), and Palmer and Brann 
(1966). 

?Conus claibornensis I. Lea, 1833: 186. 

Conus parvus H. C. Lea, 1841: 103, pl. 1, fig. 24. 

Conus tortilis Conrad 1855: 260, pl. 15, fig. 5. 

Conus alveatus Conrad, 1865: 148, pl. 11, fig. 4. 

Conus subsauridens Conrad, 1865: 148, pl. 11, fig. 9. 

Conus jacksonensis Meyer, 1585: 466. 

Conus deperditus var. subdiadema de Gregorio, 1890: 20, pl. 1, 
figs. 56-58. 

Conus (Leptoconus) santander Gardner, 1945: 
figs. 5, 9, 10, 14. 

?Conus cracens Hoerle, 1976: 14, 16, pl. 3, figs. 1-3. 

Conus alveatus spiralis Dockery in Mz acNeil and Dockery, 
1984: 165, pl. 59, figs. 3, 4. 

2Conus nocens Garvie, 1996: 90, pl. 19, figs. 9, 10. 


251, pl. 26, 


Diagnosis: Early postnuclear whorls tuberculate, later 
teleoconch whorls smooth; shell often widest below 
shoulder; shoulder typically ridge-like; sutural ramps 
with raised spiral threads; incised spiral grooves at base 
of shell. 


Description: Shell medium to large-sized (up to 116 
mm). Last whorl conical to broadly and ventricosely coni- 
cal; outline typically convex near shoulder, nearly straight 
below. Shell often widest below shoulder. Shoulder ty pi- 
cally broadly carinate and forming pronounced ridge, less 
often angulate: smooth. Spire of low to moderate he ight, 
spire angle ( SA) typically obtuse; outline typically straight 
to concave in smaller specimens, usually sigmoid in 
larger shells. Larval shell multispiral, with at Teatt three 
whorls. Early postnuclear whorls tuberculate. Subsutural 
flexure symmetrically curved, depth about 2.5x width. 
Teleoconch sutural ramps typically sigmoidal (rarely flat 
or convex), with raised spiral threads (typically three to 
seven; deine ‘ry increases with shell size) and interve ning 
grooves; ornamentation begins on earliest postnuclear 
whorls as a single incised spiral groove. Aperture opening 
of approximately uniform width from base to shoulder. 
Some large specimens bear a pronounced siphonal fas- 
ciole. Last whorl with incised spiral grooves. at base, 


sometimes extending weakly to near center of whorl and 
occasionally to shoulder in small shells; spiral grooves 
obsolete in some large shells. 


Shell Morphometrics: Thirteen shells of type and 
non-type specimens of C. sauridens were measured and 
morphometric ratios were computed from these mea- 
surements (Table 2): relative diameter (RD) ranges from 
0.57-0.74 (average = 0.67); position of maximum diam- 
eter (PMD) ranges from 0.83-0.95 (average = 0.91); and 
relative spire height (RSH) ranges from 0. 090.21 (aver- 
age = 0.17 ). 


Type Specimens: Type specimens examined include: 
ee 14854, lectotype of Conus sauridens (Figures 2, 

; ANSP 535813, three paralectotypes of C. eee 
ica 4-6); ANSP 53812, holotype of C. subsauridens 
(Figure 7); ANSP 13161, holotype of C. parvus (Figure 
8); and ANSP 13196, holotype of C. tortilis (Figures 9, 
10). See Table 2 for measurements of type specimens. 


Type Locality and Occurrence: Conrad (1833) de- 
scribed the species from specimens collected at Clai- 
borne, Alabama. Palmer (1937: 10) designated the type 
locality and stratum of Conus sauridens as PRI station 
104: “ ‘Ferruginous sand’ bed at Claiborne, on the Ala- 
bama River, Monroe County, Ala. Gosport sand” 
(Palmer, 1937: 10). The Gosport Sand is the uppermost 
formation of the Claiborne Group and is late middle 
Eocene (Bartonian) in age (Dockery, 1980). This species 
has been previously reported from numerous Paleogene 
U.S. Gulf Coast strata, including the middle Eocene 
Claiborne Group, as C. tortilis in the upper Eocene 
Jackson Group, and as C. alveatus in the lower Oli- 
gocene Vicksburg Group (the reader is directed to the 
follow ing sources for detailed discussions of specific oc- 
currence records: Palmer, 1937; Harris and Palmer, 
1947; Palmer and Brann, 1966; and MacNeil and Dock- 
ery, 1984). Conus sauridens also occurs in the upper 
Eocene Ocala Limestone of Alachua, Suwannee, and 
Jackson counties, Florida. This species also questionably 
occurs as C. cracens in the lower Miocene Chipola For- 
mation of northern Florida. 


Other Material Examined: In addition to the type 
specimen lots listed above, 25 specimen lots containing 
C. emer ns we re examined se r SO specimens). ). These 


UF ee UF 16661, U F 18874, UF 57713, UF 57733, 
UF 100693, UF 101995, UF 114376, UF 115875, UF 
119912, UF 119913, UF 119918, UF 119962, UF 
120021-UF 120027, UF 120032, UF 122384, and UF 
126927. 


Discussion: Conus sawridens was the third fossil Co- 
nus species to be described from North America, and the 
first from the Paleogene (Green described the Neogene 
species C. deluvianus and C. marylandicus in 1830; see 

Kohn, 1992). As such, this species has received much 
attention in the literature, particularly with regards to its 


J. R. Hendricks and R. W. Portell, 2008 


Figures 2-18. Specimens of Conus sauridens Conrad, 1833 (2-10, 15-18), C. cracens Hoerle, 1976 (11, 12), and C. nocens 
Garvie, 1996 (13, 14). See text and Table 1 for locality information. All scale bars equal 1 cm. Scale bar above Figure 5 pertains to 
Figures 2-7 and 11-14. Scale bar below Figure 8 pertains only to that figure. Scale bar between Figures 9 and 10 pertains to Figures 
1 15-18. 2-3. Lectotype (ANSP 14854) of C sauridens, shell length 33.4 mm, maximum diameter 21.1 mm. 4. Paralectotype 
ANSP 53813-3) of C. sawridens, shell length 31.9 mm. 5. Paralectotype (ANSP 53813-1) of C. sauwridens, maximum diameter 22.3 
mm. 6. Paralec totype (ANSP 53813-2) of C sauridens, shell length 30.2 mm. 7. Holotype (ANSP 53812) of C subsauridens Conrad 
1865, shell length 33.4 mm. 8. Holotype (ANSP 13161) of C. parvus Lea, 1841, shell length 6 mm, 9-10. Holotype (ANSP 13196 
of C. tortilis Conrad, 1855, shell length 90.0 mm, maximum diameter 53.0 mm, 11-12. Paratype (UF 119560-1) of C. cracens Hoerk 
1976, shell length 32.2 mm, maximum diameter 17.8 mm. 13-14. Holotype (TMM-962TX22) of C. nocens Garvie, 1996, shell length 
19.2 mm, maximum diameter 8.7 mm. 15. R.T.V. silicone rubber cast from external mold of C. sauridens (UF 120026). 16. R.T.\ 
silicone rubber cast from external mold of C. sauridens (UF 122384). 17-18. Highly leached, heavily pitted, and slightly silicified 
shell of C. sauridens (UF 120027), shell length 27.4 mm, maximum diameter 16.5 mm 


Page 84 


THE NAUTILUS, Vol. 122, No. 2 


Table 2. Measurements (in mm) and morphometric ratios of type and non-type specimens examined. Some measurements could 
not be collected from casts made from molds; other measurements (and associated ratios) are not accurate because of shell damage 


(indicated by an asterisk, °). Morphological abbreviations: SL, shell length; MD, maximum diameter; AH 


AH, aperture height; HMD, 


height of maximum diameter; SA, spire angle (in degrees); RD, relative diameter; PMD, position of maximum diameter; and RSH, 
relative spire height. Specimen abbreviations: H, holotype; L, lectotype; P, paratype; and PL, paralectotype. 


Specimen 


SL MD AH HMD SA RD PMD RSH 


Type Specimens of Conus sauridens Conrad, 1833 

ANSP 14854 (L, C. sauridens, Figures 2, 3) 
ANSP 53813- : (PL, C. sauridens, Figure ! 5) 
ANSP 53813-2 (PL, C. sauridens, Figure 6) 
ANSP 53813- 3 (PL, C. sauridens, Figure 4) 
ANSP 53812 (H, C. subsauridens, Figure 7) 
ANSP 13161 (H, C. parvus, Figure §) 

ANSP 13196 (H, C. tortilis, Figures 9, 10) 


Non-Type Specimens of Conus sauridens Conrad, 1833 
UF 283 
UF 290-1 
UF 8511-1 
UF S647 
UF 16726 
UF 115875 
UF 120022 
Conus palmerae new species 
UF 108858 (HL, C. palmerae, new species, Figures 23, 24) 
UF 15586 (P) 
UF 18599 (P, Figure 25) 
UF 18711 ( 
UF 18719 ( 
UF 18737 (P, Figure 26) 
( 
 ( 
> ( 


a 7 57018 


UF 74473 3 (P, Figures 29, 30) 
UF 110360 (P) — 

UF 111327 (P) 

UF 112981 (P) 


Conus alleni new species 
UF 119920 CH, Figures 31, 
UF 119919 (P, Figure 34) 
UF 119976 (P, Figure 35) 
UF 119977 (P, Figure 33) 

Other Type Specimens 
ANSP 16145 (H, C. cormacki, Figure 39) 

UF 119560-1 (P, C. cracens, Figures 11, 12) 

UF 76798-1 (P, C. cracens) 

TMM-BEG 35656 (H, C. smithvillensis, Figures 36-38) 
TMM-962TX22 (H, C. nocens, Figures 13, 14) 
TMM-9627TX23 (P, C. nocens) 


43.6 

30.2° 
31.9° 
33.4° 


48.83 27.38 40.5 
41.44 24.3 33.1 
22.95 12.36 19.5 18.0 110 0.63 0.92 0.15 
33.45 19.86 28.4 25.7 119 0.70 0.90 0.15 
116.11 69.15 95.2 84° 121 0.73 0.88° 0.18 


22.0° 11.5° 146° 13.7° 77 
— 5 

00.1° 
12.0° 
24.9° 
56.1° 


Wwwww 
C C2 DD 


UU GO 


21.1 30.5 28.9 145 0.70 0.95 0.09 
22.3° 36.5 33.6 106 0.61 0.92 0.16 
15.0 26.2 24.5 110 0.57 0.94 0.13° 
16.7 25.9 23.5 101 0.64 0.91 0.19° 
17.3 28.2 26.2 115 0.61 0.93 0.16° 
90.9 53.0 75.1 62.1 100° 0.71 0.83 0.17 
37.59 22.4 30,3 27. 110 0.74 0.91 0.19 
24.43 1187 19.3 a 90 0.62 0.90 0.21 


109 0.68 0.58 0.17 
109 0.73 0.92 0.20 


wwe 
on 
ARwn 


22.3° — _— 103 
21.5° 
— 14.2° — — 101 
— 19.3 — —_ 9S 
os 19.0° — 9S 
65.0% 284°  53,8° 
11.3 a 10S 
— 12.4 — — LOS 
2 20.7° = 31.3 28.6 121 0.66" 0.91 0.14 
2° = =17.3° — — 92 
Oo”  L7.2° 
1° 16.8° — — 112° 
ue 22.3 
2 17.8 27.5 24.0 113 0.65 0.87 0.15 
0 37.3° 
145° 28.6 26.9° 53 0.52° 0.94° 0.32 
2 8.7° 15.1 14.3 S7 0.58° 0.95 0.21 
19.4 8.7° 15.4 14.5 ST 0.56 0.94 0.2] 


morphological variation and probable synonyms. Much 
of the following was derived from discussions in Palmer 
(1937), Harris and Palmer (1947), and Palmer and Brann 
(1966). 

Timothy Conrad (1533) described C. sauridens from 
shells that he collected at Claiborne, Alabama. According 
to Palmer (1937: 461), the 
sauridens consists of 5 specimens” which were appar- 
ently glued to one card, cataloged as ANSP 14854. 


Palmer (1937: 461) goes on to state that the 


“Conradian collection of 


“type of 


Conus subsauridens Con. |see below] is also on the card 
with the Conus sauridens collection.” These two passages 
suggest that Conrad’s Claiborne collection originally con- 
siste ae of six specimens. The collection now consists of 
five specimens: the lectotype (ANSP 14854; Figures 2, 3) 
of C. sauridens, three paralectotypes (ANSP 53813: Fig- 
ures 4-6), and the holotype of C. subsauridens ( ANSP 

S12; Figure 7); the location of any possible er speci- 
men is not iano One low- =e ‘d specimen in the type 
series (ANSP 14554; Figure 2) closely resembles Con- 


J. R. Hendricks and R. W. Portell, 2008 


Page 85 


rad’s (1835, pl. 15, fig. 7) original figure of C. sawridens. 
cave (1937) alluded to this specimen in her text as 
iets Conrad’s figure, but did not formally designate 
it as the lectotype. Kohn (1992)— following Moore 

(1962) and Palmer and a (1966)—considered this 
tae, to be the lectotype and we accept his conclu- 
sion. 

Palmer (1937) argued that the matrix filling the aper- 
tures of the shells in Conrad’s Conus sawuridens series 
suggests that the shells do not likely share the same geo- 
logical provenance. The matrix filling the lectotype is 
orange and appears to be sand from the Gosport For- 
mation (Palmer, 1937). The matrix filling the three para- 


lectotypes is light-gray. The matrix f filling the holotype of 


C. subsauridens is orange, resembling that of the lecto- 
type of C. sauridens (Palmer [1937: : 461] described the 
matrix filling the holotype of C. subsauridens as “a red, 
silicified matrix resembling that of the Orangeburg” 
material from South Carolina). It is important to note, 
however, that these differences in matrix color may be 
due to diagenetic weathering of the original matrix ma- 
terial. 

The lectotype (ANSP 14854; Figures 2,3) of C. sau- 
ridens differs from the paralectotypes (ANSP 53813; Fig- 
ures 4-6)—which, with the holotype of C. subsauridens 
(see below), are of the much more common morphol- 
ogy—in the following respects: it has flat to slightly con- 
vex sutural ramps with more raised spiral threads (about 
6) than is typical, the shoulder is angulate rather than 
forming a carinate ridge, the widest point of the last 
whorl is nearly at the shoulder rather than beneath it, 


and the spire is lower than in most other specimens of 


the species (see Table 2); further, the lectotype does not 
provide definite evidence of tubercles on the early post- 
nuclear whorls, though these are highly eroded on this 
specimen, 

Isaac Lea (1833) described Conus claibornensis from 
materials sent to him by Judge Charles Tait of Claiborne, 

Alabama, but lost his only specimen before it could be 
figured ( Kohn, 1992). His description suggests that his 
specimen was a C. sauridens, a name that has priority by 
three months (Kohn, 1992). Harris (1895), Palmer 
(1937), and Palmer and Brann (1966) synonymized C. 
claibornensis with C. sauridens, while de Gregorio 
(1890), Dall (1896), and Kohn (1992) regarded this taxon 
as a nomen dubium. Given that no available evidence 
suggests that more than one fossil Conus species is 
present at Claiborne. Alabama, we agree with the former 
authors that C. claibornensis is equivalent, while ques- 


tionably. to C. sauridens. For a historical overview of 


interactions between I. Lea, T. A. Conrad, and C. Tait, 
see Wheeler (1935). 

In 1841. H. C. Lea (I. Lea’s father) described Conus 
parcvus on the basis of a single, small, damaged shell 
ANSP 13161; Figure §) from the Gosport sand at Clai- 
borne, Alabama. We agree with Dall (1896), Palmer 
(1937). and Palmer and Brann (1966) that C. parvus is a 
juvenile C. sauridens. Features uniting the holotype of C. 


parvus with C. sauridens include: tuberculate early post- 
nuclear whorls, raised spiral threads on the sutural 
ramps, and incised spiral grooves on the anterior half of 
the last whorl. 

Conrad (1855) described Conus tortilis from one large 
specimen (ANSP 13196; Figures 9, 10) from Jackson, 
Mississippi and differentiated it from C. sauridens by its 

‘more prominent and convex spire, in the large twisted 
callus at base, & c.” (p. 260). Dall (1896) and Palmer 
(1937) both considered C. tortilis synonymous with C 
sauridens and Palmer (1937) described C. tortilis as rep- 
resenting “the maximum growth of the species” (p. 459). 
We agree with these authors that specimens of C. tortilis 
are large C. sauridens. 

Conrad (1565) described two additional species of 
Eocene Conus: C. subsauridens and C. alveatus. Conrad 
stated that C. subsauridens was from “the Burrstone, 
probably, of Alabama”; we assume (see above) that the 
holotype (ANSP 535812) is from Clairborne. We in- 
spected the type of C. subsauridens (Figure 7) and agree 
with Dall (1896) and Palmer (1937) that it is a junior 
synonym of C. sauridens. 

The type locality for Conus alveatus is Vicksburg, Mis- 
sissippi; MacNeil and Dockery (1984) suggested that the 
lectotype (ANSP 13446; MacNeil and Dockery, 1954, pl. 
38, fig. 26) and paratype (ANSP 13494) are probably 
from the Byram Formation. Conrad (1865: 148) differ- 
entiated C. alveatus from C. sauridens by its “less el- 
evated and... more profoundly carinated spire, and the 
revolving lines on the spire are less numerous than in the 
former [C. sauridens]|.” Dall (1896) and Palmer (1937) 
recognized Conus tortilis and C. alveatus, respectively, as 
the Jacksonian (upper Eocene) and Vicksburgian 
(lower Oligocene) forms of the older, Clairbornian 
(middle Eocene), C. sauridens. MacNeil and Dockery 
(1954), however, continued to recognize C. alveatus as a 
distinct Oligocene species occurring in Mississippi and 
Mexico; they did not compare C. alveatus (or Dockery’s 
subspecies C. alveatus spiralis; in MacNeil and ee kery, 
1984) with C. sauridens. MacNeil and Dockery’s 1984) 
figures of C. aleveatus (including the lectotype: al 35 
fig. 26) appear consistent with C. sauridens as circum- 
scribed here. 


Meyer (1885) described—but did not figure—Conws 


jac eee from Jackson, Mississippi, and de scribed the 


species as similar to Conus protracta Meyer, 1S85 (an 
Oligocene taxon from Vicksburg and Red Bluff, Missis- 
sippi that we accept, but do not consider further here: 
see MacNeil and Dockery, 1954 for details), but “with 
revolving lines on the spire” (p. 466). Meyer [1886] pre- 
sented C. protracta as C. protractus and the latter 
spelling is the one most commonly seen in the literature. 
Harris and Palmer (1947: pl. 62, fig. 17) figured a speci- 
men (unnumbered, but said to be from the collections of 
the Geology Department of Johns Hopkins University) 
that they regarde das the holotype of C. jac ksonensis. 
This specimen “consists of the apical whorls” and is 3.5 
mm in size (Harris and Palmer, 1947: 446). Harris and 


Page 86 


THE NAUTILUS, Vol. 122, No. 2 


Palmer (1947) considered this taxon a junior synonym of 


C. sauridens and we agree, particularly because of the 
presence of vubensulater early postnuclear whorls and the 
presence of raised spiral threads on the sutural ramp. 

De Gregorio (1890) described Conus deperditus var. 
subdiadema from Claiborne, Alabama. Palmer and 
Brann (1966) considered this subspecies equivalent to C. 
sauridens and, based upon our inspection of de Grego- 
rio’s (1890) figures, we agree. We have not, however, 
viewed the holotype of C. deperditus var. subdiadema, 
which Palmer and Brann reported as PRI 26436. 

Gardner (1945: 252) described Conus santander as 
including “those species from the western Gulf that have 
formerly been included under Conus sauwridens Conrad, 
described from Claiborne.” Gardner's figures of the ho- 
lotype (USNM 495181; Moseleys Ferry, Burleson 
County, Texas) of C. santander appear consistent with C, 
sauridens and we consider this taxon a probable syn- 
onym. 

Hoerle (1976) described Conus cracens (see paratype 
in UF 119560, Figures 11, 12) from the lower Miocene 
Chipola Formation of northern Florida and noted its 
strong similarity to C. sauridens: “C. cracens appears to 
be a descendant of the v widespread (Alabama, Missis- 
sippi, Texas) ) middle Eocene. to Oligocene species, C. 
sauridens Conrad” (p. 16). She differentiated C. cracens 
from C. sauridens on the basis of several characters: 
the “nodes on the spire whorls persist for a greater num- 
ber of tums on C. cracens, also it is larger, more slender, 
with stronger and more opisthocyrt growth lines and 
more pronounced basal ornament” (p. 16). We examined 
Hoerle’s paratypes at the FLMNH and could not find 
any discrete morphological characters separating the two 
taxa. Given the vast amount of geological time separating 
the species (over 12 million years), however, we ques- 
tionably synonymize C. cracens with C. sauridens. Re- 
solving the relationship between these two taxa will re- 
quire additional study. 

Finally, Garvie (1996) described Conus nocens on the 
basis of two small specimens from the Reklaw Formation 
of Bz istrop County, Texas: TMM-962TX22 ( (holotype; er- 
roneously publishe das UT-TMM 84822; Figures 13, 14) 
and TMM-962TX23 (paratype, erroneously published as 
UT-TMM 54823). In addition to these two specimens, 
Garvie (1996) reported that he examined 25 additional 
specimens of C. nocens from the Weches Formation and 
over 100 specimens from the Cook Mountain Formation. 
While we did not observe these additional, stratigraphi- 


cally younger specimens, the holotype and paratype of 


C. nocens appear consistent with Conus sauridens. They 
each have raised spiral threads on the sutural ramp, sym- 
metrically curved subsutural flexures, and incised spiral 
srooves on the anterior half of the last whorl. Further, 
both have last whorl shapes similar to C. sauridens. The 
early postnuclear whorls of the holotype are tuberculate, 
but most appear smooth on the paratype. Garvie (1996: 
90) stated that four characters separate C. nocens from 
C. sauridens: “the flat sides, the sharp unridged carina, 


the coarse, strong, spiral basal lines, and the lack of 
strong growth lines on the ramp.” We do not consider 
these qnrapteristies=an isolation or combination— 
sufficient to discriminate C. nocens from C. sawridens (as 
circumscribed above), especially because TMM-962TX22 
and TMM-962TX23 are both shells of juveniles. Since, 
however, we have not seen the other specimens of C. 
nocens mentioned by Garvie (1996), we consider our 
synonymy of C. nocens with C. sauridens tentative. Re- 
gardless of their taxonomic identity, TMM-962TX22 and 
TMM-962TX23 are important specimens because of 
their likely stratigraphic position in the Reklaw Forma- 
tion. The Reklaw Formation is thought to span the early- 
middle Eocene (or, Ypresian- Lutetian) boundary and has 
a relative age equivalent to nannoplankton biochrono- 
zone NP14 (Zachos and Molineux, 2003: fig. 2), which 
has an absolute age of about 49.7 to 47.3 Ma (Berggren 
and Pearson, 2005). This age would make these two 
specimens the oldest known Conus fossils from the 
United States Coastal Plain, and only slightly younger 
than the oldest Conus in general ( (Ypresian of England 
and France; Kohn, 1990). This oldest regional occur- 
rence record is tempered, however, by the fact that the 
position of the original collection locality of TMM- 
962TX22 and TMM-962TX23 (“Devil’s Eye, Colorado 
R.”; station 11 of the Geological Survey of Texas and 
locality 11-T-36 of the Texas Bureau of Economic Geol- 
ogy) is uncertain (Zachos et al., 2005) and may no longer 
exist (Garvie, 1996). Further, no additional specimens of 
Conus from the Reklaw Formation have yet been found 
in the collections of the Texas Natural Science Center (A. 
Molineux, personal communication to JRH, July 11, 
2007). 

Most of our knowledge of Eocene C. sauridens in 
Florida is from R.T.V. silicone rubber casts of external 
molds in limestone (e.g., UF 120026, Figure 15; UF 
122354, Figure 16), though UF 120027 (Figures 17, 
18)—which is a highly leached, heavily pitted, and 
slightly silicified dieil-ag an exception, Conus sauridens 
is present in upper Eocene Ocala Limestone of Jackson 
(UF 18874, UF locality JA002; UF 120026, UF locality 
a UF 119912. UF locality JA027 - and UF 119918, 

UF locality JA031), Alachua (UF 120027. UF locality 
ALOOL), and Suwannee (UF 122384 and UF 120032, 
both UF locality SU003) counties. F inally, one additional 
specimen (UF 119913) that may be C. sauridens is from 
the ‘aa Oligocene Bumpnose Uiraestone of Jackson 
County (UF locality JA025). Besides these records, the 
only pe known record of this taxon (as recognized 
here) in Florida is Dall’s (1916: 4489) account of C. 
tortilis in the “Ocala.” Conus sauridens co-occurs in the 
Eocene of Florida with the new species C. palmerae and 
C. alleni; characteristics that distinguish the new species 
from C. sauridens are discussed below. 

A complete review of the fossil record of Conus sau- 
ridens is beyond the scope of this paper, though the 
preliminary observations we have made here support 
earlier demonstrations (Palmer, 1937) that C. sauwridens 


J. R. Hendricks and R. W. Portell, 2008 


was a morphologically variable, geographically wide- 
spread, and temporally persistent species. While the old- 
est known (early Eocene) Conus fossils were small (<35 
mm in shell length), larger species (ca. 70 mm in shell 
length) have previously been reported from the middle 
Eocene (Kohn, 1990). During this study, we recognized 
a very large (shell length, 116.1 mm) specimen (UF 
120022: Figure 19) of C 
Eocene Moodys Branch Formation of Grant Parish, 
Louisiana (UF locality ZLO04). This specimen may be 
the largest Conus yet known from the Eocene of the U.S. 
Coastal Plain. The large geographic range of C. sauridens 


sauridens from the upper 


could be related to its developmental mode, which was 
likely planktotrophic based upon its multispiral larval 
shell (Figures 20-22), though testing this hypothesis 
within the context of Shuto’s (1974) model of the rela- 
tionship between developmental mode and larval shell 
morphology (also see Kohn and Perron, 1994) will re- 
quire additional study. While not necessarily useful as a 
cuide fossil, shells of C. sauwridens (as circumscribed 
here) have recently been utilized in several isotopic stud- 
ies (Kobashi et al., 2001; Kobashi and Grossman, 2003; 
Kobashi et al., 2004) of Paleogene climate and have 
proven to be geologically useful in this regard. 


Conus palm rae new species 

Figures 23-30, Table 2) 

Conus sp. B, Palmer in Richards and Palmer, 1953: 40, pl. 2, 
hig. 14 


Diagnosis: Teleoconch whorls stepped; early post- 
nuclear whorls smooth; sutural ramps typically smooth: 
last whorl smooth. 


Description: Shell small to moderately large-sized (up 
to about 65 mm in length). Last whorl conical: outline 
slightly concave. Shoulder sharply angulate, smooth. 
Spire of moderate height; outline concave to straight. 
Teleoconch whorls stepped; spire angle of early whorls 
typically obtuse relative to later whorls. Larval shell un- 
known. Early postnuclear whorls smooth. Subsutural 
flexure symmetrically curved. Teleoconch sutural ramps 
concave and typically smooth, though occasionally 2or3 
weak spiral threads are present. Aperture morphology 
unknown. Last whorl typically smooth, though fine spiral 
lines May cover the last whorl ot some small specimens. 


Type Specimens: Holotype UF 105855 (Figures 23, 
24). a specimen originally described as “Conus sp. B” by 
Palmer in Richards and Palmer (1953: 40, pl. 2, fig. 14). 
The holotype is preserved as a calcite-replaced shell. All 
yaratypes are moldic (consisting of just external or exter- 
ial and internal molds) and include: UF 15886, UF 
18599 (Figure 25). UF 18711, UF 18719, UF 18737 
Figure 26), UF 57018, UF 66735 (Figure 27), UF 68306 
Figure 28), UF 74473 (Figures 29, 30), UF 110360, UF 
111327. and UF 112981. See Table 2 for measurements 


of these specimens 


2 
7 


Figures 19-22. Specimens of Conus sauridens Conrad 
1833. 19. Largest known specimen (UF 120022) of C. sau- 
ridens, shell length 116.1 mm, Moodys Branch Formation, UF 
locality ZLO04 (Montgomery Landing 01), Grant Parish, Loui- 
siana; scale bar equals 1 cm. 20. Juvenile shell (UF 126927 


shell length 3.4 mm, Moodys Branch Formation, UF locality 
ZLOO04 (Montgomery Landing 01), Grant Parish, Louisiana 
scale bar equals 1 mm. 21-22. Shell (GSA.2007.005), shell 
length 60.5 mm, Moodys Branch Formation, Montgomery 
Landing, Grant Parish, Louisiana; Figure 21 shows the larval 
shell and early postnuclear whorls of this specimen, which is 


also shown in Figure 22 (both scale bars equal 1 cm 


Page 88 THE NAUTILUS, Vol. 122, No. 2 


Figures 23-30. Specimens of Conus palmerae new species. All scale bars equal 1 cm. 23-24. Holotype (UF 108858), preserved 


shell length 22.0 mm, preserved maximum diameter 11.5 mm, lower member of the Ocala Limestone (formerly Inglis Formation), 
UF locality LVO14 (Gulf Hammock 02), Levy County, Florida. 25. R.T.V. silicone rubber cast of paratype (UF 18599), preserved 
shell length 50.1 mm, preserved maximum diameter 23.4 mm, Ocala Limestone, UF locality ALO16 (S.M. Wall Quarry 01), Alachua 
County, Florida. 26. R.T.V. silicone rubber cast of paratype (UF 18737), preserved shell length 56.1 mm, preserved maximum 
diameter 24.1 mm, Ocala Limestone, UF locality LFOO1 (Dell Limerock Mine), Lafayette County, Florida. 27. R.T.V. silicone rubber 
cast of paratype, (UF 66738), maximum diameter 14.2 mm, Ocala Limestone, UF locality ALO28 (Newberry 03), Alachua County, 
Florida. 28. R.T.V. silicone rubber cast of paratype (UF 68306), maximum diameter 19.3 mm, Ocala Limestone, UF locality ALO04 
Dickerson Limerock Mines), Alachua County, Florida. 29-30. R.T.V. silicone rubber cast of paratype (UF 74473), preserved 
maximum diameter 19.0 mm, Ocala Limestone, UF locality ALO04 (Dickerson Limerock Mines), Alachua County, Florida; the 
topmost portion (indicated by arrow) of Figure 29 is magnified 3.5x in relation to Figure 30 to show details of the sutural ramps 


Type Locality and Occurrence: The holotype (UF 18737, UF 57018, and UF 110360, UF locality LFOO1, 
LOSS58; Figures 23, 24) is from the lower member of the Dell Limerock Mine, Ocala Limestone). Thus, all speci- 
Ocala Limestone (formerly the Inglis Formation) at UF mens of C. palmerae are from the upper Eocene, Jack- 
locality LVO14, Gulf Hammock 02, Levy County, sonian Ocala Limestone of Florida 
Florida. Richards and Palmer (1953: 5) described the 
locality R. O. Vernon’s L-93) as a “road metal pit 2.9 Etymology: This species 1s named in honor of 
miles south of the north limits of the town of Gulf Ham- Katherine V. W. Palmer (1895-1982). second director of 
mock just southwest of State Road 55 in the southwest the Paleontological Research Institution (Ithaca, NY), 
quarter of Section 34, Township 14 South, Range 16 who was the first to recognize this form as a new, unde- 
East.” The varatype specimens are from \lachua County scribed species and for her important contributions to 
UF 68306, UF 74473, UF 111327, and UF 112981, UF Cenozoic paleontology (see Caster, 1983) 
locality ALO04, Dickerson Limerock Mines, Ocala Lime- 
tone; UF 18599, UF locality ALO16, S.M. Wall Quarry Discussion: Palmer (1953) did not describe UF 
Ol, Ocala Limestone; UF 15886, UF locality ALO17, LOSS5S as a new taxon because she did not consider this 
Newb« Corporation Pit 01, Ocala Limestone; and UF single damaged shell adequate for this purpose. Newly 
6735, UF locality ALO28, Newberry 03, Ocala Lime collected specimens (all molds) are consistent with the 
| Lafayette County (UF 18711, UF 18719, UF gross morphology of Palmer's fossil, but offer new mor- 


J. RB. Hendricks and R. W. Portell, 2008 Page 89 


Figures 31-39. Specimens of Conus alleni new species (31-35), C. smithvillensis Harris, 1895 (36-38), and C. cormacki Har- 
bison, 1944 (39). Scale bar equals 1 cm. 31-32. Holotype (UF 119920) of C. alleni, shell length 36.2 mm, preserved maximum 


| 
( 


liameter 20.7 


mm, Moodys Branch Formation, UF locality ZLO04 (Montgomery Landing), Grant Parish, Louisiana. 33. Paratype 
UF 119977) of C. alleni, preserved shell length 35.1 mm, Ocala Limestone, UF locality LF002 (Mill Creek Quarry), Lafayette County 
Florida. 34. Paratype (UF 119919) of C alleni preserved shell length 33.2 mm, Ocala Limestone, UF locality SUO14 (Suwannee American 
Cement), Suwannee County, Florida. 35. Paratype (UF 119976) of C. alleni, preserved shell length 34.5 mm, UF locality LF002 (Mill 
Creek Quarry), Lafayette County, Florida. 36-38. Holotype TMM-BEG 35656) of C. smithvillensis, shell length 42.1 mm, pre 

served maximum diameter 14.8 mm, Weches Formation, Colorado River at Smithville, Sastrop County, Texas. 39. Holotype (ANSP 
16145) of C. cormacki (not a Conus; see text), preserved shell length 28.7 mm, Santee Cooper Canal, South Carolina 


phological details that now justify description of this ridens Conrad, 1833, and C. alleni new species. Conus 
species palmerae may be distinguished from both of these species 
Conus pale rae co-occurs in the Ocala Limestone by its stepped teleoconch whorls smooth early postnuclear 


with two other late Eocene Conus in Florida: C. sau- whorls, smooth sutural ramps and smooth last whorl 


Page 90 


THE NAUTILUS, Vol. 122, No. 2 


Conus alleni new species 
(Figures 31-35, Table 2) 


Diagnosis: Shoulder undulate; subsutural flexure di- 
agonal; last whorl with raised spiral cords on anterior 


half. 


Description: Shell medium-sized (up to about 36 mm 
in length). Last whorl conical; outline straight to slightly 
sigmoidal (convex near shoulder). Shell widest at shoul- 
der. Shoulder angulate and with large tubercles resulting 
from weak undulations. Spire of moderate height; outline 
straight to slightly concave. Larval shell unknown. Early 
postnuclear w horls tuberculate. Subsutural flexure di- 
agonal, depth about 1.5x width. Teleoconch sutural 
ramps concave with several raised spiral cords. Aperture 
opening about as wide at base as at shoulder. Last whorl 
with pronounced raised spiral cords on anterior half, 
sometimes extending weakly as threads to shoulder. 


Type Series: Holotype, UF 119920 (Figures 31, 32). 
The three paratypes consist of external molds and in- 
clude UF 119977 (Figure 33), UF 119919 (Figure 34), 
and UF 119976 (Figure 35). See Table 2 for measure- 
ments of these specimens. 


Type Locality and Occurrence: The holotype (UF 

119920, Figures 31, 32) was collected by J. E. Allen from 
the Jacksonian Minas Branch Formation at Montgom- 
ery Landing (UF locality ZL004), Grant Parish, Louisi- 


ana. The paratypes are all from the Ocala Limestone of 


Florida, including two specimens from Lafayette County 
(UF 119976, U F 119977, UF locality LF002, Mill Creek 
Quarry) and one specimen from Suwannee C ounty (UF 
119919, UF locality SU014, Suwannee American Ce- 
ment), 


Etymology: This species is named in honor of James 

E. Allen (1914-1997) of Alexandria, Louisiana, who was 
an enthusiastic collector and scholar of Gulf Coast 
Eocene mollusks. 


Discussion: 
Louisiana with C. sauridens and in the Ocala Limestone 
of Florida with C. palmerae new species and C. sau- 
ridens. Conus alleni can be readily differentiated from 
both species by its undulate shoulder and spiral cords on 
the anterior half of the last whorl (raised spiral threads on 
the last whorl may also be present on small shells of C. 
palmerae, but if so are much weaker). 

Conus alleni shares some resemblance with a moldic 
Oligocene fossil (USNM 166720) from Decatur County, 
Georgia that Dall ( rs described as C. vaughani. Dall’s 

(1916: pl. 56, fig. 1) figure of the cast shows a specimen 
(partially obsei re a be matrix) with an obtuse spire angle, 


undulate shoulder, and raised spiral threads on the su- 
tural ramps that are similar to the teleoconch morphol- 
ogy of C, alleni. The presence of rows of spiral beads on 
the last whorl, the fact that the shell is widest below the 
shoulder (rather than at the shoulder, as in C. alleni), and 
the fact that the anterior end of the shell appears com- 


Conus alleni co-occurs in the Eocene of 


pletely obscured by matrix prevents us, however, from 
considering these two forms e quivalent. 

The only known shell material of Conus alleni is the 
holotype (from Grant Parish, Louisiana); the other three 
specimens are from Florida and all consist of external 
molds. This taxon was apparently rare, especially outside 
of Florida. We recognized this new form from the moldic 
Floridian material belo we—by chance—discovered 
the similar shell from Louisiana in the FLMNH collec- 
tions. We chose to designate the shell as the holotype, 
rather than one of the paratype external molds, because 
of its greater number of characters available for obser- 
vation. 


OTHER RECORDS OF EOCENE CONUS FROM THE U.S. 
COASTAL PLAIN 


Unidentifiable internal molds of Conus are common in 
the Eocene Ocala Limestone of northern Florida. We 
examined 54 such lots (over 440 specimens) from Ala- 
chua, Citrus, Jackson, Lafayette, Marion, and Suwannee 
counties. These include: UF 15884, UF 15892, UF 
15905, UF 17831, UF 17832, UF 17879, UF 17947, UF 
17950, UF 17967, UF 18423, UF 18759, UF 18848, UF 
18864, UF 18896, UF 18955, UF 18962, UF 19140, UF 
19174, UF 19204, UF 19215, UF 20744-20746, UF 
46435, UF 68270, UF 107265, UF 119900-UF 119904, 
UF 119906-119911, UF 119914119917, UF 119921, 
UF 120027, UF 120032-120040, UF 120047, and UF 
126926. 

During the course of this work, we became aware of 
several other Eocene Coastal Plain Conus species that 
are likely distinct, but are not known to occur in Florida 
and will require additional investigation; these include: 
Conus smithvillensis Harris, 1895: C. smithvillensis var. 
ee sry, 1980 (also see C. smithvillensis var. Dockery in 

Campbell, 1995); and Conus haighti Gardner, 1945. 

Harris (1895) described C. sniiudlonss from the 
Colorado River at Smithville, Bastrop County, Texas 
(Weches Formation according to TMM records). The 
holotype (TMM-BEG 35656; Figures 36-35) shares 
some characteristics with C. sauridens (including a mul- 
tispiral protoconch, tuberculate early postnuclear whorls, 
spiral threads on the sutural ramp, growth lines showing 
a deep and symmetrically curved cubsuual flexure, aan 
incised spiral g erooves near the base of the last whorl), but 
has a very different overall shell shape: the spire is much 
higher (spire angle: 53°, Table 2; in mature individuals of 
C. sauridens, the spire angle is typically over 100°), the 
conical last whorl has straight sides, and the sutural 
ramps are flat. Dockery (1980) figured a shell (MGS 590) 
from the slightly younger Cook Mountain Formation of 
Mississippi this tt he described as a varie ty of C. smithvil- 
lensis. His figured shell appears to bear many of the 
discrete characteristics of C. smithvillensis described 
above, though has a lower spire (ca. 76°) and the last 
whorl is sigmoidal in profile. Conus smithvillensis is not 
at all similar in form to C. palmerae. While it bears some 
of the shell characteristics of C. alleni, it lacks C. alleni’s 


J. R. Hendricks and R. W. Portell, 2008 


Page 91 


distinctive raised spiral cords at the base of the last whorl. 
Campbell (1995) attributed an external mold (UNC 
15448) from the Santee Limestone near Cross, South 
Carolina to Dockery’s (1980) ae of C. smithvillensis. 
Campbell (1995: 146) stated that this form is the “most 
common Conus in the Santee Limestone” and that it “has 
a nodose shoulder and a taller, more tabulate spire than 
the widespread Conus (Lithoconus) sauridens.” Further 
study will be required to determine whether these vari- 
eties are indeed consistent with Harris’s taxon. 
Gardner (1945) described C. haighti from the Laredo 
Formation of Zapata County, Texas. She did not figure 
her a (USNM 495182), but her figured (pl. 26, 
fig. 7) paratype specimen (USNM 495183) has a convex 
spire profile and rounded shoulder that appear distinct 
from those characters in C. sawridens, C. palmerae, and 
C. alleni. Gardner’s other figured (pl. 26, fig. 2) specimen 
of C. haighti lacks this distinctive spire form. 


NOMINA DUBIA 


The following species of Conus described from the 
Eocene of the U. S. Coastal Plain should be regarded as 
nomina dubia: Conus gyratus Morton, 1834; Conus mu- 
tilatus Tuomey, 1852; Conus improvidus de Gregorio, 
1890; and Conus cormacki Harbison, 1944. Conus gyra- 
tus (holotype, ANSP 211) is an internal mold of a shell of 
uncertain provenance from South Carolina (also see 
Kohn, 1992); Campbell (1995) suggested that it could be 
an internal mold of the purported variety of C. smithvil- 
lensis presented by Dockery (1980). Conus multilatus 
was described (but not fieured) by Tuomey from casts 
found near Wilmington, North Carolina. De Gregorio 
(1890) described C. improvidus from Claibome, Ales 
bama, but his holotype is reportedly lost (Palmer and 
Brann, 1966) and Palmer (1937: 465) considered the 
taxon “of doubtful status as an American species.” Conus 
cormacki Harbison, 1944 was described from the Eocene 
Santee Formation of South Carolina, but the holotype 
(ANSP 16415, Figure 39, Table 2) is clearly not a Conus. 
Campbell (1995: 146) stated that C. cormacki “is actually 
a broken volutid.” 

Several other Eocene Conus species are also problem- 
atic. Conus pulcherrimus eee 1879 (type, AMNH- 
FI 10175) was recognized by Harris (1895) as a turrid 
(see Palmer and Brann, 1966). The type specimen of 
Conus (Conospirus) granopsis de Gregorio (1890) is lost 
(Palmer and Brann, 1966) and Dall (1896) and Palmer 

1937) suggested that the small shell figured by de Gre- 
gorio (1890) may be the juvenile of “another species. 
Given that the type specimen is lost and that the shell is 
likely a juvenile ania C. granopsis is a name that 
should probably be disregarded. 

Finally, along with Conus sp. B (described here as the 
new species C. palmerae; holotype, UF 108858), — 
in Richards and Palmer (1953) also noted a Conus sp. 
She said, “species A... is a broad (21 mm.), low- Ree 
(7 mm.) shell with sharp angulation of the s houlder of the 


whorls; the surface was apparently smooth. The speci- 
men is a fragment, 25 mm. high” (Palmer, 1953: 40). We 
located this specimen (UF 108683), which—like UF 
10S858—is also a calcite pseudomorph from UF locality 
LV014, Gulf Hammock 02, Levy County, Florida (lower 
member of the Ocala Limestone). UF 108683 is poorly 
preserved and is too fragmentary (more than half of the 
spire is eroded away and much of the last whorl is miss- 
ing) to either assign to a known species or to describe as 
new species. 


ACKNOWLEDGMENTS 


Sable Allen (Alexandria, Louisiana) graciously donated to 
the FLMNH the large, well-documented Eocene Gulf 
Coast mollusk collection of her late husband, James E, 
Allen. Larry Rogers (Limestone Products), Wayne Bea- 
ver (Denali Limerock Mine), Joe Horton (Suwannee 
American Cement), and Leon Brooks (Hi-Cal Quarry) 
kindly allowed access to collect in their Florida mines. 
Paul Callomon (Academy of Natural Sciences, Philadel- 
phia) and Ann Molineux (Texas Memorial Museum) are 
thanked for allowing access to type specimens. Ann Mo- 
lineux also pr ovided helpful discussions on Eocene stra- 
tigraphy in Texas. Bushra Hussaini and Sandy Ebersole 
furnished information pertaining to an American Mu- 
seum of Natural History specimen and a Geological 
Survey of Alabama specimen, respectively. Alan Kohn 
provided helpful discussions of and suggestions for ter- 
minology related to the “subsutural Fexre” of cone 
shells. Sean Roberts (Florida Museum of Natural His- 
tory) assisted us a some of the digital photography 
using a Sony DSC RIL camera (10.3 megapixel resolu- 
tion). We thank Alan Kohn and an anonymous reviewer 
for comments that improved the quality of this manu- 
script. JRH’s contributions to this work were supported 
by the Brayfield Award of the South West Florida Fossil 
Club and NSF EAR 0518976. RWP’s contribution to this 
work was supported by the McGinty Endowment of the 
Florida Museum of Natural History, by Barbara and 
Reed Toomey, and by the late David ‘Nicol whose $1000 
grant allowed RWP to permanently transfer the Florida 
Geological Survey Eocene mollusk collection from the 
Paleontological Research Institution (where it was under 
study by Katherine V.W. Palmer) to the Florida Museum 
of Natural History. This is Univer sity of Florida Contri- 
bution to Paleobiology 602. 


LITERATURE CITED 


Applin, P. L. and E. R. Applin. 1944. Regional subsurface stra- 
tigraphy and structure of Florida and southern Ge orgia 
American Association of Petroleum Geologists Bulletin 28: 
1673-1753. 

Berggren, W. A. and P. N. Pearson. 2005. A revised tropical to 
subtropical Paleogeneplanktonic foraminiferal zonation 
Journal of Foraminiferal Research 35: 279-298. 

Campbell, D. 1995. New molluscan faunas from the Eocene of 


Page 92 


THE NAUTILUS, Vol. 122, No. 2 


South Carolina. Tulane Studies in Geology and Paleontol- 
ogy 27: 119-151. 


Caster, K. E. 1983. Memorial to Katherine Van Winkle Palmer 


(1895-1982). Journal of Paleontology 57: 1141-1144. 

Chaney, D. S. 1989. Mold making with room temperature vul- 
canizing silicone rubber. In: Feldmann, R. M., R.E. Chap- 
man, and J. T. Hannibal (eds.). Paleotechniques. The Pa- 
leontological Society Special Publication No. 4, Depart- 
ment of Geological Sciences, University of Tennessee, 
Knoxville, PD 975-331. 

Conrad, T. A. 1833 
North America, illustrated by figures drawn on stone, from 
nature. Volume 1(3). Private ly published by J. Dobson. 
[Reprinted by Harris, 1893.] 


Conrad, T. A. 1835. Fossil shells of the Tertiary formations of 


North America. Eocene fossils of Claiborne, with obser- 
vations on this formation in the United States, and a geo- 
logical map of Alabama. Volume 1(3): 29-56. {Reprinted 
by Harris, 1893. ] 
Conrad, T. A. 1855. 
Jackson, Mississippi, with descriptions of thirty-four new 


species of shells and corals. Proceedings of the Acade omy of 


Natural Sciences of Philadelphia 7: 257-263. 

Conrad, T. A. 1865. Descriptions of new Eocene shells of the 
United States. American Journal of Conchology 1: 142- 
149. 

Cooke, C. W. 1915. The age of the Ocala Limestone. United 
States Geological Survey Professional Paper 95-1: LO7— 
LT. 

Correlation of Stratigraphic Units of North America (COSUNA) 
Project. 1988. Gulf Coast Region. The American Associa- 
tion of Petroleum Geologists, 

Dall, W. FH. 1896 (1895). Diagnoses of new Tertiary fossils from 
the southern United States, Proceedings of the United 
States National Museum 18; 21-46. 


Dall, W. H. 1916. A contribution to the invertebrate fauna of 


the Oligocene beds of the Flint River, Georgia. Proceed- 
ings of the United States National Museum 51(2162): 
487-524. 

Dall, W. H. and G. D. Harris. 1892. Correlation papers: 
Neocene. United States Geological Survey Bulletin 84, 
349 pp. 

De Gregorio, A. M. 1590. Monographie de la faune Koceénique 
de T Alabama et surtout de celle de Claiborne de l'Etage 
Parisien. Annales de 
1-316. 

Dockery, D. T. IIE. 1980. The invertebrate macropaleontology 
of the Clarke County, Mississippi, Area. Mississippi De- 
partment of Natural Resources Bureau of Geology Bulle- 
ae os 1-387. 

Duda, T. F. and A. J. Kohn. 2005. Species-level phylogeogra- 
i ae evolutionary history of the hyperdiverse marine 
gastropod genus Conus. Molecular Phylogenetics and 
Evolution 34: 257-272. 

Fenk, E. M. 1979. Sedimentology and stratigraphy of middle 
and upper Eocene carbonate rocks, Lake, Hernando, and 
Levy counties, Florida. Unpub. M.S. Thesis, University of 


Géologie et de Paléontologie 7: 


Florida, Gainesville, 133 pp: 

Fleming, J. 1522. The Philosophy of Zoology or a General View 
of the Structure, Functions, and C Taso te: ation of Animals. 
Archibald Constable, Edinburgh, 565 pp. 

Gardner, J. 1945. Mollusca of the Tertiary formations of north- 
eastern Mexico. Geological Society of America Memoir 
Ll: 1-332 


3. Fossil shells of the Tertiary formations of 


Observations on the Eocene deposit of 


Garvie, C. L. 1996. The mollusean macrofauna of the Reklaw 
Formation, Marquez Member (Eocene: Lower Clai- 
bornian), in Texas. Bulletins of American Paleontology 
111(352):; 1-177. 

Green, J. 1530. Monograph of the cones of North America, 
including three new species. Transactions of the Albany 
Institute 1: 121-125. 

Harbison, A. 1944. Mollusks from the Eocene Santee Lime- 
stone, South Carolina. Notulae Naturae of the Academy of 
Natural Sciences of Philadelphia 143: 1-12. 

Harris, G. D. 1893. Republication of Conrad’s fossil shells of 
the Tertiary formations of North America. Rufus H. 
Darby, Washington, 121 pp. [Republished by the Paleon- 
tological Research Institution, 1963. | 

Harris, G. D. 1595. New and otherwise interesting Tertiary 
Mollusca from Texas. Proceedings of the Academy of 
Natural Sciences of Philadelphia 47: 45-88. 

Harris, G. D. 1951. Preliminary notes on Ocala bivalves. Bul- 
letins of American Paleontology 33(138): 1-55. 

Harris, G. D. and kK. V. W. Palmer. 1947. The Mollusca of the 
Jackson Eocene of the sega tt Embayment (Sabine 
River to the Alabama River), Part I, Univalves and Index. 
Bulletins of American Paleentclo) 30(117); 209-563, 

Heilprin, A. 1879. On some new Eocene fossils from the Clai- 
borne marine formation of Alabama, Proceedings of the 
Academy of Natural Sciences of Philadelphia: 211-216, 

Hendricks, J. R. In press. The genus Conus (Mollusca: Neo- 
gastropoda) in the Plio-Pleistocene of the southeastern 
United States. Bulletins of American Pz deontology. 

Hoerle, S. E. 1976. The genus Conus (Mollusca: Gastropoda) 
from the Alum Blufl Group of northwestern Florida. Tu- 
lane Studies in Geology and Paleontology 12; 1-31. 

Kobashi, T., E. L. Grossman, T. FE. Yancey, and D. T. Dockery 
IIT. 2001, Reevaluation of conflicting Eocene tropical tem- 
perature estimates: molluskan oxygen isotope evidence for 
warm low ae Geology 29: 983-986. 

Kobashi, T. and E. Grossman. 2003. The oxygen isotopic 
re ie of season: as in Conus shells and its application to 
understanding late middle Eocene (38 Ma) climate. Pale- 
ontological Research 7:343-355. 

Kobashi, T., E. L. Grossman, D, T. Dockery IH, and L. C. 
Ivany. 2004. Water mass stability reconstructions from 
greenhouse (Eocene) to icehouse (Oligocene) for the 
northern Gulf Coast continental shelf (USA). Paleocean- 
ography 19: PA1022. 

Kolin, A. J. 1990. Tempo and mode of evolution in Conidae. 
Malacologia 32; 55-67, 

Kohn, A. J. 1992. A chronological taxonomy of Conus, 1758— 
1S40. Smithsonian Institution Press, W ‘ashington, 315 pp. 

Kohn, A. J. and T. R. Anderson. 2008. The Conus Biodiversity 
Website. URL: http://biology.burke.washington.edu/ 
conus/. 

Kohn, A. J. and F. E. Perron. 1994. Life History and Biogeog- 
raphy: Patterns in Conus. Clarendon Press, Oxford, 106 
Pp- 

Lea, H.C, 
from the Eocene, at Claiborne, Alabama. The American 
Journal of Science and Arts 40: 92-104, 

Lea, 1. 1833, Contributions to Geology. Carey, Lea and Blan- 
chard, Philadelphia, 227 pp 

Linnaeus, C. 1758. Systema Naturae per Regna Tria Naturae, 
10! agi 1. Holmiae, $24 pp 

MacNeil, F. S. and D. T. Dockery TH. 1984. Lower Oligocene 
G: cp Se eee da, and Ce phalopoda of the Vicks- 


1541. Descriptions of some new species fossil shells, 


J. R. Hendricks and R. W. Portell, 2008 


Page 93 


burg Group in Mississippi. Mississippi Department of 


Natural Resources Bureau of Geology Bulletin 124: 
1415. 

Mever, O. 1885. The genealogy and age of the species in the 
southern old- Tertiary. The American Journal of Science 
(Series 3) 29: 457-AGS. 


Meyer, O. 1886. Contributions to the Eocene paleontology of 


Alabama and Mississippi. Author's Edition, Part II. Geo- 
logical Survey of Alabama Bulletin 1: 61-85. 
Moore, E. J. 1962. Conrad’s Cenozoic fossil marine mollusk 


type specimens at the Academy of Natural Sciences of 


Philadelphia. Proceedings of the Academy of Natural Sci- 
ences ae Philade Iphia 114: 23-120. 

Morton, 1834. Synopsis of the organic remains of the 
C — group of the United States. Key and Biddle, 
Philadelphia, SS pp. 

Muiiiz-Solis, R. 1999. El género Conus L., 1758 (Gastropoda, 
Neogastropoda) del Plioceno de Estepona (Malaga, 
Espana). Iberus 17: 31-90. 

North American Commission on Stratigraphic Nomenclature. 
1983. North American Stratigraphic Code. American As- 
sociation of Petroleum Geologists Bulletin 67: 841-875. 


Oven, C. W. 1995. Biostratigraphy and paleobiogeography of 


Eocene comatulid crinoids from Florida. Tulane Studies 
in Geology and Paleontology 27: 153-164. 

Oyen, C. W. and R. W. Portell. 2001. Diversity patterns and 

biostratigraphy of Cenozoic echinoderms from Florida. 
Palaeogeography, Palaeoclimatology, Palaeoecology 166: 
193-218. 

Palmer, K. V. W. 1937. The Claibomian Scaphopoda, Gas- 
tropoda and Dibranchiate Cephalopoda of the southern 
United States. Bulletins of American Paleontology 7(32): 
1-730. 

Palmer, k. V. W. and D. C. Brann. 1966. Catalogue of the 
Paleocene and Eocene Mollusca of the southern and east- 
ern United States. Part I, Gastropoda. Bulletins of Ameri- 
can Paleontology 48(218): 471-1057. 

Portell, R. W. and E. H. Vokes. 1997. A new species of Pteryno- 
tus (Gastropoda: Muricidae) from the Eocene Ocala 
Limestone of Florida. Tulane Studies in Geology and Pa- 
leontology 30: 203-206. 


Puri, H. S. 1953. Zonation of the Ocala Group in the peninsular 
Florida. Journal of Sedimentary Petrology 23: 130 (ab- 
stract). 

Puri, H. S. 1957. Stratigraphy and zonation of the Ocala Group. 
Florida Geological Survey Bulletin 38, 248 pp. 

Richards, H. G. and K. V. W. Palmer. 1953. Eocene mollusks 
from Citrus and Levy counties, Florida. Florida Geological 
Survey Bulletin 35; 1-95. : 

Roéckel, D., W. Korn, and A. J. Kohn. 1995. Manual of the 
Living Conidae, Volume 1: Indo-Pacific Region. Verlag 
Christa Hemmen, Grillparzerstr, Germany, 517 pp. 

Scott, T. M. 1991. A geological overview of Florida. In: Scott, 
T. M., J. M. Lloyd, and G. Maddox (eds.) Florida’s ground 
water quality monitoring program — hydrogeological 
framework. Florida G eological Survey Special Publication 
32: 5-14. 

Shuto, T. 1974. Larval ecology of prosobranch gastropods and 
its bearing on biogec ography and paleontology. Lethaia 7: 
239-256. 

Smith, B. 1930. Some specific criteria in Conus. Proceedings of 
the Academy of Natural Sciences of Philadelphia 82: 279- 
25S. 

Toulmin, L. D. 1977. Stratigraphic distribution of Paleocene 
and Eocene fossils in the eastern Gulf Coast region. Geo- 
logical Survey of Alabama Monograph 13, 602 pp. 

Tuomey, M. 1852. Description of some fossil shells from the 
Tertiary of the southern states. Proceedings of the Acad- 
emy of Natural Sciences of Philadel Iphia 6: 192-194. 

Vernon, R. O. 1951. Geology of Citrus and Levy counties, 
Florida. Florida Geological Survey Bulletin 33, 256 pp. 

Wheeler, H. E. 1935. Timothy Abbott Conrad, with particular 
reference to his work in Alabama one hundred years ago. 
Bulletins of American Paleontology 23(77):I- 157. [Re- 
printed by the Paleontological Research Institution, 1977. ] 

Zachos, L. G. and A. Molineux. 2003. Eocene echinoids of 
Texas. Journal of Paleontology 77: 491-508. 

Zachos, L. G., C. L. Garvie, and A. Molineux. 2005. Definitive 
locations of Paleocene and Eocene marine fossil localities, 
Colorado River, Bastrop County, Texas. Texas Journal of 
Science 57: 317-328. 


THE NAUTILUS 122(2):94-98, 2008 


Page 94 


Synonymization of Neohyalimax Simroth, 1896, and Omalonyx 


dOrbigny, 


1837, with a redescription of Omalonyx brasiliensis 


(Simroth, 1896) (Gastropoda: Succineidae ) 


Janine O. Arruda José W. Thomé 


Laboratorio de Malacologia 

Museu de Ciéncias e Tecnologia 
Pontificia Universidade Catolica do Rio 
Grande do Sul 

Avenida Ipiranga 6681, 

90619—900 Porto Alegre, BRAZIL 
arrudajo@gmail.com 


josewthor 1 1e@pq.cnpq.| or 


ABSTRACT 


Neohyalimax was established as a monotypic genus for N. bra- 
siliensis. Characters of the reproductive system prompted au- 
thors to subsequently treat the taxon as a subgenus of Omal- 
onyx. Examination of the holotype and discussion of these char- 
acters allowed Neohyalimax to be synonymized with Omalonyx. 
Omalonyx brasiliensis remains a valid species based solely on 
the holotype. 


Additional Keywords: Molhusca, Pulmonata, Brazil, Rio Grande 
do Sul 


INTRODUCTION 


The genus Omalonyx @Orbigny, 1837, comprises succi- 
ne ge. slugs of West Indies, Central and South American 
and Juan Fernandez, with a reduced, flat and fingernail- 
like shell. Omalonyx has commonly been classified with 
two subgenera (e.g., Patterson, 1971: Tillier, 1981), 
namely Omalonyx sensu stricto and Neohyalimax Sim- 
roth, 1896. Zilch (1959-1960) and Thiele (1992) indi- 
cated that the separation between the subgenera Oma- 
lonyx and Neohyalimax is based on the mantle covering 
either only the margin or the entire shell and the spire 
being small or scarce ‘ly indicated. The monotypic subge- 
nus Neohyalimax, type-species Neohyalimax brasiliensis 
Simroth, 1896, was described from a unique specimen 
collected from Rio Grande do Brazil. In revising the 
family Succineidae Beck, 1837, Patterson (1971) dis- 
cussed the similarities of Didone felina (Guppy, 1872) 
O. matheroni (Potiez and Michaud, 1835), according 
to Tillier, 1981] and N. brasiliensis, and brought atten- 
tion in partic silar to details of the origins ul dese ription of 
the latter species (name ‘ly the number of seminal vesicles 
and c¢ phalic tentacles) that suggested recognition of 
Veohyalimax at the subgeneric level. It should be noted, 


however, that Patterson (1971) did not examine the type 
of N. brasiliensis, nor made a direct comparison with O. 
unguis (dOrbigny, 1837), the type species of Omalonyx. 
In a revision of the South American and Juan Fernandez 
succineid slugs, Tillier (1981) similarly did not examine 
the type of N. brasiliensis, but accepted Patterson (1971) 
assessment of taxonomic status based on the original de- 
scription prov ided by Simroth (1896). In support of rec- 
ognition at the subgeneric rank, Tillier (1981: 144) noted 
‘an altogether diflere nt way of life (terrestrial), a plesio- 
morphic radula (marginal teeth still longitudinally elon- 
gated) and slug-like ae wacters more apomorphic than in 
Omalonyx sensu stricto’. 

Simroth (1896) mentioned that the type specimen of 

N. brasiliensis was deposited with the “Berlin Museum”. 
ss d there is a specimen in Museum fiir Naturkunde 
der Humboldt Universitit zu Berlin (ZMB 45.913) la- 
beled Neohyalimax brasiliensis, but it bears no indication 
of type status. That the specimen corresponds well with 
the de scription given by Simroth ( including most details 
of the internal anatomy) and that no other specimen of 
this species has been located, the ZMB specimen is here 
considered to be the holotype. 

With the recent rediscove sry of Simroth’s type speci- 
men, we take the opportunity to redescribe N. brasilien- 
sis and reassess its taxonomic status. 


MATERIALS AND METHODS 


The following lots were examined: GUADELOUPE, 
Pointe-a-Pitre, Omalonyx matheroni, 4 spec., 
27.1X.1972, Pointier leg. (MNHN); O. matheroni, 11 
spec., S. Tillier det. 1979. (MNHN); FRENCH GUI- 
ANA, along the road Cayenne-Kourou, 17 km SE 
Kourou bridge, O. matheroni, 5 spec., Tillier leg. 


(MNEIN); PARAGUAY, O. ungitis (lectotype designated 


J. O. Arruda and J. W. Thomé, 2008 


‘ 2 O5 
Page 95 


by Tillier, L9S1), 1 shell. (MNHN); Asuncion, Asuncién, 
O. unguis, 1 spec., canje F.H. Schade. (MACN 199658, it 
is labeled as Hom. unguis @Orb.); ARGENTINA, For- 
mosa, Villafane (26°14.17' S 59°07.08' W, Arroyo Bel- 
laco, piece La Marcela), unguis, 2 spec., 
30.X.2004, Cristian Ituarte leg. (MLP 11878); BRAZIL, 
Rio G eae do Sul, Neohiyainer brasiliensis Simroth, 1 
spec. (ZMB 45.913) Cachoeirinha, O. convexus, 34 
spec., 29.1V.2006, A. Palatin and J. O. Arruda leg., J. O. 
Arruda det. (MCP. 8839): Pelotas (Estrada Pelnies hie 
Grande, 31°45'53.1" S 52°22'48.2” W), O. convexus, 48 
spec., 18.11.2006, A. Paladini and J. O. Neruda leg., J. O. 
Arruda det. (MCP 8836). 

Based on examination of the holotype of Neohyalimax 
brasiliensis, we redescribed the reproductive system 
(Figure 1). The specimen had been previously dissected, 
with a sagittal cut from its head to the mantle edge, 
which then continues towards the left side of animal’s 
tegument. The reproductive system was separated in two 
parts: one including the vagina (previously dissected and 
with the papillae on its internal surface exposed), the 


Ss —- f x & ae 
fc — io 3 —— pr 
a % 4M 
. { . YS ie 
2 Lay * 
a —— Ne 
7 — Yee 
vi 
S) e- 


1mm 


Figure 1. 
Omalonyx brasiliensis (Simroth, 1896) (ZMB 45.913). Abbre- 
viations: ag, albumen gland; b, epiphallial bulb; dd, deferent 
duct: ep, epiphallus fe, fertilization complex: fo, free oviduct; 
gg, gametolitic gland: 0, ovariotestis; ov, ovulispermduct; p- 
evertophallus: po, pallial oviduct; pr, prostate; ps, evertophal- 
lus sheath: rm, retractor muscle; sv, seminal vesicles; v, vagina. 


Reconstruction of the reproductive system of 


gametolitic gland duct and its gland and the free oviduct: 
the other including the pallial oviduct, prostate, albumen 
gland, fertilization “complex, ovulispermduct, ovariotestis, 
deferent duct, epiphallus and evertophallus (= penis). 

Institutional abbreviations: MACN, Museo Argentino 
de Ciencias Naturales “Bernardino Rivadavia” e Instituto 
Nacional de Investigacion de las Ciencias Naturales 
(Buenos Aires, Argentina), MCP, Museu de Ciéncias e 
Tecnologia da Pontificia Universidade Cat6lica do Rio 
Grande do Sul (Porto Alegre, Brazil); MLP, Museo de La 
Plata (La Plata, Argentina), MNHN, Muséum national 
dHistoire naturelle (Paris, France) and ZMB, Museum 
fiir Naturkunde der Humboldt Universitit zu Berlin 
(Germany). 


SYSTEMATICS 


Order Heterurethra Pilsbry, 1900 
Superfamily Succineoidea Beck, 1837 
Family Succineidae Beck, 1537 


Genus Omalonyx d Orbigny, 1537 


Type Species: Succinea (Omalonyx) unguis @Orbigny, 
1837: 229 (year according Sherborn and Woodward, 
1901), by original designation. 


Diagnosis: Shell very flat. fingernail-shaped, obtuse 
and small spire. body w vhorl and aperture very large and 
elongate-oval, animal body twice the shell length: ever- 
tophallus exceptionally long. 


Omalonyx brasiliensis (Simroth, 1596) 


Neohyalimax brasiliensis Simroth, 1896: 39-45; Morretes, 
1949: 130; Simone, 2006: 237. 

Omalonyx (Neohyalimax) brasiliensis —Zilch, 1959-1960: 202; 
Patterson, 1971: 187; Tillier: 1981. 

Homalonyx (Neohyalymax) brasiliensis.—Thiele, 1992. 

Omalonyx brasiliensis —Salgado and Coelho, 2003: 153. 


Diagnosis: A species of Omalonyx with a thick-walled 
vagina with robust diamond shaped papillae on its inter- 
nal surface. 


Holotype: ZMB 45.913 


Type Locality: Brazil, Rio Grande do Sul. 


Redescription: The shell is thin, oval and highly flat- 
tened (it is fragmented, which precludes its measure- 
ment). 

The jaw and radula are missing (see Discussion be- 
low). Within the reproductive system, the epiphallus is 
approximately one fourth of the le ngth of the evertophal- 
lus. The former possesses a thick wall, a smooth outer 
face and a bulb, and its internal surface has transverse 
folds. The opening of the epiphallus into the evertophal- 
lus is a simple opening, without any papilla. 

The evertophallus has a thick, wide wall, and the in- 
ternal surface is covered with aligned diamond shaped 
papillae. At its proximal extremity, the papillae are 


Page 96 


THE NAUTILUS, Vol. 122, No. 2 


smaller and discoid. At its distal extremity there are small 
longitudinal folds. The evertophallus sheath, thick and 
aiuseular: covers practically the entire distal half of the 
evertophallus. It gradually becomes a thin and transpar- 
ent membrane, w ich covers the other half of the ever- 
tophallus and the epiphallus. The evertophallus retractor 
muscle inserts at the junction of the deferent duct with 
the epiphallus. 

The prostate is long and the section of the deferent 
duct that is embedded in the prostate has small orifices 
throughout its length within the gland. 

The vagina is approximately half of the length of the 
ev ertophallus. It has a thick wall and aligned, diamond 
shaped papillae on its internal surface. These papillae are 
similar to those on the evertophallus, but more robust. 
The gametolitic gland is spherical, with a diameter ap- 
proximateh ly half the length of its duct. The gametolitic 
gland duct is slim and about two thirds the length of the 

vagina. 

The free oviduct is long, thick, and convoluted with 
rippled longitudinal folds on its internal surface. The 


pallial ov iduct, approximately half the length of the lower 


oviduct, is very inflated and convoluted. 

The albumen gland is small, bean-like and about one 
third the length of the prostate, The ovulispermduct 1S 
slightly convoluted and more inflated in its distal portion. 
The ovariotestis is spherical and large. 

The pallial cavity and the nervous and digestive sys- 
tems are damaged. 


DISCUSSION 


Simroth (1896: 40, pL. V, Fig. 3) characterized the shell as 
‘a small flat plate of 9 mm in length and 6 mm in breadth. 
Beneath the light-yellowish periostracum the lime was 
deposited in concentric rings, with an excentric nucleus 
to the left of the posterior end. The nearer the nucleus, 
the thicker is the shelly matter. The periphery is solid, 
surrounded, near the poste rior right-hand margin, by « 

brown conchiolin line. The nucleus is somawliat ade 
ened; the shell throughout is perfectly flat, and therefore 
does not enclose any of the soft parts’. The shell, accord- 
ing to Simroth (op. cit. ), would pl we Ae specie s close to 
the genus Hyalimax H. & A. Adams, 1855. However, that 
author re cognize ‘d that the shell ‘In ee it is slightly 
. Simroth (1896; 41) described the jaw 
as ‘brown and lit] | 1as a sharp edge, with a slight median 
process (oxygnath) and a lighter -coloured palatal plate 


arched (Fischer) 


(elasmognath), the two posterior corners of which are 
somewhat prominent. This plate is finely sculptured, the 
thread-like markings converging towards the median 
projection’ 
Simroth (1896: 4 
lows: ‘the median tooth is tricuspid, the middle cusp 
being very large. In the lateral teeth the median cusp is 


) described the radular teeth as fol- 


still larger, and directed oblique ‘ly towards the middle 
line of the radula; the inner cusp is reduced in size, whilst 
the outer one is duplicated. In the marginal teeth the 


inner cusp increases in size, whilst the exterior outer 
cusp subdivides, giving rise to three outer cusps’. 

In the description of O. felina, Guppy (1872) indicated 
that the shell was usually covered by the mantle, but 
when retracted, it occasionally exposed the central region 
of the shell. Gibbons (1879), however, repeatedly ob- 
served living specimens of Omalonyx and never noticed 
more than a limited portion of the shell border covered 
by the mantle. In observations of live specimens of Oma- 
lonyx from the states of Rio Grande do Sul and Minas 
Gerais, Brazil, we found that the mantle covered the 
shell in different degrees, and although the shell was well 
covered in some, in none presented its shell completely 
covered by the mantle. Based on Guppy (1872) and on 
our observation of live specimens of Omalonyx, the char- 
acter “mantle covering partially or totally the shell”, in- 
dicated by Zilch ( 1959-1960) and Thiele (1992) for the 
separation of the subgenre Omalonyx and Neohyalimax. 
are unjustifiable. The other character, “small spire or 
extremely small spire” is not a systematic character. 

Patterson (1971), using Simroth’s de scription of Neo- 
hyalimax brasiliensis as reference, indicated that it has 
similarities with Omalonyx felina concerning the general 
body's shape, the mantle pigmentation and the anatomy 
of the reproductive system. Because the reproductive 
system of Neohyalimax brasiliensis is similar to that in 
Omalonyx s.s and different from that in Hyalimax, which 
resembles Succinea Draparnaud, 1801, Patterson consid- 
ered Neohyalimax as a subgenus of Omalonyx. However, 
Patterson also noted that “Neohyalimax brasiliensis has 
one seminal vesicle, whereas there are two in O. felina, 
and that N. brasiliensis has only one pair of tentacles, 
while O. felina has two. We verified the presence of two 
vesicles and one fecundation pouch in N. brasiliensis, 
which contradicts its original description by Simroth 
(1896) and concurs with the observations of Patterson 

(1971) for rH sensu stricto. 

Simroth (1896) observed only one pair of tentacles in 
N. brasiliensis, the ommatophores. Patterson (1971), 
however, stated that Omalonyx sensu stricto has two 
pairs of tentacles. Through an external ventral view exam 
of the anterior region of N. brasiliensis, we could observe 
that the lower tentacles are present, very close to the 
animal's mouth opening, 

Simroth (1896) did not mention the presence of an 
epiphallus in N. brasiliensis. However, we verified the 
presence of this structure, which is slightly narrower than 
the evertophallus and is also present in the other species 
of Omalonyx s.s. We also observed small longitudinal 
folds at the distal extremity of the eve rtophallus, which 
are present in all Omalonyx species with known anatomy. 

Barker (2001) noted the placement of the genital ap- 
erture in Neohyalimax immediately adjacent to the right 
tentacle. We confirmed that this does not differ from its 
position in other species of Omalonyx sensu stricto. 

In the state of Rio Grande do Sul, type locality of O. 
brasiliensis, no specimens with the characteristics of this 
species were found, despite extensive investigation and 
collecting. In the municipalities of Cachoeirinha, Pelotas 


J. O. Arruda and J. W. Thomé, 2008 


Page 97 


and other localities in Rio Grande do Sul, only Omalonyx 
convexus (Heynemann, LS6S) was found, suggesting its 
overlapping geographical area with O. brasiliensis. The 
reproductive systems of these species are distinct. In O. 
convexus, the ovariotestis is hemispherical or discoidal: 
the lower oviduct is generally half the length of the upper 
oviduct; half of the proximal region of the evertophallus 
is generally a little wider than the rest, and its internal 
surface has longitudinal folds in the wider proximal re- 
gion and cordiform papillae in the distal region; the in- 
ternal surface of the vagina has smooth longitudinal 
folds. In O. brasiliensis, the ovariotestis is spherical and 
large; the lower oviduct is approximately twice the length 
of the upper oviduct; the internal surface of the ever- 
tophallus has diamond shaped papillae and the vagina 
has similar but more robust papillae. 

The reproductive system of O. brasiliensis resembles 
that in the studied specimens of O. matheroni (Potiez 
and Michaud, 1835), which agrees with the meticulous 
description of Tillier (1981), with the exception of the 
internal surface of the vagina. According to our observa- 
tions and to Tillier, the vagina in O. matheroni has ir- 
regular longitudinal folds, inconsistent in number or 


shape, whereas O. brasiliensis has papillae instead of 


folds, similar to the ones in its evertophallus. Despite the 
reproductive system similarities of these species, they do 
not have overlapping ranges. Tillier (1951) considered 
the probable range of O. matheroni to be the entire 
Amazon basin, the eastern coastal regions of South 
America from Caracas (Venezuela) to Rio de Janeiro 
(Brazil) and part of the Lesser Antilles. Neohyalimax bra- 
siliensis has been reported only once, from Rio Grande 
do Sul (Brazil). 

When compared to the O. unguis (@Orbigny, 1837) 
type species, O. brasiliensis presents specifics different 
characteristics such as a smooth external surface of the 
epiphallus, the internal surface of the evertophallus and 
vagina has diamond shaped papillae, more robust in the 
latter, the length of the gametolitic gland corresponds to 
approximately a half the length of its duct and the lower 
oviduct, which is little Goavolite: is generally two times 
the length of the upper oviduct. In O. unguis, according 
) Arruda et al. (2006), the external surface of the 
epiphallus has a looped fold, the internal surface of the 
evertophallus has discoidal papillae, the internal surface 


of the vagina has inflated elliptic papillae, the length of 


the gametolitic gland corresponds to approximately a 
quarter the length of its duct and the lower oviduct, 
which is convolute, is approximately three times the 
length of the upper oviduct. 


CONCLUSION 


Considering the foregoing discussion regarding the num- 
ber of seminal vesicles, the number of tentacles and the 
position of the genital aperture, as well as the observa- 
tions of the reproductive system, we here consider Neo- 


hyalimax Simroth, 1596 as a junior synonym of Omalo- 

nyx @Orbigny, 1837. Omalonyx brasiliensis (Simroth, 
1896) remains a valid species, based on the only known 

specimen, which we consider to be the holotype. 


ACKNOWLEDGMENTS 


We cae the reviewers, Dr. Gary Barker and Dr. Robert 
Cowie, for their great contribution with very good criti- 
cisms and improving the English; Alejandro. Tablado 
(MACN), Cristidn Ituarte (MLP), Lticia Maria Z. Rich- 
initti (MCP), Virginie Héros (MNHN) and Matthias 
Glaubrecht (ZMB) for loaning material; Guacira Gil for 
comments and suggestions; Luiz Augusto F. Aratijo for 
the translation from Portuguese to English; to Conselho 
Nacional de Desenvolvimento Cientifico e Tecnolégico 
for financial support (CNPq 131716/2005-0 for JOA, 
500032/2003-3 for JWT). 


LITERATURE CITED 


Arruda, J. O., S. R. Gomes, R. Ramirez, and J. W. Thomé 2006. 
Morfoanatomia de duas espécies do género Omalonyx 
(Mollusca, Pulmonata, Succineidae) com novo registro 
para Minas Gerais, Brasil. Biociéncias 14 (1): 61-70. 

Barker, G. M. 2001. Gastropods on land: phylogeny, diversity 
and adaptive morphology. In: Barker, G. M. (ed.) The 
biology of terrestrial molluscs. CABI Publishing, New 
York, PP 1-146. : 

Beck, H. 1837. Index Molluscorum praesentis aevi Musei Prin- 
cipis Augustissimi Christiani Frederici. Copenhagen, 124 
pp. 

Gibbons, J. S. 1879. Comparison of Omalonyx unguis d’Orb., 
with O. felina Guppy. Journal of Conchology 2: 98-101. 

Guppy, R. J. L. 1872. Third series of additions to the catalogue 
of the land and freshwater Mollusca of Trinidad, with a 
revised list of all species. Proceedings of the Scientific 
Association of Trinidad 2: 17-25. 

Morretes, F. L. 1949. Ensaio de Catalogo dos Moluscos do 
Brasil. Arquivos do Museu Paranaense 7 (1): 1-216. 
dOrbigny, A.D. 1837 [1835-1847]. Voyage dans ! Amerique 
Méridionale: Mollusques. Paris: Pitois-Levraut 5(3): 

1-758. 

Patterson, C. M. 1971. Taxonomic studies of the land snail 
family Succineidae. Malacological Review 4: 131-202. 
Pilsbry, H. A. 1900. On the zoological position of Partula and 
Achatinella. Proceedings of the Academy of Natural Sci- 

ences of Philadelphia 3: 561-567, pl. 17. 
Salgado, N. C. and A. C. 8. Coelho, 2003. Moluscos terrestres 
do Brasil (Gastropodes operculados ou nao, exclusive 
Veronicellidae, Milacidae e Limacidae). Revista de Bi- 
ologia Tropical (International Journ il of Tropical Biology 
and Conservation) 51 ( (suppl. 3: Malacologia Latinoameri- 
cana): 149-189. 

Sherborn, C. D. and Woodward, B. B. 1901. Notes on the dates 
of publication of the Natural History portions of some 
French Voyages—Part I. ‘Amérique méridionale’; ‘Indes 
orientales’; ‘Pole Sud’; (‘Astrolabe’ and “Zélée’); ‘La Bon- 
ite’; “La Coquille’; and ‘L’Uranie et Physicienne’. Annals 
and Magazine of Natural History (series 7) 7 (40): 385— 
392. 


Page 95 


THE NAUTILUS, Vol. 122, No. 2 


Simone, L. R. 2006. Land and freshwater molluses of Brazil. 
Fapesp, Sao Paulo, 390 pp. 

Simroth, H. 1896. On Neohyalimax brasiliensis, n. gen., n. sp. 
(allied to Hyalimax), from Brazil. Proceedings of the Mal- 
acological Society of London 2: 39-45, 

Thiele, J. 1992. Handbook of Systematic Malacology. Part 2 
(Gastropoda: Opisthobranchia and Pulmonata). In: Bieler, 
R and P. Mikkelsen (translators), SmithsonianInstitution 


Libraries. The National Science Foundation, Washington 
627-1189 pp. 
Tillier, S. 1981. South American and Juan Femandez succineid 
slugs (Pulmonata). Journal of Molluscan Studies 47; 125— 
146. 

Zilch, A. 1959-60. Euthyneura. In: Wenz, W, (ed.). Handbuch 
der Paliiozoologie, Gebriider Borntraeger, Berlin, vol. 2, 


834 pp. 


THE NAUTILUS 122(2):99-106, 2008 


Page 99 


Rehabilitation of Ergalatax martensi (Schepman, 1892) 
(Gastropoda: Muricidae), senior synonym of Ergalatax obscura 
Houart, 1996, and description of Ergalatax junionae, new name 


for Morula martensi Dall, 1923 


Roland Houart 

Institut royal des Sciences naturelles de Belgique 
Rue Vautier, 29 

B-1000 Bruxelles, BELGIUM 

roland. houart@skynet.be 


ABSTRACT 


Ergalatax junionae nomen novum is proposed as replacement 
name for Morula martensi Dall, 1923, from the Gulf of Oman 
and the Persian Gulf. The latter name is a junior secondary 
homonym of Pentadactylus martensi Schepman, 1892, from the 
Red Sea, of which E. obscura Houart, 1996, is here considered 
as a junior synonym. Ergalatax junionae nomen novum is com- 
pared with Ergalatax martensi (Schepman) and E. margariti- 
cola (Broderip, 1833). The geographical distribution of E 
junionae and E. martensi is updated and the introduced Medi- 
terranean species is identified as E. jumionae. 


Additional Keywords: Mollusca, Gastropoda, Muricidae, Gulf 


of Oman. Persian Gulf, Red Sea, Mediterranean Sea, hom- 
onymy, synonymy, new name. 


INTRODUCTION 


Some vears ago (Houart, 1996: 13), I described Ergalatax 
obscura, a muricid from the Red Sea. Shortly after that 
paper was published, R. G. Moolenbeek (in litt.) told me 
about a name introduced by Schepman (1892), Penta- 
dactylus martensi, which he also described from the Red 
Sea. Having then examined a syntype of that species 
from ZMA (Figure 14), I concluded that P. martensi 
Schepman, 1592, was a junior synonym of Ergalatax 
margariticola (Broderip, 1833). 

However. after recently requesting additional infor- 
mation about the other syntypes of P. martensi in 
NMNL, I received digital images that illustrate the re- 
maining three specimens (Figures 1S—23), among them 
the 17 mm long specimen that probably served for the 
description (Figures 15-19). Those syntypes are un- 
doubtedly conspecific with E. obscura, while the ZMA 
syntype represents E. margariticola. 

For many years (Houart, 2001), I had identified an 
introduced eastern Mediterranean species as Ergalatax 
obscura Houart, 1996 [= E. martensi (Schepman, 1$92)]. 


However, this concerns yet a third species, which occurs 
primarily in the Gulf of Oman and in the Persian Gulf. 
The species was first wrongly illustrated as Purpura 
(Ricinula) siderea Reeve (a buccinid) by von Martens 
(1874: 95, pl. 5, fig. 49). Later, Dall (1923: 304) ) provided 
anew name for that species, naming it Morula martensi. 
Unfortunately, both species, Pe ntadactylus marte nsi 
Schepman, 1892. and Morula martensi Dall, 1923. are 
now included in Ergalatax, and D 
ipso facto a junior secondary homonym and needs to be 
replaced. 

I personally mixed both species because of their close 
enters and the oo of some related forms 
(compare Figures 30 and 35). First (in litt.), I provision- 
ally identified the Meditenanean specime nas E. mar- 
tensi (Dall, 1923), but afterwards | erroneously consid- 
ered it as conspecific with E. obscura [= E. martensi 
(Schepman, 1892)] from the Red Sea and the Gull of 
Aden (Houart, 1996: 12). 

Abbreviations used herein are: AMS: Australian Mu- 
seum. Sydney, Australias MNHN: Muséum national 
(Histoire naturelle, Paris, France; NMNL: National 
Museum of Natural History Naturalis, Leiden, The 
Netherlands: RH: collection of the author; ZMA: Zoolo- 
gisch Museum, University of Amsterdam, The Nether- 
lands: ZMB: Museum fiir Naturkunde der Humboldt 
Universitit zu Berlin, zoologisches Museum, Germany. 
Terminology for shell spiral ae (after Merle, 1999 and 
2001) given in Figure 1. 


all’s name oem 


SYSTEMATICS 


Subfamily Muricoidea Rafinesque, 1$15 

Family Muricidae Rafinesque, 1S15 
Subfamily Ergalataxinae Kuroda, Habe and Oyama, 1971 
Genus Ergalatax Iredale, 1931 
Ergalatax recurrens Iredale, 1931, Aus- 
1846)] by 


Type Species: 
tralia (Figure 17) [?= E. contracta (Reeve, 
original designation. 


Page 100 


THE NAUTILUS, Vol. 122, No, 2 


TELEOCONCH WHORLS 


SP | Subsutural cord 

‘TP Infrasutural primary cord (primary cord on shoulder) 

adis : adapical infrasutural secondary cord (shoulder) 

abis : abapical infrasutural secondary cord (shoulder) 

P Primary cord 

's secondary cord - 
t tertiary cord 

Pl ] Shoulder cord 

P2-P6 : Primary cords of the convex part of the teleoconch whorl 

sl-s5 : : secondary cords of the convex part of the teleoconch whorl 


example: s] = secondary cord between P| and P2; s2 = secondary cord between P2 and P3, 


etc 

SIPHONAL CANAL 

ADP: adapertural primary cord on the siphonal canal 7 
“MP median primary cord on the siphonal canal 

APERTURE 

ID | Infrasutural denticle 

D1 to DS Abapical denticles 


Figure 1. Terminology used to describe the spiral cords (af- 
ter Merle, 1999 and 2001) (sporadic sculpture is shown in pa- 
rentheses). 


Ergalatax martenst (Schepman, 1892) 
(Figures 3, 5, 6, 11, 13, 18-30, 46) 


Pentadactylus (Morula) martensi Schepman, 1892: 104. 
Cronia martensi Dall—Sharabati, 1954: pl. 19, fig. 9, 9a, 9b; 


Singer and Mienis, 1991b: 58, fig. 19; Coulombel, 1994: 
73, text fig. (not Morula martensi Dall, 1923); Verbinnen 
and Dirkx, 2000: 69, fig. 9 (not Morula martensi Dall, 
1923). 

Drupella rugosa.—Singer and Mienis, 1991a: 18, fig. 6 (not 
Murex rugosus Born, 1778). 

Ergalatax obscura Houart, 1996: 13, figs 1, 3-6; Houart, 2001: 
108 (in part), figs 17, 31, 106 and 449 only; Heiman and 
Mienis, 2003: 22-23 (text fig.). 

Not Morula martensi (Schepman, 1892) —Tan, 1995: 160, figs 
52, 192 g, h (= Ergalatax junionae nom. nov.). 

Not Ergalatax obscura.—Houart, 2001 (in part): 108 (in part), 
figs 31 and 450-451 only; Delongueville and Scaillet, 2007: 
57, fig. 31 (= Ergalatax junionae nom. nov.) 


Description: — Shell medium sized for the genus, up to 
25.5 mm in length at maturity. LengthAvidth ratio 1.39- 
1.92. Heavy, stout. Spire high with 3+ protoconch whorls 
(tip partially broken), and up to 7 broad, strongly shoul- 
dered teleoconch whorls. Suture adpressed. Protoconch 
conical, acute, whorls smooth. Axial sculpture of teleo- 
conch whorls consisting of high, rounded, nodose ribs: 10 
or 11 from first to penultimate whorl, 6 to 8 on last whorl, 
rarely 5 or 9. Spiral sculpture of high, strong, primary, 
secondary and tertiary cords. Last teleoconch whorl with 
adis, IP, abis, P1, immediately followed by small P2, s2, 
P3, (t), s3, P4,(t), s4, P5, (s5), P6, ADP, MP. PL and P2 
small, P3, P4 and P5 similar in strength, P6 very small or 
obsolete. 

Sculpture forming high, nodose knobs at intersection 
of spiral cords and axial ribs. Aperture relatively small, 
ovate. Columellar lip with 2 or 3 weak knobs abapically, 
rim adherent. Anal notch broad, moderately deep. Outer 
lip weakly crenulate, with 7 strong elongate denticles 
within [I[D, D2 (D1-D2 fused), D3 split, D4, D5 split]. 
Siphonal canal short, broad, broadly open. Milky white, 


Figures 2-5. Ergalatax species. Spiral sculpture and apertural denticles. 2, 4. Ergalatax junionae nomen novum (lectotype ZMB). 
3,5. £. martensi (Schepman, 1892) (Red Sea, RH). 


R. Houart, 2008 Page 101 


13 


Figures 6-13. Ergalatax species. 6. Radula of Ergalatax martensi (Schepman, 1893), Gulf of Aden. Scale bar = 150 jum. 7. Radula 
of E. junionae nomen novum, Gulf of Iskenderun, Turkey. Scale bar = 120 jm. 8. Radula of E. margariticola (Broderip, 1833), Palau. 
Scale bar = 120 pm. 9-10. Radula of E. contracta (Reeve, 1846). 9. Aden, juvenile specimen. Scale bar = 60 pm. 10. Korea, large 
adult specimen (lateral denticles worn). Scale bar = 150 pm. 11. Operculum of E. martensi Scale bar = 1 mm. 12. Operculum of 
E. junionae. Scale bar = 1 mm. 13. Protoconch of E. martensi. Scale bar = 0.5 mm. (Figures 6-11, photos A. Warén). 


Page 102 


THE NAUTILUS, Vol. 122, No. 2 


i 


16 


Figures 14-17. Ergalatax species. 14. Ergalatax margariticola (Broderip, 1833). Lectotype of Pentadactylus (Morula) martensi 


Schepman, 1893. ° ‘Red Sea, coll. Forskal”, ZMA Moll. 2 


93.005, 24.8 mm. 15-16. Er: galatax junionae nom. noy. Lectotype of Morula 


martensi Dall, 1923, ZMB 21596, 21.4 mm. 17. Ergalatax recurrens Iredale, 1931. Holotype AMS C€57761, 25.5 mm (photo courtesy 


IE. H. Vokes). 


creamy white, or tan, usually with some light to dark 
brown colored spiral cords on shoulder dad on more 
prominent nodes. Aperture cream or pale yellow within. 

Operculum ergalataxine (Fig. 11), D- shaped with lat- 
eral nucleus in lower right. 

Radula (Fig. 6) with a rachidian bearing a long, narrow 
central cusp, a small, narrow, triangular ace, denticle 
anda broad, long lateral cusp on each side. Sickle-shaped 
lateral teeth broad at base and narrow at their end. 


Type Material Examined: 3 syntypes NMNL, Red 
Sea: RMNH.MOL.57165, here selected as lectotype and 
paralectotypes, 1 syntype ZMA Moll. 2.93.005, Red Sea, 
here selected as paralectotype; Ergalatax obscura 
Houart, 1996: Perim, Strait of Bab a Mandeb, holo- 
type and 4 paratypes MNHN, | paratype RH; Djibouti, 
Obock, Gulf of Aden, 9 paratypes MNHN; Yemen, 
Aden, 4 paratypes MNHIN, 1 paratype RH. 


Othe Material Examined: — Djibouti: Obock, Gulf of 
Aden, 2 RH; Gulf of Aden: 2, RH. Red Sea: (no other 
data), 1 RH; Massawa, Taulud Is., 1 RH; Egypt, Sinai, 
under stones at low tide, 10 RH; Egypt, Sinai, Sharm E] 
Sheik, 1 RH; Egypt, Sinai, Shark’s Bay (marsa umm 
mureihha), under stones, low tide, 2 RH. Gulf of Aqaba: 
Israel, Eilat, 0.5-1.0 m, 3 RH: Gulf of Aqaba, Israel, 
Eilat, under stones, 1-2 m, 4 RH. 


Distribution: 
Aqaba 0-2 m, on and under stones (Fig. 46). 


Remarks: For a comparison with Ergalatax junionae 
e that species below. 


Ergalatax junionae nomen novum 
Figures 2, 4, 7, 12, 15-16, 31-40, 46) 


From the Gulf of Aden to Eilat, Gulf of 


Morula siderea Reeve.—von Martens, 1874: 95, pl. 5, fig. 49 
(not Ricinula siderea Reeve, 1846). 

Morula martensi Dall, 1923: 304, new name for Morula siderea 
von Martens, 1874, not Reeve, 1546. 

Cronia konkanensis.—Bosch and Bosch, 1982: 95, text fig; 
Smythe, 1982: 60, pl. 1, fig. i; Bosch and Bosch, 1989: 60, 
text fig. (not Ricinula konkanensis Melvill, 1893). 

Cronia cf. konkanensis—Giunchi and Tisselli, 1995: 8, text 
figs. 

Ergalatax martens.—Buzzurro, Engl and Tiimtiirk, 1995: (no 
pag.), text fig.; Engl, 1995: 46, fig. 10. 

Cronia cf. hovieonensitBocdh et al, 1995: 121, fig. 480, 

Morula martensi (Schepman, 1$92)—Tan, 1995: 160, figs 52, 
192 ¢, h (not Pentadactylus (Morula) martensi Schepman, 
1892). 

Ergalatax obscura.—Houart, 2001; 108 (in part), figs. 450-451 
only; Delongueville and Scaillet, 2007: 57, fig. 31 (not 
Ergalatax obscura Houart, 1996). 


Description: Shell medium sized for the genus, up to 
29 mm in length at maturity. LengthAvidth ratio 1.91- 
2.03, Slender, lanceolate, heavy, nodose. Shoulder 
weakly sloping, concave. 

White or creamy white with dark brown or blackish 
colored primary spiral cords and occasionally s2 or s3. 
Aperture glossy white. 

Spire very high with 3.5 protoconch whorls and teleo- 
conch up to 7 weakly convex, strongly shouldered, no- 
dose whorls. Suture adpressed. Protoconch a coni- 
cal, acute; terminal lip raised, of sinusigeral 

Axial sculpture of teleoconch whorls ee vof high, 
strong, broad, nodose ribs and erratically placed strong 

varices. Last teleoconch whorl with 8-11 ribs, oceasion- 
ally with one or two erratically placed, broad varices. 
Spiral sculpture of high, strong, nodose and squamose 


R. Houart, 2008 Page 103 


Figures 18-30. Ergalatax martensi (S¢ hepman 1892). 18-23. Lectotype and paralectotypes RMNH.MOI 97 165 photo | 
Goud). 18-19. 17.4 mm; 20-21. 15.7 mm; 22-23. 17.3 mm. 24-25. Holotype ot E. obscura Houart, 1996. Perim, Strait of Bab el 
Mandeb, MNHN 0159, 24.2 mm photo MNHN 26. Eilat, Israél, RH, 20.8 mm; 27. Obock, Gulf of Aden, RH, 21.6 mm; 28. Eila 
Israél, RH, 25.1 mm; 29. Red Sea (no other data), RH, 21.4 mm; 30. Aden, Gulf of Aden, RH, 21 mm 


Page 104 THE NAUTILUS, Vol. 122, No. 2 


Figures 31-45. Ergalatax species. 31-40. Ergalatax junionae nomen novum, 31-33. Doha, Wahra, Qatar, RH. 31-32. 26.5 mm; 
33-34. 22.9 mm: 35. Yalikent, Iskenderun, Turkey, RH, 17.2 mm; 36-37. Lebanon, Bay of Jounich, 19.6 mim; 38-40. Iskenderun, 
Bay of Iskenderun, Turkey, RH. 38. 25.6 mm; 39. 22.2 mm; 40. 22.9 mm, 41-45. Ergalatax margariticola (Broderip, 1833), 41. 


927 


Rar Pumaotu Archipelago, RH, 28.4 min; 42. Kai Is., Mollucas, RH, 20.5 min, 43, South of New Caledonia, RH, 23.7 mm, 44. 


Beach, East coast of Singapore, RH, 30.4 min; 45. Kwajalein Atoll, RH, 25.9 mm 


R. Houart, 2008 


primary, secondary and tertiary cords. Shoulder of last 
teleoconch whorl with broad SP, adis, IP, abis, P1, P2, s2, 
(t), P3, s3, P4,(s4), P5, (s5), P6, ADP, MP: Pl and P2 
narrow, P3, P5 and P6 broad, similar in strength, P4 
smaller. 

Aperture large, narrow, ovate; columellar lip smooth, 
entirely adherent, with low parietal tooth at adapical ex- 
tremity; anal notch oe broad; outer lip weakly erect, 
with ID, D2-D5. ID] argest, broad; D2-D4 decreasing i in 
strength abapically, D5 split; denticles elongate within 
aperture. Siphonal canal short, broad, dorsally recurved, 
broadly open. 

Operculum (Fig. 12) dark brown, D-shaped, with lat- 
eral nucleus in lower right; attached surface with about 8 
growth lines and broad, callused rim, about 30-40 % of 
opercular width. 

Radula (Fig. 7) with a rachidian bearing a long, slender 
central cusp, a small, short lateral denticle and a broad, 
long lateral cusp on each side. Lateral teeth sickle- 
shaped, with broad base and narrow end. 


Type Material Examined: Morula martensi Dall, 
1923, lectotype (selected by Houart, 1996) and 6 syn- 
types ZMB 21596. 


Other Material Examined: Persian Gulf: Kuwait, 
Kuwait City, 4 RH; Qatar, Doha, under rocks, 30 RH; 
Sharjah. 25°20" N, 55°21’ E, on rocks, 4 RH; Abu Dhabi, 
0.5-1.0 m, 3 RH. Gulf of Oman: Al Hamra, near Qurm, 
10 km NW ‘of Muscat, 16 RH; Al Bustan, under rocks, 1 
RH. Lebanon: Beirut, harbour entrance, breakwater, 
max. 15 m, 2 RH: Bay of Jounieh (N), 10-25 m, 2 RH; 
Batroun, 1-4 m, under stones, 2 RH. Turkey: Gulf of 
Iskenderun, Iskenderun, under rocks, harbor, 1 m, 8 RH; 
Kale, beach, 6 RH; Yumurtalik, rock pools, 6 RH; Ya- 
likent, shallow water, on rocks, 1 RH; Bay of Antalya, 6 
km off Kemer, 36°39’ N, 30°33’ E, on rocks with mussel 
banks, 0.5-1.0 m, 7 RH; Fethiye-Oludeniz, on rock at 2 
m, | RH. 


Distribution: Gulf of Oman, Persian Gulf and eastern 
Mediterranean Sea, intertidal to 4 m, on and under rocks 
and stones (Figure 46). 


Etymology: This species is named in honor of Mrs. 
Marie-Louise Buyle-Junion (1916-2003), former librar- 
ian of the Belgian Malacological Society. She was one of 
the mainstays of the Society, together with her husband, 
Jean Buyle. 

Remarks: Ships docking at oil terminals in the Gulf of 
Iskenderun (eastern Turkey) could have introduced the 
species into the eastern Mediterranean Sea. As noted by 
Delongueville and Scaillet (2007), the transport via ship 
hulls or ballast water may be suspected. 

The shell morphology of some specimens of E. junio- 
nae (Fig. 35) is nearly the same to E. martensi (Fig. 30), 
however E. martensi differs in having a more strongly 

shouldered, broader shell, and a yellowish aperture in- 
stead of white, with 7 denticles within instead of 6 in E. 
junionae. Moreover, the spiral cords differ in number 
and strength, as described above and illustrated in Fig- 


Page 105 


50_0 


Ses a ; 
’ “4, ; 
9° 2 FY Cupp. 
O Yo, 
oO ikea 
mo 
\e 
° 
“ 
“e@@ Aden 
® Gos of” 

@ Ergalatax martensi (Schepman, 1893) ° 
O Ergalatax junionae nomen novum — Ps 


Fig. 46. Distribution of Ergalatax martensis and E. junionae. 


ures 2-5. Twenty-five specimens of E. martensi and a 
few more of E. junionae were examined to confirm the 
stability of these differences. 

Ergalatax martensi usually has 6-8 axial ribs on the last 
teleoconch whorl, rarely 5 or 9, compared to E. junionae 
which bears 8-11 ribs and varices on the last whorl. 

Ergalatax margariticola (Figures 8, 14, 41-45), a very 
common Indo-West Pacific species is also related to E. 


junionae, however it is generally stouter and broader, 


with a wider shoulder, a more squamose spiral sculpture, 
more uniformly colored shell and different aperture 
color, being bluish-white, occasionally with a tinge of 
pink or mauve on the columellar lip vs. completely white 
in E. junionae. The shell morphology and color of E. 
margariticola are highly variable, however it is always 
easily distinguishable from E. junionae by one or more 
differences cited above. 

The three species are related to the ergalataxine Er- 
galatax contracta (Reeve, 1845) (Figures 9-10), a prob- 
ale senior synonym of Ergalatax recurrens Iredale, 1931 
(Fig. 17), the type species of Ergalatax. 

Buzzurro, Engl and Tiimtiirk (1995) were the first to 
mention the presence of Ergalatax junionae [as Erga- 
latax martensi (Dall, 1923)] in the eastern Medien: 
nean Sea. 

Tan (1995:147) in his Ph D. thesis also selected a lec- 
totype for Morula martensi Dall, 1923. However, he des- 
ignated a specimen of Ergalatax margariticola (Tan, in 
litt.) from Kingsmill Is (USNM 52472). Nevertheless, 
this designation being published in a thesis that does not 
satisfy Articles 8.1.2 and 8.1.3 of the International Code 
of Zoological Nomenclature, it is not available as such. 


ACKNOWLEDGMENTS 


I am very grateful to Jeroen Goud, National Museum of 
Natural History Naturalis, Leiden for the digital images 


Page 106 


THE NAUTILUS, Vol. 122, No. 2 


of the type material of Pentadactylus martensi and for his 
comments, to David Reid, Natural History Museum, 
London, for providing me useful photocopies and for 
comments, to Robert G. Moolenbeek, Zoologisch Mu- 
seum, University of Amsterdam for the loan of the lec- 
totype of P. martensi, to the late Prof. R. Kilias, Museum 
fiir Naturkunde der Humboldt Universitat zu Berlin, for 
the loan of the specimen illustrated by Martens (1874), 
lectotype of Morula martensi Dall, to Anders Warén, 
(Natural History Museum, Stockholm) for radula prepa- 
ration and SEM work of radula and operculum, to E. H. 
Vokes (Prof. Emeritus, Tulane Univ ersity, Louisiana) for 
the photograph of E. recurrens Iredale, 1931, to Greg 
Herbert (University of South Florida, Tampa, USA) for 
his useful comments and suggestions, and to John Wolff, 
Lancaster, Philadelphia, for checking the English text. 


LITERATURE CITED 


Bosch, D. and Bosch, 1982. Seashells of Oman. Longman 
Croup, England: an pp. 

Bosch, D. and Bosch, E. 1989. Seashells of Southern Arabia. 
Motivate, Dubai: 95 pp. 

Bosch, D. T., Dance, S. P., Moolenbeek, R. G. and Oliver, 
P.G. 1995. Seashells of Eastern Arabia. Ed. P. Dance, 
Motivate Publishing: pp. 1-296. 

Buzzurro, C., Engl W. and Tiimtiirk, I. 1995. Bivalven und 
a n der Europiiischen Meere (4): Ergalatax mar- 
tensi (Dall, 1923) (Muricidae), Ein neuer Lesseps’schere 
Einwanderer von der Tiirkischen Siidkiiste. Club Con- 
chylia Informationen 27(1): 17-18. 

Coulombel, A. 1994. Coquillages de Djibouti. Edisud, La 
Calade, Aix-en-Provence: 143 pp. 


Dall, W. H. 1923. Notes on Drupa and Morula. Proceedings of 


303— 


the Academy of Natural Sciences of Philadelphia 75: 
306. 
Delongueville, C. and Scaillet, R. 2007. Les espéces invasives 
de mollusques en Méditerranée. Novapex 8(2): 47-70. 
Engl, W.1995. Specie prevalentemente Lessepsiane attestate 
lungo le coste Turche. Bolletino Malacologico 31(1-4): 
43-50, 


Giunchi, L. and Tisselli, M. 1995. Cronia cf. konkanensis 
(Melvill, 1893), new Indo-Pacific host in the Mediterra- 
nean Sea. La Conchiglia 27(275): S—9. 

Heiman, E. L. and Mienis, H. K. 2003. Shells of East Sinai, an 
illustrated list. Muricidae (2), Triton: 22-23. 

Houart, R. 1995, The Ergalataxinae (Gastropoda, Muricidae) 
from the New Caledonia region with some comments on 
the subfamily and the description of thirteen new species 
from the Indo-West Pacific. Bulletin du Muséum national 
d Histoire naturelle, Paris, 4e sér., 16A(2-4):197—245. 

Houart, R. 1996. On the identity of Morula martensi Dall, 1923 
and description of a new species of Ergalatax from the 
Red Sea (Gastropoda: Muricidae: Ergalataxinae). The 
Nautilus 110(1): 12-16. 

Houart, R. 2001. A review of the Recent Mediterranean and 
Northeastern Atlantic species of Muricidae. Evolver: pp. 
1-227. 

Martens, E. von 1874. Ueber Vorderasiatische Conchylien 
nach den Sammlungen des Prof. Hausknecht. Cassel, T. 
Fischer: 127 pp, 9 pl. 

Merle, D. 1999. La radiz den des Muricidae (Gastropoda; Neo- 
g gastropoda) au Paléogene: approche phylogénétique et 
évolutive, Thése du Muséum national d'Histoire naturelle, 
Paris: 499 pp. 

Merle, D. 2001. The spiral cords and the internal denticles of 
the outer lip of the Muricidae: terminology and method- 
ological comments. Novapex 2 (3): 69-91. 

Schepman, M. M., 1892. Note VI. Two supposed new species 
of Pentadactylus. Notes from the Leyden Museum, Vol. 
XV: LOS—104. 

Sharabati, D. 1954. Red Sea Shells, KPI, London: 128 pp. 

Singer, B.S. and H. K. Mienis. 1991a. Shells of the Red Sea. 
The family Thaidididae (sic) (1). La Conchiglia 27(260): 
16-19. 

Singer, B.S. and H. k. say 1991b. Shells of the Red Sea. 

The family Thaididae . La Conchiglia 27(261): 54-60. 

Smythe, K. 1982. Seashe ils of es Arabian Gulf. George Allen 
& Unwin, London: 122 pp. ; 

Tan, K-S. 1995. Taxonomy of Thais and Morula (Mollusca: 
Gastropoda: Muricidae) in Singapore and vicinity. Ph.D. 
thesis, National University of Singe apore. 

Verbinnen, G. and Dirkx, M. 2000. Red Sea Mollusca, Part 6. 
Gloria Maris 38(4—-5): 64-76. 


THE NAUTILUS 122(2):107-114, 2008 


Page LOT 


Two new deep-sea muricids (Gastropoda) from Argentina 


Guido Pastorino 

o Argentino de Cie ea Naturales 
Angel Gallardo 470, ° piso lab. 57 

Cl40sDIR Buenos ives: ARGENTINA — Antropologia 

gpastorino@macn.gov.ar CoG, 399 =.C.'P. 


Fabrizio Scarabino 


Direccion Nacional de Recursos Acuaticos 
and Museo Nacional de Historia Natural y 


11.000 


Montevideo, URUGUAY 


fscara@gmail.com 


ABSTRACT 


Two new species of muricids belonging in the genus Trophon 


are described from the upper slope off the Atlantic coast of 


Argentina. Both species have a small size for the genus. The 
radulae show similarities with those of Antarctic species of the 
same genus. Trophon columbarioides new species has a smooth 
shell with spines pointed adapically and was collected at 37-38° 

S, in 209-382 m. Trophon fasciolarioides new species has 
prominent spiral cords and was collected at Burwood Bank in 
286-292 m depth and off Bahia Blanca in ca. 1000 m depth. 


Additional Keywords: Mollusca, Muricidae, Trophon, South- 
western Atlantic, Gastropods, Taxonomy, Patagonia 


INTRODUCTION 


The genus Trophon includes a group of species of un- 
doubtedly austral origin. The older species can be traced 
as far back as the Oligocene, from Patagonian deposits 
(Griffin and Pastorino, 2005). Pastorino (2005) re- 
described all known living species of Trophon from 
southern South America, and also some new species be- 


longing in the genus. While the subfamilial affinities of 


this genus are discussed by several authors (e.g. Kool, 
1993, 1993a: Vermeij and Carlson, 2000, among others), 
the genus is firmly established for all fusiform and/or 
lamellate gastropods—usually with spiral ornamenta- 
tion—tfrom the southwestern Atlantic. Two of the species 
characterize the shallow waters along the Patagonian 
coast, i.e., Trophon geversianus and T. plicatus. Both are 
sympatric in the southern part of Argentina; however 
only T. geversianus could be collected intertidally as far 
North as Buenos Aires province. All the other species are 
mostly subtidal. Trophon species, as far as it is known, 
are all predators. feeding on the mussel banks and bar- 
nacles that are fairly common along the coast of the 
southern part of South America. 

Living in moderately deep waters are Trophon acan- 
thodes Watson, 1882, and the recently rediscovered T. 
clenchi (Carcelles, 1953), a rare lamellose and spiny spe- 
cies originally assigned to the genus Murex (Pastorino, 
2005). Both of them undoubtedly belong in Trophon. In 


addition, Houart (1991) and Pastorino (1999) described 
Trophon mucrone and T. veronicae from deep waters off 
South America. The two species were included in Tro- 
phon despite some minor but distinctive differences in 
radulae, protoconchs and penises. Recently Houart 
(2003) and Houart and Sellanes (2006) described new 
species from deep waters off Chile. 

In this paper two new deep water species from the 
southwestern Atlantic are described and compared with 
the other related ones living around the region. 


MATERIALS AND METHODS 


Specimens of T. columbarioides new eae studied 
herein were collected by one of us (FS) on board the 
Uruguayan R/V ALDEBARAN. The other specimens are 
from the 2002 cruise to Antarctica of the German R/V 
POLARSTERN. Additional material was collected by Uru- 
guayan fishing boats, Dissections were performed on 
ec ethanol-preserved specimens to study radulae and 
male reproductive system when it was av ailable. Radulae 
were cleaned with commercial bleach and ultrasound, 
and observed using a Philips XL 30 scanning electron 
microscope (SEM) at the Museo Argentino de Ciencias 
Naturales (MACN). Critical point drying of the penises 
(when available) was prepared at the MACN. Radular 
terminology follows Kool (1993: fig. 6B). Shell photo- 
graphs were taken using a digital camera. All images 
were digitally processed. The material is hagsed at the 
MACN one the Museo Nacional de Historia Natural y 
Antropologia, Montevideo (MNIINM). 


SYSTEMATICS 


Class Gastropoda Cuvier, 1797 

Order Neogastropoda Wenz, 1935 
Family Muricidae Ratfinesque, 1515 
Subfamily Trophoninae Cossmann, 1903 
Genus Trophon Montfort, 1810 


Type Species: Murex mage llanicus Gmelin, 1791, 
Trophon geversianus ( (Pallas, 1774) by original designa- 
tion. 


Page LOS 


THE NAUTILUS, Vol. 122, No. 2 


Trophon columbarioides new species 
(Figures 1-11, 15-18) 


Diagnosis: Shell very small, thin, fusiform, chalky; 
axial ornamentation of 30-35 regular axial lamellae grow- 
ing along the entire whorl surface, attached to the ‘shell, 
producing open, long, regularly spaced spines along pe- 
riphery, pointing upwar ds. Siphonal canal very long. 


Description: Shell small (up to 14 mm), slender, thin 


profile, chalky, white, somewhat bright; protoconch of 
two whorls, slightly globose, symmetrical: teleoconch of 


five oblique, shouldered whorls: spire height less than 4 
of total shell height. Spire angle about 40°: suture im- 
pressed; subsutural shelf well defined, oblique; aperture 
semicircular, interior chalky white; anterior siphonal ca- 
nal very long (longer than aperture height), narrow, 
straight, open; outer lip sharp, rounded, inner lip ad- 
pressed. Axial ornamentation of 30-35 regular axial 
lamellae growing along the entire whorl surface, but at- 
tached to shell and producing open, long, regularly 


spaced spines along periphery (almost in the middie of 


the whorl), pointing adapically; last three whorls with ten 
lamellae each . A second obsolete series of spines ap- 
pears to rise at the periphery of older specimens (Figure 

1). Spiral omamentation lacking. Growth lines present 
throughout shell. 

Operculum subtriangular, nucleus terminal. External 
surface covered by concentric, irregular, growth lines. 
Inner surface attachment area reaching upper side or 
center, with horseshoe-shape scars (Figures 17-18). 

Rachidian teeth of radula with thin, small central cusp; 
lateral cusps wider and larger than central cusp; denticle 
between central and aeenl cusp rising from base. Base 
of rachidian tooth slightly curved. Lateral teeth with 
single, long cusps; attachment area thick (Figures 15- 
16). 


Type Material: Holotype MACN-In 37380 (Figures 
1-3, 7, 9-10) and two paratypes, MACN-In 37381 (Fig- 
ures 4-6, 8) and MNHNM 15540 (Figure 11). 


Type Eason R/V ALDEBARAN cruise 2003/01, sta- 
tion 37, 37°43’ S, 55°00" W, 209 m, October 26 2003, 
5.3°C of bottom : ne rature, Piccard dredge; (holotype 
aad one paratype); between 37°05" S, 54°12! W in 255 m 
and 37°02’ S, 54°02’ W in 382 m (one paratype). 


Etymology: The general shell morphology reminds 
some species of the genus Columbarium (Caenogas- 
tropoda: Turbinellidae), to which it has no close relation- 
ship. 


Distribution: Known only from three specimens from 
the type locality and vicinity. 


Remarks: At first glance the general morphology of 
the shell shows some similarities with juveniles of Tro- 
phon acanthodes Watson, 1882. However the typical 
spiral cords of the latter appear early in ontogeny (see 
Figures 12-13) and are completely absent in the new 
species. In addition both species have clear radular dif- 
ferences (see Pastorino, 2005:69). In addition, T. plicatus 
and T. clenchi are comparable species. The latter has 
unmistakable early developed spiral ornamentation; the 
former has complete lamellae and a shorter siphonal ca- 
nal, besides radular differences. 

Despite the geographic distance separating them, Tro- 
phon scolopax and T. septus described by Watson (1882) 
are comparable species. They live around Kerguelen Is., 
in the southernmost Indian Ocean. As a main diferente 
Trophon columbarioides new species has a higher spire 
and only one series of large and open spines pointing 
upwar ds, instead of the three series shown by T. scolopax 
or the triangular upturned ones of T. septus. 

Trophon echinatus (Kiener, 1840), an extremely vari- 
able species (according to Bouchet and Warén, 1985: 
141), from deep waters off Northeastern Atlantic and 
Mediterranean shows a remarkable shell similarity with 
T. columbarioides new species Despite this, we do not 
support phylogenetic affinities between these and we 
made the comparison ist for showing the existence of 
specific contrasting differences, particularly at radular 
level. As far as we can see in the material available there 
are no signs of spiral cords in the new species which are 
common in the deeper specimens of the northern one. 
However, some porcellaneous (not chalky) smooth speci- 
mens are in fact comparable. All morphs of T. echinatus 
have a shorter protoconch. Some grown specimens of T. 
columbarioides shows the apparently presence of a sec- 
ond rows of spines while T. echinatus has only one. In 
addition, the radulae (illustrated by Bouchet and Warén, 
1985 figs. 333, 335, 336) shows a pair of almost obsolete 
intermediate denticles between the lateral cusps while in 
T. columbarioides new species they are slightly smaller 
than the central and lateral cusps. The base of the rachid- 
ian is sinuous in T. echinatus and somewhat straight in 
the new species. The intermediate denticles of the 
rachidian teeth rise from the internal side of the lateral 
cusps in a way that is typical of the southwestern Atlantic 
Trophon species. The attached portion of the marginal 
teeth are also different. 

Houart (2001) considered T. echinatus as belonging in 
to the genus Pagodula Monterosato, 1884 despite the 
differences that Bouchet and Warén (1985) pointed out 
with the protoconch of the type species of Pagodula: the 
Pleistocene species Murex vaginatus Cristofori and Jan, 
1832. 


Figures 1-11. 


view of the par: atype 9-10. Protoconch of the holotype, scale bar = 


Trophon columbarivides new species . 1-3. Holotype, MACN-In 37380, us ALDEBARAN cruise 2003/01, station 37, 
37°43" S, 55°00! W in 209 m. 4-6. Paratype, MACN-In 37381. Same locality as holotype 
Linm. 11. Par: itype, MNHNM 15540, between 37°05! S 


. Apical view of the holotype. 8. Apical 
54°19 


W in 255 m and 37°02" S, 54°02’ W in 382 m. 12-14. Trophon acanthodes Watson, 1882, MACN-In 25165-2, 37°35’ S, 54°55’ W, 


192 m. Seale bar = 1 em for all figures except 9-10. 


G. Pastorino and F. Scarabino, 2008 Page 1LO9 


Page 110 


THE NAUTILUS, Vol. 122, No. 


bo 


Figures 15-18. 


Unfortunately both dissected specimens of the new 


species here described were females so nothing can be 
said about the morphology of the penises. 


Trophon fasciolarivides new species 
(Figures 19-31) 


Diagnosis: — shell very small, thin, translucid or chalky; 
very weakly develope :d axial ornamentation of regular, 
very low, varices attached to the shell. Spiral ornamen- 
tation of 2, 3 and § or 9 spire al rounded cords in the 
second, ae and last whorl respectively; cords of similar 
width; intersection of spiral cords with axial varices con- 


veying a slightly cancellate appearance to shell surface 


Description: 
profile translucid white or chalky; protoconch of 2 
vhorls, symmetrical; teleoconch of 412 tabular whorls: 
spire height less than 5 of total shell height. Spire angle 


shell small (up to 12 mim), slender, thin 


about 45°: suture impressed, subsutural shelf very 


Trophon columbarivides new species. 15. Radula frontal view, scale bar = 10 jm. 16. Radula lateral view, scale 
bar= 20 wm. 17. Operculum of the holotype and, 18. Paratype in figs. 4 


~6. Scale bars= 1 mm. 


oblique; aperture suboval; anterior siphonal canal long 
but never longer than aperture height, slightly curved 
adaxially, open; outer lip sharp, rounded, inner lip ad- 
pressed, Axial ornamentation of poorly developed va- 
rices, regular, very low, attached to the shell, growing 
along the entire whorl surface except the siphonal canal. 
Spiral ommamentation of 2, 3 and 8 or 9 spiral rounded 
cords in the second, third and last whorl respectively; 
cords of similar width; intersection of spiral cords with 
axial varices conveying a slightly cancellate appearance to 
shell surface; growth lines present throughout shell, be- 
coming scaly at intersection with spiral cords. 

Operculum suboval, nucleus terminal. External sur- 
face covered by concentric, irregular, extremely thin 
growth lines. Inner surface attachment area reaching up- 
per side or center, with horseshoe-shaped scars (Figure 
31) 

Rachidian teeth of radula with thin central cusp, 
higher than laterals; denticle between central and lateral 


G. Pastorino and F. Scarabino, 2008 


Figures 19-26. Trophon fasciolarioides new species 19-21. Holotype MACN-IN 37382. Banco Burwood, Polarstern St. 150 
between 54°30.22' S, 56°08.58’ W in 286 m and 54°29.64' S 56°08.09’ W in 292 m. 22. Paratype MACN-IN 37383, same locality 


as holotype. 23-24. Protoconch of the paratype of figure 22, arrow head transition to teleoconch, scale bar = 1 mm. 25-26. Paratype 
MACN-In 37354, off Bahia Blanca ca. 1000 m depth Seale bar = 5 mm for all figures except 23 and 24 


cusp thin and long, rising from the base. Base of rachid- 
ian tooth curved. Marginal area with single cusp. Lateral 
teeth with single long cusps attachment area thick Fig- 
ures 27-29 

The penis shows an unusual morphology among the 
Patagonian species of the genus: it is wide slightly 
curved and laterally flattened with a small slit at the tip 
Figure 30 


Type Material: Holotype MACN-IN 37382 and two 
paratypes MACN-IN 37383 and MACN-In 37384 


Type Locality: Banco Burwood. Polarstern station 
150, 6 Apr. 2002, AGT net, between 54°30.22' S 
56°08.58' W in 286 m and 54°29.64' S$, 56°0S.09’ W in 


292 m. 


Etymology: The general shell morphology reminds 
some species ol the genus Fasciolaria (Caenogastropoda 


Fasciolariidae 


Distribution: known from type locality and approxi 
mately off Bahia Blanca in ca. 1,000 m depth (MACN-In 
37384 


Page 112 


THE NAUTILUS, Vol. 122, No. 2 


Figures 27-31. Trophon fasciolarioides new species. 27. Radula of a paratype, MACN-In 37354, frontal view, scale bar = 40 jum. 


an 


28. Lateral view of the same radula as 27, scale bar = 50 ym, 29. Radula of the paratype MACN-In 37353, scale bar = 20 um. 30. 


Critical point dry of the penis, scale bar = 500 jzm. 31. Two views of the operculum of the paratype of figures 2 


lL mm. 


Remarks: The morphology of the shell and the radula 


match some of the Patagonian and Antarctic species of 


the genus Trophon. From the first group, T. ohlini 
Strebel, 1904, a rare species from the Magellanic area, 
has a similar profile. However, its distinctive protoconch 
points out a clear difference. In addition, the morphology 
of the penis is far from the typical Patagonian represen- 
tatives of the genus. Trophon emilyae Pastorino, 2002, T. 
declinans Watson, 1882, and T. cuspidarioides Powell, 
1951, are comparable Antarctic species (Pastorino, 
2002a). The first two species differ from the new species 
in having well developed axial sculpture only. Also T. 


cuspidarioides has 5} 


» whorls including the protoconch 
anc five blunt spiral cords in the body whorl, while the 


new species has a larger spire and 62 whorls, and 8 or 9 


26, scale bar = 


rounded and well defined cords in the last whorl. The 
axial ornamentation in the new species is very irregular 
but higher in number than the 25 axials present in T. 
cuspidarioides, 


DISCUSSION 


In previous papers two groups of species belonging in the 
genus Trophon—i.e., from South America and Antarc- 
tica—were pointed out. The main differences between 
these two groups are several radular and anatomical fea- 
tures (Pastorino, 2002b, 2005). The shell morphology of 
the two new species described herein agrees with that of 
Trophon living off the South American coast. However, 


G. Pastorino and F. Scarabino, 2008 


Page 113 


Golfo 
San Jorge 


© 
S 
= 
xv | 
Ss 
GB 


, Malvinas Is. | 
“Faas o- 
| 


Staten Is. 


70 


Figure 32. | Map showing the type localities of Trophon 
columbarioides new species @ (filled circle) and Trophon fas- 
ciolarioides new species [(_] (blank square). 


the radula has some features found in the Antarctic rep- 
resentatives of the genus—albeit the taxonomic status of 
the Antarctic species still needs confirmation. Like in the 
Antarctic group of species, the central cusp of the rachid- 
ian teeth is shorter and thinner than the laterals, the 
internal denticle rise from the base of the rachidian in- 
stead of the upper third of the internal side of the lateral 
cusp, and the lateral teeth have a wide attachment area. 

The geographic ranges of species of these two groups 
do not overlap. Therefore. the morphological features 
seem to be clearly separated geographically too. How- 
ever. in the species described herein, this seems not to be 
the case. While there is no geographic overlapping of 
species. the morphological features that characterize the 
Antarctic group do appear in these two new Patagonian 
species. Contrarily, this does not occur in the case of 


Patagonian species, the morphological features of which 
are restricted to South America. Moreover, they are bet- 
ter developed in species from the Magellanic among 
those described from southern South America (see Pas- 
torino, 2005). 

Recently, Houart (2003) introduced three new species 
he assigned to Trophon from off Chile: T. ceciliae; T. 
condei aad T. vangoethemi. The morphology of the 
radula, only known for the last two species, together with 
that of T. mucrone Houart, 1991, T. veronicae Pastorino, 
1999 and both new species described here, is that of the 
Antarctic group, despite the geographic distance sepa- 
rating them. All mentioned species were collected from 
more than 300 m depth and in most cases they reach 
more than 1,000 m (see Table 1). Most of the Patagonian 
species of Trophon were collected alive from the inter- 
tidal zone to about 300 m depth. Exceptions are some 
specimens of T. acanthodes, T. clenchi, and T. bahamon- 
dei that come from deeper areas. However, there is no 
Antarctic species—or with their characteristic radular 
and anatomical features—that live at shallower depths or 
closer to the continent. Both new species described here 
were collected from about 300 m depth, which is the 
shallowest for a Trophon with Antarctic features at South 
American latitudes. 

Generic assignment in the whole Trophoninae group 
is actually under revision. The concept of the genus Tro- 
phon sensu stricto is easy to apply to the shelf species. 
However, when deeper species are studied, the presence 
of several features in common among the shallower spe- 
cies appear somewhat contradictory. Nevertheless, we 
think that biogeographic boundaries are hard to surpass 
for species without free larvae as both new species ap- 
pears to be according to the shape and whorl number of 
the protoconch. Contrasting historical biogeographic 
processes must be considered when suggesting affinities 
between species and therefore comparison with species 
living far away from the Southwestern Atlantic area can 
be considered an empty exercise. Some exempli are cases 
like T. columbarioides new species—T. echinatus which 
have very few characters that clearly split both species. In 


Table 1. Recently described South American species of Trophon. 


Depth (m) Type locality 


T. condei Houart, 900-1350 Ancud, Chile 
2003 

T. ceciliae Houart, 434-1000, 1300 Antofagasta, Chile 
2003 


Itata, North of 
Concepcion, Chile 


T. vangoethemi About 350 


Houart, 2003 


T. mucrone Houart, 790-1575 Off Rio de 
1991 Janeiro, Brazil 

T. veronicae Pastorino, 298-1272 Subantarctic 
1999 

T. columbarioides 209-382 37°43'S, 55°00'W. 
new species 

T. fasciolarioides 286-292 Burwood bank 


new species 


Page 114 


THE NAUTILUS, Vol. 122, No. 2 


the future molecular characters could add to the under- 
standing of the evolution of morphological features in 
this group. 


ACKNOWLEDGMENTS 


We are grateful to L. Paesch and R. Bird (DINARA) 
for facilities in the collecting of Trophon specimens, as 
well as to J. L. Viggiano and H. Racz-Lorenz (Monte- 
video) for the material provided. We thank Miguel Grif- 
fin (UNLPam) who helped to improve the English ver- 
sion of the manuscript. R. Houart (Belgium) reviewed 
this paper; his comments and discussions undoubtedly 
contributed to enhance the original version despite dis- 
agreement on some points. This contribution was sup- 
ported in part by the project PICT No, 03-14419 from 
the National Agency for Scientific and Technical Promo- 
tion, Argentina. We acknowledge funding by the Consejo 
Nacional de Investigaciones C ientificas y Técnicas 
(CONICET) of Argentina, which sponsors the research 
of G.P 


LITERATURE CITED 


Bouchet, P. and A. Warén. 1985. Revision of the northeast 
Atlantic bathyal and abyssal Neogastropoda excluding Tur- 
ridae (Mollusca, Gastropoda). Bolletino Malacologico 
supplemento 2: 123-296. 

Carcelles, A. 1953. Nuevas especies de gastropodos marinos de 
las Reptiblicas Oriental del Uruguay y Argentina. Comu- 
nicaciones Zooldgicas del Museo de Historia Natural de 
Montevideo, 4(70):1—18. 

Griffin, M. and G, Pastorino, 2005. The genus Trophon Mont- 
fort, 1810 ( Gastropoda: Muricidae) in the Tertiary of Pa- 
tagonia. Journal of Paleontology 79: 296-311. 

Houart, R. 1991. The southeastern Brazilian Muricidae col- 
lected by R V Marion-Dufresne in 1987, with the descrip- 
tion of three new species. The Nautilus 105; 26-37. 


Houart, R. 2001. Ingensia gen. nov. and eleven new species of 


Muricidae (Gastropoda) from New Caledonia, Vanuatu, 
and Wallis and Futuna Islands. In: P. Bouchet and B. A. 
Marshall (eds.) Tropical Deep-Sea Benthos, volume 22. 
Mémoires du Muséum national d’Histoire naturelle 1S5: 
243-269. 

Houart, R. 2003. Description of three new species of Trophon 
s. |. Montfort, 1810 (Gastropoda: Muricidae) from Chile. 
Novapex 4: 101-110. 

Houart, R. and J. Sellanes. 2006. New data on recently de- 
scribed Chilean trophonines (Gastropoda: Muricidae), de- 
sc ription of anew species and notes of their occurrence at 
a cold seep site. Zootaxa 1222: 53-68. 

Kool, S. P. 1993. Phylogenetic analysis of the Rapaninae (Neo- 

gastropoda: Muricidae): Mal: wcologia 35: 155-260. 

Kool. S. P. 1993a. The systematic position of the genus Nucella 
(Prosobranchia: Muricidae: Ocenebrinae): The Nautilus 
107: 43-57. 

Monterosato, A. di, 1554. Nomenclatura generica e specifica di 
alcune conchiglie Mediterranee. Palermo, 152 pp. 

Pastorino, G. 1999. A new species of gastropod of the genus 
Trophon Montfort, 1810 (Mollusca: Gastropoda: Muri- 
cidae) from subantarctic waters. The Veliger 42: 169-174. 

Pastorino, G. 2002a. Two new Trophoninae (Gastropoda: Mu- 
ricidae) from Antarctic waters. Malacologia 44: 353-361. 

Pastorino, G. 2002b. Systematics and phylogeny of the genus 
Trophon Montfort, 1810 (G astropoda: Muricidae) from 
Patagonia and Antarctica: morphological pattems: Bollet- 
tino Malacologico 38; 127-134. 

Pastorino, G. 2005, A revision of the genus Trophon Montfort, 
1810 (Mollusca: Muricidae) from southern South 
America: The Nautilus, 119: 55-82. 

Powell, A. W. B. 1951. Antarctic and subantarctic mollusca: 
Pelecypoda and Gastropoda. Discovery Reports, 26: 47— 
196. 

Vermeij, G. and S. Carlson, 2000. The muricid gastropod sub- 
family Rapaninae: phylogeny and ecological history. Pa- 
leobiology 26; 1946. 

Watson, R. B. 1882. Mollusca of H. M.S. ‘Challenger’ Expedi- 
tion. Part 13. The Journal of The Linnean Society. Zool- 
ogy, 16: 358-392. 


Sponsored in part by the State of Florida, Department 
of State, Division of Cultural Affairs, the Florida Arts 
Council and the National Endowment for the Arts 


pRI> ¢ 
S 


XQ 
e) 
aa 
9 
2 

% 


v ce 


NATIONAL 
ENDOWMENT 
FOR THE ARTS 


gH 


INSTRUCTIONS TO AUTHORS 


THE NAUTILUS publishes articles on all aspects of the 
biology, Baleontolenie and systematics of mollusks. 
Manuscripts describing original, unpublished research 
and review articles will be Re Gaara Brief articles, not 
exceeding 1000 words, will be published as notes and do 
not require an abstract. Notices of interest to the mala- 
cological community will appear in a notices section. 


Manuscripts: Each original manuscript and accompa- 
nying illustrations should be submitted to the editor pref- 
erably via e-mail or as hardcopy in triplicate. 

Text must conform to the dimensions of 84 x 11-inch 
paper, double-spaced, and single-column throughout (in- 
cluding literature cited, tables, and figure captions). Au- 
thors should follow the general recommendations of Sci- 
entific Style and Format—The CSE Manual for Authors, 
Editors, and Publishers, available from the Council of 
Science Editors at www.councilscienceeditors.org. The 
first mention of a scientific name in the text should be 
accompanied by the taxonomic authority, including year. 
Latinized names and other words to be printed in italics 
must be underlined; leave other formatting indications to 
the editor. Metric, not English, units are to be used. Geo- 
chronologic modifiers should be capitalized only when 
units are formally recognized: for instance, use Late Cre- 
taceous but early Miocene. Likewise, only modifiers of 
formally recognized chronostratigraphic units are capi- 
talized: use Lower Jurassic but upper Oligocene. 

The sequence of sections should be title page, ab- 
stract, introduction, materials and methods, results, dis- 
cussion, acknowledgments, literature cited, tables, figure 
captions, figures. The title page should include the title, 
author’s name(s) and address(es). If corresponding au- 
thor is not the senior author, please indicate. The ab- 
stract should summarize in 250 words or less the scope, 
main results, and conclusions of the article. Abstracts 
should be followed by a list of additional key words. All 
references cited in the text must appear in the Literature 
Cited section and vice-versa. Please follow a recent issue 
of THE NAUTILUS for bibliographic style, noting that 
journal titles must be unabbreviated. Information on 
plates and figures should be cited only if not included 
within the pagination of cited work. Tables must be num- 
bered and each placed on a separate page. If in doubt, 
please follow a recent issue of the journal for sequence of 
sections and other style requirements. 


Illustrations: Illustrations are rendered either at full- 
page width (maximum width 17 cm) or column width 
(maximum width 8.2 cm). Please take these dimensions 
into consideration when preparing illustrations. Page- 
width illustrations ideally should span the entire width of 

rinted page (17 cm). “Tall” page-width illustrations 
should be avoided, square or “landscape” formats work 
better. Please design plates accordingly, such that there 
will be enough space Heft at the bottom of printed page 
for plate caption. (Digital technology has made this task 
much easier.) 


All line drawings must be in black, clearly detailed, 
and completely labeled. Abbreviation definitions must be 
included in the caption. Line drawings must be high 
resolution files at least 600 dpi (dots per inch) resolution 
at actual size. Standard digital formats for line drawings 
include .tif, .bmp, .psd, .eps, and .pdf. 

Photographs may be submitted in black-and-white or 
color, preferably in RGB mode if in color. Standard digi- 
tal formats for photographs include .tif, .psd, .jpg, or .pdf. 
Photographs must be high resolution files at least 300 dpi 
resolution at actual size. 

If more than one figure is included in an illustration, 
all figures are to be consecutively numbered (Figures 1, 
2, 3;,..., NOT Figures 1A, 1B, 1C,..., NOR Plate 1, 
Figure 1, . . .). In illustrations with more than one figure, 
make sure that blank areas between figures is kept toa 
minimum, thereby allowing for more area for each indi- 
vidual figure. 

Compressed files (e.g., jpg) may be used to facilitate 
transmission of files during original submission, but may 
not be acceptable at final submission (see below). 


Voucher Specimens: Deposition of the holotype in a 
recognized institutional, public collection is a require- 
ment for publication of articles in which new species- 
level taxa are described. Deposition of paratypes in in- 
stitutional collections is strongly encouraged, as is the 
deposition of representative: voucher specimens for all 
other types of research work. 


The Editorial Process: Upon receipt, all manuscripts are 
assigned a number and acknowledged. The editor reserves 
the right to return manuscripts that are substandard or 
not appropriate in scope for THE NAUTILUS. Manu- 
scripts deemed appropriate for the journal will be sent 
for critical review to at least two reviewers. The review- 
ers’ recommendations will serve as basis for rejection or 
continuation of the editorial process. Reviewed manu- 
scripts will be sent back to authors for consideration of 
the reviewers’ comments. The revised version of the 
manuscript may at this point be considered accepted for 
publication by the journal. 


Final Submission: Authors of accepted manuscripts are 
required to submit a final version via e-mail to the editor 
at jleal@shellmuseum.org. Please do not send low-resolu- 
tion or compressed illustration files at this stage. Send any 
files larger than 20 Mb on a CD or DVD to the editor. 


Proofs: After typesetting, proofs will be sent to the au- 
thor. Author should read proofs carefully and send cor- 
rections to the editor within 48 hours. Changes other than 
typesetting errors will be charged to the author at cost. 


Offprints: An order form for offprints will accompany 
the proofs. Offprints will be ordered through the Zt ae 
pao with institutional, grant, or other research sup- 
port will be asked to pay for page charges at the rate of 
$60 per page. 


@ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper) 


hee any an 


Y . " i 


SY We VAAN, 
i ny ) cn 


: ie ; th 


aM 


! nn bisnyy the ne ne 7 ° 
va i" ae i i ie Wa 


if ae Ie ign a ne ur i if 


BG i a i, hans rs ont his Ash a yi 


on 


ae 
on pa 


i ie WAL: 
‘ 


le | 


eu ye oy NY | etter: 1 o 
i y pe if VEL ia he | "{ a i ne if ' Mi 7 th i Lae» ary i 
nig haa ite a) 14 eh Ay ‘ , uae PASSED Ae Hye Ab wilt ey 
ital ! ey) Wi oi i a ee ue co y Bay ls eh Ke Da 
. 7s mae . +i mt wi | 


——— 
= 

ee 
2 = 
-_ 

— 

5 

— 
— 


ca 
H ; 
mnt & ae | 
‘ a Ml ite a ; 
Ne Vi : 1 oa es og i i sf 
The (ha i> Ph 
Be j a Te ey Le Ww ; idl 
| VR i j f i Sf if | ' Abas | Ayes 
heed || BY Lye ne fi) i net 
i De, er 82 ey gy Jib. as hy 
0 15 wae ea At’ 
y iy ee ne iy i “? 
mal ; un ee Pre 
Hy | seent ; fl ' 
i lh : Af : 4 
bo ty os 
v at (ifs aint i 
. ¥ ‘i 
We Ay Al | ay 4 
Abh Ti) ! ‘ : 
ta : A ‘1 { 
{y a ij i | ¥ 
; Lt) 7 if 1 y i! ha 1) 
! fi 
/ ‘ ae l Pe 
1 aa] 7 
“— t 
i ! t’ ' | 
' 
i | ay | ! ' ‘ 1) i y Y 
4 i= 
1) roe yal ; ' i 
= ) - ' . ‘ - ! 
1! ' ii ' A 
1 Chrys 
a iF i : r pei 
yay A : 4 | i 
: " ', : : iV i aan 
‘ 4 1% i ' : 
i ; } {> / / 
q 7 toy [ ' i : 
: i , 4 : j ; 
é y ' 1 : i 1 _ 
3 : = a 
ie i )? 
7 : } : 
e 4 . ‘ 
i 1 i 1 - 1 1 1 ' 
y' = ? : ] 
1 ‘ 
t ' <é ; t 
| - ' 


; : ; ' ‘ ey he ie 
; rs cot he 
yt ' i : " fase capen ale 
, 7 ; ; | : ‘ ‘ i ug i) : Dis cae 
i. . | ; ly a 
; : : a ' b AGE ail 
AY = i a, 7 ee 


. ; po 
' : 4 
f oe Ya 
« f os 

; : ; i : a5) A 4 \ 
' : uy ace 
A! ye ae 
r% i 1 1a 

' ; c Fyintile 

: | ei" 
oe rp ttce 
i ' i i x 
i ' 1 l { : | goed : 
t Hid te 
! ‘i } 
: : ; ’ 14 i hat 
7 y° ‘ - 
1 : : ; 1 ! . i ay 
| Fr i _t 7 HN . 
| jl - ; - A ' J 
; =" ; ; 
‘ ) oe » 
' , 
- io : rr | : _ - {| - 


THE NAUTILUS 


Volume 122, Number 3 
September 26, 2008 
ISSN 0028-1344 


A quarterly devoted 
to malacology. 


EDITOR-IN-CHIEF 


Dr. José H. Leal 

The Bailey-Matthews Shell Museum 
3075 Sanibel-Captiva Road 
Sanibel, FL 33957 


BUSINESS MANAGER 


Mary Jo Bunnell 

The Bailey-Matthews Shell Museum 
3075 Sanibel-Captiva Road 
Sanibel, FL 33957 


EDITOR EMERITUS 


Dr. M. G. Harasewych 
Department of Invertebrate Zoology 
National Museum of 

Natural History 
Smithsonian Institution 
Washington, DC 20560 


CONSULTING EDITORS 


Dr. Riidiger Bieler 
Department of Invertebrates 
Field Museum of 

Natural History 
Chicago, IL 60605 


Dr. Arthur E. Bogan 

North Carolina State Museum of 
Natural Sciences 

Raleigh, NC 27626 


Dr. Philippe Bouchet 

Laboratoire de Biologie des 
Invertébrés Marins et Malacologie 

Muséum National d’Histoire Naturelle 

55, rue Buffon 

Paris, 75005 France 


Dr. Robert H. Cowie 

Center for Conservation Research 
and Training 

University of Hawaii 

3050 Maile Way, Gilmore 409 

Honolulu, HI 96822 


Dr. Robert T. Dillon, Jr. 
Department of Biology 
College of Charleston 
Charleston, SC 29424 


Dr. Eileen H. Jokinen 
8234 E. North Shore Road 
Sault Ste. Marie, MI 49783 


Dr. Douglas S. Jones 

Florida Museum of Natural History 
University of Florida 

Gainesville, FL 32611-2035 


Dr. Harry G. Lee 
4132 Ortega Forest Drive 
Jacksonville, FL 32210 


Dr. Charles Lydeard 

Biodiversity and Systematics 
Department of Biological Sciences 
University of Alabama 

Tuscaloosa, AL 35487 


Bruce A. Marshall 
Museum of New Zealand 
Te Papa Tongarewa 
P.O. Box 467 
Wellington, NEW ZEALAND 


Dr. James H.,.McLean 
Department of Malacology 
Natural History Museum 
of Los Angeles County 
900 Exposition Boulevard 
Los Angeles, CA 90007 


Dr. Paula M. Mikkelsen 

Paleontological Research 
Institution 

1259 Trumansburg Road 

Ithaca, NY 14850 


Dr. Diarmaid O F oighil 

Museum of Zoology and Department 
of Biology 

University of Michigan 

Ann Arbor, MI 48109-1079 


Dr. Gustav Paulay 

Florida Museum of Natural History 
University of Florida 

Gainesville, FL 32611-2035 


Mr. Richard E. Petit 
P.O. Box 30 
North Myrtle Beach, SC 29582 


Dr. Gary Rosenberg 

Department of Mollusks 

The Academy of Natural Sciences 
1900 Benjamin Franklin Parkway 
Philadelphia, PA 19103 


Dr. Angel Valdés 
Department of Malacology 
Natural History Museum 
of Los Angeles County 
900 Exposition Boulevard 
Los Angeles, CA 90007 


Dr. Geerat J. Vermeij 
Department of Geology 
University of California at Davis 
Davis, CA 95616 


Dr. G. Thomas Watters 
Aquatic Ecology Laboratory 
1314 Kinnear Road 
Columbus, OH 43212-1194 


SUBSCRIPTION INFORMATION 


The subscription rate for volume 
123 (2009) is US $50.00 for 
individuals, US $85.00 for 
institutions. Postage outside the 
United States is an additional US 
$10.00 for regular mail and US 
$25.00 for air delivery. All orders 
should be accompanied by payment 
and sent to: THE NAUTILUS, P.O. 
Box 1580, Sanibel, FL 33957, USA, 
(239) 395-2233. 


Change of address: Please inform 
the publisher of your new address at 
least 6 weeks in advance. All 
communications should include both 
old and new addresses (with zip 
codes) and state the effective date. 


THE NAUTILUS (ISSN 0028-1344) 
is published quarterly by The Bailey- 
Matthews Shell Museum, 3075 
Sanibel-Captiva Road, Sanibel, FL 
33975. 


Periodicals postage paid at Sanibel, 
FL, and additional mailing offices. 


POSTMASTER: Send address 
changes to: THE NAUTILUS 
P.O. Box 1580 

Sanibel, FL 33957 


CONTENTS 


NAUTILUS 


Volume 122, Number 3 
September 26, 2008 
ISSN 0028-1544 


LouElla R. Saul 
Richard L. Squires 


Samuel S. Espino 


Cretaceous trichotropid gastropods from the Pacific slope of North 
America: Possible pathways to calyptraeid morphology ........ . 115 


Feeding behavior, phylogeny, and toxinology of Conus furvus Reeve, 


Alan J. Kohn 1843 (Gastropoda: Neogastropc nda: Conidae) 2... 200200000... 148 

James A. Villanueva 

Frank M. Heralde HI 

Patrice Showers Corneli 

Gisela P. Concepcion 

Baldomero M. Olivera 

Richard Duerr Two new species of Mitrella (Gastropoda: Neogastropoda: 
Columbellidae) from the lower Miocene Chipola Formation of 
northwestern Florida 2.0.0.0... 0000000000 eee 15] 

Fresia Villalobos-Rojas Catalogue of the type material of mollusks deposited at the Zoology 

Ana. G. Guzman-Mora Museum, University of Costa Rica... 0... ee ee 155 

Yolanda E. Camacho-Garcia 

Takenori Sasaki Dilemma japonicum new species (Bivalvia: Anomalodesmata: 

José H. Leal Poromyidae): A new record of the genus from the 
Northwest Pacific 2... 000000000 ce ee ees 166 

Victor Scarabino On the genus Heteroschismoides Ludbrook, 1960 (Scaphopoda: 

Carlos Henrique Soares Caetano Gadilida: Entalinidae), with descriptions of two new species .... . 171 

Norma Emilia Gonzalez-Vallejo Parasitism of Monogamus minibulla (Olsson and McGinty, 1958) 
(Gastropoda: Eulimidae) on the red sea-urchin Echinometra lucunter 
(Linnaeus, 1758) (Echinodermata: Echinometridae) on the Caribbean 
coast of Mexico 2.0... 0. ee ee 17S 

EE AUUAMNE oo, ya fe anh Gia ec al st less es sh oe ey tard vy 4h dd haat aa ds dich Gp wes decd ete, Bh Suc ae Pate 182 

INQUGES! <2 bo. Fie loa jo dod a Be ROR, Oh Be Solin he Ge & Geb ead FH ed dad beg BOS bee ah ea eee oo ae Oe 183 


, 
oto *s 
' 
' 

cm - 

' 

' . 
‘ 


THE NAUTILUS 122(3):115-142, 2008 


Page 115 


Cretaceous trichotropid gastropods from the Pacific slope of 
North America: Possible pathways to calyptraeid morphology 


LouElla R. Saul 


Invertebrate Paleontology Section 
Natural History Museum of 

Los Angeles County 

900 Exposition Boulevard, 

Los Angeles, CA 90007 USA 
lousaul@earthlink.net 


Richard L. Squires 
Department of Geological Sciences 
California State Univ ersity, 
Northridge, CA 91330-8266 USA 
richard.squires@csun.edu 


ABSTRACT 


Late Cretaceous gastropods belonging to genus Lysis Gabb, 
1864, from the Pacific slope of North America, bridge the mor- 
phologic gap between turbiniform trichotropids and limpet-like 

calyptraeids. Development of the depressed and broadened 
inner lip/columella of Lysis resulted in a larger aperture that 
allowed more space for the foot to grasp a hard substrate. 

Pacific slope species of Lysis are represented by five species 
that collectively span an interval from late Coniacian to late 
Maastrichtian. They stem from two lineages of the trichotro- 
pine genus Ariadnaria Habe, 1961. The first lineage, which 
includes Ariadnaria ainikta new species of late Albian to Cen- 
omanian age, Ariadnaria stibara new species of Cenomanian 
age, and Ariadnaria obstricta (White, 1889) of late Coniacian? 
and Santonian age, gave rise to the Lysis duplicosta group of 
neritiform to haliotiform, coarse-ribbed Lysis, including Lysis 
mickeyi new species (earliest Lysis in the world), Lysis dupli- 
costa Gabb, 1864, Lysis jalamaca new species, and Lysis lo- 
maensis new species. The second lineage of Ariadnaria consists 
of the Turonian Ariadnaria aldersoni new species, which gave 
rise to the Lysis suciensis (Whiteaves, 1879) group. Morpho- 
logically, this group, which show crepiduliform and fine ribbed 

shells, appears likely to be a stem group from which Cenozoic 
Crepidula-like genera evolved. Garzasia new genus, which 

ranges from late Campanian or early Maastrichtian age to the 

mid Maastrichtian, evolved from the Lysis duplicosta group 
and includes Garzasia intermedia (Cooper, 1894) and Garzasia 
diabla new species. The very broad, depressed spiraling inner 
lip of Garzasia is suggestive of Calyptraea Lamarck, 1799. We 
— placement of Lysis and Garzasia in Lysinae new sub- 
family of the Trichotropidae. In addition to their occurrence 
along the Pacific Slope of North America, Lysis or Lysis-like 
gastropods are known from middle Santonian to lower Cam- 
panian strata in South Africa, upper Campanian in the Congo, 
and Maastrichtian strata in Mozambique and Japan. 


Additional Keywords: Trichotropidae, Lysinae, Calyptraeidae, 
evolution, paleogeographical occurrence 


INTRODUCTION 


This study deals with the fossil record of the extinct ge- 
nus Lysis Gabb, 1864, a small group of enigmatic gas- 


tropods which has received little or no study regarding its 
ancestry, point of origin in terms of time and geogr aphic 
locale, and ev. olutionary history. Specimens have been 
found in shallow-marine Cretaceous deposits from 
southern Vancouver Island and neighboring Gulf Is- 
lands, British Columbia, Canada to northem Baja Cali- 
fornia, Mexico (Figure 1) and, although Lysis-like gas- 
tropods have been reported at a few locales | in the world, 
its familial placement has been uncertain. This study 
brings new information about all of these items. 

Material for this study included type specimens, addi- 
tional collections from their type localities, and speci- 
mens of Late Cretaceous age (Coniacian to Maastrich- 
tian) from previously unreported- upon localities. Figure 
il provides an index to areas which yielded specimens 
used in this study. 

During the study we discovered undescribed species 
of the trichotropine g genus Ariadnaria Habe, 1961, which 
appear to have given rise to two groups of Lysis, a ner- 
itiform coarse- nabbed group and a crepiduliform fine- 
ribbed group. As will be discussed under “Evolutionary 
Implications,” we propose that the neritiform group 
evolved into Garzasia new genus, which appears to be a 
precursor to calyptraeiform | genera, 

A total of three genera (one of them new) and 11 
species (seven of them new) make up this study. The taxa 
are: four species of Ariadnaria (three of them new), five 
species of Lysis (three of them new), and two species of 
Garzasia (one of them new). The ranges in time of all 
these species, as well as two recognizable groups of Lysis, 
are plotted on Figure 2. The Lysis duplicosta group con- 
sists of four species, spanning a total interval of late Co- 
niacian to late Maastrichtian. This group was also the 
most geographically widespread, with specimens col- 
lected from British Columbia to Baja California. The 
Lysis suciensis group is known only from the Campanian 
to possibly early Maastrichtian, with specimens known 
from British Columbia to Baja California Sur. Lysis per- 
sisted for a total of approximately 18 million years in the 
study area. 

The specimens studied here are mainly from fine- 


Page 116 THE NAUTILUS, Vol. 122, No. ‘ 


LATITUDINAL DISTRIBUTION OF SPECIES 


13 - Agua Caliente Canyon. 
Santa Barbara Co., CA 
14 - Warm Springs Mtn., 


1 -Vancouver and adjacent 
islands, British Columbia 
2 - Sucia Island, San Juan Co., 


Washington Los Angeles Co., Calif. 
3 - South Cow & Bear creeks, 15 - Live Oak Canyon, 
Shasta Co., Califomia 1 1 e Kem Co., California 
4 - Ono area, Shasta Co., Calif. 2 e 2 16 - Simi Hills, Los Angeles 
5 - Chico Creek, Butte Co., @ ® and Ventura counties, 
California California 
6 - Pentz and Dry Creek, Butte 17 - Santa Monica Mtns, 
Co., Califomia Los Angeles and 
7 - Texas Flat & Granite Bay, Ventura counties, Calif. 
Placer Co., California 18 - Santa Ana Mtns., 
- Pigeon Point, San Mateo Co., Orange Co., California 
California 19 - near Carlsbad, San 


9 - Garzas Creek, Stanislaus Co., Diego Co., California 


California 20 - Point Loma, San 
10 -south of Garzas Creek, Diego Co., California 
Merced Co.,Califomia 21 - East of La Misién, 


Baja California, Mexico 
22 - Cafion San Fernando, 

Baja California, Mexico 
23 - Punta Abreojos, Baja 


California Sur, Mexico 


11 -Cooper Canyon, Fresno 
Co., Califomia 
2 -Jalama area, Santa Barbara 
Co., Califomia 


e ‘e 3@ °e 
5@ -® 
s § (97% 68 
'S = b= 7@ 708 
a. = = 
| i} 
— 
N 8 9 
® 108 
11 = 
@ 3s 
re = 
§ SU 159 
= e 2e 12g “e S 
< 16@ 16 S 
= 19@ 19@ i 
= Ss 20@ 20@ 
iS 2105 
= ‘Se 8 
js s 8s 
§ ys § 
S SS 
~ YY 
22 
22@ <@ 
S 
=. & 
: Kee 3 BAT, Ss 
‘ ~ 


50 100 500 kn} CALIFORN), 


a 23geN 23@ 


Figure 1. Index map of collecting localities 


L. R. Saul and R. L. Squires, 2008 Page 117 


35 33.8 —t_ cia | foe 3 
ae PRIABONIAN IC16 | jaSo%| : 
— 372 — Hse 

40} 1.3 BARTONIAN = FRS|CI8) es} 
| YQ es eo | Pa = 
ed m 4c20| Peres) 

45 en 5 LUTETIAN ee| & : 
— IE a Ee: 
=] Ol ae Tl fest S, 

50-4 oO a C22 | SS a 
=== | —|—_ i = = 
4 YPRESIAN a3 »~ aS 

ox) 4 = » SS 
= < 55.8 3 8 < = 
=4 Z THANETIAN x SS & = ss 
— we 58.7 = ££ & te = Sy 

0) |} 2 | SELANDIAN es £ € € &§ s § BH 
= DANIAN - .& 5 8 :. 8 £ oS 

—— = 65.5 ——— (29) L rR eS. & S Pa > 
= MAASTRICHTIAN jet —— : Jw 4 OS 

70 — Pcs oi J fet = 
= TOG re aa —— ee 

2 7 
|——— | 

15 = ~ | : 
= CAMPANIAN & : 

C33 ~ vee 

SS if 8 S| 
= < so : | 
= eal $35 || ; ae es 

oe SANTONIAN = FES 4 | 
a | CONIACIAN | 

90 =] O 89.3 —— 

7 TURONIAN ce | 
| 4 93.5 3 

95 —} ©) 4 
—j CENOMANIAN S 
— mS 

100 —| 99.6 oe = 
4] | S yy 

105—| & = 
0 ALBIAN su 

110 be ~ 
— — = 
Sel e 112.0 - 

a M1 1 
ed APTIAN 

Zo 

255 | — 125.0 — ii 


Figure 2. Age ranges of species of Ariadnaria, Lysis, Garzasia, Calyptraea, and Crepidula discussed in text. Lines connecting the 
species indicate inferred descent based on morphologic simiarity. Lysis duplicosta gave rise to two species one of which, L. lomaensis, 
produced proto-calyptraeform Garzasia and the other, L. jalamaca, is crepiduliform. An additional crepiduliform line was founded 
by L. suciensis. No known intermediate forms conne . these Cretaceous calyptraeiform and cre piduliform g gastropods to Tertiary 
species. The earliest known Pacific Slope calyptraeid, ‘C. diegoana lacks the strong ribbing of Garzasia but crepiduliform Spiro- 
crypta pileum differs largely from L. suciensis in its a size. Time scale after Gradstein et al. (2004). 


Page 118 


THE NAUTILUS, Vol. 122, No. 3 


grained argillaceous sandstone or siltstone that constitute 
shelfal deposits that accumulated at depths near or just 
below wave base. 

Specimens are low in number and almost always in- 
complete. Protoconchs are rarely preserved, the larval 
shell is in part missing, its shape partially represented by 
its inner cast in all available specimens. Apertures are 
usually missing their anteriormost area. Adhering matrix, 
commonly consisting of well-cemented s sinidstone usually 
plugs the aperture, “thereby necessitating careful clean- 
ing. 

‘As will be discussed under “Evolutionary Implica- 
tions,” the shape of Lysis, with the exception of L. mick- 
eyi new species, approaches that of Crepidula, resulting 
from flattening of the aperture, development of a Bioad 
shelf, and lateral coiling of the spire. The crepiduliform 
shape, as well as the Calyptraea-like shape of Garzasia, 
pose problems for terminology used for describing cer- 
tain figures of specimens, as well as for describing certain 
shell dimensions. For some specimens, a full view of the 
aperture could only be depicted by having the the shell 
tipped out of the plane of its axis. The true height of 
these shells, therefore, is not shown; hence, the ‘ ‘vertical 
dimension” of the view is given instead, and perpendicu- 
lar to it is the “horizontal dimension.” In most cases, the 
abapertural view is in the plane of the shell axis; hence, 
the terms “height” and “diameter” apply. 

Abbreviations used in the text are: ANSP: Academy of 
Natural Sciences of Philadelphia; CAS: California Acad- 
emy of Sciences, San Francisco; CGS: Geological Survey 
of Canada, Ottawa: CIT: California Institute of Technol- 
ogy, Pasadena (collections now housed at LACMIP); 
CSMB: California State Mining Bureau (specimen at 
CAS): IGM: Instituto de Geologia, Universidad Nacional 
Aut6noma de México; LACMIP: Natural History Mu- 
seum of Los Angeles C ounty; SDNHM: San Diego 
Natural History Museum: UCLA: University of Califor- 
nia, Los Angeles (collections now at LACMIP); UCMP: 
University of California, Berkeley, Museum of Paleon- 
tology: USGS: United States Geological Survey, Menlo 
Park (collections now housed at UCMP): USNM: Na- 
tional Museum of Natural History, Smithsonian Institu- 
tion. 


PALEOBIOGEOGRAPHIC DISTRIBUTION OF 
ARIADNARIA, LYSIS, AND GARZASIA 


Ariadnaria ranges from late Albian to Recent with its 
earliest appearance in Pacific slope deposits of North 
America, where it is found in strata ranging from late 
Albian to Santonian age. From Japan, Kase (1990) fig- 


ured, as Tric hotropis?, a possible Ariadnaria species of 


earliest Maastrichtian age. 
Lysis ranges from the late Coniacian to late Maastri- 


chitan, and that is also its range for the Pacific slope of 


North America. The genus apparently originated in Cali- 


fornia, with five species known from the Pacific slope of 


North America. Four additional pre »bable species of Ly- 


sis are known elsewhere in the world. They are: Lysis 
capensis Rennie, 1930, from the middle Santonian to 
lower Campanian of South Africa (Kiel and Bandel, 
2003); Lysis congolensis (Brébion, 1956) from upper 
Campanian of fic Congo, Africa; Lysis africana (Cox, 
1925) from the Maastichtinn (undifferentiated) of Mo- 
zambique; and Lysis izumiensis Kase, 1990, from the 
earliest Maastrichtian of Japan. 

Rennie (1935) reported Lysis caffra Rennie, 1935, 
from the Upper Cretaceous near the eastern border of 
the Eastern Cape Province (Pondoland), southeast Af- 
rica, but its swollen naticiform shape with a very large, 
inflated body whorl and broad, non-depressed inner lip/ 
columellar region, that appears to have a thin callus, are 
features not found in Lysis. 

Garzasia ranges from late Campanian or early Maas- 
trichtian to the anid Maastrichtian and is endemic to the 
Pacific slope of North America. 


MODE OF LIFE OF LYSIS 


Lysis has been found attached to a few specimens of 
large volutid gastropods. One specimen of Longoconcha 
aumneka Saul and Squires, 2008, from the Point Loma 
Formation near Carlsbad, California has two specimens 
of crepiduliform Lysis jalamaca new species on its shell, 
near the outer lip margin (Figure 34), as well as an at- 
tachment scar on the abapertural side of the shell. Speci- 
mens of Lysis suciensis from the Chatsworth Formation 
near Chatsworth, California, although not found in situ 
as are the younger Carlsbad specimens, have aperture 
shapes that sould fit on the exterior of a specimen of 
Volutoderma Gabb, 1877. Indication of such an associa- 
tion is absent prior to middle Campanian. Whether the 
specimens of Lysis species were using dead shells as a 
substrate or had developed an association with living Vo- 
lutoderma is undetermined. 

Comparison to probable family members indicates 
that Lysis was a sedentary faculative ciliary feeder. Tri- 
chotropines, as well as calyptraeids, are ciliary feeders 
that live a sedentary life on hard substrates and are pro- 
trandrous hermaphrodites among whom brooding their 
young is common (Graham, 1954; Yonge, 1962). "These 
mode. of-life characteristics have historically been used 
to classify trichotropines as being close to calyptraeids. 


SYSTEMATIC PALEONTOLOGY 
Superfamily Calyptraeoidea Lamarck, 1S09 


Discussion: The taxonomy of calyptraeoids, as with 
most gastropods, was traditionally based on shell mor- 
phology and later modified by increasing anatomical 
knowledge. The inclusion of patelliform Capulidae 
Fleming, 1822, turbiniform Trichotropidae, and coiled 
limpet- ee aped Calyptraeidae in the superfamily Calyp- 
traeoidea (e.g¢., Thiele, 1929: Wenz, 1940), has provided 


L. R. Saul and R. L. Squires, 2008 


Page 119 


calyptraeoids a considerable morphological range and a 
somewhat complicated classification history (Bz wndel and 
Riedel, 1994). In addition to the above three families, 
Bandel and Riedel (1994), included Hipponicidae 
Troschel, 1861, in Gchonaestdee but Collin (2003: 632) 
rejected hipponicids from a close relationship with ca- 
lyptraeids. Ponder and Warén (1988) and Ponder (1998), 
equated family Capulidae with family Trichotropidae 
Gray, 1850, on anatomical grounds, reducing the in- 
cluded families to two. Capulids, as exemplified by 
Capulus Montfort, 1810, have limpet-shaped shells; tri- 
chotropids, as exemplified by Trichotropis Broderip and 
Sowerby, 1529, usually have coiled shells; and calyp- 
traeids, as exemplified by Calyptraea Lamarck, 1799, 
Crepidula Lamarck, 1799, and Crucibulum Schumacher, 
1817, have limpet-shaped shells with an internal shelf of 
variable shape. 

In her analysis of calyptraeids, Collin (2003a, 2003b) 
utilized shell characters, anatomical characters, and mo- 
lecular characters. Collin (2003a) argued that although 
the so-called echinospira larva of Capulus and of Tricho- 
tropis do not appear to be “true” echinospira, the thick- 
ened and elaborate larval shell of these two groups is 
clearly different from the simple larval shell of extant 
calyptraeids, and she (Collin, 2003a, 2003b, 2005) has 
continued to recognize the families Capulidae Fleming, 
1822, Calyptraeidae, and Pai ae Collin (2003a) 
mentioned that despite the detailed studies done on ca- 
lyptraeids, their taxonomy remains contentious and un- 
certain. Because specimens studied here show no resem- 
blance to capulids but do, in part, resemble trichotro- 
pids, and calyptraeids such as Calyptraea and Crepidula, 
we follow Collin in recognizing families Calyptraeidae 
and Trichotropidae. 


Family Trichotropidae Gray, 1850 
Subfamily Trichotropinae Gray, 1850 


Description: Small to medium sized (usually 15 to 25 
mm, but up to 40 mm in height), coiled, high- -conic to 
broad, low-conic, or nearly cap-shaped; spiral sculpture 
usually better developed than collabral sculpture and 
represented by distinct cords and/or keels; umbilicus 
broadly open to slit-like or completely closed; aperture 
variable in shape, ranging from irregularly triangular and 
broadly oval to elongate- -oval: some forms ath more or 
less attenutated canal: operculum small, horny; radula 
taenioglossate: periostracum forming combs, bristles, 
spines ‘usually in places of intersection of spiral and col- 
labral sculpture (from Egorov and Alexeyev, 1998). 


Discussion: Trichotropids provide the geologically 
oldest representatives of the three families Trichotropi- 
dae, Capulidae, and Calyptraeidae. We did not follow 
Ponder and Warén (1988), Ponder (1998), and Bouchet 
and Rocroi (2005) in placing Trichotropidae in Capul- 
idae because trichotropids, such as the high spired Ari- 
adnaria spp. discussed herein, differ distinctly from cap— 
shaped capulids. Collin (2003b) referred to trichotropids 
plus capulids as the closest outgroup to calyptraeids, sug- 


gesting that she recognized these three as separate fami- 
lies. At present the geologic record finds trichotropids 
occurring earlier than capulids and also earlier than ca- 
lyptraeids, and we suggest that based on species de- 
scribed herein, trichotropid-like gastropods gave rise to 
calyptraeids and that trichotropid specimens reported 
upon herein demonstrate a progressive development to- 
ward either a crepiduliform or a calyptraeiform shell. 
Atresius Gabb, 1869, of Early Cretaceous (Valanginien 

to Hauterivian) age is the earliest trichotropine recog- 
nized by Wenz (1940), but its sole named species A. 
liratus Gabb, 1869, is a prominent constituent of chemo- 
synthetic paleocommunities in northern California and 
probably not a trichotropid. Linpsa Stephenson, 1952, of 
Cenomanian age from the Woodbine Formation of Texas 
may be the earliest previously known trichotropine. The 

earliest reported cap- shaped sears is Capulus verus 
(B6hm, 1885) of Late Cretaceous (early Campanian) age 
from Aachen, Germany. 


Genus Ariadnaria Habe, 1961 
[= Ariadna Fischer, 1864; non Audouin, 1826]. 


Type Species: Trichotropis borealis Broderip and 
Sowerby, 1829, by monotypy; boreal Arctic circumpolar 
(Egorov and Alexeyev rev, 1998) and cool temperate seas: 
norhern North Aglenue south to Northumbria and all 
Scottish coasts (Fretter and Graham, 1962) and Massa- 
chusetts Bay (Emerson and Jacobson, 1976); the Bering 
Sea and north Pacific south to British Columbia (LACM 
collection, Forrester Island). 


Description: Shell turbiniform (oval-conic) with 
raised spire; spiral sculpture well developed and consist- 
ing of raised cords separated by interspaces of variable 
width: collabral sculpture consisting of raised growth 
lines; aperture wide; inner lip slightly concave; canal 
short and straight; umbilicus. slit- like: operculum thin; 
periostracum for ming long, closely spaced bristles on spi- 
ral ribs. 


Discussion: No prior records of Ariadnaria as a fossil 
were found by us. Ariadnaria differs from Trichotropis 
Broderip and Sowerby, 1829, by having an umbilicus. 
Turbinopsis Conrad, 1860, a turbiniform trichotropid of 
late Campanian or early Maastrichtian age from Missis- 
sippi, differs from Ariadnaria by having a wider umbili- 
cus (profound according to Conrad, 1860), a last whorl 
that is more inflated and is tabulate, and having a very 
oblique fold near the basal margin of the columella. 


Ariadnaria ainikta new species (Figures 3-4) 

Diagnosis: Medium-size Ariadnaria with sturdy shell, 
rounded whorls, strong sculpture with nine spiral cords 
widely spaced; collabral sculpture thickly foliate; umbili- 


Page 120 


THE NAUTILUS, Vol. 122, No. 3 


Figures 3-10. 
CAS 61794.00, CAS loc. 61794, height 20 mm, diameter 17 mm. 3. Apertural view. 4. Abapertural view. 5-6. Ariadnaria stibara new 
species, holotype LACMIP 13371, LACMIP loc. 23464, height 19 mm, diameter 13.5 mm. 5. Apertural view. 6. Abapertural view. 


7-8. Ariadnaria aldersoni, holotype LACMIP 13372, LACMIP loc. 


Ariadnaria species. Specimens coated with ammonium chloride. 83-4. Ariadnaria ainikta new species, holotype 


26370, height 12 mm, diameter $.5 mm. 7. Apertural view. 8. 


Abapertural view. 9-10. Ariadnaria obstricta (White, 1889), hypotype LACMIP 13373, LACMIP loc. 28717, height 21 mm, diameter 


15 mm. 9. Apertural view. 10. Abapertural view. 


cus elliptical; inner lip broad, expanded anterior and pos- 
terior to umbilicus. 


Description: Shell medium (approximately 22 mm 
height), sturdy, turbiniform, spire moderately high, ap- 
proximately 36% of total shell height: apical angle | 110°: 
most of protoconch missing, remanent low and appar- 
ently smooth; teleoconch whorls three, whorls well 
rounded and enlar ging very rapidly; suture appressed but 


appearing channe ip d: umbilicus open, deep, and ellipti- 


cal; grow th | ine prosocline; spiral sculpture consisting of 
regularly spaced and equally narrow cords; four éoide. on 
penultimate whorl: nine cords on last whorl, becoming 
stronger and more raised near umbilicus; spiral cords on 
last Se widely spaced and occasionally with spiral 
thread in interspaces, especially anterior of periphery: 
collabral sculpture consisting of f thickly foliate ridges co- 
incident with growth heel. especié ally near outer lip; 
aperture D-shaped; inner lip broad, expande -d anteriorly 
and posteriorly of umbilicus; abapertural edge of inner 
lip delineated ‘by sharp ridge; basal lip broadened. 


Holotype: CASG 61794.00, height 20 mm, diameter 
17 mm, spire height 7.5 mm. 
CASG loc. 61794 


dpe ] ocality: {=CASG loc. 
134€ 


Distribution: Basal Bald Hills Member of the Budden 
Canyon Formation, (area 4) Ono area, Shasta Co., Cali- 
fornia. 


Geologic Age: Late Albian. 


Discussion: Only the holotype is known. It evidently 
had a very foliate, thick shell. What remains is riddled 
with endobiont borings. Remnants of the protoconch are 
present, there is no clear evidence of an anterior sinus to 
the aperture, and the shell does not appear to have been 
nacreous. 

The new species most resembles the trichotropid 
Turbinopsis hilgardi Conrad, 1860 (Conrad, 1560: 289, 
pl. 46, fig. 29; Sohl, 1960: 91, pl. 10, figs. 17, 18) from the 
upper Campanian/lower Maastrichtian Ripley Formation 
of Tippah Co., Mississippi, except A. ainikta has a 
smaller umbilicus and narrower cords with — wider 
interspaces, Additionally, the inner lip of A. ainikta is 
more expanded both anterior and posterior ee the um- 
bilicus. 

Ariadnaria ainikta and A. stibara new species are simi- 
lar in that there is no ridge separating the umbilicus from 
the inner lip. Arindnaria ainikta differs from A. stibara 
by having a less elongate shell, wider pe Geran whorl, 
stronger spiral cords with much wider interspaces, a 
shorter umbilicus that is oval rather than slit-like, foliate 


L. R. Saul and R. L. Squires, 2008 


Page 12] 


collabral sculpture, and no parietal swelling on the inner 
lip. Ariadnaria ainikta differs from A. aldersoni new spe- 
cies by being larger, having much stronger spiral cords 
with much wider interspaces, a well rounded last whorl 
(not angulate), shorter umbilicus that is oval rather than 
slit-like, and no ridge separating the umbilicus from the 
inner lip. Ariadnaria ainikta differs from A. obstricta 
(White, 1SS9) by having a lower spire, wider penultimate 
whorl, round last whorl (not angulate) more spiral cords, 
foliate collabral sculpture, shorter umbilicus that is ov al 
rather than slit-like, and no ridge separating the umbili- 
cus from the inner lip. The strong spiral ribbing of A. 
ainikta resembles that of A. obstricta. 


Etymology: Ainiktos, Greek, meaning: baffling, ob- 
scure, or enigmatic. 


Ariadnaria stibara new species 
(Figures 5-6) 


Diagnosis: A medium size sturdy Ariadnaria with 
rounded whorls, medium strong sculpture with many 
spiral cords moderately closely spaced: collabral sculp- 
ture very fine and lattice-like on spire whorls; umbilicus 
chink-like: inner lip with low parietal swelling. 


Description: Shell medium small (height approxi- 
mately 20 mm), sturdy, turbiniform, somewhat elongate, 
spire high and approximately 50% of total shell height: 
apical angle approximately 67°; protoconch not present; 
teleoconch whorls four, whorls smell rounded and enlarg- 
ing rapidly: last whorl tapering anteriorly; suture ap- 
pressed, appearing channeled, and rapidly descending 
near aperture; umbilicus narrow, chink-like and present 
only adjacent to medial and posterior parts of inner lip; 
growth line prosocline, with several irregularly spaced 
crowth checks near outer lip; spiral sculpture consisting 
of numerous fine subequal rounded cords; approximately 
ten closely spaced cords on penultimate whorl with in- 
terspaces s of nearly equal width; approximately 18 cords 
on last whorl with interspaces slightly wider than cords: 

cords strongest, most widely spaced, and occasionally 
with spiral thread in interspaces on medial and anterior 


portions of last whorl; collabral sculpture consisting of 


thin, raised growth lines, forming nearly microscopic are 
tice-like pattern on spire iia aperture D-shaped, 
moderately large, oblique, narrowed at posterior end by 
low parietal swelling: inner lip broad, somewhat exca- 
vated (concave) medially and flattened ante riorly; aba- 
pertural edge of inner lip delineated by low but distinct 
ridge: b asal lip broadened. 


Holotype: Holotype LACMIP 13371, height 19 mm 
(incomplete), diameter 13.5 mm, spire height 9 mm. 
Type Locality: LACMIP loc. 23464 is also type local- 
ity of Turrilites dilleri Murphy and Rodda, 1960. 
Distribution: Bald Hills Member (unit [IV of Matsu- 


moto, 1960) of the Budden Canyon Formation, (area 4) 
Ono area, Shasta Co., California. 


Geologic Age: Middle Cenomanian. 


Discussion: Only the holotype is known, and it lacks 
the protoconch and the anterior end of the teleoconch. 
Whether or not an anterior siphonal notch was present is 
unknown, but the shape of the last whorl suggests that at 
least a small one was present. The species is assigned to 
Ariadnaria based on shell shape, but it differs from typi- 
cal Aradnaria in its very sturdy shell, the fineness of its 
sculpture, and the presence of a parietal swelling at the 
posterior end of the aperture. 

Ariadnaria stibara differs from A. ainikta by having a 
more elongate shell, narrower penultimate whorl, much 
weaker spiral cords with much narrower interspaces, 
longer and narrower umbilicus, parietal swelling on inner 
lip, and absence of foliate collabral sculpture. Ariadnaria 
stibara differs from A. aldersoni by larger size, less elon- 
gate shell, more rounded whorls that are not lax in their 
coiling, coarser spiral ribs, lattice-like very fine collabral 
sculpture on spire whorls, parietal swe lling on inner lip, 
less delineated abapertural edge of inner lip, and no fas- 
ciole-like flange aaa the umbilicus. Ariadnaria 
stibara differs from A. obstricta by having rounded 
whorls, many more spiral ribs that are much weaker and 
much more closely spaced, and a less well demarked 
abapertural edge of the inner lip. 


Etymology: Named for its sturdy shell, stibaros, 
Greek, meaning strong or sturdy. 


Ariadnaria aldersoni new species 
(Figures 7—S) 


Diagnosis: Small Ariadnaria with elongate shell and 
last whorl medially subangulate, coiling stretched axially, 
whorls bearing many very fine and closely spaced ribs, 
umbilicus chéniks like, spiral sculpture very fine, abaper- 
tural edge of inner lip well delineated and raised, umbi- 
licus bordered by fasicole-like flange. 


Description: Shell small (approximately 13.5 mm 
height, estimated); elongately turbiniform, with medially 
subangulate whorls; upper spire missing; teleoconch 
whorls enlarging rapidly; suture apparently appressed on 
earlier whorls, becoming channeled on penultimate 
whorl, rapidly desce nding: umbilicus chink-like, 
bounded by strong rounded “fasciole- like ridge; growth 
line prosocline, we aT marked with numerous growth welts 
on last whorl; spiral sculpture consisting of fine, well 
spaced ribs of unequal strength crossing irregular growth 
welts; interspaces between ribs wider than ribs; aperture 
D-shaped, narrowed posteriorly but not angled, with 
short, broad anterior sinus: outer lip thin, simple; inner 
lip moderately narrow, abapertural edge raised and well 
demarked, 


Holotype: LACMIP 13372 
plete), diameter 8.5 mm. 


- 2 WeOTne 
height 12 mm (incom 


Type Locality: LACMIP 26370 
Distribution: Panoche Formation, (area 11) Alcalde 
Hills. Fresno Co., ¢ 


Jalifornia. 


Dace 199 
Page 122 


THE NAUTILUS, Vol. 122, No. 3 


Geologic Age: Late Turonian. 


Discussion: Only the holotype is known. It is incom- 
plete, consisting only of the last two whorls, and its small 
size may iaciente that it is not mature. The elongate 
shape is a distinctive characteristic of this species. Ari- 
adnaria aldersoni apparently had a relatively high spire, 
and the coiling is lax and reminiscent of Lirpsa Ste phen- 
son, 1952. It somewhat resembles Lirpsa teres Stephen- 
son, 1952, but the new species has a narrow, chink-like 
umbilicus. 

Ariadnaria aldersoni is most similar to A. stibara and 
differs by being smaller, having a more elongate shell, 
angulated ions that are lax in their coiling: much 
weaker spiral ribs, abapertural edge of inner lip better 
delineated, fasciole-like ridge bounding the umbilicus, 
absence of lattice-like very fine eollabral sculpture on 
spire whorls, and absence of parietal swelling on inner 
lip. Ariadnaria aldersoni differs from A. dindleta by being 
smaller, having much weaker spiral cords with much nar- 
rower interspaces, angulate last whorl, longer umbilicus 
that is slit-like rather than oval, and having a ridge sepa- 
rating the umbilicus sige the inner lip. Ariadnaria al- 
dersoni differs from A. obstricta by being smaller and 
having fewer and eae weaker spiral ae with much 
narrower interspaces. 

In some respects A. aldersoni is similar to Lysis sucien- 


sis. Both have fine ribbing, a relatively high spire, rather 


lax coiling, and a somewhat slower increase of whorl 
diameter. Whereas the suture of A. obstricta and Lysis 
mickeyji is very close to or at the perimeter of the previ- 
ous whorl, in A. aldersoni and Lysis suciensis the suture 
is usually abapical to the previous whorl’s perimeter. 


Etymology: The species is named for John M. Alder- 
son who collected the holotype from Cooper Canyon. 


Ariadnaria obstricta (White, 1889) 
(Figures 9-10) 
Stomatia obstricta White, 1889: 15-19, pl. 4, figs. LO-11. 


Diagnosis: Medium size Ariadnaria with high spire 
and angulate last whorl, sculpture of a few widely spaced 
strong cords, umbilicus chink-like or covered. 


Description: Shell medium size (approximately 21.4 
mm height), elongate turbiniform; whorl profile rounded 
with slight angulation at third strong cord on last whorl; 
spire high and approximately 50% of total shell height: 
apical angle approximately 67°; protoconch missing; 
teleoconch whorls four, enlarging rapidly and last whorl 
tapering anteriorly; suture appressed, anterior to suban- 
gulate periphery; umbilicus chinklike or covered by inner 
lip expansion; bounded abaperturally by strong ridge; 
growth line prosocline; sculpture of three strong, wide ‘ly 
spaced cords on spire, five or six on body whorl; inter- 
spaces commonly with mid thread; aperture large and 
ovate with abape rtural edge raised and_ shi ply de- 
marked; outer lip appare mntly simple; inner lip somewhat 

‘xspanded and standing high along umbilical chink; basal 
lip barely drawn out into slight spout-like sinus. 


Holotype: USNM 20124. 


Type Locality: Little Cow Creek but additional speci- 
mens have not been found there. The species is abun- 
dant at some localities along South Cow Creek, Shasta 
Co., California. 


Hypotype: LACMIP 13373, height 20 mm, diameter 
15 mm from LACMIP loc. 28717. 


Distribution: Redding Formation, Bear Creek Sand- 
stone Member, especially ( area 3) along South Cow 
Creek and Bear Creek, Shasta Co.; Chico Formation, 
Musty Buck Member, (area 5) Chico Creek, Butte Co., 
California. 


Late Coniacian? to Santonian. 


Geologic Age: 


Discussion: The above description is based on 16 
specimens; all but one are from LACMIP loc. 28717. 
Most specimens are poorly preserved, and specimens 
with the shell surface preserved are difficult to find. The 
primary cords are strong and almost flange-like. 

White’s species is here assigned to Ariadnaria based 
on shell shape, sculpture, umbilicus, and presence of a 
small spout-like sinus in the aperture. This would be the 
earliest unquestioned occurrence of this genus that pre- 
viously was known only from the Recent (Wenz, 1940). 

In shape and probably sculpture (preservation makes 
comparison difficult) A. obstricta resembles illustrations 
of Trichotropis? sp. in "se (1990: 568, figs. 2.26, 2.27). 
Kase’s specimen was from the Izumi Group of Japan of 
early Maastrichtian age. 

Ariadnaria obstricta is very similar to Lysis mickeyi 
new species, but A. obstricta has more regular spiral ribs, 
a slighly higher spire, the strong spiral delimiting an um- 
hhilical chink, and a free st: nding inner lip. Aviadnane 
obstricta differs from Ariadnorin ainikta by having a 
higher spire, narrower penultimate whorl, angulate last 
whorl, fewer spiral cords, shorter umbilicus that is slit- 
like rather than oval, ridge separating the umbilicus from 
the inner lip, and an absence of foliate collabral sculp- 
ture. Ariadnaria obstricta differs from A. aldersoni by 
being larger, having fewer spiral cords that are much 
stronger and much more widely spaced, and lacking a 
fasciole-like ridge bounding the chink like umbilicus. 
Ariadnaria obstricta differs from A. stibara by having 
angulate whorls, fewer spiral ribs fee are much stronger 
sad much more widely spaced, and a more demarked 
abapertural edge of the inner lip. 


Subfamily Lysinae new subfamily 


Description: Small to moderately large (15 to 80 mm 
in height), low turbiniform to almost haliotiform, barely 
siphon ite shells with spiral ribbing. Final whorl some- 
what to greatly enlarged; spire very heets aperture large, 
nearly croiilar toe longate oval: columella and inner lip 
flattened, expanded, and de »pressed to form a shelf within 
the aperture; some with shelf that spirals into an “umbi- 
licus.” 


Discussion: The subfamily Lysinae includes Lysis 


L. R. Saul and R. L. Squires, 2008 


Page 123 


Gabb, Garzasia new genus, and probably Spirogalerus 
Finlay and Marwick, 1937. These gastropods are inter- 
mediate in form between trichotropines and calyp- 
traeids. If their characteristics were better known, some 
other species such as those discussed under Global Dis- 
tribution of Cretaceous Lysiform Gastropods, probably 

could be included here, some as Lysis or Garzasia others 
in as yet undescribed genera. 


Genus Lysis Gabb, 1864 
Tropidothais Cox, 1925: 213-214. 


Type Species: Lysis duplicosta Gabb, 1864, by mono- 


typy (Stewart, 1927: 345): Campanian of Pacific slope of 


North America. 


Description: Turbinate to crepiduliform gastropods 
with a rapidly expanding whorl diameter having the col- 
umella/inner lip flattened, expanded, and submerged to 
form a narrow to broad shelf or deck. Shell sculptured by 
spiral cords or smooth. Nonumbilicate. Aperture with 
very slight anterior siphonal notch. 


Discussion: Lysis differs from Trichotropis and Ari- 
adnaria in having the inner lip completely appressed to 
the columella. Typical Lysis (i.e., the group of L. dupli- 
costa) has a carinated whorl in the juvenile stage and a 
few moderately strong to strong spiral cords. The stron- 
ger spirals are typic Sie scaly. Ineludeds in this group is L. 
duplicosta and the following new species: Lysis mickeyi, 
L. jalamaca, and L. lomaensis. The group of Lysis sucien- 
sis has a more rounded whorl profile and more subdued, 
finer spiral sculpture; included in it is L. suciensis. 


Group of Lysis duplicosta 


The genus Lysis was proposed by Gabb (1864) for a 
low-spired, turbiniform gastropod with a depressed inner 
lip. He had only immature specimens of a single species 
(ie., the type species) and did not recognize thew simi- 
larity to genus Crepidula Lamarck, 1799. He gave no 
indication of the familial affinities of Lysis, other than 
stating the general form is like genus Stomatia Helbling, 
1779. 

During the last 127 years, Lysis has been placed in at 
least 11 inde scattered among “archaeogastropods” to 
the neogastropods. A review of this placement history is 
given here. Species that were eventually placed in Lysis 
were originally placed in Stomatia of the Stomatellidae 
Gray. 1840, by Whiteaves (1879, 1903) and White 

(1889). Stoliezka (1867-1868: 157-158) suggested ae 
Lysis should be placed near Separatista Gray, 1847, 
the Trichotropidae Gray, 1850, in the event that ae 
does not belong in either the Naticidae Guilding, 1834, 
or the Velutinidae Gray, 1840. Fischer (1885) placed 
Lysis in the Naticidae near Eunaticina Fischer, 1885. 
Tryon (1554: 112) did not hesitate to refer Lysis to the 
Muricidae Rafinesque, 1815 (as Murexia) [= Purpuradae 

Children, 1823], but on page 208 Tryon suggested a re- 
lationship to Velutina Fleming, 1821. of nes Lamellari- 
idae dOrbigny, 1841. Cossmann (1903) wrote that Lysis 
could not be a muricid but must be placed near Fossarus 


Philippi, 1841, presumably in the Fossaridae Adams, 
1860, where Stewart, 1927, Rennie (1930), Wenz, 1940, 
and Anderson (1958) also put it. Cossmann (1925) con- 
sidered Lysis to be a subgenus of Micreschara Cossmann 
(1891) in the family Vanikoroidae Gray, 1840. Dall in 
Eastman (1913) and Packard (1922) placed Lysis in the 
“Thaisiidae” (=Thaididae) Suter, 1909. Saul (1959) and 
Saul and Alderson (1981) placed Lysis in Calyptraeidae, 
and, in 1990, Saul included it in superfamily Calyptrae- 
oidea. Kase (1990) discussed previous taxonomic treat- 
ments of Lysis and suggested that, based on its inner lip 
shelf, the genus should be placed within the Calyp- 
traeidae Lamarck, 1799. Bandel and Riedel (1994) and 
Kiel and Bandel (2003) supported this placement. Cox 
(1925) named and described genus Tropidothais Cox, 
1925, which is a junior synonym of Lysis. He based 
Tropidothais on T. africana Cox, 1925, from the Maas- 
trichtian of Mozambique [formerly Portuguese East Af- 
rica] and tentatively placed his genus in ‘the Thaididae 
Jousseaume, 1588. Upon realizing its similarity to Lysis, 
Cox (in Rennie, 1935) synonomized the two genera and 
placed Lysis in the Stomatellidae. In this present paper, 
we place Lysis in the family Trichotropidae, subfamily 
Lysinae because Lysis appears to have evolved from tri- 
chotropids by expanding the columella/inner lip area 
(width and length) t to form an interior shelf suggestive of 
the calyptraeid Crepidula. 


Lysis mickeyi new species. 
(Figures 11-16) 


Diagnosis: A relatively high spired Lysis with eight or 
nine strong cords on last w horl: shelf moderately broad, 
somewhat depressed, and shallowly concave. 


Description: Shell medium small (height approxi- 
mately 20 mm), turbiniform: whorl profile overall 
rounded with medial angulation on last whorl; spire mod- 
erately low and approximately 30% of total shell height: 
apical angle approximately 90°; protoconch 1.5 whorls, 
low and sacoth: teleoconch whorls 3.5, Pee ex- 
panding and last whorl tapering anteriorly; suture abut- 
ting and becoming laxly channeled on later whorls: sculp- 
ture of strongly raised cords, either moderately closely 
spaced or widely spaced; penultimate whorl with two to 
six and last whorl with eight or nine strong spirals, with 
variable number (two to five) of finer spirals in inter- 
spaces: both cords and interspaces crossed by fine colla- 
bral ribs, producing beaded appearance; angulation mod- 
erately strong on last whorl, coincident with strongest 
spiral ‘cord: spiral cord anterior to angulation nearly same 
strength, thereby oS bicarinate appearance to 
medal part of last whorl: aperture large, oblique, barely 
notched anteriorly; outer lip simple; shelf moder rately 
broad, somewhat depressed, and shallowly concave. 


Holotype: LACMIP 13374, 
9.5 mm, spire height 5 mm. 


Paratypes: LACMIP 13375 from LACMIP loc. 23617 
and 13376 and 13377 from LACMIP loc. 10757. 


height 13 mm, diameter 


Page 124 THE NAUTILUS, Vol. 122, No. 3 


Figures 11-28. Lysis species. 11-16. Lysis mickeyi new species. L1-12. Paratype LACMIP 13375, LACMIP loc. 23617, height 
21 mm, diameter 16 mm. 11. Apertural view. 12. Right-lateral view. 13. Paratype LACMIP 13376, LACMIP loc. 10757, left- later 
view, height 9 mm, diameter 11 mm. 14. Holotype LACMIP 13374, LACMIP loc. 10757, abapertural view, height 13 mm, diameter 
9.5 mm. 15-16. Paratype LACMIP 13377, LACMIP loc. 10757, height 5 mm, diameter 9 mm. 15, Left-lateral view. 16. Apical view. 
17-28. Lysis duplicosta Gabb, 1864. 17. Plasto-lectotype of Stomatia suciensis carinifera Whiteaves, 1879, CGS 5772, height 10 mm, 
diameter 19.5 mm. 18-19. Hypotype LACMIP 13378, LACMIP loc. 24128. 18. Apertural view, vertical Aienetision 27 mm, 
horizontal dimension 21 mm. 19. Abapertural view, height 22 mm, diameter 26.5 mm, 20-21. Hypotype LACMIP 13379, LACMIP 
loc. 24340, 20. pected view, vertical dimension 28 mm, horizontal dimension 25 mm. 21. Lateral view, vertical dimension 13 mm, 
horizontal diameter 29.5 mm, 22-23. Hypotype LACMIP 13380, LACMIP loc. 24340, height 19 mm, diameter 18 mm. 22. 
Apertural view. 23. Aan’ rtural view. 24. Hypotype LACMIP 13381, LACMIP loc. 24340, abe ipertural view, vertical dimension 28 
mm, diameter 37.5 mm. 25. Hypotype LACMIP 13382, LACMIP loc. 24349, abapertural view, height 17 mm, diameter 22 mm. 26. 
Hypotype LACMIP 13383, LACMIP loc. 10095, crushed specimen, abapertural view, height 24 mm, diameter 22.5 mm. 27. 
Hypotype LACMIP 13354, LAC MIP loc. 26951, abapertural view, height 21 mm, diameter : 25 mm. 28. Hypotype LACMIP 13355, 
LACMIP loc. 24340, abapertural view, height 7.5 mm, diameter 12 mm. 


L. R. Saul and R. L. Squires, 2008 


Page 125 


Type Locality: LACMIP loc. 10757. 


Distribution: Redding Formation, Bear Creek Sand- 
stone Member of Haggart (area 3) on Bear Creek, Shasta 
Co.; Chico Formation, top of Ponderosa Way Member 
and Musty Buck Member (200 m to 650 m above the 
base of the section) (area 5) on Chico C oa Butte Co.; 
basal Tuna Canyon Formation (area 17) at head of Ga- 
rapito Creek, Santa Monica Mountains, Los Angeles Co., 
California. 


Geologic Age: Early Coniacian to Santonian. 


Discussion: The above description is based on SO 
specimens; most of these are from LACMIP locs. 10846 
and 23617. Most specimens are internal molds. Many 
show endobiont boreholes, especially on the spire 
whorls. Two specimens show the protoconch. The oldest 
specimen. is from LACMIP. loc. 26967 in the Santa 
Monica Mountains. 

Lysis mickeyi resembles Lysis suciensis (Whiteaves, 
1879) in height of spire but is closer to L. duplicosta in 
sculpture. Lysis mickeyi differs from L. duplicosta in 
having a higher spire, less expanding last whorl, much 
less e xpanded shelf, more oval aperture, and thinner spi- 
ral cords. Lysis mickeyi greatly resembles Trichotropis 
obstricta (White, 1889), but on L. mickeyi the spiral ribs 
are less regular, and its spire is slightly lower. In addition, 
L. mickeyi lacks an umbilical chink and a free- standing 
inner lip. 

Lysis mickeyi is the earliest known Lysis from any- 
where in the world. 


Etymology: Named for Mickey of Mickey's House on 
Chico Creek near the locality, LACMIP 23617, from 
which the species was first recognized. 


Lysis duplicosta Gabb, 1864 
(Figures 17-28) 


Lysis duplicosta Gabb, 1864: 138, pl. 21, fig. 9Sa—98c; Tryon, 
1883: 112, pl. 44, fig. 25-26; Cossmann, 1903: 70; Stewart, 
1927: 345-346, pl. 21, figs. 7, 7a; Anderson, 1955: 169. 

Stomatia suciensis variety carinifera Whiteaves, 1879: 128-129, 
pl. 16, fig. 5. 

Lysis oppansus White, 1889: 17, pl. 4, 
1958: 169. 

Lysis suciensis var. carinifera (Whiteaves)—Whiteaves, 1903: 
367, pl. 45, fig. 4. 

Micreschara (Lysis) duplicosta (Gabb).—Cossmann, 1925: 173 
pl. 9, figs. 6, 21. 

Lysis duplicostata Gabb.— Wenz, 1940: 880, fig. 2587 (reprint 
of Stewart, 1927); Elder and Saul, 1993: pl. 2, figs. 14-15. 

Lysis carinifera (Whiteaves) .—Anderson, 1955S: 170. 

? Lysis duplicosta carinifera (Whiteaves).—Dailey and Pope- 
noe, 1966: 6. 

Not Lysis duplicosta Gabb.—Saul and Alderson, 1951: 35—36, 


. 14-15; Anderson, 


pl. 3, figs. 3-4 [= Lysis suciensis (Whiteaves) fide Saul, 
1990]. 
Diagnosis: Variably sculptured Lysis, with many fine 


cordlets or with six to eight strong cords, including 
prominent (often flange-like) carina on periphery; col- 
umella and inner lip depressed and expanded to form 


crescentic shelf, occupying at least one third of aperture 
in larger specimens. 


Description: Shell medium size (height up to approxi- 
mately 26 mm), neritiform to oe spire mod- 

erately low, approximately 20% of total shell height; pro- 
toconch 1.5 whorls, low and smooth: teleoconch approxi- 
mately two whorls, overall rounded, enlarging very 
rapidly, and medially carinate; sculpture consisting of 
spiral ribs, gener: ally six to eight prominent ones, “but 
highly atebles in number, spacing, and strength; periph- 
ery alw ays demarked by very strong (occasionally flange- 
like) carina, located anteriorly of hodial position on spire 
whorl and located medially on last whorl; remainder of 
whorls covered by spiral sculpture, 1 ranging from numer- 
ous closely spaced fine cordlets to several widely spaced 
moderately strong bien alternate in strength) cords, with 

interspaces smooth or bearing many sordlé ts or threads; 

cords just anterior and, to a lesser degree, just posterior 
of medial carina on last whorl commonly approaching 
strength of medial carina, thereby imparting either a bi- 
carinate or tricarinate appearance to whorl profile; aper- 
ture circular with a scarcely discernable anterior canal 
notch; abapertural edge of aperture sharply demarked by 
raised edge; inner lip and columella flattened, and ex- 
panded to form shelf; shelf moderately wide (occupying 
at least one third of aperture in larger specimens), sub- 
merged within the aperture, wrapping far past suture and 
attached to inside of outer lip; medial part of outer lip 
digitate. 


Lectotype: Of Lysis duplicosta UCMP 11975, height 
10 mm, diameter 19.5 mm. Gabb (1864) did not indicate 
a holotype. Merriam (1895) recognized UCMP 11975 as 
the figured specimen, Stewart's (1927) statement that 
this is the type specimen is taken as designation of lec- 
totype. 


Paralectotype: Of Lysis duplicosta ANSP 4242. 


Syntypes: Stomatia suciensis variety carinifera 
Whiteaves, 1879, CGS 5772, a-d (Bolton, 1965). 
Whiteaves (1903: py 16, fig. 5) figured one of the five 
syntypes CGS 5772. 


Holotype: Of Lysis oppansus White, 1589, USNM 


20115. 


Hypotypes: Of Stomatia suciensis carinifera, CGS 
5939 (Whiteaves, 1903); Of Lysis duplicosta LACMIP 
13375-13385; USNM 465555, 4685586. 


Type Locality: Of Lysis duplicosta, Texas Flat, near 
Rock Corral, from a mine shaft at a de pth of 12 m [40 
ft.|, near the Placer-Sacramento Co. ae Placer Co., 
northern Califormia; Of Stomatia suciensis carinifera, Su- 
cia Island, San Juan Co., Washington. Of Lysis oppansus, 
Pentz Ranch, Butte Co., northern California. 


Distribution: Cedar District Formation, Nanaimo Ba- 
sin, (area 1) Vancouver Island area, southern British Co- 
lumbia and (area 2) Sucia Island, San Juan Co., Wash- 
ington; Chico Formation, Ten Mile Member on (area 5) 


Page 126 


THE NAUTILUS, Vol. 122, No. 3 


Chico Creek and Musty Buck Member along (area 6) 
Dry Creek, near Pentz, Butte Co., California; Chisd For- 
mation, (area 7) Granite Bay and Texas Flat, Placer Co. 
California; Pigeon Point Formation, southern sequence, 
(area 8) north of Pigeon Point, San Mateo Co., Califor- 
nia; Jalama Formation, (area 12) western Santa Ynez 
Mountains, Santa Barbara Co., California; Ladd Forma- 
tion, uppermost Holz Shale Member and Williams For- 
mation, Schulz Member, (area 18) Santa Ana Mountains, 
Orange Co., California. 


Geologic Age: 


Campanian, 


Discussion: The above description is based on 101 
specimens; most of these are from the Pentz area (LAC- 
MIP loc. 24340). At any locality, most of the specimens 
are internal molds. Some show endobiont boreholes. 
One specimen, LACMIP 13385 (Figure 28), shows the 
shape of the protoconch. 

The holotype of Lysis duplicosta is a juvenile and had 
not yet developed the wider deck of an adult. The largest 
specimens of this species thus far found are from the 
Musty Buck Member of the Chico Formation at LAC- 
MIP loc. 24340 near Pentz (area 6). The specimens are 
from a matrix-supported pebble conglomerate richly fos- 


siliferous in places. The fauna, which includes scraps of 


cypraeids, suggests warm, shallow water. Lysis duplicosta 
is very rare in the coeval Ten Mile Member on Chico 
Creek (area 5) which probably represents deeper water 
than at LACMIP 24340, and the specimen from LAC- 
MIP loc. 23639 on Chico Creek may have been trans- 
ported downslope. 

The strength of the cords, especially of the strongest 
one, varies greatly between individuals. Gabb’s specific 
name refers to a doubled appearance of each major cord, 
but Gabb’s (1564) sharp, deep channel along ihe strong 
ribs results from the wearing or breaking off of the scales 
on the ribs. 

Dailey and Popenoe (1966: fig. 3) listed Lysis dupli- 
costa carinife ra from the Jalama bres at LACMIP 
loc. 24128. This somewhat distorted specimen (Figures 
18, 19) has some stronger ribs as in L. duplicosta, but 


represent a strong ribbed variant of L. jalamaca. If 


L. duplicosta, it is the ‘geologic ally youngest specimen of 
fxs spe cles. 

Rennie (1930: 1935) described two species of Lysis 
from the Umzamba Saat of the ola Cape 


Province (formerly Pondoland), South Africa, one of 


which Lysis capensis Rennie, 1930, is very similar to L. 


duplicosta. According to Klinger and Kennedy (1980), 
the lower Umzamba Formation at its type locality is lat- 
est Santonian or earliest Campanian in age and, there- 
fore, similar in age to the Chico Formation near Pentz, 
Butte Co., California, where L. duplicosta is common. 
Rennie apparently had only two specimens of L. capensis 
which he said had six stout, sharp, spiral ribs. His holo- 
type is small, similar in size to the holotype of L. dupli- 
costa, and both apparently have a relatively narrower 
deck than is found in large specimens of L. duplicosta. 

Kase’s (1990) report of L. duplicosta in the Chats- 
worth Formation is based on a misidentification of L. 
suciensis in Saul and Alderson (1981). 


Lysis jalamaca new species 

(Figures 29-36) 

Lysis duplicosta Gabb.—Dailey and Popenoe, 1966: 6. 
Not Lysis duplicosta Gabb, 1864. 


Diagnosis: A Lysis with sculpture of fine spiral 
cordlets of alternating strength and within the aperture 
the shelf wrapping considerably past suture. 


Description: Shell medium size (height up to 26 mm), 
crepiduliform, with rapidly expanding whorls; spire low, 
and approximately 30% of total shell height; protoconch 
1.5 whorls, low and smooth; teleoconch approximately 
two whorls, well rounded; whorls flattened adjacent to 
suture; periphery near mid-whorl height; suture abutting 


just below periphery; growth line prosocline, occasionally 


forming collabral wrinkles; sculpture consisting of nu- 
merous Closely spaced fine spiral cordlets alternating in 
strength and somewhat scaly; cordlets strongest at aud 
near periphery where two or three can be stronger be- 
coming cords and the scales bead-like, with approxi- 
mately three cordlets in interspaces; aperture ovoid, its 
abapertural margin sharply demarked and steeply de- 
scending; inner lip and columella flattened and expanded 
to form shelf; shelf narrow anteriorly becoming moder- 
ately wide medially, submerged w ithin aperture, wrap- 
ping past suture and attachec | to inside of outer lip. 


Holotype: LACMIP 13386, height approximately 35 
mm, diameter 34 mm. 


Paratypes: LACMIP 13387 (LACMIP loc. 24137) 
-13388 (LACMIP loc. 24122) and SDNHM 114595, 
114596 (SDNHM loc. 3405). 


Type Locality: LACMIP loc. 24137, Jalama Forma- 


tion. 


Figures 29-39. 


Lysis species. Specimens coated with ammonium chloride. 29-36. Lysis jalamaca new species. 29-30. Holotype 


LACMIP 13386, LACMIP loc. 24137. 29. Apertural view, vertical dimension 34 mm, horizontal dimension 30 mm. 30. Abapertural 


view, vertical dimension 27 mm, pies 34 mm. 31-32. Paratype LACMIP 13387, LACMIP loc. 24137 


_ 31. Left-lateral view, 


height 18 mm, horizontal dimension 19 mm. 32. Lateral view, vertical dimension 11.5 mm, diameter 28 mm. 33. Paratype LACMIP 
13358, LACMIP loc. 24122, abapertural view, height 11 mm, diameter 13 mm, 34. Paratypes SDNHM 114595 and 114596, SDNHM 
loc. 3405 on outer lip of volute gastropod Longoconcha eumeka Saul and Squires, 2008 (SDNHM 70974), SDNHM loc. 3405, height 
133 mm, diameter 33 mm. 35-36. Posteriormost paratype SDNHM 114596 on volute shown in previous figure. 35. Abapertural 
view, height 13 mm, diameter 20 mm. 36. Oblique apical view, vertical dimension 17 mm, diameter 20 mm, 37-39. Lysis lomaensis 


new species, holotype SDNHM 67150, SDNHM loc. 3403, height 22 mm, diameter 23 mm, 37. 
view. 39. Apical/lateral view, vertical dimension 8 mm, diameter 2 


Apertural view. 38. Abapertural 


3 mm. 


L. R. Saul and R. L. Squires, 


2008 


Page 127 


Distribution: Jalama Formation, (area 12) western 
Santa Ynez Mountains, Santa Barbara Co.; Debris Dam 

Sandstone, (area 13) Agua Caliente Canyon, San Rafael 
Mountains, Santa Barbara Co., California; Chatsworth 
Formation, upper part (area 16) at Lang Ranch, Ventura 
Co., California; Point Loma Formation, (area 19) near 
Carlsbad, San Diego Co., California; Rosario Formation, 
(area 22) Cafion San Fernando, northwestern Baja Cali- 
fornia, Mexico. 


Geologic Age: Late Campanian and early Maastrich- 
tian. 


Discussion: The above description is based on 13 
specimens. The best preservation occurs in the Point 
Loma Formaton near Carlsbad. 

Lysis jalamaca is similar to L. suciensis but has 
coarser, more scaly costae than L. suciensis, and available 
specimens of L. jalamaca are smaller and less elongate 
than are large specimens of L. suciensis. Lysis jalamaca 
is also cimiilar to L. duplicosta in having variable sculp- 
ture and in having the strongest cords on the periphery, 
but L. jalamaca is less angulate at the periphery and has 
weaker cords there. In addition, L. jalamaca ditters from 
L. duplicosta by being more elongate, having finer sculp- 
ture, and having the shelf broader posteriorly. The speci- 
men (LACMIP 13375, Figures 18, 19) from the Jalama 
Formation has ribbing similar to L. duplicosta, but its 
shelf appears to wrap farther onto the interior of the 
outer whorl than is usual for L. duplicosta. 


Etymology: The specific epithet, a name in apposi- 
tion, reflects the new species occurrence in the Jalama 
Formation, Santa Barbara Co., California. 


Lysis lomaensis new species 
(Figures 37-39) 
g 


Diagnosis: Lysis with low spire and last whorl bearing 
numerous flanged carinae separated by wide interspaces 
bearing several spiral threads. 


Description: Shell medium size (up to 17 mm height 
and 30 mm diameter, same specimen), neritiform (last 
whorl rapidly expanding); spire very low, approximately 
10 to 15% of shell height; protoconch 1.5 whorls, low and 
smooth; teleoconch 2 to 2.5 whorls, very carinated; su- 
ture abutting just anterior to periphery; sculpture con- 
sisting of up to 11 strong cords; cord at periphery and 
next three cords posterior of periphery very thin, flanged, 
and protruding (with cord at periphery protruding most); 

posteriormost part of last whorl (in vicinity of suture) 
with approximately three low scaly cordlets: up to three 
beaded to lowly spinose cordlets anterior to medial ca- 
rina at periphery; interspaces between all cords wide and 
bearing up to seven spiral threads (occasional thread can 
locally dev elop into small flanged cord); area anterior to 
medial carina can be nee | with only fine cordlets; 
aperture circular with margin sharply demarked by 
raised edge; shelf moderately wide, wrapping past suture 
and attached to inside of outer lip; posterior half of outer 

lip digitate. 


Page 125 


THE NAUTILUS, Vol. 122, No. 3 


Holotype: SDNHM 67150, height 22 mm, diameter 


23 mm. 
Paratype: SDNHM 67152 from SDNHM loc. 4071. 


Type Locality: SDNHM loc. 3403. 


Distribution: Basal San Francisquito Formation, 
(area 14) Warm Springs Mountain, Los Angeles Co.; 
Point Loma Formation, (area 19) San Diego Co., Cali- 
fornia: Rosario Formation, (area 21) five miles east of La 
Mision, northwestern Baja California, Mexico. 


Geologic Age: Latest Campanian to late Maastrich- 
tian. 


Discussion: Five specimens were studied. Preserva- 
tion is generally excellent for three specimens from the 
Point Loma Formation near Carlsbad, although the pro- 
toconch of the holotype is imperfectly preserved. The 
incomplete, crushed specimen from Warm Springs 
Mountain is from beds of latest Maastrichtian age at the 
base of the San Francisquito Formation (LACMIP loc. 


Figures 40-50. 


14310). Although the abapertural side is not available, 
the flatness of the apical whorls suggests that it is prob- 
ably L. lomaensis. 

The new species is most similar to Lysis duplicosta but 
available specimens are smaller than a large L. dupli- 
costa. Lysis lomaensis also has a lower spire, more cari- 
nate last whorl (especially posterior to the medial carina), 
and spiral threads on interspaces between carinae. 

Abaperturally, L. lomaensis is similar in shape and 
sculpture to Garzasia intermedia, but L. lomaensis has a 
lower spire and in the apertural view the shelf margin is 
less arcuate, not sigmoid, and the shelf does not spiral 
into the umbilicus. 


Group of Lysis suciensis 


The group of L. suciensis differs from that of L. dupli- 
costa in havi ‘ing a more elongate aperture and finer Scal 
sculpture. 


Lysis suciensis (Whiteaves, 1879) 
(Figures 40-50) 


Lysis suciaensis (Whiteaves, 1879), Specimens coated with ammonium chloride. 40-41. Hypotype LACMIP 
13359, LACMIP loc. 26020. 40. Apertural view, vertical dimension 44 mm, horizontal dimension 40 mm. 41. Abapertural view, 


height 25 mm, diameter 43 min, 42-43. Hypotype LACMIP 13390, LACMIP loc. LO711. 42. Abapertural view, height 1S mm, 
diameter 26.5 mm. 43. Lateral view, vertical dimension $8 mm, diameter 26.5 mm, 44. Hypotype LACMIP 13391, LACMIP loc. 
10095, lateral view, vertical dimension 5 mm, diameter 12.5 mm. 45. Hypotype LACMIP 13392, Cafon San Fernando, 32 km 
southeast of El Rosario, Baja California, Mexico, abape rtural view, height 1S mm, diameter 21 mm. 46-50. Hypotype LACMIP 
10495, LACMIP loc. 26020. 46-47. Vertical dimension 70 mm, horizontal dimension 38 mm. 46. Apertural view. 47. Abapertural 
view. 48. Right-lateral view, vertical dimension 70 mm, horizontal dimension 25 mm. 49. Slightly oblique right-lateral view, vertical 


dimension 70 mm, horizontal dimension 27 mm. 50. Lateral view, vertical dimension 25 mm, horizontal dimension 36 mm. 


L. R. Saul and R. L. Squires, 2008 


Dace 126 
Page 129 


Stomatia suciensis Whiteaves, 1S79: 128-129, pl. 16, fig. 4. 

Lysis suciensis (Whiteaves) —Whiteaves, 1903: 367, pl. 45, fig. 
3. Stecheson, 2004: 60-62, pl. 2, figs. 4-5. 

Lysis californiensis Packard, 1922: 431, pl. 37, figs. 2-3; Sta- 
dum, 1973: pl. 2, fig. 12. 

Lysis duplicosta Gabb.—Saul and Alderson, 1981: 36, pl. 3 
figs. 8-4. Not. Lysis duplicosta Gabb, 1864. 


Diagnosis: Medium to large Lysis, shell elongate with 
fine, usually wavy cordlets, sculpture obsolete on large 
specimens. 


Description: Medium to large (height up to $4 mm), 
crepiduliform, elongately exp: ae d; spire moderately 
high to low, approaching ‘horizontal coiling, and approxi- 
mately 30 to 35% of total shell height; protoconch ap- 
proximately 1 to 1.5 whorls, low and smooth; teleoconch 
approximately 1.5 to 2 whorls, whorls subcarinate to 
rounded; periphery near one-third whorl height; suture 
abutting just anterior to periphery, descending; crowth 
line slightly to moderately parasigmoid il, w ah sinus at 
posterior end of outer lip on large, smooth individuals; 
sculpture consisting of fine narrow cordlets, usually 
wavy, alternating in strength, with narrow interspaces; 
aperture elongate ovoid, its abapertural margin sharply 
demarked and steeply descending; inner lip and col- 
umella flattened and expanded to form crescentic and 
moderately wide deck submerged within aperture, pos- 
terior end attached to inside of outer lip; inner deck 
margin concavely curved. 


Syntypes: Of Stomatia suciensis CGS 5771, ad. 


Lectotype (here chosen): Of Stomatia suciensis CGS 
5771, height 66 mm. 


eon 


Paralectotypes: Of Stomatia suciensis CGS 5771a— 
CGS 5771d. 

Holotype: Of Lysis californiensis UCMP 12287. 
Paratype: Of Lysis californiensis UCMP 12288 
Type Locality: Stomatia suciensis, Sucia Island, San 
Juan Co., Washington. Of Lysis californiensis, UCMP 
loc. 2167, Santa Ana Mountains, Orange Co., California. 
Figured Specimens: LACMIP 10495, 13359, from 
LACMIP loc. 26020; LACMIP 13390 from LACMIP 


loc. LOT11; LACMIP 13391 from LACMIP loc. 10095; 
LACMIP 13392 from the Rosario Formation, (area 22) 


26 km east of coastline at elevation 200 m on west side of 


Cafion San Fernando, 15 km north of Mesa San Carlos, 
approximately 32 km. southeast of El Rosario, Baja Cali- 
fornia, Mexico. 


Distribution: Upper Cedar District Formation, (area 
1) Denman Island, Gulf Islands, Prien Columbia; 
Lower Cedar District Formation, (area 2) Sucia Island, 
San Co., Washington; Chico Serna (area 7) 
Granite Bay, Placer Co.; Jalama Formation, (area 12) 
western Santa Ynez Mountains, Santa Barbara Co.; Chats- 
worth Formation, (area 16) Bell Canyon and Dayton 
Canyon, Simi Hills, Ventura Co., California; Ladd For- 
mation. upper Holz Shale Member, (area 15) Santa Ana 


Mountains, Orange Co., California; Rosario Formation 
(area 22) at Canon San Fernando, 26 km SE : El Ro- 
sario Baja California; Valle Formation, (area 23) 10 kin 
north from Punta Abreojos, Baja California Sur, Mexico. 


Geologic Age: Late early Campanian and early Maas- 
trichtian. 


Discussion: The above description was based on LOT 
specimens. The deck of smaller specimens is rather nar- 
row and attaches to the base of the previous whorl. Small 
specimens are rounded, larger specimens more elongate. 
In larger specimens the shelf is broader and its posterior 
end attaches to the inside of the outer lip. Several speci- 
mens, especially the larger specimens, have an obtusely 
biangulate whorl abapical to the periphery. Most speci- 
mens from Sucia Island are of small size and badly 
weathered, but Whiteaves’s largest specimen (1879: 129, 
pl. 16, fig. 4), here designated as lectotype, was indicated 
by him to be “two inches and a half in length” (i.e., 66 
mm). The specimen from Bell Canyon (area 16) of Fig- 
ures 46— 50, lacking its spire and w ith its outer lip broken. 
is 75 mm high. 

The ribbing on larger specimens from the Chatsworth 
Formation in Bell and Dayton canyons, Simi Hills (area 
16) tends to become fainter toward the outer lip espe- 
cially on the posterior slope. The specimen figured (Fig- 
ures 40, 41) retains ribbing around the pe riphe ry, but the 
ribs fade posteriorly toward the aperture. Height range 
of the smoother specimens is 45 to 76 mm (incomplete) 
making these the largest specimens thus far found of 
Lysis. The largest of these specimens are larger than 
representatives of most species of Crepidula; only Gran- 
dicrepidula princeps (Conrad, 1856), attains a larger size. 


Genus Garzasia new genus 


Type Species: Garzasia diabla new species from the 
“Garzas Sand” Member of the Moreno Formation, Stani- 
slaus Co., California; mid Maastrichtian. 


Description: Very low to moderately high spired, Ca- 
lyptraca-like shell with weak to strong spiral cords, ap- 
erture very broadly expanded and forming base of shell, 
shelf attached marginally, surrounding the umbilicus, 
and spiralling into it. 


Discussion: Garzasia is most similar to Lysis but dif- 
fers from the latter in that the deck surrounds and spirals 
into an umbilicus somewhat as in Calyptraca but with a 
wider more open umbilicus. It differs from Calyptraea in 
having a sturdier shelf attached marginally as in Trochita. 
It differs from Tro thita (Figure 71) in having the shelf 
crescentic with an arcuate to slightly sigmoidal shelfal 
edge, the axis of spirialing of its shelf off center, and 
external ribbing spiral rather than protractive. 
Etymology: The genus Garzasia is named for the 
“Garzas Sand” and Garzas Creek, Stanislaus Co., Cali- 
fornia. 

Garzasia intermedia (Cooper, 1894) 

(Figures 51-54) 


Page 130 


Figures 51-58. 


Stomatia intermedia Cooper, 1894: 46, pl. 3, fig. 43 [re- 
figured in Yates, 1903: pl. 3, fig. 43]. 

Lysis intermedia (Cooper).—Anderson, 1958: 170; Coan, 
1981: 165, fig. 12 [reprint of Cooper's pl. 3]. 


Diagnosis: Large Garzasia, haliotitorm, carinate on 
periphery, sculpture elsewhere consisting of several low 
spiral cords separated by wide interspaces bearing up to 
10 spiral threads, shelf very broad, with arcuate to 
slightly sigmoidal margin. 


Description: Large size (height up to 37 mm, diam- 
eter 65 mm, same specimen); haliotiform with very rap- 
idly expanding last whorl; protoconch unknown; spire 
highly variable in elevation, ranging from 50 to 85% of 
total shell height; suture abutting below periphery; 
growth line prosocline; sculpture consisting of commonly 
five spiral cords; periphery carinate and bearing stron- 
gest cord; peripheral carina located anteriorly or medial 
position on spire whorl but medially on last whorl: adapi- 
cal to medial carina two to three medium- strong cords 
with wide interspaces bearing up to 10 spiral heeade just 
abapical to medial carina several closely spaced and scaly 
medium strong cords; aperture circular; inner lip and 
columella flattened to slightly concave and e ae d to 
form very wide crescentic shelf submerged within aper- 
ture; posterior end of shelf surrounds win biicis and spi- 
rals into it. 


THE NAUTILUS, Vol. 122, No. 3 


Species of Garzasia new genus. 51-54. Garzasia intermedia (Cooper, 1894). 51-52. Plasto-lectotype of CASG 
609, Point Loma Formation. 51. Apertural view, vertical dimension 21.5 mm. 52. Apical view, diameter 19 mm. 53-54. Hypotype 
SDSNH 67149, SDSNH loc. 3403. 53. Apertural view, vertical dimension 61.5 mm, diameter 64 mm. 54. Left-lateral view of spire 
above posterior part of aperture, height 20 mm, diameter 64 mm. 55-58. Garzasia diabla new species. 55-57. Holotype LACMIP 
13393, LACMIP loc. 22588. 55. Apertural view, vertical dimension 36 mm, diameter 37 mm. 56. Abapertural view, vertical 
dimension 34 mm, diameter 37 mm. 57. Side view of spire above posterior part of aperture, height 1S mm. diameter 37 mm, 58. 
Paratype LACMIP 13394, LACMIP loc. 26353, slightly oblique left-lateral view (partial specimen), height 29 mm, diameter 9 mm. 


Lectotype: CAS 609 (formerly CSMB 13742), height 
6.4 mm, diameter 22.6 mm (incomplete). 


Type Locality: Point Loma, San Diego Co., Califor- 


nia. 


Figured Specimen: SDSNH 67149 from SDSNH 
3403. 


Distribution: Point Loma Formation, (area 19 and 20) 
San Diego Co., California. 


Geologic Age: Latest Campanian and early Maas- 
trichtian, Baculites lomaensis zone. 


Discussion: The above description was based on six 
specimens, of which preservation is generally good to 
excellent. 

Cooper's (1894) figure 43 line drawing is actually a 
combination of three specimens. The lectotype (CASG 
609) is designated by us and is photographed here (Fig- 
ure 51) for the first time. 

Garzasia intermedia difters from G. diabla new spe- 
cies in achieving larger size, having weaker carinae with 
the medial carina being most prominent, and having spi- 
ral threads in interspaces between carinae. 

Garzasia intermedia is somewhat similar to Sigapatella 
Lesson, 1830, some species of which [e.g., Sigapatella 
novaexelandiae (Lesson, 1830)| have spiral sculpture, an 
off-centered apex, and a well-developed “false” umbili- 


L. R. Saul and R. L. Squires, 2008 


Page 13] 


cus. The umbilicus of Garzasia appears to form as the 
widely expanded, flattened columella is attached medi- 
ally around the axis of coiling forming a broad shelfal 
area. In the holotype of G. intermedia and some other 
specimens this area is an open, funnel shape, but in oth- 
ers the area is more or less filled, resulting in shelves with 
differing “umbilical” development from a moderate to 
deep depression. The shelfal edge of Garzasia differs 
from that of Sigapatella in being sigmoidal rather than 
arcuate. 

A large specimen SDNHM 67149/3403 has faint 
markings at each end of the shelf which resemble muscle 
scars. 


Garzasia diabla new species 
(Figures 55-58) 


Diagnosis: Medium sized, Calyptraca-like, with mod- 
erately high spire, having three to four equal-strength 
carinate ribs with wide interspaces. 


Description: Medium to moderately large size (up to 
30 mm in height), Calyptraeca-like shell with rapidly ex- 
panding last w vhorl: spire moderately high, approximately 
40 to 50% of total shell height: protoconch missing; te- 
leoconch approximately 1.5 whorls, carinate; pe riphery 
near one-half whorl height; suture abutting at periphery; 
sculpture consisting generally of three to four « equal- 
strength and equally spaced carinae with very wide in- 
terspaces; aperture circular with its ab yapertural margin 
sharply demarked and steeply descending; inner lip mar- 
gin slightly sigmoidal; posterior end ateched to inside of 
outer lip directly beneath suture; shelf spiraling into um- 
bilicus. 


Holotype: LACMIP 13393, height 18 mm, diameter 
37 mm. 


Paratype: LACMIP 13394 from LACMIP loc. 26353. 
Type Locality: LACMIP loc. 22558. 


Distribution: Moreno Formation, “Garzas Sand” 
Member. (area 9 and 10) Merced and Stanislaus coun- 
ties, California. 


Geologic Age: “Mid” Maastrichtian. 


Etymology: The species is named for its occurrence in 
the eastern foothills of the Diablo Range, Merced and 
Stanislaus counties, California. The specific epithet dia- 
bla is used as a name in apposition. 


Discussion: The above description is based on eight 
specimens. Preservation of the known specimens of this 
species is poor. Except for the holotype, all the speci- 
mens are internal molds. 

This species is very similar to G. intermedia from 
which it differs in having a generally lower but more 
roundly inflated, spire that is appare ently not variable in 
height. higher last whorl, much stronger sculpture, and 
the basal flange wrap into the aperture to meet the ap- 
ertural edge of the shelf. 


EVOLUTIONARY IMPLICATIONS 


The presence of both fine-ribbed and coarse-ribbed Ari- 
adnaria and Lysis suggests that species of Lysis may have 
been derived from Ariadnaria. During the e volution of 
Lysis, the inner lip broadened (as did the columella) and 
apparently moved deeper into the aperture, thereby al- 

lowing more room for the foot to grasp hard substrate. 

These changes, assumed to be a function of the rapid 
expansion of the aperture, were necessary in order that 
the aperture could accommodate a larger foot for attach- 
ment to a hard substrate. The resultant shell shape was 
crepiduliform. Modern trichotropids studied by Yonge 
(1962) in Puget Sound thrive on unstable shell beds. ay 
some Cretaceous trichotropids resembled the modern 
trichotropids in (1) being filter feeders, (2) living on firm 
substrates, and ( 4 being protandrous hermophodrites, 
they would have had ehan acteristics that allowed them to 
evolve toward the less vagile calyptraeids or crepidulids. 

The inner lip of trichotropids is homologous to the 
internal deck or shelf of calpytraeids, and the develop- 
ment of the shelf in Lysis serves as a pattern for the 
development of the flat shelf in Crepidula Lamarck, 
1799, and of the shelf in Garzasia for the spiral shelf in 
Calyptraea Lamarck, 1799. Trichotropid and crepidulid 
sculpture is predominantly spiral (i.e., in the direction of 
coiling), but some calyptraeid sculpture is radial or pro- 
tractive. 

The shell exterior of Lysis suciensis group (i.e., Lysis 
suciensis lineage) is fine-ribbed to relatively smooth; in- 
teriorly the columella and inner lip are flattened and 
broadened into a crescentic shelf which, as it lengthened 
posteriorly, came to resemble a shelf of ” Crepidula. Ex- 
cept for its large size, Lysis suciensis resembles the ear- 
liest species from the Pacific slope usually assigned to 
Crepidula; namely, ‘Crepidula’ pileum (Gabb, 1864), 
which ranges from middle Eocene to Oligocene in Cali- 
fornia to Washington. Stewart (1927) and Hoagland 
(1977) slightly extended the range of “C.’ (Spirocrypta) 
pileum by synonymizing with it the late early Eocene 
ae rypta inornata ( (Dickerson, 1916) and the middle 
to late Eocene Spirocrypta dickersoni (Weaver and 

Palmer, 1922), but Vokes (1939) considered Spirocrypta 
inornata of ‘ eee age to have “a funnel-like pro- 
cess due to the upward curving of the posterior portion 
of the septa” (Vokes, 1939, pl. 13, fig. 7) rather than the 
less apparent “umbilicus” of S. pilewm (see discussion 
below). 

Gabb (1864) originally placed ‘Crepidula’ pileum in 
the invalid genus Crypta Humphrey, 1797 (a synonym of 
Crepidula), and subgenus a ice Gabb, 1864, of 
which ‘C.’ (S.) pileum is the type species. ‘Crepidula’ (S.) 
pileum, ihich is common in the Tejon Formation of 
middle Eocene age in Live Oak Canyon, Kern Co., Cali- 
fornia, has been figured several times (e. g., Gabb, 1864: 
pl. 29, figs. 233, 233a—b; Stewart, 1927: pl. 29, figs. 2-3; 
Anderson and Hanna, 1925: pl. 13, fig. 7; Clark, 1935: pl. 
4, fig. 19; Wenz, 1940: figs. 2660a, b [reprint of Stewart]. 
A large specimen from the late Eocene is figured by 


Page 132 


THE NAUTILUS, Vol. 122, No. 3 


ae and Weaver (1963, pl. 24, fig. 11). Weaver's 
(1943: 724, pl. 71, fig. 16) ilustaton of Calyptraea 
pee (Conrad) is a lapsus and is a posterior-end-up, 
apertural view of ‘Crepidula’ pileum. On page 356 he 
correctly lists the figure as Crepidula parent Gabb’s 
(1864: pl. 29, fig. 233a) and Stewart’s (1927, pl. 29, fig. 3) 
show the shelf. ren s figure is a fascimile and Stewart's 
is a photograph that has een reproduced in other dis- 
cussions of Spirocrypta (e.g., Wenz, 1940: 903, fig. 
2660a). Gabb’s and Stewart's figures are based on lecto- 
type ANSP 4221, but, unfortunately, the shelf of this 
specimen is broken. Both figures create the false impres- 
sion that there is a sinus near both ends of the shelf and 
that the middle part protrudes and is concave. An addi- 
tional representative specimen (hypotype LACMIP 
13395) was cleaned by the senior author and is illustrated 
in Figures 59-61. Its shelf (Figure 59), which is unbro- 
ken, is slightly sigmoidal and long on the left side (or 
anterior end) ands shorter on the right side (or posterior 
end). There is also a slight conv exity of the somewhat 
sinuous shelf as it approaches the posterior end of the 
aperture and the shelf margin spirals over slightly to form 
an indication of an aaniheue: In addition, the shelf is also 
narrower and closer to the shell margin on the left/ 


Figures 59-73. 
(Gabb, 1864), hypotype LACMIP 13395, LACMIP loc. 22 
Abapertural view. 61. Lateral view. 62-64. ‘Calyptraea’ diegoana (Conrad, 1855), hypotype LACMIP 13458, LACMIP loc. 22340. 
62. Apertural view, vertical dimension 29 mm, diameter 31 mm. 63. Apical view, diameter 31 mm. 64. Right lateral view, height 13 
mim, diameter 31 mm, 65-67. ‘Calyptraca’ aperta Solander in Brander, 1766, hypotype LACMIP 13396, LACMIP loc. 7333. 65. 
Apertural view, vertical dimension 11 mm, diameter 11 mm, 66. Apical view, diameter 11 mm. 67. Right-lateral view, height 7 mm, 
diameter 11 mm. 68-70. Calyptraca chinensis (Linnaeus, 1758), Recent, type-species, hypotype LACM 161651, Cherbourg 


(Manche 


anterior and noticeably farther from the shell margin 
(deeper into the aperture) on the right/posterior. The 
shelf of °C. pileum thus, as noted by Gabb (1864), spirals 
inward toward the apex. Gabb’s figure (pl. 29, fig. 233b) 
in part illustrates this, as does Figure 59. Although Stew- 
art (1927) synonomized Spirocrypta with Crepidula, 
Gabb’s description of Spirocrypta recognizes this very 
important characteristic, which helps to distinguish Spi- 
rocrypta from Crepidula, In modern Crepidula forni- 
cata, the shelf does not spiral into the whorl apex. 

Aperture/shelf features of Spirocrypta pileum and S. 
inornata resemble those of the early Paleocene Spirogal- 
erus lamellaria Finlay and Marwick, 1937, from New 
Zealand, in that the shelf of $. lamellaria is also narrower 
and closer to the shell margin on the left/anterior and 
noticeably farther from the shell margin (deeper into the 
aperture) on the right/posterior. 

Our proposed evolutionary relationship between tri- 
chotropines and calyptraeids differs from any previous 
author's proposal. Hoagland (1977), for example, in her 
study of Crepidula Lamarck, 1799, rejected trichotropids 
as direct ancestors of calytraeids and crepidulids. Hoag- 
land (1977) ee that although Trichotropis Broderip 
and Sowerby, 1829, and Crepidula had a common an- 


Fossil Calyptraeidae. 59-64. Comparative Eocene Crepidula and Calyptraea species. 59-61. Spirocrypta pileum 
356, vertical dimension 3 mm, diameter 9 mm. 59. Apertural view. 60. 


), Normandie, France. 68. Apertural view. vertical dimension L5 mm, diameter 14.5 mm, 69. Apical view, diameter 14.5 


min. 70. Right-lateral view, height 6 mm, diameter 14.5 mm, 71-73. Trochita trochiformis (Born, 1778), Recent, type-species, 
LACM loc, lot 75-41, Purmalin, west of Isla Teleon, Gulf of Corcovado, Chiloe Province, Chile — intertidal. 71, Apertural view, vertical 
dimension 40 mm, diameter 38 mm. 72. Apical view, diameter 38 mm. 73. Right-lateral view, height 18 mm. 


L. R. Saul and R. L. Squires, 2008 


Page 133 


cestor, Trichotropis has no direct relationship to Crep- 
idula because Trichotropis is boreal, living in cold and 
deep water and has rough sculpture, whereas Crepidula 
had its origin in the Panamic region and is relatively 
smooth sculpture od. Her implications that Trichotropis is 
a deep, cold-water gastropod and that Crepidula is a 
shallow, warm-water gastropod are misleading. Modern 
Trichotropis cancellata (Hinds, 1843) is intertidal in 


Alaska, British Columbia, and Washington to depths of 


104 m off Alaska, 165 m off British Columbia, and 274 m 
off Washington (LACM Malacology collection). In the 
southern part of T. cancellata’s range, it is in cool- 
temperate, not boreal waters. The question is, however, 
not where trichotropids live now but where they were 
during the Late Cretaceous, when there were no polar 
ice caps and the subtropical and temperate zones were 
much wider. Trichotropids and calyptraeid-like gastro- 
pods co-existed during the Late Cretaceous in the study 
area, and both forms lived in relativ ely warm, shallow 


waters. Although the Coniacian-Santonian faunas of 


northern California are noticeably cooler than Turonian 
faunas (Saul and Squires, 2008) of that area, they would 
have been temperate. Hoagland (1977) mentioned that 
anatomical features of T) ichotropis suggest affinities to 
calyptraeids, but she believed that similarities of life his- 
tory and niche between Trichotropis and Crepidula are 
convergent. She suggested that Crepidula is derived 
from some form of * ‘calyptraeid stock” that, in turn, was 
derived from Trochita Schumacher, 1817, reported by 
Wenz (1940) to range from Eocene to Recent. She did 
not provide any geologic time parameters as to when 
these derivations took place, but indicated that Shimer 
and Shrock (1959) recorded the first calyptraeids as 
“lower Cretaceous.” The “lower” seems to be a minor 
lapsus: all printings of Shimer and Shrock from first 1944 
to last 1989 list range of Calyptraca and Crepidula as “U. 
Cret—Recent” and Crucibulum as “Tert.—Recent.” Wenz 
(1940: 902), however, questionably included the peculiar 
looking genus Galericulus Seeley, 1861 from the Upper 
Greensand (Lower Cretaceous upper Albian) of England 
in calyptraeids. It does not resemble any calyptraeid we 
have studied. Seeley (1861) named Cre pidula cooksoniae 
also from the Upper Greensand, but Hoagland (1977: 
395) found it “unconvincing” as a calyptraeid. 

The Campanian to Midas mchivan Damesia Holzaptel, 
1885. of Europe and Tennessee, has been regarded as a 
calyptraeid by some workers (e.g., Sohl, 1960), but Dock- 

ery (1993) assigned Damesia to the ese elloids. 
Bandel and Riedel (1994) reviewed P lacement and 
content of Calytraeoidea, and in comparison to Hoag- 
land’s (1977) study, they arrived at a different faints 
content and different relationships between the families. 
Their Calyptraeoidea consisted of two families: Calyp- 
traeidae containing genera Calyptraea, Crucibulum, and 
Crepidula; and Hipponicidae containing Cheilea, Hip- 
ponix, Neojanaca, and Thylacus: but both families ex- 
cluded trichotropids. According to Bandel and Riedel 
(1994). all genera in Calyptraeidae lack a probosis and 


are obligatory filter feeders. However, in Hawaii Ulbrick 
(1969) reported algae grazing, in addition to filter feed- 
ing by Crucibulum spinosum Sowerby, 1824. 

‘Cretaceous trichotropines probably had several char- 
aucun that lysines would have inherited from them: 

) filter fee ding, 2) living epifaunally, probably on a hard 
nas 3) brooding egg capsules from which plank- 
totrophic larvae hatch, and 4) being protandrous her- 
maphrodites. Characteristics the lysines would pass on to 
calyptraeids. Lysines’ first recognizable difference from 
the trichotropines i is the increase in e Xpansion rate of the 
whorl, especially of the last whorl, resulting in an en- 
larged aperture in which the columella is broadened. The 
earliest bro: idening and enlargement were moderate and 
only become striking as geologically younger species be- 
gan to have a very disinetne morphology. Until con- 
nected with its ancestry, Lysis was difficult to pa 
The shell also became more flattened and limpet-like, « 
the aperture enlarged. 

Hoagland ( 1977) ie Crepidula hochstetteriana 
Wilckens (1922: 5-6, pl. 1, figs. 9a, b) as being the ear- 
liest Crepidula because it was reported from the calcar- 
eous conglomerate stratigraphic unit of the Lower 
“Amuri Group” in the “Amuri Bluff area, northeastern 
South Island, New Zealand (Wilckens, 1922: Warren and 
Speden, 1978). Modern spelling of “Amuri” is Haumuri, 
and the Cretaceous strata are referred to the Mata Se- 
ries. Woods (1917) reported that these strata at Haumuri 

Bluff contain the bivalves Inoceramus australis Woods, 
1917 and Inoceramus pacificus Woods, 1917, which ac- 
cording to Wellman (1959) are limited to the Piripauan 
Stage of latest Coniacian to Santonian age. 

Warren and Speden (1978) noted “problems” with the 
early collections from this area but nevertheless, listed 
ocala eg hochstetteriana (Wilkens, 1922) from the 

Campanian Okarahia Sandstone of the Mata Series. De- 
scribed conditions of collecting suggested a strong pos- 
sibility of co-mingling of material from different strata 
and that the only known specimens of M. hochstetteriana 
might not be from the Mata Series. In search of the type 
specimen, we contacted three extraodinarily helpful 
New Zealanders: A. G. Beu, A. Grebneff, and i D. Stil- 
well. Their email communications (2006) indicated that 
M. hochstetteriana is not from the Okarahia Sandstone, 
and is not of Cretaceous age. Beu found the type speci- 
men (GNS TM2608) in the New Zealand Geological 
Survey collections. Fortunately, the type specimen was in 
a large enough block of matrix to take a sample for mi- 
crofossil dating, He enlisted the aid of G. Wilson who 
dated the dinoflagellates as late Oligocene, at the oldest. 
Ian Raine, who looked at the rich spore-pollen assem- 
blage also from the sample, found Acacia pollen, which is 
not known earlier than Miocene in New Zealand. Mio- 
cene strata crop out above the Mata Series, and some of 
the original material sent to Wilkens was apparently from 
beach boulders derived from younger strata ove -rlying the 
Cretaceous Mata Series. This Neogene age for Wile i ns’ 
species is much more likely conside ‘ring that the shelf 


Page 134 


THE NAUTILUS, Vol. 122, No. 3 


covers nearly one-half of the aperture (Hoagland, 1977 
380). 

Hoagland published more papers on Crepidula and, in 
1986, she revised several items of her 1977 paper but did 
not design a new evolutionary pathway. We differ from 
Hoagland (1977) in that we believe it is difficult to go 
from the centrally oriented apex plan of Trochita (Figure 
71) or Calyptraea (Figure 68) with its spiraling shelf and 
get to Crepidula with its eccentric spire and apparently 
unspiraled deck. Whether or not Lysis species are an- 
cestral to any modern Crepidula, younger species of Ly- 


sis achieved a crepiduloid form with a respectable shelf 


by wrapping the posterior end of the shelf onto the inside 
of the outer whorl. Furthermore, from this, the calyp- 
traeid form appears to have evolved with the develop- 
ment of an “umbilicus” in Garzasia intermedia making 
Cooper's specific name remarkably prescient. 

At about the Campanian/Maastrichtian boundary, de- 
velopment of the spiral shelf of Garzasia resulted in a 


shelf that appears to be a pattern for development of 


shelves in Calyptraea and perhaps Crucibulum. The very 
broad, depressed spiraling inner lip of Garzasia appears 


to provide a likely pattern for development of the shelf of 


Calyptraea or Crucibulum, not for the more decklike 
shelf of Crepidula. 

The earliest reported Calyptraea on the Pacific slope 
is ‘Calyptraea’ diegoana (Conrad, 1855) which ranges 
from middle Paleocene to Oligocene and occurs from 
California to Washington and easternmost Russia 
(Squires, 1987). A representative specimen from the Te- 
jon Formation Eocene is shown in Figures 62-64. Figure 
62 me ays the shelf of this species, which is similar to 
that of ‘Calyptraea@ aperta Solander in Brander, 1766 
from the Eocene of Europe (Figures 65-67). Both of 
these species have often been referred to Trochita Schu- 
macher, 1517 (type species Turbo trochiformis Born, 

778), but their shelves (Figures 62 and 65) spiral from 
an off-center position roughly a quarter of the diameter 
in from the aperture edge. A so-called pseudoumbilicus 
and 65 
could result from reduction of an umbilicus such as that 
of Garzasia. The shelves of ‘C. diegoana and ‘C. aperta 
are narrower than shelves of Garzasia and expand across 
about one third of the aperture, They differ distinctly 
from that of Trochita (Figures 71-73) ), which has a sturdy 
spiraling internal shelf extending from the centered axis 
of coiling to the outer shell margin, giving the impression 
of dividing the circular aperture in half. 

‘Calyptraea’ diegoana does not appear to be a direct 
descendent of Garzasia. Its similarity to ‘C.’ aperta sug- 
gests an ancestor in the Old World Tethyan Sea and, like 
many of the other Early Cenozoic mollusks. ( (Squires, 
1987, 2003), it probably arrived onto the Pacific slope via 


at the upper end of the shelves in Fi igures 62 


a circum-equatorial current. These species ‘C.’ diegoana 
and ‘C. 
Calyptraea chinensis (Linnaeus, 1758), the type species 
of Calyptraea. In C. chinensis, the thin fragile shelf (Fig- 
ure 68 occupies about a quarter of the apertural circle. 


aperta do not have shelves similar to that of 


It arises from the apex with a folded-over edge that forms 
the umbilicus. At the open end of the umbilicus, the 
shelf edge abruptly veers counter to coiling direction and 
approaches the apertural margin at an acute angle. The 
total shelfal edge is sickle- shaped with a ligne handle 
(the umbilical edge) and a long curved blade (the outer 
margin of the shelf). This leaves a deep notch between 
the attachment of the shelf to the shell and a delicate, 
lobate shelf. 

Trochita is present in modern eastern Pacific slope 
faunas from Mazatlan, Mexico to Valpariso, central 
Chile. It occurs in lower to middle Miocene strata as far 
north as the La Panza Range, San Luis Obispo Co., is 
present in the Kern River section, Kern Co. (Addicott, 
1970) and the Topanga Formation, Santa Monica Moun- 
tains, Los Angeles Co., California. It has a thick shell 
with protractive ribbing and, as mentioned above, a 
sturdy shelf. 


GLOBAL DISTRIBUTION OF CRETACEOUS 
LYSIFORM GASTROPODS 


Both ‘Crepidula’ and ‘Calyptraea’ have been reported 
from Campanian and Maastrichtian age strata from 
throughout the world. Preservation of most specimens 
makes identification of them problematical. Some of 
these gastropods have proven to belong to other families 
or not to be of Cretaceous age. Others need verification. 
Our tally of occurrences is doubtless incomplete. 
Although stated above as “throughout the world,” 
these gastropods are recovered from areas that were 
probably temperate to tropical. Verified lysines are all 
younger than Turonian and older than Eocene. Classing 
Spirogalae rus as a lysine ides the only Paleosene 
record of this subfamily. Late Cretaceous occurrences 
suggest that the calyptraeids developed in several geo- 
graphic places from widely distributed trichotropids 
evolving into Lysinae. An example of this is the evolving 
shape of Lysis. Although on the Pacific Slope calyptraei- 
form calyptraeoids dev ‘eloped from a crepiduliform ca- 
lyptraeoid, the calyptraeiform has been recognized more 
widely geographically, but not earlier than Coniacian. 
Cretaceous calyptraeiform calyptraeoids are more 
widely recognized geographically than are crepiduliform 
calyptraeoids. Europe has calyptraeiform occurrences 
but no crepiduliform occurrences. Africa has calyptraei- 
form occurrences in the north and crepiduliform occur- 
rences in the south. African occurrences are close in time 
to those of the Pacific Slope. North America (exclusive of 
the Pacific Slope) has a very few reports along the Gulf 
Coast of calyptraeiform specimens. The Pacific : Slope has 
both calyptraeiform Garzasia and crepiduliform Lysis. 
South America has calyptraeiform species. Japan some- 
what mirrors the Pacific Slope. Its known lysine is of 
early Maastrichtian age, but the additional presence of a 
trichotropid ( (Kase, 1990) similar to A. obstricta suggests 
that lysines were developing there roughly synchronously 
with those along the Pacific Slope. New Zealand has no 


L. R. Saul and R. L. Squires, 2008 


105 85 80 75 70 65 60 55 350 302520 15 5 
CRETACEOUS PLA ENE gies nl 


ALBIAN CENO- |TUR-]| CO- CAMPANIAN oe PALEOCENE E 5 OLIGOCENE MIOCENE PLIOJ. 
IMANIANT ON. [NIA EN 
| 


? _ ‘Calyptraea’ a Calyptraea 
5 7 


RAEIFORM 
| Sigapatella 
Trochita. —___________» 
6 | | 


4 


G apa ae | 


| apiaaaie ts ane | 
ee ? Grandicrepidula 


7 _ Lysis >» CREPIDULIFORM/ —_———> ? cepidula 


Spirocrypta 


/ Maoricrypta 


ZZ muscle scars i 


=> = = 


1 
TRICHOTROPIDAE 
Trichotropinae 
Ariadnaria 


Figure 74. Generalized proposed evolutionary trends of calyptraeoids. Time scale after Gradstein et al., 2004. Calyptraeids and 
crepidulids probably evolved from several trichotropids. Lysis or Lysis-like fossils of Coniacian to Maastrichtian age have been 
described from California, and from Campanian to Maastrichtian age from southern Africa and Japan. 1. Turriculate gastropods 
having gill capable of filter feeding, sedentary adult life on hard eubaeats: Probably capable of copulation, protandrism, and brooding 
of young. 2. Broadening of columella and inner lip. Enlarging final whorl. 3. Atti achment of posterior shelf end to interior of outer 
lip to develop crescentic shelf. 4. Some species developed * "ana bilions” in spiraling shelf as in Garzasia. 5. “Umbilicus” closed or nearly 
so in Eocene ‘Crepidula’ and moving toward more central position in more Grcdler base. 6. Broad shelf, spiraling from near center 
of base. Shelf edge nearly straight. 7. Spiral shell with low spire, round base. Shelf edge extremely sigmoidal. * = Pleistocene. Stages 
abbreviated are Turonian, Coniacian, Santonian, Mansel ies, 


known Cretaceous lysines but does have the youngest 
known calyptraeiform lysine. 


Maastrichtian southwestern France, has a spiraling shelf 
similar to that of Eocene ‘Calyptraea’ aperta except that 


EUROPE: Crepidula mytiloidea Bellardi and Mich- 
elotti, 1840 from Villavernia near Tortona Italy was listed 
among Nomina Dubia by Hoagland (1977) as being from 
the Cretaceous. Bellardi and Miche lotti were describing 
a Tertiary fauna. Because the specimen and ‘llinsivaaone 
are so small, Hoagland could not verify that this species 
was a Crepidula. 

Calyptraea cretacea (dOrbigny, 1842: 390, pl. 234, 
figs. 1-3) [Tnfindibubur| was examined by Kollmann 

2005) who determined that d’Orbigny’s material was 
from Campanian of C harente-Maritime, southern 
France, and that the specimen identified as C. cretacea 
by Delpey (1942: 165, fig. 1) 
southwestern France. Poor preservation of dOrbigny’s 
type caused Kollmann (2005: 172, pl. 18, fig. 18) to fetes 
to it as “Calyptraea s. lato cretacea (V Orbigny, 1843), 
species dubia.” aie s illustration shows no shelf. Koll- 
mann considered both specimens to be only Calyptraea 
sensu lato. These specimens expand their w horl diameter 
much less rapidly and have much higher spires than 
specimens of Lysis and Garzasia from the Pacific Slope 
of North America. 

Calyptraea depressa Delpey, 1942 (p. 165, fig. 2) from 


was from Maastrichtian of 


its shelfal margin is concavely arcuate between rim and 
whorl center. 

Kollmann and Odin (2001: 446, pl. 1, figs. 15-19) re- 
corded Calyptraea sp. of Maastrichtian age from south- 
western France, but the preservation of the specimens 
appears to be too poor to allow positive generic identi- 
fication. 

AFRICA: As figured, ‘Calyptraea’ bouéi (Pervin- 
quiére, 1912: 10-11, pl. 1, figs. 7-11) from the Maas- 
trichtian of Tunisia externally resembles some forms as- 
signed to ‘C. aperta. The shape of the shelf of ‘C. bouéi 
shown in figure 9 is difficult to determine, but appears to 
have a straight edge and probably resembles that of ‘C. 
aperta. 

Galerus libyca Quaas, 1902, was described from upper 
Maastrichtian/ possibly D anian strata (Exogyra overwegi 
beds) from the Ammonite Hills, Great Sand Sea, Egypt 

(Quaas, 1902: 238, pl. 25, figs. 26-29). It has been re- 
ported also from the Congo (Dartevelle and ae, 
1956: 29-30, pl. 1, figs. 9-10), and from Libya, Egypt, 
Congo, and Madagascar (Bandel and Riedel, 1994: 339- 
340, pl. 7, figs. 2-3). Bandel and Riedel reported Ou: 1as’ 
original specimens lost in World War II and figure od the 


Page 136 


THE NAUTILUS, Vol. 122, No. 3 


exterior of a subsequently collected specimen of ‘Calyp- 
traea’ libyca which resembles ‘C. aperta. They did not 
figure the shelf, but described it too briefly as “a flat 
spiral shelf like that of modern Calyptraca.” 

A Crepidula chain was reported by Brébion (1956) in 
describing Crepidula congolensis Brébion, 1956, from 
the upper Campanian of the Congo, Africa. This African 
species resembles a Lysis more than it does a modern 
Crepidula, in that C. congolensis has coiling similar to 
Lysis and a depressed inner lip that barely wraps onto 
the labral side of the Ga Lysis? congole nsis (Bré- 
bion, 1956: 89, fig. 1; pl. 1, fig. 7a, 7b) is most similar to 
L. jalamaca in shape and sculpture but appears to have 
much finer ribbing than L. jalamaca. 

‘Calyptraea’ primogenita Kiel and Bandel (2003: 460, 
fig. 4.14-4.16) and Lysis capensis (Rennie, 1930) illus- 
trated by Kiel and Bandel (2003: 460, fig. 6.1-6.2) are 
from the upper Santonian/lower Campanian, Umzamba 
Formation. ‘Calyptraea’ primogenita was described trom 
a single worn and broken specimen. Its ribbing (except 
on last quarter-whorl) is protractive as in Trochita, and it 
has a thick shell as does Trochita. Its whorl shape is more 


trochiform than in Trochita or Calyptraea, it consists of 


more whorls than a Trochita or a Calyptraca, the last 
whorl lacks the notable enlargement of a Trochita or a 
Calyptraea, and unlike Trochita or Calyptraea, it has a 
small open umbilicus and “the columellar lip bears a 
strong plate” (Kiel and Bandel, 2003). 

Lysis capensis is very similar in shape and sculpture to 
L. duplicosta, but its inner lip seems narrower and more 
similar to that of L. mickeyi. Lysis duplicosta is generally 
lower spired and has a broader inner lip/columell: 1. 

Rennie (1945: 50, 116, pl. 3, fig. 10) figured a Calyp- 
traea sp. from the Upper Cretaceous Senonian of An- 
gola, Africa. 
though the shape and angle of suture in the figure re- 
semble Calyptraea, the base and aperture are not 
illustrated, thereby making generic assignment indeter- 
minate. 

NORTH AMERICA (exclusive of the Pacific Slope): 
Crucibulum? sp. of Sohl (1960: pl. LO, fig. 2 
mature, incomplete specimen from the upper part of the 
Ripley Formation (Maastrichtian) in Mississippi. Sohl in- 
dicated that its incomplete shell suggested a close rela- 
tionship to Crucibulum, and that it definitely appeared to 
belong in the Calyptroidea. The specimen is too incom- 
plete to determine its genus, its similarity to Crucibulum 
could be a result of the way it is broken, but its spire does 
suggest Calyptraeoidea. 

Soh ( (1960) classed Thylac us cretaceous Conrad, 1860 
in Capulidae, but Dockery (1993) moved it to Calyp- 
traeidae and Bandel and Riedel (1994) included Con- 
rad’s species in Hipponicidae. 


A more precise age is unknown and, al- 


1) is an im- 


It apparently lived at- 
tached to the columella within the aperture of empty 
gastropod shells. It differs from Lysis and Garzasia in the 
way Thylacus muscles were attached. At the end of ju- 
venile coiling and beginning of expansion of the last 
whorl, Thylacus deposited left and right prongs instead 


of modifying the inner lip/columella into a shelf as in 
Lysis. Sohl (1960: pl. 10, fig. 4) and (Dockery (1993: 
pl.18, figs. 1 and 4) provided good illustrations of the 
early bce which does seem more hipponicid 
than calyptraeid. When developed to maturity a horse- 
shoe-shaped muscle scar resulted. Dockery’ s (1993) 
specimens were from the Coffee Sand of Campanian age 
and Sohl’s (1960) from the Ripley Formation of Maas- 
trichtian age. 

SOUTH AMERICA: ‘Calyptraea’ aperta (Solander in 
Brander, 1766), a European Eocene species (see Figures 
65-67 for a representative late Eocene specimen from 
the Paris Basin, France), was identified by Olsson (1944: 
245-249, pl. 9, figs. LO-13) from northern Peru. Olsson’s 
placement of f the Tortuga fossil beds below his Radiolite 

sandstones with Baculites suggests a Maastrichtian age. 
As he provided only exterior views and no description of 
the shelf, this identification needs further verification. 
Specimens identified as ‘Calyptraea’ aperta from Europe 
and the American Gulf Coast range widely as to whorl 
height and sculpture which varies from smooth to spi- 
nose. The Peruvian specimens increase in diameter at a 
slower rate and they have more strongly impressed su- 
tures than Eocene specimens, suggesting that the Peru- 
vian specimens are probably not °C. aperta. 

Calyptraea laevis (Philippi, 1887) from Maastrichtian 
of central Chile was described as Trochita laevis Philippi 
(1887: 92, pl. 11, fig. 3: referred to Galeropsis by Wilck- 
ens (1904: 195-196, pl. 17, figs. 9a, b) because of its high 
spire, and to Calyptraeca (Trochita) by Bandel and Stin- 
nesbeck (2000: 763-764, pl. 1, fig. C). Bandel and Stin- 
nesbeck provided a view of the exterior only, and de- 
scribed the shelf as “a flat concave shell like that of mod- 
erm Calyptraca.” Although “flat concave shell” might 
partially describe a Troc hita shelf, it is not an acurate 
description of the shelf in Cann ytraea chinensis (Figure 
6S). Wilckens (1907: 13, pL 3, fig. 6) also reported Ca- 
lyptraca aff. laevis Philippi of Maastrichtian age from 
southern Patagonia, but the preservation of the specimen 
appears to be too poor to allow positive generic identi- 
fication. 

Calyptraea pileolus COrbigny, 141 was indicated by 
Hoagland (1977: 354) to have been recorded from Lower 
Cretaceous strata of Argentina by von thering (1907), but 
the species was listed By Fe ruclio ( 1937: 187) foi the 
Patagoniano (Tertiary) ) of Lago Argentino, Argentina. 

JAPAN: As noted earlier, Lysis izumiensis Kase, 1990 
has been described from the earliest Maastrichtian of 
Japan. It resembles the group of Lysis duplicosta and 
appears from the illustrations (Kase, 1990: figs. 2.11- 
2.12) to be most similar to L. duplicosta in strength of 
sculpture and development of innerlip/columellar expan- 
sion. Kase (1990) also recorded a trichotropid similar to 
Ariadnaria obstricta from the Izumi Group in Japan. 

NEW ZEALAND: The early Paleocene (Danian) 
Spirogalaerus Finlay and Marwick. 1937, from New Zea- 
land is Lysis-like, especially as to growth line, and this 
similarity requires further study to de ‘termine whether it 


L. R. Saul and R. L. Squires, 2008 


Page 137 


results from common ancestry, conv ergent evolution, or 
parallel evolution. Finlay and Marwick (1937) assigned 
Spirogalerus to family C Calyptraeidae and based their ge- 
nus on Spirogalerus lamellaria Finlay and Marwick, 
1937, which resembles L. suciensis except that L. sucien- 
sis lacks the “pseudoumbilicus” described for Spirogal- 
erus. Finlay and Marwick (1937) and Boshier (1960) 
opined that Spirogalerus lamellaria could represent the 
evolutionary link between crepidulids and calyptraeiform 
Sigapatella Lesson, 1830. 

Classification of Spirogalerus has been inconsistent. 
Wenz (1940) made it a subgenus of Calyptraea Lamarck, 
1799. Beu and Maxwell (1 1990) made Spirogalerus a sub- 
genus of the ied Sigapatella Lesson, 1830, but 
Stilwell and Zinsmeister (1992) separated Sigapatella 
from Spirogalerus because the latter has a strongly ex- 
erted spire. Collin (2003a) noted that Sigapatella has a 
shell and anatomy ate to the eee genera Ca- 
lyptraea, Trochita Schumacher, 1817, and Zegalerus 
Finlay, 1926. Collin (2003b) considered Sigapatella to be 
monophyletic but ‘id not mention Spirogalerus. Mar- 
shall (2003) recognized Sigapatella (=Zegalerus) as a dis- 
tinct genus based on its shelf edge being broadly and 
evenly concave instead of sigmoidal. This is a very dif- 
ferent shelf than that of Calyptraea chinensis (Figure 
68). The differences between Spirogalerus and Sigapa- 
tella are similar to those found between Lysis and Gar- 
zasia, and the New Zealand forms also record an evolu- 
tionary pattern of enlargement of the last whorl and the 
aperture, with broadening of the columella/inner lip area 
into a shelf within a limpetifor m shell. Thus, these two 
genera were probably not links from actual crepidulids, 
but represent links from lysines to calyptraeids. 


ACKNOWLEDGMENTS 


We mention again the extraordinary assistance of the 
New Zealanders in documenting a Tertiary age for a 
specimen mistakenly recorded as Cretaceous. Malcolm 
G. Laird, University of Canterbury, Christchurch, New 
Zealand, and Jeffrey D. Stikhvell, Monash University, A. 
Grebneff, Otago University, Dunedin, New Zealand, and 
A. G. Beu, GNS Science, New Zealand, provided help in 
researching the age of the New Zealand Cretaceous 
units. They recruited micropaleontologists Graeme Wil- 
son, GNS Science, New Zealand, who provided an age 
range based on the diatoms, and Ian Raine, GNS Sci- 
ence, New Zealand, who recognized the Acacia pollen 
and reported its maximum age in New Zealand. Ray- 
mond Graham, Royal British Columbia Museum, Victo- 
ria, Canada provided information on the occurrence of 
Lysis duplicosta in the Gulf Islands area. Lindsey 
Groves, LACM Malacology, provided specimens of Ca- 
lyptraea chinensis and Trochita trochiformis and assis- 
tance in finding several obscure references. We also 
thank an anonymous reviewer and S. Kiel of University of 
Leeds, U.K. for i improvements in the text. 


LITERATURE CITED 


Adams, A. 1860. On some new genera and species of Mollusca 
from Japan. Annals and Magazine of Natural History, Se- 
ries 3, 5: 405-413. 

Addicott, W. O. 1970. Miocene gastropods and biostratigraphy 
of the Kern River area, California. U.S. Geological Survey, 
Professional Paper 642, 174 pp., 21 pl. 

Anderson, F. M. 1955. Upper Cretaceous of the Pacific Coast. 
Geological Society of America Memoir 71, 378 pp. 

Anderson, F. M. and G. D. Hanna. 1925. Fauna and stratigra- 
phy relations of the Tejon Eocene at the type locality in 
Kern County, California. California Academy of Sciences, 
Occasional Papers 11: 1-249. 

Audouin, V. 1826. Explication sommaire des planches de mol- 
lusques de /Egypte et de la Syria, publiées par J. C. Sav- 
igny. In: V. Audouin, Description de Egypte ... , His- 
toire Naturelle. Commission des Sciences et Arts d’ 
Egypte, Paris, 1(4): 7-56. 

Bandel K. and F. Riedel. 1994. Classification of fossil and re- 
cent Calyptraeoidea (Caenogastropoda) with a discussion 
on neomesogastropod phylogeny. Berliner Geowissen- 
schaftliche Abhandlungen Reihe E 13: 329-637. 

Bandel, k. and W. Stinnesbeck. 2000. Gastropods of the 
Quiriquina Formation (Maastrichtian) in central Chiles: 
paleobiogeographic relationships and the description of a 
few new taxa. Zentralblatt fiir Geologie und Palaontologie 
Teil 1 (Heft 7/8): 757-788. 

Bellardi, L. and G. Michelotti. 1840. Saggio orittographico sulla 
classe dei Gasteropodi fossili dei terreni Terziarii del 
Piemonte. Memorie Accademia della Scienze de Torino, 
(2) 3: 80 pp.. 8 pls. 

Beu, A. G. and P. A. Maxwell. 1990. Cenozoic Mollusca of New 
Zealand. New Zealand Geological Survey Paleontological 
Bulletin 58: 1-518. 

Bohm, J. 1855. Der Griinsand von Aachen und seine Mol- 
luskenfauna. Verhandhimgen des naturhistorischen Ver- 
eines des Preussischen Rheinlands, Westtalens und des 
Reg.-Bezirks Osnabriick 42: 1-152. 

Bolton, T. E. 1965. Catalogue of type invertebrate fossils of the 
Geological Survey of Canada. Vol. II. Geological Survey of 
Cia. Ottawa, 344 pp. 

Born, I. 1778. Index Rerum Naturalium Musei Caesarei Vin- 
dobonensis. Part I: Testacea. Vindobonae, xlii+458 pp., 
1 pl. 

Boshier, D. P. 1960. The fossil history of some New Zealand 

Calyptraeidae noua: New Zealand Journal of Ge- 
ology and Geophysics 3: 390-399. 

Bouchet, P. and J. P. Rocroi. 2005. Classification aut Nomen- 
clator of Gastropod Families. Malacologia 47 (1-2): 1-397. 

Brander, G. 1766. Fossilia Hantoniensia collecta, et in uiticaee 
Britannico deposita. London, 43 pp., 9 pls. 

Brébion, P. 1956. Gasteropoda Cretacés du Bas-Congo, recol- 
tes par H. Pierard. Annales du Musée Royal du Congo 
Belge Tervuren, Belgique, Sciences Géologiques 17: 51— 
93. 

Broderip, W. J. and G. W. Sowerby. 1829. Observations on new 
or interesting Mollusca contained, for the most part, in the 
Museum of the Zoological Society. Zoological Journal 4: 
359-379. 

Children, J. G. 1822-1824. Lamarck’s genera of shells. Quar- 
terly Journal of Science, Literature and the Arts, 14(27) 
64-86 [October 1822]; 14(28): 298-322 [January 1523]; 
15(29):; 23-52 [April 1823]; 15(30): 216-258 [July 1823]: 


Page 138 


THE NAUTILUS, Vol. 122, No. 3 


16(31): 49-79 [October 1823]; 16(32): 241-246 [January 
1824]. 

Clark, B. L. 1938. Fauna from the Markley Formation (upper 
Eocene) on Pleasant Creek, California. Bulletin of the 
Geological Society of America 49: 683-730 

Coan, E. V. 1981. James Graham Cooper, pioneer western 
naturalist. Moscow, Idaho: University Press of Idaho, 255 
pp: 

Collin, R. 2003a. The utility of morphological characters in 
gastropod phylogenetics: an example from the Calyp- 
traeidae. Biological Journal of the Linnean Society 78: 
541-593. 

Collin, R. 2003b. Phylogenetic relationships among calyptraeid 
gastropods and their implications for the biogeography of 
marine speciation. Systematic Biology 52(5): 618-640. 

Collin, R. 2005. Dev elopment, phylogeny, and taxonomy of 
Bostrycapulus (Caenogastropoda: “Calyptraeidae), an an- 
cient cryptic radiation. Zoological Journal of the Linnean 
Society, 144: 75-101. 

Conrad, T. A. 1855. Report on the fossil shells collected in 
California by W. P. Blake, geologist of the expedition un- 
der the command of Lieut. R. S. Williamson, U.S. Topo- 
graphical Engineers, 1852. In: The preliminary geological 
report of W. P. Blake, U.S. 33"! Congress, 1 session, 
House Exec. Doc. No. 129 (July, 1855: 5-20. (Reprinted, 
1909, U.S. Geological Survey Professional Paper 59: 163— 
M71. 

Conrad, T. A. 1856. Descriptions of the fossil shells. U.S. Ex- 
plorations and surveys for a railroad route from the Mis- 
sissippi River to the Pacific Ocean, 5: 317-329. pls. 2-9. 
(Appendix to the preliminary geological report.) 

Conrad, T. A. 1860. Descriptions of new species of Cretaceous 
and Eocene fossils of Mississippi and Alabama. Journal of 
the Academy of Natural Sciences of Philadelphia, Series 2, 
4: 275-297. 

Cooper, J. G. 1894. Catalogue of Californian fossils, parts 2-5. 
California State Mining Bureau, Bulletin 4, 65 pp. 
Cossmann, M. 1891. Revue de paléontologie pour l'année 
1889, dirigée par M. H. Douvillé. Mollusques: Gastero- 

podes. Annuaire Géologie 6: 855-896. 

Cossmann, M. 1903. Essais de Paléoconchologie Comparée, 
Tome 5. Privately published, Paris: 1-216. 

Cossmann, M. 1925. Essais de Paléoconchologie Comparée, 
Tome 13. Presses Universitaires de France. Paris, 345 pp. 

Cox, L. R. 1925. Cretaceous Gastropoda from Portuguese East 
Africa. Annals of the Transvaal Museum 11: 201-216. 

Dailey, D. H., and W. P. Popenoe. 1966. Mollusca from the 
Upper Cretaceous Jalama Formation, Santa Barbara 
County, California. University of California Publications in 
Geological Sciences 65, 40 pp. 

Darteville, E. and P. Brébion. 1956. Mollusques fossiles du 
Crétacé de la Cote occidentale d'Afrique du Cameroun a& 
Angola. I. Gasteropodes. Annales du Musée Royal du 
Congo Belge Tervuren, Sciences Géologiques 15: 1-128. 

Delpey, G. 1942. Gastéropodes du Crétacé Supérieur dans le 
sud-ouest de la France. Bulletin de la Société d'Histoire 
Naturelle de Toulouse 77: 161-197 

Dickerson, R. E. 1916. Stré itigraphy and fauna of the Tejon 
Eocene of California Unive srsity of California, Publica- 
tions in Geological Sciences 9(17): 363-524. 

Docke sry, D. T., III. 1993. The stre ptoneuran gastropods, ex- 
Be of the Stenoglossa, of the Coffee Sand (Campa- 
nian) of northeastern Mississippi. Mississippi Department 


of Environmental Quality Office of Geology Bulletin, 129, 
191 pp. 

dOrbigny, A. D. 1841. Mollusques. In: R. de la Sagra, Histoire 
physique, politique et naturelle de ile de Cuba. Arthus 
Betrand, Paris, Volume 1, 208 pp. 

d'Orbigny, A. D. 1842-1843. Paléontologie Frangaise. Terrains 
Crétacés. Volume 2, Gastéropodes. Arthus Bertrand, 
Paris. 456 pp. 

Egorov, R., and D. Alexeyev. 1998. Trichotropidae. Treasure of 
Russian Shells. Vol. 2. Moscow. 36 pp. 

Elder, W. P. and L. R. Saul. 1993. Paleogeographic implica- 
tions of molluscan assemblages in the Upper Cretaceous 
(Campanian) Pigeon Point Formation, California. In: G. 
Dunne and McDougall (eds.) Mesozoic Paleogeography of 
the Western United States-I. Pacific Section, Society of 
Economic Paleontologists and Mineralogists 71: 171- 186. 

Emerson, W. K. and M. K. Jacobson. 1976. The American 
Museum of Natural History Guide to Shells. Alfred A. 
Knopf, Inc., New York, 482 +xviii pp. 

Feruglio, E. 1936. Palaeontographia Patagonica—Part I. R. 
Universita di Padova, Istituto di Geologico, Memorie 11: 
1-192, 20 pl. 

Finlay, H. J. 1926. A further commentary on New Zealand 
molluscan systematics. Transactions of the New Zealand 
Institute 57: 320-485. 

Finlay, H. J., and J. Marwick. 1937. The Wangaloan and asso- 
ciated molluscan faunas of eonenie Green Island sub- 
division. New Zealand Geological Surv ey, Palaeontological 
Bulletin 15, 140 pp. 

Fischer, P. 1864. Note sur le genre Fossarus, suivie du cata- 
logue des espéces. Journal de Conchyliologie, Paris 12: 
252— pe 

Fischer, P. 1880-1887. Manuel de conchyliologie et de palé- 
eee conchyliologique, ou histoire naturelle des mol- 
lusques vivants et fossiles, Savy, Paris, 1369 pp. 

Fleming, J. 1821. Mollusca. In New Edinburgh Encyclo- 
paedica. American edition, Philadelphia, 13, pt. 2: 656— 
696. 

Fleming, J. 1522. Philosophy of Zoology. Volume 2. Edinburgh, 
618 pp. 

Fretter, V. and A. Graham. 1962. British Prosobranch Mol- 
luses. Ray Society, London, 755 pp. 

Gabb, W. M. 1864. De scription of the Cretaceous fossils. Geo- 
logical Survey of California, Palaeontology 1: 57-243. 
Gabb, W. M. 1869. Cretaceous and Tertiary fossils. Geological 

Survey of California, Palaeontology 2: 1-299. 

Gabb, W. M. 1877. Notes on American Cretaceous fossils, with 
descriptions of some new species. Academy of Natural 
Sciences Proceedings 28: 276-324, pl. 17. 

Gradstein, F. M., J. G. Ogg, and A. Smith. 2004. Geologic Time 
Scale, Cambridge University Press, Cambridge, 589 pp. 

Graham, A. 1954. The anatomy of the prosobranch Trichotro- 
pis borealis Broderip and Sowerby, and the systematic 
position of the Capulidae. Journal of the Marine Biology 
Association of the United Kingdom, 33: 129-144. 

Gray, J. E. 1840. Shells of molluscous animals. In: Synopsis of 
the contents of the British Museum, 42 edition, C. Wood- 
fall and Son, London: 105-156. 

Gray, J. E. 1847. A list of the genera of Recent Mollusca, their 
synonyma and types. Procee dings of the Zoological Society 
of London 15: 129-219. 

Gray, J. E. 1850. Figures of molluscous animals selected from 
various authors. Etched for the use of students | ny M. E. 


L. R. Saul and R. L. Squires, 2008 


Page 139 


Gray. Volume 4. Longman, Brown, Green and Longmans, 
London, 219 pp. 

Guilding, L. 1834. Observations on Naticina and Dentalium, 
two genera of molluscous animals. Transactions of the Lin- 
nean Society of London 17: 29-36. 

Habe, T. 1961. Coloured illustrations of the shells of Japan, 
Volume 2. Osaka, Hoikusha, 148 pp., Appendix 42 pp. 

Helbling, G. S. 1779. Beytriige zur Kenntniss neuer und 
seltener Konchylien. Abhand lungen einer Privatgesell- 
schaft in Boéhmen zur Antoehane der Mathematik, der 
vaterlindischen Geschichte und der Naturgeschichte 4 
102-131. 

Hinds, R. B. 1844-1845. The zoology of the voyage of H. M. S. 
Sulphur, under the command of Capt. Sir Edward 
Belcher, ... during the years 1836-42, Vol. 2 [Mollusca]. 
Smith, Elder & Co.: London, 72 pp., 21 pl. 

Hoagland, K. E. 1977. Systematic review of fossil and recent 
Crepidula. Malacologia 16: 363-420. 

Hoagland, k. E. 1986. Patterns of encapsulation and brooding 
in the Calyptraeidae (Prosobranchia: Mesogastropoda). 
American Malacological Bulletin 4: 173-183. 

Holzapfel, L, E. 1888. Die Mollusken der Aachener kreide. 

Palaeontographica 34: 29-180, pl. 4-21. 

Humphrey, G. 1797. Museum Calonnianum. Catalogue. Lon- 
don, 84 pp. 

Thering, H. von. 1907. Los Mollusques fossiles du Tertiaire et 
du Crétace supérieur de l’Argenitine. Anales del Museo 
Nacional de Buenos Aires (3)7: 1-611. 

Jousseaume, F. 1888. Description des mollusques recueillis 
par M. le Dr. Faurot dans la Mer Rouge et le Golfo 
d’Aden. Mémoires de la Société Zoologie de France 1: 
165-223. 

Kase, T. 1990. Late Cretaceous gastropods from the Izumi 
Group of southwest Japan. Journal of Paleontology 64: 
563-5758. 

Keen, A. M. 1971. Sea Shells of Tropical West America. Stan- 
ford University Press, Stanford, California, 1064 pp. 
Kiel, S.. and K. Bandel. 2003. New taxonomic data for the 
gastropod fauna of the Umzamba Formation (Santonian- 
Campanian), South Africa based on newly collected ma- 

terial. Cretaceous Research 24: 449-475. 

Kleinpell, R. M., and D. W. Weaver. 1963. Oligocene Biostra- 
tigraphy of the Santa Barbara Embayment, California. 
University of California Publications in Geological Sci- 
ences 43: 250 pp. 

Klinger, H. C. and Kennedy, W. J. 1980. The Umzamba For- 
mation at its type section, Umzamba Estuary (Pondoland, 
Transkii), the ammonite content and palaeontographical 
distribution. South African Museum, Annals 81: 207-222, 

Kollmann, H. A. and G. S. Odin. 2001. Gastropods from 
the Upper Cretaceous eee site at Tercis les Bains 
(SW France). In: Odin, G. (ed.) The Campanian- 
Maastrichtian Stage es Amsterdam: Elsevier Sci- 
ence: 437-451. 

Kollmann, H. A. 2005. Révision Critique de la Paléontologie 
Francaise d’Alcide d’Orbigny [translated et adapti par 
Jean-Claude Fischer], V olume III Gastropodes Crétacés 
Museum d'Histoire naturelle de Vienne, Autriche, Back 
huys Publishers Leiden, Pays-Bas, France, 239 pp. + Réé- 
dition du Tome Second (Terrains Crétacés, Gastropodes 
de la Paléontologie Francaise d’Alcide d’Orbigny (édition 
original : 1842-1843) 456 pp. + Atlas of plates 149-236. 

Lamarck, J. B. 1799. Prodrome d'une nouvelle classification 


des coquilles. Mémoires de la Société d'Histoire Naturelle 
de Paris 1(1): 63-91. 

Lamarck, J. B. 1809, Philosophie zoologique. Volume 1. Dentu, 
Paris, 428 pp. 

Leach, W. E. 1852. Molluscorum Britanniae synopsis. John Van 
Voorst, London, 376 pp. 

Lesson, R. P. 1830, Mollusques, annélides et vers. Zoologie 2 
(1), Chapter 11. In: L. J. Duperry, Voyage autour du 
monde, exéuté par ordre du Roi, sur la corvette de S, M. 
La Coquille pendant les années 1822, 1824 et 1825. 
Arthus Bertrand, Paris: 239-488. 

Linnaeus, C. 1758. Systema naturae per regna tria penne 
Editio duodecima, reformata, Regnum animale, vol. 
Holmiae, 824 pp. 

Marshall, B. A. 2003. A review of the Recent and Late Ceno- 
zoic Calyptraeidae of New Zealand (Mollusca: Gas- 
tropoda). The Veliger 46: 117-144. 

Matsumoto, T. 1960. Upper Cretaceous ammonites of Califor- 
nia, Part HW. Kyushu University, Memoirs of the Faculty of 
Science, Series D, Geology, Special Volume II, 204 pp. 

Merriam, J. C. 1895. A list of type specimens in the Geological 
Museum of the University of California, which have 
served as originals for figures and descriptions in the Pa- 
leontology of the State “Geological Survey of California 
under J. D. Whitney, Compiled for the use of workers in 
California geology. University of California Berkeley, Ge- 
ology Department: 3 unnumbered printed pages. Re- 
printed by A. W. Vogdes, 1896, California State Mining 
Bureau, Bulletin 10: 21- 23; Second printing, 1904, Cali- 
fornia State Mining Bureau, Bulletin 30; 39-42. 

Montfort, P. D. 1810. Conchyliologie systématique et classifi- 
cation méthodique des coquilles. Conchyliologie Systéma- 
tique Vol. 2. F. Schoell, Paris, 676 pp. 

Murphy, M. A., and P. U. Rodda. 1960. Mollusca of the Bald 
Hills Formation of California. Journal of Paleontology 34: 
835-58. 

Olsson, A. A. 1944. Contributions to the paleontology of north- 
em Peru, Pt. VIE The Cretaceous of the Paita region. 
Bulletins of American Paleontology 28: 1-146. 

Packard, E. L. 1922. New species from the Cretaceous of the 
Santa Ana Mountains, California. University of California 
Publications, Department of Geological Sciences Bulletin 
13: 413482. 

Pervinquiére, L. 1912, Etudes de paléontology tunisienne; I. 
Gastéropodes et lamellebranches des terrains crétacés. 
Direction Générale des Traveaux Publics. Carte 
Géologique de la Tunisie, 352 pp., 23 pls. 

Philippi, R. A. 1841. Zoologische Bemerkungen. Archive fiir 
Naturgeschichte 7(1): 42-49. 

Philippi, R. A. 1887. Die Tertiiren und Quartiren ver- 
steinerungen Chiles. F. A. Brockhaus, Leipzig, 266 pp. 

Pilsbry, H. A. (with the emendations of W. H. Dall). 1913. 
Gastropoda. In: C. R. Eastman (ed.), Text-book of Pale- 
ontology, Volume 1, adapted from the German of kK. A. v. 
Zittel, Second edition, MacMillan and Co., Limited, Lon- 
don: 514-583. 

Ponder, W. F. 1998. Superfamily Capuloidea. In: P. L. Beesley, 
G. J. B. Ross, and A. Wells (eds.), Mollusca: The Southern 
Synthesis. Fauna of Australia. Woe _ CSIRO Publish- 
ing, Melbourne, Part B: 774-775. 

Ponder, W. F. and A. Warén. 1988. Classification of the Cae- 
nogastropoda and Heterostropha—A list of Bs family- 
group names and higher taxa. In: Ponder, F. (ed.) 


Page 140 


THE NAUTILUS, Vol. 122, No. 3 


Prosobranch Phylogeny. Malacological Review, Supple- 
ment 4; 288-328. 

Quaas, A. 1883-1903. Fauna der obersten Kreidebildungen in 
der Libyschen Wiiste. Palaeontographica 30: 153-336, 

Rafinesque, C. S. 1815. Analyse de la nature ou tableau de 
univers et des corps organisés. Palerme, 223 pp. 

Rennie, J. V. L. 1930. New Lamellibranchia and Gastropoda 
from the Upper Cretaceous of Pondoland (with an appen- 
dix on some species from the Cretaceous of Zululand). 
Annals of the South African Museum 28(2): 161-260. 

Rennie, J. V. L. 1935. On a new species of Lysis (Gastropoda) 
from the Cretaceous of Pondoland (with a note by L. R. 
Cox on the identity of Tropidothais Cox with Lysis Gabb). 
Records of the Albany Museum (Grahamstown, South Af- 
rica) 4: igs! 

Rennie, J. V. L. 1945. Lamelibranquios e gastrépodos do 
Cretacico Superior de Angola. Republica Portuguesa, 
Ministério das Coldnias; Junte i das Missdes Geogr dficas e 
de Investigagdes Colonias, Memorias, Série Géologica 1: 
l- ae 

Riccardi, A. C. 1988. The Cretaceous system of southern South 
te Geological Society of America Memoir 168, 161 
pp. 

Saul, L. R. 1959. Senonian mollusks from Chico Creek. Un- 
published M.A. thesis, University of California Los Ange- 
les, 170 pp. 

Saul, L. R. 1990. From Trichotropidae to Calyptraeidae by way 
of Lysis Gabb, 1864. The Western Society of Malacolo- 
gists Annual Report 22: 12-13. 

Saul, L. R., and J. M. Alderson. 1981. Late Cretaceous Mol- 
lusea of the Simi Hills: an introduction. In; Link, M. H., 
R. L. Squires, and I. P. Colburn (eds.) Simi Hills Creta- 
ceous turbidites, Southern California. Society of Eco- 
nomic Paleontologists and Mineralogists, Pacific Section, 
Fall 1981 Field Trip Volume and Guidebook: 29-41. 

Saul, L. R., and R. L. Squires. 2008. North American Pacific 
slope species of Volutoderminae (Gastropoda, Volutidae) 
of Coniacian through Maastrichtian age. Journal of Pale- 
ont logy 82: 213-237, 

Schumacher, C. F. 1817. Essai dun nouveau systeme des habi- 
tations des vers testacés. Schulz, Copenhagen, 287 PP. 
Seeley, H. G. 1861. Notes on Cambridge palaeontology 2. On 
some new gastropods from the Upper Greensand. Annals 
and Magazine of Natural History (3) 7: 251-295, pl. 11. 

Shimer, H. W., and R. R. Shrock. 1944 (19 printings, 1944— 
1989). Index Fossils of North America. New York, John 
Wiley & Sons, Inc., ix + $37 pp. 

Sohl, N. F. 1960. Archaeogastropoda, Mesogastropoda and 


stratigraphy of the Ripley, Owl Creek, and Prairie Bluff 


formations. U.S 
331-A: 1-151. 
Sowerby, G. B. (Ist of the name 


Recent and fossil shells. 2 a jondon. 


Geological Survey Professional Paper 


Squires. R. L. 1987. Eocene molluscan paleontology of the 


Whitaker Peak area, Los Angeles and Ventura counties, 
California. Natural History Museum of Los Angeles 
County, Contributions in Science, 388, 93 pp 

Squires, R. L, 2003. Turnovers in marine gastropod faunas 
during the Eocene-Oligocene transition, west coast of the 
United States. In: D. R. Prothero, L. C. Ivany and E. A. 
Nesbitt (eds 
Eocene-Oligocene Transition. New York: Columbia Uni- 


), From Greenhouse to [ceehouse: The Marine 


versity Press: 14—35 


1821-1834. The genera of 


Stadum, C. J. 1973. A student guide to Orange County fossils. 
Chapman College Press, Orange, California. 64 pp. 

Stecheson, M. 2004. Systematic paleontology of marine gastro- 
pods from the U pper Cretaceous Chatsworth Formation, 
Simi Hills, southerm California. Unpublished M.S. thesis. 
California State University, Northridge, 142 pp. 

Stephenson, L. W. 1952. Larger invertebrate fossils of the 
Woodbine Formation (Cenomanian) of Texas. U.S. Geo- 
logical Survey Professional Paper 242, 226 pp. 

Stewart, R. B. 1927. Gabb’s California fossil type gastropods. 
Proceedings of the Academy of Natural Sciences of Phila- 
delphia 75: 287-447, 

Stilwell, J. D. and W. J. Zinsmeister. 1992. Molluscan system- 
atics and biostratigraphy, lower Tertiary Le Meseta For- 
mation, Seymour Island, Antarctica Peninsula. Antarctic 
Research Series 55: ix +xii, 1-192. 

Stoliczka, F. 1867-1868. Cretaceous fauna of South India. 
Volume 2. The Gastropoda. Memoirs of the Geological 
Survey of India, Paleontologica Indica, Series 5, 21: 1-497. 

Suter, H. 1909. The Mollusca of the subantarctic islands of 
New Zealand. In: C. Chilton (ed.), Subantartic Islands of 
New Zealand. Volume 1. Government Printer, Welling- 
ton, New Zealand, xxxv + $45 pp. 

Thiele, J. eye: Handbuch der systematischen Weich- 
tie erkunde. Vol. 1. Jena Gustav Fischer, Jena, 376 pp. 
Troschel, F. H. 1861. Das Gebiss der Schnecken, zur Begriin- 
dung einer natiirlichen Classification. Nicolai, Berlin. Vol. 

1 (4): 153-196. 

Tryon, G. W, Jr. 1882-1554. Structural and systematic conchol- 
ogy, and introduction to the study - Mollusca. Philadel- 
phia: published by i author, vol. 1, 1852, 312 pp.; vol. 2, 
1883, 430 pp.; vol. 3, 1884, 453 

Ulbrick, M. L. 1969. oe on Crucibulum spinosum (Sow- 
erby). Malacological Society of London, Proceedings 38: 
431-438. 

Vokes, H. E. 1939, Molluscan faunas of the Domengine and 
Arroyo Hondo formations of the California Eocene. New 
York Academy of Sciences, Annals 35; 1-246, pls. 1-22 

Warren, G., and I. Speden. 1978. The Piripauan and Hau- 
murian stratotypes (Mata) Series, Upper Cretaceous) and 
correlative sequences in the Haumuri Bluff District, South 
Marlborough (S56). New Zealand Geological Survey Bul- 
letin 92, 60 pp. 

Weaver, C. E. 1943. Paleontology of the marine Tertiary for- 
mations of Oregon and Washington. University of Wash- 
ington Publications in Geology 5: 759 pp., 104 pls. 

Weaver, C. E. and k. V. W. Palmer. 1922. Fauna from the 
Eocene of Washington. University of Washington Publi- 
cations in Geology 1(3): 1-56. 

Wellman, H. W. 1959. Divisons of the New Zealand Creta- 
ceous. Transactions of the Royal Society of New Zealand 
S87: 99-163. 

Wenz, W. 1935-1944. Gastropoda. Teil 1: Allgemeiner Teil 
und Prosobranchia. In O. H. Schindewolf (ed.) Handbuch 
der Palaiozoologie, Band 6, Gebriider Borntraeger, Berlin, 
1639 pp. [Reprinted 1960-1961], 

White, C. A. 1889. On invertebrate fossils from the Pacific 
Coast. United States Geological Survey Bulletin 51, 102 
pp. 

Whiteaves, J. F. 1879. On the fossils of the Cretaceous rocks of 
Vancouver and adjacent islands in the Strait of Georgia. 
Canada Geological Survey, Mesozoic Fossils 1: 93-190. 

Whiteaves, J. F. 1903. On some additional fossils from the 
Vancouver Cretaceous, with a revised list of species there- 


L. R. Saul and R. L. Squires, 2008 


Page 141 


from. Canada Geological Survey, Mesozoic Fossils 1: 309- 
415. 

Wilckens, O. 1904. Beitriige zur Geologie und Palacontologie 
von Siidamerika, NI. Neues Jahrbuch fiir Mineralogie, Ge- 
ologie und Paleontologie S: 181-284. 

Wilckens, O. 1907. Die Lamellibranchiaten, Gastropoden ete. 
der oberen Kreide Siidpatagoniens. Berichte der Natur- 
forschenden Gessellschaft zu Freiburg 15: 3-70. 

Wilckens, O. 1922. The Upper Cretaceous gastropods of New 
Zealand. New Zealand Department of Mines Geological 
Survey Branch Palaeontological Bulletin 9: 1-42. 


Woods, H. 1917. The Cretaceous Faunas of the North-Eastern 
Part of the South Island of New Zealand. New Zealand 
Geological Survey, Paleontological Bulletin No. 4, 41 pp.., 
19 pls. 

Yates, L. G. 1903. Prehistoric California. Bulletin of the South- 
ern California Academy of Sciences 2(7): 86-90. 

Yonge, C. M. 1962. On the biology of the mesogastropod Tri- 
chotropis cancellata Hinds, a benthic indicator species. 
Biological Bulletin 122: 160-181. 


APPENDIX 1 


LOCALITIES CITED 


Localities are LACMIP, unless otherwise noted. All quadrangle maps 
are U. S. Geological Survey maps. Bracketed numbers are areas shown 
on Figure 1. 


[4] CASG 61794. [= CAS 1346-A]. In conglomeratic sandstone 1.6 km 
(1 mi.) above mouth of Huling Creek, North Fork Cottonwood Creek, 
Ono Quadrangle (15 minute, 1952), Shasta Co., California, Budden 
Canyon Formation, Bald Hills Member. Late Albian. 


7333. Le Fayel, Paris Basin, France. Late Eocene (Bartonian Stage). 


[18] 10095. [=CIT 83]. Fine sandstone just above shale, sectionline 
fence gate on old road 0.4 km (1/4 mi.) W of Schulz Ranch, 122 m (400 
ft.) S. of northeast corner of section 19, T. 5 S, R. 7 W, El Toro 
Quadrangle (7.5 minute, 1968), south side of Williams Canyon, Santa 
Ana Mountains, Orange Co., California. Coll.: B. N. Moore, 1 January, 
1926. Ladd Formation, uppermost Holz Shale Member. Early Cam- 
panian. 


[16] 10711. [=CIT 1158]. Approximately 1.84 km (1.5 mi.) due west of 


Los Angeles-Ventura Co. line on the boundary (extended) between T. 
1 N and T. 2 N. north bank of Bell C Canyon, southeast slope of Simi 
Hills, Calabasas Quadrangle (7.5 minute, 1952), Ventura Co., Califor- 
nia. Coll: W. P. Popenoe, 18 July, 1935. Chatsworth Formation. 
Middle Campanian. 


(3] 10757. [=CIT 1593]. Massive sandstones in bed of North Fork Bear 
Creek, approximately 777 m (2550 ft.) south and 533 m (1750 ft) east 
of northeast corner of section 5, T. 31 N, R. 1 E, Whitmore Quadrangle 
(15 minute, 1956), Shasta County, California. Coll.: W. P. Popenoe and 
W. M. Tovell, 12 Sept., 1941. Redding Formation, Bear Creek Sand- 
stone. Coniacian. 


[5] 10846. [=CIT 1014]. Concretions in sandstone, right bank of Chico 
eae about 1.6 km (1 mi.) upstream from the little bridge across creek 
below Mickey house and about 4.16 km (2.5 mi.) N6°W of 14-mile 

house on Humboldt Road, NW 1/4, SE 1/4 of section 1, T. 23 N, R. 2 
E, Paradise Quadrangle (15 minute, 1953), Butte Co., California Coll: 
W. P. Popenoe and D. W. Scharf, 16 August, 1931. Chico Formation, 
top of Ponderosa Way Member. Late Coniacian or early Santonian 


[23] 11944. Approximately 10 km N of Punta Abreojos (SW of San 
Ignacio). in first ridges N of arroyo that crosses Punta Abreojos road 


(dirt) just S of Campo Rene tumoff, approx. 2 km NW of road. Approx. 
2-3 km up ravine, hill .5 km to east is pachydiscid loc., Viscaiano Penin- 
sula, Baja California Sur, Mexico, Coll.: R. Demetrion, 1987, 1989. Valle 
Formation. Middle? Campanian, with Hoplitoplacenticeras? 


{14] 14310. About 450m north and 70m west of lookout at summit of 
Warm Springs Mountain at elev. of 1052 m (3450 ft.), at base of Kirby’s 
(1991, M. A. Thesis) measured section no. 2, Warm Springs Mountain 
Quadrangle (1958), Los Angeles County, California. Coll.: M. X. Kirby. 
Basal San Francisquito Formation. Late Maastrichtian. 


22340. Gritty conglomeratic sandstone lenses in fine-grained sandstone 
and shales, east side Grapevine Canyon about 0.8 km (0.5 mi.) south of 
its north end, about 0.4 km (0.25 mi.) east of, and 30.48 m (100 ft.) 
above the abandoned highway roadbed on east side of canyon, at about 
648 m (2100 ft.) contour, about 91 m (300 ft.) northwest of Tejon/ 
granite fault contact. Locality is 3048 m (10,000 ft.) N24°W of 3174 ft. 
BenchMark at old Fort Tejon, Tejon Quadrangle, Kern County, Cali- 
fornia. Coll: W. P. Popenoe, 9 December, 1946. Tejon Formation. 
Middle Eocene. 


22386. Prominent shell bed at crest of ridge on east side of Live Oak 
Canyon, about 0.4 m (0.25 mi.) south of its mouth, T 10 N, R 19 W, 
Pastoria Creek Quadrangle, 7.5, 1958, photorevised 1974, Kern 
County, California. Coll; W. P. Popenoe, 13 March, 1947. Tejon For- 
mation, Metralla Sandstone Member. Middle to upper Eocene (“Tejon 
Stage”). 


[10] 22588. About 2/3 of the way to the top of a gully on southwest 
slope of a northwest-trending hill on south side of Garzas Creek, where 
the creek enters the San Joaquin Plain, approximately 610 m south and 
183 m west of northeast comer of section 19, T 8 S, R 8 E, Howard 

Ranch Quadrangle, 7.5', 1953, photorevised 1971, west side of San 
Joaquin Valley, Stanislaus Co., California. Coll: W. P. Popenoe and T. 
Susuki, April, 1950. Moreno Formation, “Garzas Sand” member. 
Middle Maastrichtian. 


[4] 23464. [PR1] Up small creek from Sulivan Ranch Rd. crossing, and 
1.28 km (0.8 mi.) north of ranch, near Gas Point Rd., 701 m (2300’) N 
75°E from mouth of Huling Creek, 579 m (1900°) S, 488 , (1600°) E of 
NW comer of section 16, T. 30 N, R. 6 W, Ono Quadrangle (15 minute, 
1952), Shasta Co., California. Coll: P. U. Rodda and M. A. Murphy, 
May 1955. Budden Canyon Formation, Bald Hills Member, unit iv in 
Matsumoto, (1960). Middle Cenomanian, probably Turrilites costatus 
Zone. 


[5] 23617. Fossil in hard, blue-gray concretion in gray-weathering buff 
sandstone approximately 15.2 m (50 ft.) below highest conglomerate, 
approx. 0.8 km (0.5 mi.) upstream from Mickey house on west side of 
Chico Creek, 1.52 m (5 ft.) above stream, 716.28m (2350 ft.) north, 
609.6 m (2000 ft.) west of southeast comer of section 1, T. 23 N, R. 2 
E, Paradise Quadrangle (15 minute, 1953), Butte Co., California. Coll.: 
R. B. Saul, 14 August, 1955. Chico Formation, top of Ponderosa Way 
Member. Late Coniacian or early Santonian. 


5] 23639. In concretions in massive, greenish-gray sandstone, east 
bank of Chico Creek, west of meadow eh large flat- topped, lava block 
at north edge near road, 373.38 m (1225 ft.) ) south and aoe 6 m (960 ft.) 
west of northeast corner of section 23, T. 23 N, R. 2 E, Paradise 
Quadrangle (15 minute, 1953), Butte County, cere California, Col- 
lectors: L. R. Saul and R. B. Saul, 20 August, 1952. Chico Formation, 
owermost part of Ten Mile Member. Early C ampanian. 


12] 24122. Fine- to coarse grained buff sandstone; 76.2 m (250 ft.) 
north of jeep trail in Jalama Canyon; elevation 190 m (625 ft.), 6.58 km 
(4.11 mi.) east and 1.1 km (0.69 mi.) south of Jalama Ranch Headquar- 
ters: 0.93 km (0.58 mi.) west and 0.66 km (0.41 mi.) north of southeast 
corner of topo, Lompoc Hills Quadrangle (7.5 minute, 1959), Santa 
Barbara Co., California. Coll.: D. Dailey, August, 1959. Jalama Forma- 
tion. Late Campanian—early Maastrichtian 


[12] 24128. Dark gray conglomerate in first small canyon east of Ra- 
majal Canyon, elevation 167.6 m (550 ft.), 0.54 km (0.34 mi) south, 3.25 
km (2.03 mi.) east of Jalama Ranch Headquarters, 1.22 km (0.76 mi.) 
north 4.27 km (2.67 mi.) west of southeast corner of Lompoc Hills 
Quadrangle (7.5 minute, 1959), Santa Barbara Co., California. Coll.: D 


Page 142 


Dailey, August 1958. Jalama Formation, Late Campanian—early Maas- 
trichtian. 


{12] 24237. Medium-grained, buff, arkosic sandstone, 396.2 m (1300 
ft.) north of Jalama Creek, elevation 160 m (525 ft.), 0.48 km (0.30 mi.) 
south, 3 km (1.88 mi.) east of the Jalama Ranch Headquarters, 1.28 km 
(0.80 mi.) north, 4.59 kin (2.87 mi.) west of southeast corner of Lompoc 
Hills Quadrangle (7.5 minute, 1959), Santa Barbara Co., California. 
Coll.: W. P. Popenoe, September 1938, Jalama Formation. Late Cam- 
panian-early Maastrichtian. 


[6] 24340, Penz vicinity, conglomerate beds cropping out just below a 
drainage canal, southeast side of new Oroville Hwy, about 1.2 km (0.75 
mi.) northeast of intersection with Pentz- Magalia- Oroville road, 426.7 
m (1400’) S., 182.9 m (600) W of the northeast corner of section 36, T. 
21 N, R.3 E, Cherokee quad. (7.5 minute, 1949), Butte Co., California. 
Coll.: W. P. Popenoe, 1960. Chico Formation, Musty Buck Member. 
Early Campanian. 


[8] 24349. [=USGS M8601 and USGS M8745]. In place? large angular 

block of sandstone surrounded by sand at shoreline in covelet on north 

side of elongate seaward-pointing rock; approx. N30°W of Pigeon Point 

lighthouse, just south of Bolsa Point, Pigeon Point Quadrangle (7.5 

minute, 1952), San Mateo County, Califomiia. Coll: and R. B. 
Saul, October 11, 1960. Pigeon Point Fm. Middle C mes 


{16] 26020. [=CIT 1158]. Hard sandstone slabs in fine-grained sand- 
stone, cropping out on high bare cliff, north bank of Bell Canyon, just 
east of mouth of large gully, and 152.4 m (500’) S, 2743.3 (9000") west 
of northeast corner of section 4,T. 1 N, R. 17 W, Calabasas Quadrangle 
(7.5 minute, 1952), Simi Hills, Ventura Co., California. Coll.: W. P. 
Popenoe, 11 Feb., 1972. Chatsworth Formation. Middle Campanian 


{10] 26353. Approximately 1.2 km south of Garzas Creek, 671 m south 
and 114 m east of northwest corner of section 20, T 8 S, R 8 E, Howard 
Ranch Quadrangle, 7.5°, 1953, photorevised 1971, Stanislaus County, 
California. Coll; R. B. Stewart and W. P. Popenoe, 1944. Moreno 
Formation, “Garzas Sand” member. Late early to early late Maastrich- 
tian, 


{11] 26370. Reworked fossiliferous Turonian blocks in upper Campa- 
nian conglomerate lens in shale, northeast side of Cooper Canyon, 


approx. 411.5 m (1350°) n, 670.6 m (2200°) W of southeast corner of 


section 2, T, 21 S, R. 14 E, Alcalde Hills Quadrangle (7.5 minute, 
1969), \, Alcalde Hills, Fresno Co., California. Coll.: J. Alderson, 18 June, 


THE NAUTILUS, Vol. 122, No. 3 


1977. Panoche Formation, “Alcalde Shale” Member. 
(with juvenile Subprionocyclus sp.). 


Late Turonian 


[18] 26951. Small sandstone lens approx. 6.1 m (20 ft.) above road in 
roadcut on north side of Silverado Truck Trail, 274.3 m (900 ft.) south 
of northeast comer of section 18, T. 5 S, R. 7 W, El Toro Quadrangle 
(7.5 minute, 1949), Orange Co., California. Coll; A. A. Almgren, 4 
Dec., 1981. Ladd Formation, uppermost Upper Holz Shale Member. 
Late early Campanian, 


[17] 26967. Small exposure of coarse-grained, poorly sorted sandstone 
at bottom of northwest-flowing tributary to main fork of Garapito 
Creek, 449.6 m (1475 ft.) and 2835 m (9380 ft.) east of northwest 
corner of section 5, T. 1S, R. 16 W, Topanga Quadrangle (7.5 minute, 
1952, photorevised, 1981), Santa Monica Mountains, Los Angles 
County, California. Coll.: J. M. Alderson, 31 Dec., 1981. Tuna Canyon 


Formation. Coniacian., 


[3] 28717. South Cow Creek Valley, sandstone associated with con- 
glomerate, lower South Cow Creek Valley, about 152.4 m (500 ft.) 
downstream from old bridge site across creek, and about 1.6 km (1 mi.) 
due west of buildings on Hunt Ranch, NE 1/4 of section 17, T. 31N, R. 
2W, Millville Quadrangle, $ Shasta Co., California. Coll.; W. P. Popenoe, 
27 Oct., 1971. Redding Formation, Oak Run Conglomerate Member of 
Haggart, 1956. Late Santonian. 


[19] SDNHM 3403. Taylor Made Golf Facility at Salk Drive and Col- 
lege Blvd., elevation 45.1 m (148 ft.) along College Blvd., lat. 33°8°25' 
N, long. 117°1656' W, [in general = SDNHM 3402, 3404, 4071, 4073], 
Carlsbad Research Center, Site 29, San Luis Rey Quadrangle 7.5° 
(1968), San Diego Co., Califormia. Coll: B. O. Riney, 26 April, 1987, 
Point Loma Formation. re Campanian/?early Maastrichtian. 


{19] SDONHM 3405. Carlsbad Research Center, 1.6 km north and 2 
km west of southeast corner of San Luis Rey Quadrangle 7.5° (1 968), 
indicated area along west side of College Blvd, starting about 0.32 km 
from intersection with El Camino Real extends southward for 0.15 km, 
San Diego Co., California. Coll: B. O. Riney, 1987. Point Loma For- 
mation, ?Late Campanian/early Maastrichtian. 

{18] UCMP 2167. 3.2 km (2 mi.) N 10°W of BM 1271, Corona Quad- 
rangle (1902), at a gate about 0.5 km (0.5 mi) below Modjeska Springs 
in Williams Canyon, Santa Ana Mountains, Orange Co., California. 
Ladd Formation, uppermost Holz Shale. Middle? Campanian. 


THE NAUTILUS 122(3):148-150, 2008 


Page 143 


Feeding behavior, phylogeny, and toxinology of Conus furvus 
Reeve, 1843 (Gastropoda: Neogastropoda: Conidae) 


Samuel S. Espino 


Institute of Chemistry 

University of the Philippines 

Diliman, ‘Quezon City 1101, 
PHILIPPINES 

and 

Department of Chemistry 

University of the Philippines in the Visayas 
Miag-ao, Iloilo, PHILIPPINES 


Alan J. Kohn 


Frank M. Heralde HI 


National Institute of Molecular Biology 
and Biotechnology 

University of fie Philippines 

Diliman, Quezon City 1101, 
PHILIPPINES 


University of Utah 


Baldomero M. Olivera! 
Department of Biology 
University of Utah 

Salt Lake City, UT 84112 USA 


olivera.biology@gmail.com 


and Biotechnology 


Department of Biology 
University of W. ashington 
Seattle, WA 98185 USA 


Patrice Showers Corneli 
Department of Biology 


Salt Lake City, UT $4112 USA 


James A. Villanueva 
Institute of Chemistry 
University of the Philippines 
Diliman, Quezon City 1101, 
PHILIPPINES 


Gisela P. Concepcion 
Marine Science Institute 
University of the Philippines 
Diliman, Quezon City 1101, 
PHILIPPINES 


Ameurfina D. Santos 
National Institute of Molecular Biology 


University of the Philippines 
Diliman, Quezon City, PHILIPPINES 


ABSTRACT 


The Indo-Pacific cone snail Conus furvus Reeve, 1843, is 
shown to be molluscivorous, based on four lines of evidence: (1) 
Specimens of Conus furvus maintained in aquaria were directly 
observed to sae devour other gastropods, including 
other cone snails; (2) Its radular teeth share morphometric 
characteristics with ce known molluscivorous Conus and dif- 
fer from those of vermivorous and piscivorous species; (3) Mo- 
lecular phylogeny places Conus furvus within a clade of mol- 
luscivorous species; (4) Sequences of two peptide toxins are 
most similar to conotoxins prey jiously characterized from mol- 
luscivorous Conus species; one of these closely related peptides 
was previously shown to block molluscan L-type Ca channels. 
Similar to molluscivorous species such as Conus omaria and C. 
victoriae, C. furvus varies widely in shell shape and color pat- 
term, even within its relatively restricted geographic range. 


Additional Keywords: Conoidea, snail-hunting cone snail, radu- 
lar morphology, conotoxins, RT-PCR, phylogeny. 


INTRODUCTION 


Cone snails are venomous marine gastropods (Kohn, 
1959; Réckel, Korn, and Kohn, 1995), and the major 
biologically active components of their venoms are the 
conotoxins (Olivera, 2006), typically small (12-35 amino 


' Author for correspondence 


acid residues), highly constrained peptides. Many cono- 
toxins dices between closely related molecular 
isoforms of a particular ion channel family (Terlau and 
Olivera, 2004); this property makes them valuable tools 
in neuroscience, and confers their therapeutic potential. 
The estimated ~100,000 different pharmacologically ac- 
tive peptides present in living Conus venoms comprise a 
very substantial neuropharmacological resource. The sys- 
tematic study of conotoxins is a se -emingly daunting task 
considering the enormous number of peptides. As the 
conotoxin sequence database expands, it is becoming 
clear that certain types are likely to be produced by spe- 
cies that specialize on particular prey and that belong to 
particular Conus clades (Olivera, 2002, 2006). Thus, es- 
tablishing both the prey and the phylogenetic relation- 
ships of Conus species provides an important guidepost 
for searching for conotoxins with certain types of physi- 
ological activities. 

Traditional cone snail taxonomy is based on shell mor- 
phology; this has proved difficult ‘because of the absence 
of resolution and possible convergence of shell charac- 
teristics. None of the infragenic classificativis schemes 
primarily based on shell shape and sculpture, shell color 
patterns, and radular morphology has received wide- 
spread acceptance (Duda, Kohn, and Palumbi, 2001). 
Thus, these traditional criteria need to be complemented 
by other types of data. 

The specific focus of this study, Conus furvus (“the 


Page 144 


THE NAUTILUS, Vol. 122, No. 3 


dark cone”) is unusual in several respects. Although it is 
found over a relatively restricted area of the tropical 
Indo-Pacific (almost all specimens have been collected in 
the Philippines, where it is quite common in some lo- 
calities), its shell shape and color pattern vary widely 
(Figure 1). In most localities, the typical form of Conus 
furvus has a brown and white shell. However, in the 
Southwestern Philippines, 
forms occurs, particularly along the island of Palawan 
and in the Sulu Sea (Figure 1). As new localities are 
explored, additional variations are being discovered. In 
the comprehensive treatment of Indo-Pacific Conus spe- 
cies by Réckel et al. (1995), the number of taxa synony- 
mized with Conus furvus was probably among the high- 
est for any species. The authors discussed abang a dozen 
distinct forms which they regard as geographic variations. 
Rockel et al. (1995) concluded that Conus furvus * ‘ap- 
pears: to be characterized by isolation of local popula- 
tions.” In general, the occurrence of sinistral spe ciumens 
in any Conus is exceedingly rare; another notable feature 
of C. furvus is that most sinistral Conus specimens col- 
lected in the Philippines belong to this species (an ex- 
ample is shown in Figure 1). 

In a study using the calmodulin intron sequence as a 
phy logenetic mar Tae Duda and co-workers (Duda et al., 
2001: “Duda and Kohn, 2005; Duda and Palumbi, 2004) 
grouped C. — with C. litteratus in their phylogenetic 
tree. They inferred that the two species Ne erged only at 
the beginning of the Pliocene; since litteratus is a 
well- gorbliened vermivorous species i feeds on capi- 
tellid polychaetes (Kohn and Nybakken, 1975), with a 
fossil record going to the Lower Miocene, this strongly 
suggested that C. furvus is a vermivorous Conus. 

Purchasers of Conus furvus in the Guimaras Islands 
public market, in the Philippines, make a soup by boiling 
the snails with tomatoes, onions, and lemongrass. Only 
the portions of C. furvus near the foot are eaten; appar- 
ently the sections near the hepatopancreas have a bitter 
taste and ere uny texture. An alternative culinary use of 
the snails is as an ingredient for a local dish prepared 
with noodles. Various other snails are ty pically sold with 
it: C. radiatus, a piscivorous spe cles, is the most common 
other Conus found for sale with C. furvus at this locality. 

In this paper we present data on prey capture ena 
by C. furvus, the morphometric characteristics of i 
radular teeth, a phylogenetic analysis based on stand: rd 
mitochondrial sequences, and the first conotoxin se- 
quences obtained from this species. These data consis- 
tently support the conclusion that Conus furvus is a mol- 
luscivorous species. 


MATERIALS AND METHODS 


Specimen Collection and Dissection: Specimens of 


C. furvus for DNA extraction were purchased ina public 
market in Guimaras Island, Philippines. The snails were 
dissected to remove the hepatopancreas for total geno- 
mic DNA extraction and the venom duct for RNA ex- 


a remarkable diversity of 


traction. A few specimens were maintained in an 
aquarium. Specimens for radular tooth analyses were 
collected in Coron Harbor and Bobok Island, Palawan, 
Philippines, and preserved in alcohol. The shells are 
deposited in the Field Museum of Natural History, 
Chicago, FMNH 300120, 300121, and 300122. The body 
of specimen No. 300121 is preserved in the Bernice P. 
Bishop Museum, Honolulu, Hawaii, BPBM 248751. 


Analysis of Radular Teeth. The radular sacs of four 
specimens were dissected and 4-6 teeth from each 
mounted permanently in polyvinyl lactophenol medium 
on slides. Barb and blade lengths were measured from 
the tip of a tooth as described | xy Kohn et al. (1999), 
using an ocular micrometer and Nomarski differential 
interference contrast optics. 


Genomic DNA Extraction: The total genomic DNA 
was extracted using the Xanthogenate DNA Extraction 
Protocol (Tillet and Neilan, 2000), a method based on 
the use of the polysaccharide solubilizing chemical xan- 
thogenate. The method is non-toxic and requires no en- 
zymatic or mechanical steps to break cells. 


RNA Extraction: Frozen venom duct in RNAlater 
(Ambion, Austin, Texas) was thawed in ice prior to RNA 
extraction, and a sample of the tissue (10-l5mg) was 
homogenized in 1 mL Trizol reagent (Invitrogen, Carls- 
bad, California). The total RNA was extracted according 
to the manutacturer’s recommendation. 


Amplification of the Mitochondrial 12S and 16S 
rRNA and mtCOI Genes: For PCR amplification, 
primer pairs designed to hybridize to a segment of 12S, 
16S rRNA genes and mtCOl genes were used. The 
primer sequences used for the amplification are listed in 
Table 1. PCR amplifications of the three mitochondrial 
genes were carried out in a 10 wh reaction mixture con- 
taining 1X reaction buffer (200mM Tris HCI, 500mM 
KCI, pH 8.4), 0.7 mM of each dNTPs, 0.7 mM of both 
primers, 0.4 unit Taq DNA Polymerase, 3.0 mM MgCl, 
and approximately 50.0 ng of template DNA. PC R am- 
plification was carried out for 40 cycles. Denaturation 
was carried out at 94 °C, annealing at 48°-55 °C and 
extension at 72 °C. The PCR products were visualized on 
1% agarose gel containing 0.1 mg EtBr/mL. Multiple or 
smeared PCR products were further purified using the 
WIZ Prep DNA purification kit (Promega, Madison, 
Wisconsin), otherwise, the terminated reaction was di- 
rectly sequenced, These sequences have been deposited 
at GenBank. 


12S Accession Numbers: — aimmiralis EU682274, auli- 
cus EU682275, bandanus EU682277, dalli EU682281, 
episcopatus EU682283, flavus EU794315, furvus 
[EU682254, geographus EU794316, kintoki EU794317, 
litteratus EU 784318, magus EUT94319, marmoreus 
EU682288, monachus EU794320, obscurus KU794321, 
omaria EU682289, textile EU682296, tulipa EU794322, 
virgo EU794323:; 16S Accession numbers: ammiralis 


Espino et al., 2008 


Figure 1. Some of the different variations of Conus furvus. The brown colored specimens shown in the bottom row, center, are 
similar to those used for this research. Note the sinistral specimen in the lower left hand corner. 


Page 146 


THE NAUTILUS, Vol. 122, No. 3 


Table 1. Primer pairs used for the amplification of the 12S, 16S rRNA gene and mtCO1 gene. 


12S 16S mtCOl 
Upstream primer 5’ AGAG(C/T)G(A/G) 5’ GTTTACCAA 5’ GGTCAACAA 
CGGGCGATGTGT 3’ AAACATGGCTTC 3’ ATCATAAAGATATTGG 3’ 
Downstream primer 5’ TGCCAGCAG 5’ CCGGTCTGA 5’ TAAACTTCAGGG 
(C/T)CGCGGTTA 3’ ACTCAGATCACGT 3’ TGACCAAAAAATCA 3’ 


EU682299, aulicus EU794324, bandanus Ps vate 
dalli EU078935, episcopatus EU078937, flavus 
EU794326, furvus EU682301, geographus RUF 794327, 
kintoki EU794328, litteratus EU794329, magus 
EU078939, marmoreus EU794330, monachus 
EU078938, obscurus EU794331, omaria EU794332, tex- 
tile EU078936, tulipa EU794333, virgo EUT94334. 


RT-PCR of M and O-superfamily Conotoxin 
Genes: Single-stranded cDNA was synthesized from 
the total RNA extracted from the snail’s venom duct 
using oligodT primers and Superscript IH (Invitrogen, 
Carlsbad, California) and following the manufacturer's 
protocol. Double-stranded cDNA was synthesized using 
primers targeted to the conserved regions of the M- and 
O- superfamily genes. The product of the synthesis was 
visualized in a 1% agarose gel containing EtBr (0.1 mg/ 
mL). The terminated reaction cont: aining the PCR prod- 
uct of the correct size (~0.4Kb) was used for the cloning 
reaction. The cloning reaction followed the manufactur- 
ers protocol for the TOPO-TA cloning kit (Invitrogen 
Life Technologies, Carlsbad, California). The clones 
were screened for inserts of the correct size and were 
cultured on a selective medium and incubated overnight 
with shaking (200 rpm) at 37 °C. The plasmids from 
these clones were extracted and sequenced. 


Phylogenetic Analysis: 12S and 16S rRNA sequences 
were aligned using ClustalX, v1.8 ( (Jeanmougin, Thomp- 
son, Gouy, Higgins, and Gibson, 1998). The alignments 
were refined by eye using MacClade 4.08 ( Maddison and 
Maddison, 2001). Trees were optimized using the indi- 
vidual rRNA alignments and the sonicated align- 
ments (prese snted herein). To account for the complexity 
of sequence evolution, final analyses were restricted to 
model-based maximum likelihood (PAUP4b10, (Swof- 
ford, 2002)) and Bayesian inference (Huelsenbeck and 
Aonquist, 2001). Each method used maximum likelihood 
parameters describing sequence evolution that were op- 
timized with a GTR+I+G model that includes six pos- 
sible substitution types (GTR) and allows some sites to 
be invariant (1), allows across-site rate heterogeneity (G) 
and allows unequal base frequencies. 

The Bz esian ani lysis ran for 100,000 generations 
(sampled every 100 generations) with the first 25,000 
generations discarde a. as burn-in trees. Two MCMCMC 
runs (metropolis-coupled Monte-Carlo markoy-chain), 
using four chains each, were used to thoroughly explore 
tree space Convergence of the likelihoods was judge od 
adequate by monitoring the MrBayes output describing 
(ASED) in 


the average standard error of the difference 


split frequencies between the two runs and by comparing 
plots of the tree log-likelihood trees from generation 
50,000 to LOO. By the! last generation, the ASED was less 
than 0.004: the plot of likelihoods versus generation had 
stabilized. Furthermore, the PSRF (Potential scale re- 
duction factor) reached 1.00 for the total tree length and 
for each model parameter. 


RESULTS 


Feeding Observations: Individuals of Conus furvus 
were frequently observed to prey on other cone snails in 
the aquarium. Most observations were made with C. ra- 
diatus as the prey. Upon introduction of prey into an 
aquarium tank with C. furcvus, immediate behavioral re- 
sponses were observed both from predator and prey. The 
prey immediately crawled away from C. furvus and at- 
tempted to bury itself in the sand. At almost the same 
time, C. furvus pointed its siphon toward the retreating 
snail and crawled in its direction. While still relativ ely far 
away from the prey, C. furvus extended its proboscis, and 
inserted it into the body of its prey with one jerking 
motion. Immediately the prey was observed to stop mov- 
ing. Conus furvus then maneuvered the immobilized 
prey so that its aperture was turned upwards, away from 
the sand, after which C. furcus commenced feeding on 
the upturned C. radiatus. 


Analysis of Radular Teeth: Radular tooth length 
ranged from 6.4 to 8.9% of shell length (mean = 7.5%) in 
the four specimens examined. Each tooth is armed with 
a barb that extends for 5% of the tooth length from the 
tip (range 4.6-5.4%), and a blade that extends for 9% of 
the length from the tip (range $.9-9.3%). This morpho- 
metric data clearly identifies the teeth as those of a mol- 
luscivorous species, as noted ie Nybakken (1990; see also 
Kohn, et al., 1999: Nishi and Kohn, 1999). However, 
these workers did not present details of C. furvus tooth 
morphometry, which differs markedly from previously 
examined molluscivorous Conus species. Nishi and Kohn 
(1999) divided the eleven species they studied into three 
groups based on both discrete and quantitative charac- 
ters. The former, i.e. presence of one barb, one blade, 
and continuous serration clearly place C. furvus in Group 
A, with C. araneosus, C. bandanus, and C. marmoreus. 
Groups B and C have teeth with two barbs and no blade 
(Table 2). However, C. furvus differs from the Group A 
species in its quantitative characters. Its teeth are more 
than twice as long relative to shell length, thus matching 
the members of Group C rather than Group A (Table 2). 
The C. furvus teeth also differ from Group A and more 


Espino et al., 2008 


Page 147 


Table 2. 


Comparison of Conus furvus radular teeth with those of other molluscivorous species according to qualitative (discrete) and 


quantitative (continuous) characters. N, Number of specimens measured. Qualitative characters: B1, First barb; B2, Second barb; BL, 


Blade; SE, Serration; +, present; —, absent; C, Continuous; D, Discontinuous. $,, Shell length; 


B,, Blade length; SE, 
C from Nishi and Kohn (1999). 


Qualitative 
characters 


T,, Tooth length; B1,, Barb length; 


. Serration length; BA\, Base width; SH\,, Shaft width. All dimensions are in mm. Data on Groups A, B, and 


Quantitative characters 


Group Species N BL B2 BL SE T;/Sy, Bly/T,., B2,/T,. BL,/T;, SE,/T, BA\/T,, SH\\/Ty, 
C furvus 4 + - + G 0.07 0.05 —- 0.09 0.70 0.06 0.04 
(0.04—0.05) 
A C. araneosus > Ss = “:, JC 0.03 0.08 — O13 0.49 O11 0.05 
C. bandanus > (0.026—.033) (O.07—0.08) (O.13-0.14) (0.33-0.68) (0.09-0.12) (O0.04—0.5) 
C. marmoreus 15 
BC. ammniralis A, a oe HE 0.05 0.04 0.06 = 0.74 0.05 0.03 
C. canonicus 7 (0.03-0.08) — (O.03—0.05)  (0.05—0.09) (0.62—0.81) (O0.04—0.08) (0.02—0.05) 
C. nodulosus 5 
C. textile 15 
C. victoriae 4 
Cc C. episcopatus 11 + + - D 0.07 0.05 0.08 0.65 0.07 0.05 
C. omaria 9 (O.07—0.08) — (O0.04-0.03) (O.07—0.09) (0.67—0.69) (0.06—0.08)  (0.04—0,05) 
C. pennaceus 30 


closely resemble Groups B and C with respect to all 
four other quantitative characters that they share (Ta- 


ble 2) 


PCR Amplification and Phylogenetic Reconstruc- 
tion: A phylogenetic tree based on two standard mark- 
ers, 12S and 16S rRNA was constructed as described 
under Methods. Conus furvus clustered on the same 
branch with species previously shown to feed on other 
mollusks (Figure 2). In contrast, Conus furvus was well- 
resolved from C. litteratus, the worm-hunting species 
postulated previously to be C. furvus’s closest relative. 
The hypothesis that C. furvus is closer to the C.litteratus 
than to the other mollusk hunting snails is significantly 
rejected (p<0.0001) using Kishino- Hasegawa (Kishino 
and Hasegawa, 1989) and Shimpceaea- Hasegawa (Shimo- 
daira and Hasegawa, 1999) tests comparing the respec- 
tive log- likelihood scores. The hypothesis that C. furvus 
and C. litteratus are sister species is significantly rejected 
by these tests using the 12S and 16S aa alone. Because 
of the great variation of forms presently assigned to Co- 
nus furvus, we also obtained a “bar code” sec mare (CO] 
gene) from the specimens analyzed, which is deposited 


in GenBank. 


Mapping Radular Tooth Characters to the Molecu- 
lar Phylogenetic Tree: Morphometric data on radu- 
lar teeth were available for seven of the nine molluscivo- 
rous species on the molecular tree (Figure 2), and the 
values of the seven quantitative morphometric characters 
for these species shown on the right side of Table 2 were 
entered into a discriminant function analysis. A phyloge- 
netic tree generated from the proximity matrix c ee ulated 
from the centroids of the first two canonical variates 
they accounted for 99.6% of the variance) (Figure 2B) 


was completely congruent with the conotoxin gene-based 
tree (Figure 2A). 


O- and M- Superfamily Genes from C. furvus: — Primer 
pairs designed to hybridize to the conserved regions of 
the O- | M-supe family conotoxins successt fully ampli- 
fied conotoxin genes from the C. firvus venom duct. 
Translation of the amplified sequences resulted in amino 
acid sequences characteristic of the canonical arrange- 
ment of conotoxin precursors. Both of the sequences 
have a hydrophobic N-terminal signal sequence, which 
is highly conserved among conotoxins belonging to the 
same superfamily, an intervening pro region, and the 
mature toxin region at the C-terminus. 

An O- superfamily conotoxin (Fr7.1) from C. furvus 
was successfully amplified and sequenced, and a 264- 
base sequence containing an open reading frame coding 
for the conotoxin was obtained. A 302- age sequence 
coded for a second conotoxin from C. firvis (Fr3.1) that 
belongs to the M-superfamily was also cloned and se- 
quenced. 

The O-superfamily conotoxin from C. furvus (Fr7.1) 
was strikingly similar to an w—conotoxin isolated from 
Conus textile (Fainzilber et al., 1996), wTxVIT (Figure 
3A). Alignment of the amino acid sequences shows 
§2.1% silty wTxVITA has previously been shown to 
target L-type Ca channels in mollusks; given the high 
sequence similarity, Fr7.1 is likely a voltage -cated Ca 
channel ante wonist. A comparison of the M- -superfamily 
conotoxins from C. furvus (Fr3.1) with M-supert family 
conotoxins from C. textile (Tx3.5) and C. marmoreus 
(Mr3.5) also shows a high degree of similarity (Figure 
3.B). Alignment of Fr3.1 with Tx3.5 shows the sequences 
to be 84.4% identical while Fr3.1 aligned with Mr3.5 
shows the sequences to be 67.2% identical. The 


Page 148 


THE NAUTILUS, Vol. 122, No. 3 


C. ammiralis 


A 100 


C. dalli 

C. textile 

C. aulicus 

C. episcopatus 

C. omaria 

C. bandanus 

C. marmoreus 

C. furvus 

C. flavus 

C. geographus 
C. obscurus 


C. tulipa 


C. magus 
C. 


monachus 


C. kintoki 


C. virgo 


C. litteratus 
0.05 substitutions/site 


B C. ammiralis 


C. textile 


C. episcopatus 


C. omaria 


C. bandanus 


10 units 


C. marmoreus 


C. furvus 


Figure 2. A. Optimal maximum likelihood tree from a Bayesian analysis . 12SrRNA and 16SrRNA sequences estimating the 
evolutionary relationship among some Conus species. The uppermost cli ide, from Conus ammiralis to C. furvus, comprises mollus- 
civorous species. Branch labels are Bayesian support values summarizing confidence in each split. B. Ne ighbor-joining UPGMA tree 
from analysis of radular tooth morphometry, based on euclidean distance matrix of centroids of canonical variates | and 2 determined 
from discriminant function analysis of 7 radular tooth variables, measured as in Nishi and Kohn (1999). 


Fr3.1 sequence clearly belongs to the m-3 branch of the 


M-conopeptide supe fi umily, which is characteristic of 


molluscivorous Conus species (Corpuz et al. , 2005). The 
two C. furvus conotoxin sequences are most similar to 
conotoxins from other molluscivorous species, consistent 
with the behavioral observations made above and with 
the phylogenetic tree shown in Figure 2. 


DISCUSSION 


In the central Philippines, Conus furvus occurs on rocky 
intertidal habitats. At low tide, individuals can be found 
in depressions on the substrate, together with other mol- 
usks; this is where C. furvus is generally collected by 
ishermen in Guimaras Island, to be sold with other com- 
mercial shellfish at the public market. Conus furvus was 
ormerly abundant along the shores of Guimaras Island, 
yut an oil spill in the area in August 2006 threatened the 
marine life furvus became much more scarce at this 
ocality after the spill, but the population appears to be 


ecovering. 

From the results presented above, we conclude that 
Conus furvus is a molluscivorous species. This conclu- 
ion is based on direct observation of C. furous attacking 


gastropod prey, on morphometric characteristics of the 
radular teeth that are shared with other molluscivorous 
Conus, by the high degree of sequence similarity be- 
tween toxin precursors in the venom duct of Conus fur- 
vus and previously described conotoxins from other mol- 
luscivorous Conus, and by its phylogenetic proximity to 
them based on molecular markers. A previous molecular 
result suggesting that C. furvus might be vermivorous 
(Duda et al., 2001; Duda and Kohn, 2005) is probably 
erroneous. 

An unusual feature of the Conus furvus attack on C. 
radiatus is that the prey was apparently injected with 
venom only once; multiple stinging was never observed. 
All other molluscivorous Conus species previously ob- 
served attack prey using multiple injections of venom 
before starting to feed on the immobilized prey (Yoshiba, 
1983, 1987: Kohn, 2003). The reason for this behavioral 
difference is unclear, but the venom composition could 
well reflect it. A further biochemical characterization of 
Conus furvus venom is in progress (S. Espino and G. 
Concepcion, unpublished results). 

The phylogenetic relationship between C. furvus and 
other Conus species evaluated for 12S and 16S rRNA 


gene sequences (Figure 2) is consistent with C. furvies 


Espino et al., 2008 


A 
Fr7.1 SC TPRGGQCGYYNDCCSHQCNINRNLCE 


w-Txvil YCTPHGGHCGYHNDCCSHQCNINRNKCE 


Fr3.1 MSTLGVLLTICLLLFSVTALPLDGDQPVDLAAERMKAEQHPLFDQKRRCCKFPCANSCRHLCCG 
Tx3.5  MSKLGVLLTICLLLFPLTALPLD GDQPADQAAERMQAEQHPLFDQKRRCCKFPCPDSCRYLCCG 
Mr3.5 MSKLGVLLTICLLLFPLTALPLD GD QPAD QRAERTQAEKHSLPDPRMGCCPFPCKTSCTTILCCG 


Mature toxins: 

Fr3.1 RCCKFPC|ANS CRIH/ILCC * 
Tx3.5 RCCKFPC|PD/SCRIY|LCC* 
mr3.5 MGCCPIFPCK Ts Cit TILCC* 


Figure 3. A. Alignment of the toxin region of Fr7.1 from 
Conus furcus and ot from Conus textile (GenBank Nuc. 
Acc. No. DD012770). B. Alignment of Fr3.1 from Conus fur- 
vus with other M- Gee umily precursors highlighting the con- 
served amino acids. Tx3.5 and Mr3.5 were from Conus textile 
and Conus marmoreus, respectively (Corpuz et al., 2005). The 
predicted mature toxins are also shown; the asterisk denotes an 
amidated C-terminus. 


being more closely related to the mollusk hunting Conus 
species than to C. litteratus (Bayesian clade support = 
9S), a vermivorous species. The several clades of mollus- 
civorous Conus probably have a common ancestor. Using 
the tree based on molecular data in Figure 2A, the fell 
supported ¢ or 7 can be referred to, using the subgenera 
of Marsh (1964) as clade names) as the Conus clade? (with 
Conus marmoreus as the type), the Cylinder clade, (with 
Conus textile as the type), and the Darioconus clade 
(with Conus omaria as the type). Although Conus furvus 
belongs within the larger mollusk hunting clade, includ- 
ing these three groups, its position hin the clade re- 
mains unresolved. Given its lack of a tented pattern, it 
has an unusual shell pattern for a molluscivore. However, 
some specimens of smaller, usually tented species such 
as Conus barbieri and Conus victoriae occasionally have 
plain brown shell variants, not unlike typical specimens 
of Conus furvus. 

A more comprehensive molecular analysis of mollus- 
civorous Conus, including C. furvus, has recently been 
carried out; these cults will be presented elsewhere. 
Additional phylogenetic markers, such as ITS-2 se- 
quences, were aged for this study. This has led to greater 
refinement of the phylogenetic tree and supports the 
basic conclusion that C. furvus is a molluscivore. 


ACKNOWLEDGMENTS 


This work was supported in part by GM48677 from the 
U.S. National Institutes of General Medical Sanaecs (to 
BMO). The support of the University of the Philippines 
in the Visayas for a fellowship aw ard to S.E. to pursue 
graduate studies in Chemistry is gratefully acknowl- 
edged. NSF Grant 0316338 partly supported the re- 
search of AJk. The late J.E. Norton contributed the C 
furvus specimens for radular tooth morphometry, and 
Dr. M. Nishi assisted in the data analysis. 


Page 149 


LITERATURE CITED 


Corpuz, G. 5 R. B. Jacobsen, E. C. Jimenez, M. Watkins, C 
Walker, Colledge, et al. 2005. Definition of the M- 
emcee “superfi amily: characterization of novel peptides 
from molluscivorous Conus venoms. Biochemistry 44(22): 
S1L76-S1S6. 

Duda, T. F., Jr. A. J. Kohn, and S. R. Palumbi. 2001. Origins of 
diverse feeding ecologies within Conus, a genus of ven- 
omous marine gastropods Biological Journal of the Lin- 
nean Society 73: 391409. 

Duda, T.F., Jr. and A. J. Kohn. 2005. Species-level phylogenog- 
raphy and evolutionary history of the hyperdiverse marine 

gastropod genus Conus. Molecular Phylogenetics and 
Evolution 34: 257-272. 

Duda, T. F., Jr. and S. R. Balweniy 2004, Gene expression and 
feeding ecology: evolution of piscivory in the venomous 
gastropod g genus Conus. Proceednigs: Biological Scieneces 
971(1544): : 1165-1174. 

Fainzilber, M., ]. C. Lodder, R. C. van der Schors, K. W. Li, Z. 
Yu, A. L. Burlingame, et al. 1996. A novel hydrophobic 
omega-conotoxin blocks molluscan dihydropyridine- 
sensitive calcium channels. Biochemistry 35(26): 8745— 
8752. 

Huelsenbeck, J. P. and F. Ronquist. 2001. MRBAYES: Bayes- 
ian inference of * phylogenetic trees. Bioinformatics 17: 
754-755. 

Jeanmougin, F., ]. D. Thompson, M. Gouy, D. G. Higgins, and 
T. J. Gibson. 1998. Multiple sequence alignment with 
Clustal X. Trends in Biochemical Sciences 23: 403-405. 

Kishino, H. and M. Hasegawa. 1989. Evaluation of the maxi- 
mum likelihood estimate of the evolutionary tree topolo- 
gies from DNA sequence data, and the branching order in 
hominoidea. Journal of Molecular Evolution 29: 170-179. 

Kohn, A. J. 1959. The Ecology of Conus in Hawaii. Ecology 
Monograph 29: 47-90. 

Kohn, A. J. a The feeding process in Conus victoriae. In: 
Wells, D. I. Walker, and D. S. Jones (Eds.) The 
Marine ae and Fauna of Dampier, Western Australia, 
Volume 1. Western Australian Museum, Perth, pp. L01— 
107. 

Kohn, A.J., M. Nishi, and B. Pemet. 1999. Snail spears and 
scimitars: A character analysis of Conus radular teeth. 
Journal of Molluscan Studies 68: 461481. 

Kohn, A. J. and J. W. Nybakken. 1975. Ecology of Conus on 
eastern Indian Ocean fringing reefs: diversity of species 
and resource utilization. Marine Biology 29: 211-234. 

Maddison, D. R. and W. P. Maddison. 2001. MaClade version 
4.08. Sinauer Associates: Sunderland, Massachusetts. 

Nishi, M. and A. J. Kohn. 1999. Radular teeth of Indo-Pacific 
molluscivorous species of Conus: A comparative analysis. 
Journal of Molluscan Studies 68: 453-497, 

Nybakken, J. 1990. Ontogenetic change in the Conus radula, its 
form, distribution among the radula types, and signifi- 

cance in systematics and ecology. Malacologia 32: 35-53. 

Olivera, B. M. 2002. Conus venom peptides: reflections from 
the biology of clades and species. Annual Review of Ecol- 
ogy and Syste matics 33: 25-47. 

Olivera, B. M. 2006. Conus peptides: biodiversity-based dis- 
covery and exoge nomics. Journal of Biological Chemistry 
281(42): 31173-3117 

Réckel, D., W. Korn, and A. J. Kohn. 1995. Manual of the 
Living Conidae. Vol. 1, Indo-Pacific Region. Verlag 
Christa Hemmen, Wiesbaden, 517 pp. 

Shimodaira, H. and M. Hasegawa. 1999. Multiple Comparisons 


Page 150 


of Log-Likelihood with Applications of Phylogenetic 
Inference. Molecular and Biology Evolution 16: 1114— 
1116. 

Swofford, D. L. 2002. PAUP® Phylogenetic analysis using par- 
simony (* and other methods). Version 4. Sunderland, 
Massachusetts: Sinauer Associates. 

Terlau, H. and B. M. Olivera. 2004. Conus venoms: a rich 
source of novel ion channel-targeted peptides. Physiologi- 
cal Reviews 84: 41-68. 


THE NAUTILUS, Vol. 122, No. 3 


Tillet, D. and B. Neilan. 2000. Xanthogenate Nucleic Acid Iso- 
lation from Culture and Environmental Cyanobacteria. 
Journal of Phycology 36; 251-2558. 

Yoshiba, S. 1983. [Feeding behavior and diet of Conus banda- 
nus, especially a comparison with those of Conus textile]. 
Venus 42: 63. (Abstract) [in Japanese] 

Yoshiba, S. 1987. [Feeding and tooth shooting frequencies of 
Conus textile for 5 years in an aquarium.| Venus 46: 52. 
(Abstract) [in Japanese | 


THE NAUTILUS 122(3):151-154, 2008 


Page 15] 


Two new species of Mitrella (Gastropoda: Neogastropoda: 
Columbellidae) from the lower Miocene Chipola Formation of 


northwestern Florida 


Richard Duerr 

Post Office Box 1055 
Okeechobee, FL 34973 USA 
pdiegel@gmail.com 


ABSTRACT 


Two new species of Mitrella from the lower Miocene Chipola 
Formation of are described. Mitrella hayesorum new species 
and Mitrella phyllisae new species are found primarily in 
Chipola Formation exposures along Farley Creek in northwest- 
ern Florida, USA. Placement of the new species in the genus 
Mitrella is tentative. 


Additional Keywords: Neogene, ultraviolet light, gastropod, 
Astyris 


INTRODUCTION 


Species assigned to the genus Mitrella are widely distrib- 
uted in the warm and shallow waters of the Recent seas 
of the world ( 
Mitrella is not clear, but some of the earliest represen- 


tative of the genus from the southeastern portion of 


North America appeared in the Eocene (MacNeil and 
Dockery HI, 1954; Palmer, 1937). Maury (1910) de- 
scribed three Mitrella (as Astyris) from the Chipola For- 
mation. These were originally deposited in the Cornell 
University collection at lead. New York, and now de- 
posited at PRI. With more specimens than Maury had at 
her disposal, Gardner (1947) described 13 species and 
two subspecies of Mitrella from the Chipola Formation. 

Mitrella have been found in all exposed facies of the 
Chipola Formation on the Chipola River, Tenmile Creek 
and Farley Creek, in the Chipola River drainage in Cal- 
houn County, Florida. They are also present in the 
Chipola Formation exposed in the lower bed at Alum 

Bluff in Liberty County, Florida. The two new species 
described herein have been collected only in the Chipola 
Formation exposures in the Chipola River drainage, pri- 
marily at Farley Creek. Vokes (1989) stated the C hipola 
facies along Farley Creek where the two new species are 
most prevalent is a bivalve-rich miliolid lime-sand with 
many calcareous algae and coral heads, and assumed it 
was a shallow back-reef environment. 

Some specimens were photographed under ultraviolet 
light (UV) to facilitate visualization of color patterns. In- 


Gardner, 1947, 1948). The exact origin of 


stitutional abbreviations are: UF: Florida Museum of 
Natural History (FLMNH for locality records), Univer- 
sity of Florida, Gainesville; PRI: Paleontological Re- 
search Institution, Ithaca; ANSP: The Academy of Natu- 
ral Sciences, Philadelphia, AMNH: American Museum 
of Natural History, New York. 


SYSTEMATICS 


Superfamily Buccinoidea Rafinesque, 1815 
Family Columbellidae Swainson, 1940 
Genus Mitrella Risso, 1826 


Diagnosis: Small to very small, smooth, fusiform 
shells. Sculpture, if present, of incised spiral lines. Spiral 
sculpture is generally restricted to the anterior end near 
the base: axial sculpture is, with few exceptions, entirely 
absent. The outer lip is dentate in adult specimens (Di- 
agnosis according to Campbell [1993] and Keen [1971]). 


PMitrella hayesorum new species 
(Figures 1-3, 10) 


Description: Shell fusiform, narrow. Height of holo- 
type 19.Smm. Protoconch mammillated, with about two 
smooth, rounded whorls, the second expanded, with no 
perceptible protoconch/teleoconch transition. Teleo- 
conch with ten slightly convex, shiny whorls with micro- 
scopic irregular spiral grooves, axial sculpture lacking. 
Spire elev ated with somewhat concave sides. Suture dis- 
tinct. Aperture elongated, less than half the length of the 
entire shell. Outer lip sinuous, varicose externally, mar- 
gin thin and sharp. Basal lip extending slightly beyond 
pillar. Posterior canal produce sd and thickened externally. 

Denticles present on inner surface of outer lip, ten to 12 
in number, weak on anterior third, strongest on central 
third, frequently absent on posterior third. Parietal wall 
with two axial ridges, outer ridge weak with small raised 
beads reflecting underlying sculpture on pillar. Inner 
ridge prominent with median notch, ac lapical half of 
ridge strong, abapical half tapering and completely dis- 
appearing abapically. Base of pillar with ee nine 


nnn. —— Ssh —R|™?™o>s4)y wa 


THE NAUTILUS, Vol. 122, No. 3 


R. Duerr, 2008 


ol 
(ey) 


Page 15: 


oblique grooves with a rounded summit between 
grooves. Dark band about one third width of penultimate 
whorl appears below suture on last three whorls of ho- 
lotype. Band has irregular narrow, oblique, light colored 
lines, some forming u- or v-ligures. 


Type Material: Holotype: UF 119655, a ight 
19.Smm, width 5.6mm: Par: itypes: ANSP-IP $1324, one 
specimen; AMNH-FT 433 312, one specimen; PRI ye 
one specimen; Diegel-Duerr collection, one specimen. 
All from type locality. 


Type Locality: FLMNH locality Farley Creek 07 
(CA022), Farley Creek east of SR 275, Calhoun County, 
Florida (to protect privacy rights of landowners, spe cific 
locality information is avail only to qualified re- 
searchers upon written request to the author or the In- 
vertebrate Paleontology Division of the FLMNH), 
Chipola Formation. 


Distribution: Chipola Formation along Tenmile and 
Farley Creeks and the Chipola River, Calhoun County, 
Florida. 


Etymology: Named to honor the forestry-oriented 
Hayes family who have graciously granted the author, 
and others, permission to collect on their property. 


Remarks: The species currently placed in Mitrella 
comprise a complex group and may have been differe tly 
assigned to the genera Alia, Astyris, Nitidella by diff erent 
authors (Keen, 1971). The taxonomic position of M. 
hayesorum is questionable and its current assignment to 
Mitrella is tentative. Mitrella hayesorum has unique 
characters, such as the bulbous second whorl of the pro- 
toconch, extended slightly concave spire and wide sipho- 
nal canal extending below the pillar, and cannot be con- 
fused with any other known Mitrella. Al though no speci- 
mens of Mitrella dalli ( Maury, 1910), the species in the 
Chipola Formation closest in form to M. hayesorum, 
were available for study, Maury’s figure of M. dalli (1910: 
pl. 6, fig. 2) indicates a smaller she TL ( 12 mm), a shorter, 
stouter spire, and a narrower anterior canal than M. 
hayesorum. The nearest European fossil congener of Mi- 
trella hayesorum is Mitrella (Mac rurella) nassoides 
(Grateloup, 1827) (Figures 4-6) from the early Pliocene 
of Italy, which is larger, wider at the midbody, and has a 
narrow anterior canal. The maximum height of all speci- 
mens of Mitrella hayesorum exé mined is 19.93 mm, 
minimum height is 18.17 mm. Exposure of M. haye- 
sorum to ultraviolet light (Figure 10) reveals a fuores- 
cent pattern of axial flammules in addition to the narrow 
band below the suture on the anterior whorls of the new 


species. Seven specimens of the most common Mitrella 
in the Chipola Formation, Mitrella ischna Gardner, 
1947, a more robust species than M. hayesorum, were 
examined under UV light and revealed fine filamentous 
lines covering all tele oconch whorls. Three specimens of 
Mitrella asema Gardner, 1947, a shorter species than M. 
hayesorum, exhibits, under UV light, narrow axial bars 
running from suture to suture, offset in alignment from 
previous whorls, with four per whorl on most teleoconch 
whorls. 


?Mitrella phyllisae new species 
(Figures 7-9, 11) 


Desempien: Shell small, fusiform. Height 7 mm, 
width 2.2 mm. Protoconch with two smooth wlicels sec- 
ond seh enlarged. Teleoconch with five smooth, con- 
vex whorls, without axial sculpture. Spire sides. slightly 
concave. Suture impressed. Aperture less than half the 
length of entire shell. Outer lip with slight varix, margin 
in and sharp, usually dentate within. Thin parietal wi ach 
saa About seven impressed oblique grooves cross 
base of pillar, separated by narrow bands a rounded 
summits. Faint round spots, eight on last whorl of holo- 
type, appear on last three whines: Spots about one- 
quarter the height of penultimate wher! in diameter rest 


just above median line of the whorls. 


Type Material: Holotype, UF 119656, height 7.0 mm, 
width 2.2 mm; Paratypes, ANSP-IP $1325, one speci- 
men; AMNH-FT 43313, one specimen; PRI 8353, one 
specimen; Diegel-Duerr collection, one specimen. All 
from type locality. 


Type a FLMNHEL locality Farley Creek 07 
(CA022), Farley Creek east of SR 275, Calhoun County, 
Florida, (to protect privacy rights of landowners, specific 
locality information is available only to qualified re- 
searchers upon written re quest to the author or the In- 
vertebrate Paleontology Division of the FLMNH), 


Chipola Formation. 


Distribution: Chipola Formation along Tenmile and 
Farley Creeks and the Chipola River. Calhoun County, 
Florida. 


Etymology: Named for Phyllis Diegel, the author's 
companion and a knowledge able conchologist and pale- 
ontologist. 


Remarks: — As with Mitrella hayesorum, the assignment 
of M. phyllisae to Mitrella is tentative. A cursory inspec- 
tion would indicate Mitrella phyllisae to be a dwarf M. 
hayesorum. Closer examination reveals that M. phyllisae 


Figures 1-11.—WMitrella species. 1-3. Apertural, lateral, and abapertural views of the holotype of Mitrella hayesorum new species, 
U F 119655, height 19.8 mm, width 5.6 mm, Burdigalian Miocene. a 6. Apertural, lateral, and ab: apertural views of Mitrella nassoides 


Grateloup, 1827 


) UF 119657, height 25.5 mm, width 8.6 mm, Zanclean Pliocene, from Liguria County, Ceriale, Italy, for comparison 


with M. hayesorum. 7-9. Apertural, lateral, and abapertural views of the holotype of Mitrella pluyllisae new species, UF 119656, 
height 7.0 mm, width 2.2mm, Burdigalian Miocene. 10, 11. Mitrella exposed to UV light. 10. fol e of Mitrella hayesorum (same 
specimen as Figure 3) showing UV exposed pattern. 11. Holotype of Mitrella phyllisae (same specimen as Figure 9) showing UV 


exposed pattern. 


Page 154 
§ 


differs from the much larger M. hayesorum by the 
shorter spire, the slightly more impressed suture, and 
fewer oblique incised lines on the base of the shell. A 
series of dots about 1 mm in diameter are visible on M. 
phyllisae encircling all teleoconch whorls, one-third the 
length of the whorl below the suture, eight on the last 
whorl. Exposure to UV light reveals (Figure 11), in ad- 
dition to the larger dots, a field of minute dots covering 
the entire teleoconch, somewhat reminiscent of the pat- 
tern on the Recent Mitrella ocellata (Gmelin, 1791) from 
the western Atlantic. Also, the anterior portion of the 
outer lip of M. phyllisae is less developed than that of M. 
hayesorum, which is wider and extends beyond the pillar. 
The height of all specimens of M. phyllisae examined 

varies lees than 1.0 mm, from a maximum of 7.4 mm to 
a minimum of 6.8 mm. 


ACKNOWLEDGMENTS 


The author extends his appreciation to Roger Portell, 
FLMNH, for providing the plates and commenting on an 
earlier draft of the manuscript. Digital images were pre- 
pared by Sean Roberts, FLMNH. Thanks also to Mare 
Grigis, Belgium, for donation of a specimen of Mitrella 
ee to Dr. Bernard Landau, Portugal, for encour- 
agement and advice, and to Burke and Brooks Hayes 
and also Archie and Vicki Whittington for permission to 
collect on their respective properties in North Florida. 
Pamela McBride provided secretarial assistance. | am in- 


THE NAUTILUS, Vol. 122, No. 3 


debted to two anonymous reviewers whose comments 
and suggestions improved the manuscript. 


LITERATURE CITED 


Campbell, L. D. 1993. Pliocene mollusks from the Yorktown 
and Chowen River Formation in Virginia. Virginia Divi- 
sion of Mineral Resources, Publication 127: vii + 173 pp., 
43 pls.. 

Gardner, J. 1947. The molluscan fauna of the Alum Bluff 

Group of Florida. Part VII. Ctenobranchia (remainder), 
Aspidobranchia and Scaphopoda. United States Geologi- 
cal Survey. Professional Paper 142-H: i-ii + 493-638, pls. 
52-62. 

Gardner, J. 1948. Mollusca from the Miocene and Lower 
Pliocene of Virginia and North Carolina. Part 2. 
Scaphopoda and Gastropoda. United States Geological 
Survey. Professional Paper 199-B: i-iii +179-279, pls. 24— 
38. 

Keen, A. M. 1971. Sea Shells of Tropical West America. Marine 
mollusks from Baja California to Peru. Second Edition. 
Stanford University Press, Stanford, California: xiv + 1064 
pp., 22 pls. 

Mac Neil, F. S. and D. T. Dockery IIL. 1984, Lower Oligocene 
Gastropoda, Scaphopoda, and C Jephalopoda of the Vicks- 
burg Group in Mississippi. Mississippi Bureau of Geology. 
Bulletin 124: 415 pp., 72 pls. 

Maury, C. J. 1910. New Oligocene (Miocene) shells from 
Florida. Bulletins of American Paleontology 4(21): 119— 
164, pls. 1S—26. 

Vokes, E. H. 1989. An overview of the Chipola Formation, 
northwestern Florida. Tulane Studies in Geology and Pa- 
leontology 22: 13-24. 


THE NAUTILUS 122(3):155-165, 2008 


Page 155 


Catalogue of the type material of mollusks deposited at the 
Zoology Museum, University of Costa Rica 


Fresia Villalobos-Rojas 
Ana. G. Guzman-Mora 
Escuela de Biologia 
Universidad de Costa Rica 
Apdo. 2060, San Pedro 

San José, COSTA RICA 


Yolanda E. Camacho-Garcia! 

Escuela de Biologia and Museo de Zoologia 
Universidad de Costa Rica 

Apdo. 2060, San Pedro 

San José, COSTA RICA 
yeamacho_99@yahoo.com 


ABSTRACT 


This catalogue compiles the basic data for mollusk type mate- 
rial deposited at the Zoology Museum of the University of 
Costa Rica. It includes 62 holotypes and 151 paratypes, ee 
senting 68 species and 41 genera. The species, authors, date of 
publication, bibliographic references, type locality, catalogue 
number, number of specimens, state of preservation, and re- 
lated information are indicated for each taxon. In some cases, 
remarks on the current taxonomic status and other useful in- 
formation are provided. For all species, except for opistho- 
branchs, the shell of either the holotype or one of the paratypes 
is illustrated. 


Additional Keywords: Holotype, paratype, Mollusca, mollusk 
collection 


INTRODUCTION 


The Zoology Museum of the University of Costa Rica 
(MZUCR) was founded in 1960 by UCR professors 
Douglas C. Robinson and William Bussing. The collec- 
tion consists of both invertebrates and verchites: and 
includes birds, fishes, mammals, reptiles, amphibians, tu- 
nicates, decapods, sipunculids, sponges, corals, and mol- 
lusks, among other groups. The specimens come from a 
wide variety of research projects, and donations from 
national and international researchers, as well as amateur 
naturalists. Most of the material is preserved in a wet 
collection (70% ethanol), but there is also a dry collec- 
tion. 


THE MOLLUSK COLLECTION AT THE UNIVERSITY OF 
Costa RICA 


Former UCR curator Carlos Villalobos and a great num- 
ber of other researchers started the mollusk collection at 
the Museum in 1964. Since then, UCR staff and visiting 
scientists have continuously added to the collection. In 
addition, a significant amount of material has been do- 


' Author for correspondence 


nated to the Museum. Some examples include the col- 
lections of malacologists George Richard, from Univer- 
sity of La Rochelle, France (mainly specimens of the 
family Conidae and other material from the Indo- 
Pacific), Kristie L. Kaiser, associated with the Santa Bar- 
bara Museum of Natural History (material from Cocos 
Island, Costa Rica), and the Latin American collection of 
independent scientist Dwight W. Taylor (including 
terrestrial and freshwater mollusks as well as type 
material); and amateur naturalists Michael Montoya (a 
valuable collection of marine mollusks from the Me- 
soamerican region, Cocos Island, and the Caribbean), 
Mary Yost (marine mollusks from Guanacaste, Costa 
Rica), Robert Nishimoto and others in 1969 (marine 
mollusks from Puntarenas, Costa Rica), and Jerry Well- 
ington in 1972 (marine mollusks from both the Pacific 
and the Atlantic). In 2005, the collection increased in size 
considerably, having absorbed the mollusk collection of 
the Costa Rican National Biodiversity Institute (INBio), 
which closed its Malacology Department in September 
2004. Currently, the édllecten is comprehensive, in- 
cluding more than 37,670 lots and 216,587 specimens 
representing the molluscan fauna of the Pacific Coast 
of North, Central, and South America (however, there 
is also material from the Indo-Pacific, France, and 
Africa). 

The collection is divided into two main categories: a 
dry collection consisting mainly of the shells of bivalves 
and gastropods, and a wet collection comprising cepha- 
lopods, gastropods, bivalves, polyplacophorans, and ter- 
restrial and freshwater mollusks preserved in alcohol. 

Considering that this collection is the most important 
in the country and perhaps in all of Central America, and 
due to the recent increase in the amount of type mate- 
rial, it is important to compile this information and pub- 
lish a catalogue that addresses in detail the literature 
available for the type material, its location, and illustra- 
tions of the shell of each type represented. We hope to 
facilitate the future work of taxonomists and help locate 
type material, contributing in this respect to their re- 
search. 


Page 156 THE NAUTILUS, Vol. 122, No. 3 


F. Villalobos-Rojas et al., 2008 


Page 157 


TYPE COLLECTION 


This catalogue includes 62 holotypes and 151 paratypes, 
representing 6S species and 41 genera. All types are 


listed alphabetically by species epithet. Photographs of 


opisthobranch types are not included, although they are 
listed in this catalogue. For each taxon, the following 
information is included: the name of the species cite i 
exactly as it was published in the original description, the 
author(s), date of publication, bibliographic references, 
type locality, catalogue number, number of specimens, 
state of preservation, and related information (SEM 
stubs, microscopy slides, egg masses of the specimens). 
In some cases, remarks on the current taxonomic status 
and other useful information are provided. We also in- 
dicate whether the specimen is preserved complete (in- 
cluding the shell and soft parts) or when only the shell 
remains in the dry collection. 

In a few cases, the information in the original publi- 
cation about the location of the type material is not ac- 
curate, and we clarify this in the remarks. For each spe- 
cies, either the holotype or one of the paratypes is pho- 
tographed, and the figure number is properly indicated 
beside the specimen. 

Although the original publications on some type ma- 
terial (Noumea regalis Ortea, Caballer and Moro, 2001; 
Dentimargo argonauta Espinosa and Ortea, 2002; Tico- 
cystiscus iberia Espinosa and Ortea, 2002; Cratena piu- 
taensis Ortea, Caballer, and Espinosa, 2003; Milleria 
ritmica Ortea, Caballer, and Espinosa, 2003; Phidiana 
adiuncta Ortea, Caballer, and Moro, 2004; and Costoa- 
nachis cascabulloi Espinosa and Ortea, 2004) stated that 
the material was deposited at IN Bio, in fact, the material 
was never deposited there or at the University of Costa 
Rica. Although attempts have been made to clarify the 
location of this material with the authors, none has been 
successful. Abbreviation used in the text: m.a.s.l. = 
meters above sea level. 


CATALOGUE OF TYPE MATERIAL 


academica, Okenia, Camacho-Garcia and Gosliner, 
2004: 431-438, figs. 1-3. Type locality: Playa Tamarindo, 
ea Conservation Area, Puntarenas, Costa Rica 
03/58" N, $5°51'08" W), 0 m depth. HOLOTYPE: 
aca BOOO3LIS1LO2 (specimen), MZUCR- 
INB0003764988 (radula and jaw, SEM stubs), 
PARATYPE: MZUCR-INB0001496648, San Miguel, 
Reserva Natural Absoluta de Cabo Blanco, Tempisque 
Conservation Area, Puntarenas, Costa Rica (9°34'53” N, 
$5°08'26" W), 0 m depth. 
aeci, Philinopsis, Ortea and Espinosa, 2001b: 41, pl. 
2C, 3A. Type locality: Punta Mona, Manzanillo, Limon, 


Costa Rica (9°6'37" N, 82°66/08" W), 6 m depth. 
HOLOTYPE: MZUCR-INBO001495781. PARATYPE: 
MZUCR-INB0003721981. Same locality data as holo- 
type. Remarks: Paratype said to be deposited at the In- 
stituto de Oceanologia, Habana, Cuba, but was deposited 
at INBio. 

alfiopivai, Plesiocystiscus, Espinosa and Ortea, 2002: 
102-106, figs. 1-3, pl.l. Type locality: Manzanillo, 
Limon, Costa Rica (9° By N, 82°39" W), 20-24 m depth. 
HOLOTYPE: MZUCR-INB0003754713 (shell, Fig- 
ure 21). 

anulatus, Janolus, oo Garcia and Gosliner, 
2006: 1295-1305, figs. 1-7. Type locality: Isla Ballena, 
Parque Nacional Marino Ballena, Puntarenas, Costa Rica 
(9°06'24" N, $3°43'35" W), 6 m depth. HOLOTYPE: 
MZUCR-INB0001495772 (specimen), MZUCR- 
INB00037649558 (SEM stub with jaw), MZUCR- 
IN B0003764987 (SEM stub with radula), MZUCR- 
INB0003765066 (slide preparation with labial plate). 
PARATYPES: MZUCR-INB0005764915 (1 speci- 
men). Same locality data as holotype; MZUCR- 
INBOOO3S36171 (1 specimen), SW side of Isla Plata, 
Guanacaste, Cone Rica (10°26'48” N, 85°48'20" W), 10 
m depth. 

arleyi, Melanella, Espinosa, Ortea and Magara, 2001: 
123-124, fig. 3. Type locality: Punta Mona, Manzanillo, 
Limon, Costa Rica (9°37" N, §2°37" W), 10-12 m depth. 
HOLOTYPE: MZUCR-INB0003138455, (shell, Fig- 
ure 1). 

awapa, Doto, Ortea, 200 1a: 21-23, pl. 2G, fig. 9. Type 
locality: Punta Mona, Manzanillo, Limon, Costa Rica 
(9°37"” N, 82°37" W), 10 m depth. HOLOTYPE: 
MZUCR-INB0001497507 ( (specimen and egg mass). 

beatrix riopejensis, Helicina (“Gemma”), Richling, 
2004: 303-308, figs. 133-138. Type locality: SW of Liv- 
erpool (about 24 ‘em of Puerto Limén) along Rio re 
Limon, Costa Rica (90°55'46"” N, 83°13 15” W ), 13. 
m.a.s.l. HOLOTYPE: MZUCR-INB003542625 (shell 
Figure 30). PARATYPE: MZUCR-INB0003542626 
(shell). Same locality data as holotype. 

boeckereli, Alcadia (Microalcadia), Richling, 2001: 
6-7, figs. 9-12. Type teeality Guanacaste National Park, 
about 10 km S of Santa Cecilia, Volean Orosi, near 
field station Pitilla, beginning of Sendero Orosilito, 
primary forest, Guanacaste, Costa Rica (10°59'18S"” N, 
§5°25'34" W), 700 m.a.s.l., HOLOTYPE: MZUCR- 
INB0003404980 (shell, Figure 24). Remarks: species was 
assigned to Alcadia in Richling, 2004: 374-377, figs. 257— 
262: previously in Helicina. 

bramale, Hoplodoris, Fahey and Gosliner, 2003: 19S— 
201, figs. 17E, 58-30. Type locality: Puerto Escondido, 
M: ainiiel Antonio National Park, Puntarenas, Costa 
Rica (9°23’ N, 84°08’ W), 0 m depth. HOLOTYPE: 


Figures 1-12. 


1. Holotype of Mellanela arleyi, 2.1 mm. 2. Holotype of Melanella zugnigae, 1.1 mm. 3. 


Holotype of Triphora 


orteai, 2.0 mm. 4. P: uratype of Chicoreus | ee 137 mm. 5. Paratype of Typhisopsis carolskoglundae, length 19 mm 


diameter, 11mm. 6. Parz ee of Mitrella loisae, 4.9 mm. 


. Holotype of Hyalina chicoi, 9.1 mm. 8. Holotype of Prunwm cahuitaensis, 


12.5 mm. 9. Holotype of Prunum chumi, 18.2 mm. 10. Holeae of Prunum holandae, 19.5 mm. 11. Holotype of Prunum lizanoi 


13.2 mm. 12. Holotype of Volvarina socoae, 11.8 mm. 


Page 158 


THE NAUTILUS, Vol. 122, No. 3 


MZUCR-INB0003572316. PARATYPES: MZUCR- 
INB0003572306 (6 specimens), Punta Uvita, Marino 
Ballena National Park, Puntarenas, Costa Rica (9°08' N, 
83°45’ W), 0-2 m ae MZUCR-INB0001498550 (1 
specimen), 0 m depth. Same locality data as the other 
paratypes. 

bribri, Gibberula, Espinosa and Ortea, 2000: LO00- 
101, pl 1, fig. 4. Type Locality: Punta Mona, Manzanillo, 
Limon, Costa Rica (9°38' N, 82°37’ W), 10-15 m depth. 
HOLOTYPE: MZUCR-INB0003349908 (shell, Figure 
16). PARATYPES: MZUCR-IN B0003349905, MZUCR- 
IN B0003349906, MZUCR-IN B0003349907 (1 specimen 
each, shells). Locality data for all paratypes same as ho- 
lotype. 

caballeri, Philine, Ortea, Espinosa and Moro, 2001: 
38-40, figs. 9-10, pl. 2B. Type locality: Punta Mona, 
Miz aizanillo, Limon, Costa Rica (9°37' N, $2°37' W), 9m 
depth. HOLOTYPE: MZUCR- INB0003138445. 

cabecar, Doto, Ortea, 2001a: 34-37, figs. 16-17, pl. 
2M. Type Locality: Punta Mona, Manzanillo, Limon, 
Costa Rica (9°39' N, 82°37’ W), 20 m depth. 
HOLOTYPE: MZUCR-IN BO003449742. 

cahuitaensis, Prunum, Magana, Espinosa and Ortea, 
20038: 122-124, figs. 1, 2A-B, pls. 1A, 2A. Type Locality: 
1 km E from Puerto Vargas Station, Cahuita National 
Park, Limon, Costa Rica (9°43'33" N, 82°48'31” W). 
HOLOTYPE: MZUCR-INB0003718203 (shell, Figure 
8). PARATYPE: MZUCR-IN B0003718204 (shell). Same 
locality data as holotype. 

caribetica, Gibberula, Espinosa and Ortea, 2002: 
113-114, fig. 10, pL.1. Type locality: Punta Mona, Man- 
zanillo, Limon, Costa Rica (9°37' N, 82°37’ W), 8 m 
depth. HOLOTYPE: MZUCR-INB0003715205 (shell, 
Figure 17). 

carolskoglundae, Typhisopsis, Houart and Hertz, 
2006: 56-58, figs. 17-25, 47-49, 59, 63. Type locality: 
Playas del Coco, Guanac: aste, Costa Rica (10°55'53” N, 
85°69'S1" W), 24-37 m depth, on mud bottom. 
PARATYPE: MZUCR-6153, (shell, Figure 5), Boca de la 
Honda, Veraguas, Panama (7°27' N, 80°51’ W), in white 
sand. Remarks: The coordinates and the collecting local- 
ity of this paratype appear to be incorrect since these 
coordinates plot inland. 

chicoi, Hyalina, Espinosa and Ortea, 1999a: 167-169, 
figs. LA-C, 2A, 3A—D. Type locality: Manzanillo, Limon, 
Costa Rica (9°38' N, 82°39’ W), 5-12 m depth. 
HOLOTYPE: MZUCR-INB0003350839 (shell, Figure 
7). Remarks: Paratype said to be de posited at INBio in 
the original publication, but never ve to INBio or the 
University of Costa Rica. 

chiquitica, Oligyra, Richling, 2001; 1-2, figs. I- 2. 
Type Locality: 9 km W of Matina, a little wre am up the 
{io Barbilla from the crossing of the road Siquirre s to 
Limon, along a tributary of Rio Barbilla, Limon, Costa 
Rica (10°03'29” N, 83°22'24” W), 70 m.a.s.l. 
HOLOTYPE: MZUCR-INB0003404977 (soft parts 
and shell, Figure 32). PARATYPE: MZUCR- 
INB0003404981 (shell and soft parts). 


chumi, Prunum, Espinosa and Ortea, 2000: 107-108, 
figs. 8-9. Type locality: Manzanillo, Limén, Costa Rica 
(9°3 8’ N, 82°39’ W), 10-15 m depth. HOLOTYPE: 
MZUCR-INB0003349912 (shell, Figure 9). 
PARATYPES: MZUCR-INB0003349913, MZUCR- 
INB0003349914, MZUCR-INB0003349915 (1 specimen 
each, shells). Locality data for all paratypes same as ho- 
lotype. 

convenientis, Eubranchus, Caballer and Ortea, 2002: 
81-85, figs. 2, 3, 7 pl 1B. Type locality: Manzanillo, 
Limon, Costa Rica b»38) N, 82°39’ W), 0 m depth. 
HOLOTYPE: MZUCR-IN B0O003576832 

corcovadensis, Cryptostrakon, Cuezzo, 1997: 1-S, 
figs. 1-14. Type locality: Corcovado National Park, Si- 
rena Station, Sendero a Rio Los Patos, Puntarenas, Costa 
Rica (8°30' N, 83°35’ W), 10 m.a.s.l. HOLOTYPE: 
MZUCR-INB0001468087 (shell, Figure 37; SEM stub 
with radula). PARATYPE: MZUCR-INB0001468080 
(shell and soft parts). Remarks: Paratype locality data 
same as holotype. In the publication, the holotype and 
paratype catalogue numbers mistakenly written 
IN BO00468087 aad IN B000468080, respectively. 

costacubensis, Janolus, Ortea and Espinosa, 2000: 
80-83, figs. 1-2. Type locality: Miramar, N coast of La 
Habana, “Caba (23°7'21" N, 82°25'10" W), 20-25 m 
depth. PARATYPE: MZUCR-INB0001497432 (1 speci- 
men), collecting locality Manzanillo, Limén, Costa Rica 
(9°38" N, 82°39! W), 20-25 m depth. 

ehuxmoralal, Dentimargo, Espinosa and Ortea, 2000: 
110-113, fig. 11, pl. 2. Type locality: reefs from Manza- 
nillo, Limon, Costa Rica (9°38' N, 82°37’ W), 12-15 m 
depth. HOLOTYPE: MZUCR-INB0003349903 (shell, 
Figure 14). 

curere, Doto, Ortea, 2001a: 17-15, fig. 7, pl. 2C. Type 
locality: Puerto Viejo, Limon, Costa Rica (9°38' N, 
§2°39' W), 6 m depth. HOLOTYPE: MZUCR- 
IN B0001496453. 

destinyae, Cuthona, Hermosillo and Valdés, 2007: 
119-124, figs. 1C, 4, 5. Type locality: La G dou. Zi- 
huatanejo, c uerrero, México (17°37. 854’ N, 101°33.562' 
W). PARATYPES: MZUCR-INBOO03118106 (4 speci- 
mens), Playa Avellanas, Guanacaste, Costa Rica 
(10°13.583' N, 85°50.433' W). 

duao, Doto, Ortea, 2001a: 28-30. figs. 12— 13, pls. 2: 
2K. Type locality: Punta Mona, Manzanillo, Limon, 
Costa Rica (9°37' N, 82°37’ W), 10-19 m depth. 
HOLOTYPE: MZUCR-IN BO00313S803. 

echandiensis, Helicina, Richling, 2004; 271-277, figs. 
77-84. Type locality: La Amistad National Park, Las Al- 
turas Sector, Southern Cordillera de Talamanca, S of 
Cerro Echandi, Campamento Echandi, Puntarenas, 
Costa Rica (09°01'33" N, 82°40'12" W), 2840 m.a.s.b. 
HOLOTYPE: MZUCR-INB0003542520 (soft parts 
and shell, Figure 26). PARATYPES: MZUCR- 
INB0003574064 (1 specimen, sott parts and_ shell), 
MZUCR-INB0003542521 (1 specimen, soft parts and 
shell), MZUCR-IN B0003428246 (19 specimens, shell 
and soft parts). Locality data for all paratypes same as 
holotype. 


F. Villalobos-Rojas et al., 2008 


Page 159 


elizabethae, Adrana, Ortea and Espinosa, 2001e: 61— 
64, fig. 17. Type locality: in front of Gandoca beach, 
Limon, Costa Rica (9°36' N, §2°35’ W), 10-15 m depth. 
HOLOTYPE: MZUCR-INB0003449558 (shell, Figs. 
38-39). 

escondida, Helicina (“Gemma”), Richling, 2004: 348— 
357, figs. 210-218. Type locality: 9 km W of Matina, a 
little stream up the Rio Barbilla from the crossing of the 
road from Siquirres to Limén, along a tributary of Rio 
Barbilla, Limon, Costa Rica (10°03'29” N, 83°22'24” W), 
70 m.as.l. HOLOTYPE: MZUCR-INB0003542623 (soft 
parts and shell, Figure 27), PARATYPE: MZUCR- 
INB0003542624 (soft parts and shell). Locality data 
same as holotype. 

espinosai, Ancula, Ortea, 2001b: 49, pl. 2D. Type 
locality: Punta Mona, Manzanillo, Limon, Costa Rica 
(9°37' N, 82°37' W),9m depth. HOLOTYPE: MZUCR- 
IN B0003188764. 

eugeniae, Elysia, Ortea and Espinosa, 2002: 130-133, 
figs. 1-2, pl. LA. Type locality: Manzanillo, Limon, Costa 
Rica (9°38’ N, 82°39’ W), 16 m depth. HOLOTYPE: 
MZUCR-INBO001497478. 

eversonii, Phyllonotus, D’Attilio, Myers and Shasky, 
1987: 162-164, figs. 1-2. Type Locality: SW side of Isla 
Manuelita, Isla del Coco, Costa Rica (5°33’ N, 87°03’ W, 
66 m depth. PARATYPE: MZUCR 4934 (ahell. Fig- 
ure 4). 

gandocaensis, Rissoella, Ortea and Espinosa, 2001a: 
36, pl. 2A. Type locality: Punta Mona, Manzanillo, 
Limén, Costa Rica (9°37 N, 82°37 W), 9 m depth. 
HOLOTYPE: MZUCR-INB0003323831. Remarks: shell 
destroyed, only dry soft body present, not figured here. 

genecoani, Plesiocystiscus, Espinosa and Ortea, 2000: 
96-97, fig. 1, pl. 1. Type locality: Manzanillo, Limon, 
Costa Rica (9°39' N, 82°37’ W), 25 m depth. 
HOLOTYPE: MZUCR-INB0003349916 (shell, Figure 
22). PARATYPES: MZUCR-INB0003349917 (2 speci- 
mens, shells). Locality data same as holotype. 

goslineri, Cylichnella, Valdés and Camacho-Gareia, 
2004: 459-497, figs. 4D-F, 5. Type locality: Sector Playi- 
tas, Golfo Balec. Puntarenas, Costa Rica (8°44'19”" N, 
83°21'57" W), 0 m eee HOLOTYPE: MZUCR- 
IN B0001497964 (shell, Figure 23). PARATYPES: 
MZUCR-INB0003718957 (15 specimens, shell and soft 
parts, lateral gizzard plate and radular teeth in SEM 
stubs). Locality data same as holotype. 

hojarasca, Alcadia (Microalcadia), Richling, 2001: 
5-6, figs. 6-8. Type locality: cordillera de Tilaran, about 
9 km N of Santa Elena, near Mirador Gerardo, Guana- 
caste, Costa Rica (10°22’19" N, 84°48’25” W), 1450 
m.as.l. HOLOTYPE: MZUCR-INB0003404979 (shell, 
Figure 25). Remarks: species was assigned to Alcadia in 
Richling, 2004: 370-374, figs. 249- 255: it was previously 
Helicina. 

holandae, Prunum, Espinosa and Ortea, 1999b: 175— 
176, fig 1H. Type locality: Cayos Limon, Islas San Blas, 
Colén, Panama (9°33'00” N, 78°53’30” W) 
HOLOTYPE: MZUCR-5750 (shell, Figure 10). 


PARATYPE: MZUCR-INB0003350838 (shell), Punta 
Uvita, Manzanillo, Limon, Costa Rica (9°38’ N, 82°41’ 
W). Remarks: According to the original publication the 
holotype was deposited at the Instituto de Oceanologia 
de la Habana, Cuba, but in fact it was deposited at IN- 
Bio. The paratype was erroneously assigned the cata- 
logue number INBIOCR1001501498 in the original pub- 
lication. Also, the locality for the paratype, Punta Uvita, was 
misspelled in the original publication as Punta “Ubitas.” 

inbiotica, Trapania, Camacho and Ortea, 2000: 317- 

321, figs. 1-3. Type locality: San Miguel Station, Reserva 

Absoluta de Cabo Blanco, Tempisque Conservation 
Area, Puntarenas, Costa Rica (9°34'49”" N, 85°08'28”" W). 
lm depth. HOLOTYPE: MZUCR-INBOOO1500889. 

iugula, Doto, Ortea, 2001a: 26-27, fig. 11, pl. 2]. Type 
locality: Punta Mona, Limén, Costa Rica (9°39’ N, 
82°37' W), 25 m depth. HOLOTYPE: MZUCR- 
IN B0003449604. 

kekoldi, Doto, Ortea, 2001a: 18-21, fig. 8, pl. 2 E-F. 
Type locality: Punta Mona, Limon, Costa Rica (9°35' N, 
82°37’ W), 9 m depth. HOLOTYPE: MZUCR- 
INB0003449587 (soft body and egg mass). 

leopoldoi, Eubranchus, Caballer, Ortea and Espi- 
nosa, 2001: 55-56, fig. 14, pl. 2E. Type locality: Punta 
Mona, Manzanillo, Limén, Costa Rica (9°38' N, 82°37’ 
W). 6 m depth. HOLOTYPE: MZUCR- 


INB0003138799. 
lizanoi, Prunum, Magafia, Espinosa, and Ortea, 2003: 
124-126, figs. 2C, 3, pls. 1B, 2 B. Type locality: Bahia 


Junquillal Wildlife Refuge, Golfo de Santa Elena, 
Guanacaste, Costa Rica. HOLOTYPE: MZUCR- 
INB0003481195 (shell, Figure 11). PARATYPES: 
MZUCR-INB0003481192 (1 specimen, shell, including a 
slide with radula), MZUCR-INB0003476231 (1 speci- 
men, shell and soft parts). 

loisae, Mitrella, Pitt and Kohl, 1979: 467-468, figs. 
2A-B, 3A-B. Type locality: N side of Punta Coralillo, 
Bahia de Caldera, about 20 km S of the city of Puntar- 
enas, Puntarenas, Costa Rica (9°54' N, 84°44’ W). 
PARATYPES: MZUCR 2363 (2 specimens, shell, Fig- 
ure 6). 

magagnai, Dendrodoris, Ortea and Espinosa, 2001d: 
52-53, fig. 13, pl. 3E. Type locality: Manzanillo, Limon, 
Costa Rica (9°38' N, 82°39" W), 20 m depth. 
HOLOTYPE: MZUCR-INB0001497496. 

manzanilloensis, Polycera, Ortea, Espinosa, and Ca- 
macho, 1999: 161-163, fig. 3. Type locality: Manzanillo, 
Limon, Costa Rica (9°38’ N, 82°39’ W), 8 m depth. 
HOLOTYPE: MZUCR-INB0001496124. 

marioi, Gibberula, Espinosa and Ortea, 2000: 101-— 
102, fig. 5, pl. 1. Type locality: coral reefs of Manzanillo, 
Limon, Costa Rica (9°39' N, 82°39’ W), 30 m depth. 
HOLOTYPE: MZUCR-INB0003371977 (a slide prepa- 
ration with radula). 

millenae, Cuthona, Hermosillo and Valdés, 2007: 
124-128, figs. 1D, 6, 7. Oy locality: Los Arcos, Bahia 
de Randers, Jalisco, México (20°3 32.855’ N, 105°17.340' 
W), 19 m depth. PARATYPE: MZUCR- 
INB0003836263. Playa Real, NE Punta Roble, Guana- 


Q 
0d 


122, No. 


NAUTILUS, Vol. 


THE 


Page 160 


F. Villalobos-Rojas et al., 2008 Page 16] 


Figures 24-32. 24. Holotypes of Alcadia (Microalcadia) boeckereli, 2.2 mm. 25. Alcadia (Microalcadia) hojarasca, 2.4 mm. 26. 
Helicina (“Tristramia”) echandiensis, 7.2 mm. 27. Helicina (“Gemma”) escondida, 6.2 mm. 28. Helicina (“Gemma”) monteverdensis 


6.6 mm. 29. Helicina (“Gemma”) talamancensis, 9.2 mm. 30. Helicina (“Gemma”) beatrix riopejensis, 7.8 mm. 31. Helicina (“Tris- 
tramia”) punctisulcata cuericiensis, 5.9 mm. 32. Oligyra chiquitica, 4.8 mm 


Figures 13-: 


ey) 


13. Holotypes of Volvarina yolandae, 6.2 mm. 14. Dentimargo cruzmoralai, 2.5 mm. 15. Dentimargo zaidetta 


16. Gibberula bribri. 11.0 mm. 17. Gibberula caribetica, 6.0 mm. 18. Gibberula sierrai, 2.1 mm. 19. Gibberula ubitaen 


2.5 

1S mm. 20. Granulina minae, 1.0 mm. 21. Plesiocysticus alfiopivai, 1.9 mm. 22. Plesiocysticus genecoani, 2.2 mm. 2« 
J i ] o 

4.5 mm 


arf ylic hne 


Page 162 THE NAUTILUS, Vol. 122, No. 3 


F. Villalobos-Rojas et al., 2008 


Page 163 


caste, Costa Rica (10°23.200' N, 85°50. 
depth. 

minae, Granulina, Espinosa and Ortea, 2000: 103, fig. 6, 
pl. 1. pis loc: uity: Punta Mona, Manzanillo, Limon, Costa 
Rica (9°37’ N, $2°37' W), 12-15 m depth. HOLOTYPE: 
MZUCR- INBO003: 349904 (shell, Figure 20). 

monteverdensis, dlc (“Gemma”), Richling, 
2004: 334-348, figs. 183-199. Type locality: Cordillera 
de Tilaran, near Monteverde, Zona Protectora Arenal- 
Monteverde, Reserva Biolégica Bosque Nuboso Mon- 
teverde, Sendero Bosque Nuboso, Puntarenas, Costa 
Rica (10°18S'08” N, 84°47'41” W), 1550 m.a.s.l. 
HOLOTYPE: MZUCR-INB0003542627 (shell, Figure 
28). PARATYPE: MZUCR-INB0003542628S (shell). Lo- 
cality data same as holotype. 

orteai, Cyerce, Valdés and Camacho-Garcia, 2000: 
445456, figs. 1-5. Type locality: Playa Cabuya, en 
Coébano, Puntarenas, Costa Rica (9°39’ N, 85°11’ W), ( 
m depth. HOLOTYPE: eit Ce th ucGIEOUCEL. 
PARATYPE: MZUCR-INBO001500644. Locality data 
same as holotype. 

orteai, Triphora, Espinosa, 2001: 21-22, fig. 7. Type 
locality: Punta Mona, Manzanillo, Limén, Costa Rica 
(9°37” N, 82°37” W), 10-15 m depth. HOLOTYPE: 
MZUCR-INB0003138795 (shell, Figure 3). 

osae, Jorunna, Camacho-Garcia and Gosliner, 2008: 
165-167, figs. 21-22. Type locality: Playa eng 
Golfito, Osa Conservation Area, Costa Rica (8°38'33" 
$3°13'40" W), O m depth. HOLOTYPE: WaGce 
INB0003701453, MZUCR-INB0003799440 (radula and 
jaw, SEM stubs). 

pacifica, Chiapaphysa, Taylor, 2003: 170-171, fig. 
168, pl. 8.8. Type locality: Rio Tenorito, Hacienda 
La Pacffica, Guanacaste, Costa Rica (10°29.02' N, 
$9°09.58' W). 100 m.a.s.l. PARATYPES: MZUCR- 
INB0003352239 (10 specimens, shells, Figure 33). 

proranao, Doto, Ortea, 2001a: 23-25, fig. 10, pl. 2H. 
Type locality: Punta Mona, Manzanillo, Limon, Costa 
Rica (9°37' N, $2°37' W), 9 m depth. HOLOTYPE: 
MZUCBR-IN B0003 138763. 

punctisulcata cuericiensis, Helicina, Richling, 2004; 
277-283, figs. 87-96. Type locality: Cordillera de Tala- 
manca, Estacién Cuerici, 4.5 km E of Villa Mills, Sen- 
Se al Mirador, Cartago, Costa Rica (9°33'28" N, 

3°40'13" W), 2700 m.a.s.l. HOLOTYPE: MZUCR- 
cet era (shell, Figure 31). PARATYPE: 
MZUCR-IN B0003542541 (shell and soft parts). Other 
ae ity data same as holotype (09°33'19" N, 83°40'13" 

2600 m.a.s.l.). 

ee Mayabina, Taylor, 2003: 102-104, 
fig. 85, pls. 3.5, 3.6. ee locality: Barra del Colorado, 
Limon, Costa Rica (10°46.37' N, 83°35.27' W). 
PARATYPES: MZUCR-INB0003352237 (10 specimens, 


733' W), 5 m 


shell), MZUCR 69-01 (10 specimens, shell, Figure 34), 
Locality data for all paratypes same as holotype 

selva, Ercolania, Ortea and meee o00le: 45-47, 
fig. 11. Type locality: Manzanillo, Limén, Costa Rica 
(9°39' N, 82°39’ W) ), 0m depth. HOLOTYPE: MZUCR- 
INBOOO313832 PARATYPE: MZUCR- 
IN B0003449624. ae data same as holotype. 

sierrai, Gibberula, Espinosa and Ortea, 2000: 99-100, 
fig. 3, pl 1. Type locality: Punta Mona, Manzanillo, 
Limon, Costa Rica (9°38’ N, 82°37’ W), 10-15 m depth. 
HOLOTYPE: MZUCR-INB0003349909 (shell, Fig- 
ure 18). 

sinusdulcensis, Tropinauta, Taylor, 2003: 110-111, 
figs. 91-94, Type locality: small stream in pasture 3 km 
SE of Golfito, Puntarenas, Costa Rica (8°36.68’ N, 
3°8.48' W). PARATYPES: MZUCR-INB0003382246 
(5 specimens, shells, Figure 36). 

socoae, Volvarina, Espinosa and Ortea, 1999b; 171— 
172, figs. LE, 2D-E. Type locality: Manzanillo, Limon, 
Costa Rica (9°39' N, 82°45’ W), 1.0-1.5 m de »pth. 
HOLOTYPE: MZUCR-INB0003350887 (shell, Figure 
12). PARATYPE: MZUCR-INB000335088S (shell). Re- 
marks: in the original publication the holotype and 
paratype were erroneously assigned the catalogue num- 
bers INBIOCR001496127 and INBIOC R1001496] 28, re- 
spectively. 

talamancensis, Helicina (“Gemma”), Richl ing, 2001: 
3-5, figs. 3,4. Type locality: Fila Costenia, N of Bajo Bo- 
nito, N of RioC laro, Puntarenas, Costa Rica (8°44'41” N, 
$3°02'09”" W), 980 m NN. HOLOTYPE: MZUCR- 
INB0003404978 (shell, Figure 29). PARATYPES: 
MZUCR-INB0001494642 (8 specimens, shell), 
MZUCR-INBO0001487761 (2 specimens, shell and soft 
parts), 3.5 kin from Escuela de Llano Bonito Carretera a 
San Vito, Puntarenas, Costa Rica (8°44'27” N, 83°02'04" 
W), 840 m m.as.l., MZUCR-INB0001494509 (1 speci- 
men, shell and soft parts), 3.5 km NE from Escuela de 
Llano Bonito, Puntarenas, Costa Rica (8°44'54" N, 
§3°02'04" W), 920 m m.a.s.l. Remarks: species was as- 
signed to Helicina in Richling, 2004; 308-315, figs. 140, 
142- 14S; it was previously Olygira el maces The 
paratype MZUCR-INBO0003389580 (1 specimen, shell, 
locality data same as MZUCR-INB0001494509) is said to 
be deposited at INBio but is not present. 

tempisquensis, Jorunna, Camacho-Garcia and Gos- 
liner, 2008: 167-170, figs. 23-24. Type locality: Cabo 
Blanco, Costa Rica (9°34'50" N, 85°08'26" W), 0 m 
depth. HOLOTYPE: MZUCR-INBOOO 35.423 76. 
PARATYPES: MZUCR-INB0003542377 (2 specimens). 

tempisquensis, Mayabina, Taylor, 2003: 109-111, fig. 
90, pl a. 4. Type locality: edge of marshes, 100 m E of W 
end of airstrip, Palo Verde National Park, Guanacaste, 
Costa Rica (10°20.68’ N, 85°20.60' W). PARATYPES: 


Figures 33-39. 33. Paratype of Chiapaphysa pac ifica, 6.1 mm. 34. Paratype of Mayabina sanctijohannis, 6.7 mm (UCR 69-01). 35. 


Paratype of Mayabina tempisquensis, 7.5 mm. (INB0003352244). 


36. Paratype of Tropinauta sinusdulcensis, 5.5 mm. 37. Holotype 


of Cryptostrakon corcovadensis, 4 mm. 38. Holotype of Adrana elizabethae, (left valve dorsal view) 32.5 mm. 39. Holotype of Adrana 


elizabethae, (left valve ventral view) 32.5mm 


Page 164 


THE NAUTILUS, Vol. 122, No. 3 


MZUCR-INB0003382244 (10 specimens, shell, Figure 
5), MZUCR 70-01 (10 specimens, shell). 

tica, Furcilla, Espinosa and Ortea, 2000: 105-107, fig. 
7, pl. 1. Type locality: coral reefs of Manzanillo, Limén, 
Costa Rica (9°39' N, 82°39’ W), 30 m depth. 
HOLOTYPE: MZUCR-IN B0003371976 (including a 
slide with radula). 

tica, Mexichromis, Gosliner, Ortea, and Valdés, 2004: 
589-593, figs. LA, 2, 3. Type locality: Bajo del Diablo, 
Isla del Cano, Punts irenas, Costa Rica (8°42'04" N, 
§3°53'20" W), PARATYPES: MZUCR-INBO001486639 
(2 specimens). 

ubitaensis, Gibberula, Espinosa and Ortea, 2000: 97— 
99, fig. 2, pl. 1. Type locality: Punta Uvita, Manzanillo, 
Limon, Costa Rica (9°38’ N, 82°41’ W), 12-15 m depth. 
HOLOTYPE: MZUCR-INB0003349902 (shell, Figure 
19). Remarks: the Type locality, Punta Uvita, was mis- 
spelled in the original publication as “Ubita.” 

yolandae, Volvarina, Espinosa and Ortea, 2000: L0S— 
110, fig. LO, pl 1. Type locality: Manzanillo, Limon, 
Costa Rica (9°39’' N, 82°39’ W), 10-25 m depth. 
HOLOTYPE: MZUCR-INB0003350843 (shell, Figure 
13). PARATYPES: MZUCR-INB0003350541 (1 speci- 
men, shell), MZUCR-INB0003350842 (1 specimen, 
shell). Locality data same as holotype. Remarks: accord- 
ing to the original publication, the paratype MZUCR- 
INB0003350841 was de »posited at the Instituto de 
Oceanologia de la Habana, Cuba, and the paratype 
MZUCR-INB0003350842 was deposited at Museo de 
Ciencias Naturales de Tenerife, Canary Islands; how- 
ever, both are depx ysited at MZUCR. 

zaidettae, Dentimargo, Espinosa and Ortea, 2000: 
113-114, fig. 12, pl. 2. Type locality: Punta Mona, Man- 
zanillo, Limon, Costa Rica (9°38' N,$2°37' W), 12-15 m 
depth. HOLOTYPE: MZUCR-INB0003349911 (shell, 
Figure 15). PARATYPE: MZUCR-INBO003349910° (1 
specimen, shell). Locality data same as holotype. Re- 
marks: According to the original publication the paratype 
was deposited at the Instituto de Oceanologia de la Ha- 
bana, Cuba, but in fact it is deposited at MZUCR. 

sugnigae, aging Espinosa, Ortea and Magana, 
2001: 26-27, figs. D. Pe locality: Punta Mona, 
Manzanillo, ‘Limon, i“ mee Rica (9°37' N, 82°37' W), 10- 
15 m de pth. HOLOTYPE: pea nmi 


(shell, Figure 2), 


ACKNOWLEDGMENTS 


We are very grateful to Alvaro Morales and Ingo Werth- 


mann for their logistic support during the elaboration of 


this paper. The photographs were taken at the facilities 
of Centro de Investigaci6n en Ciencias del Mar y Lim- 
nologia (CIMAR) at the University of Costa Rica. Eu- 
gene V. Coan and David Butvill made constructive com- 
ments on this manuscript. 


LITERATURE CITED 
Caballer, M., J. Ortea, and J. Espinosa. 2001. Descripcién de 


una nueva especie de Eubranchus Forbes, 1834, Avicen- 
nia Suplemento 1. 55-56 


Camacho, Y. and J. Ortea. 2000. A new species of Trapania 
(Nudibranchia: Goniododididae) from the Pacific coast of 
Central America. Revista de Biologia Tropical 48: 317— 
322. 

Camacho-Garcia, Y. and T. Gosliner, 2004. A new species of 
Okenia (Gastropoda: Nudibranchia: Goniodoridae) from 
the Pacific coast of Costa Rica Proceedings of the Califor- 
nia Acade my of Sciences 55: 431-438. 

Camacho-Garcia, Y. and T. Gosliner. 2006. A new species of 
the zephyrinid nudibranch genus Janolis (Mollusca: Nudi- 
branchia) from North America and Costa Rica. Revista de 
Biologia Tropical 54: 1295-1305. 

Camacho-Gareia, Y. and T. Gosliner. 2008. Systematic revision 
and morphological phylogenetic analysis of the genus Jo- 
runna Bergh, 1876 (Nudibranchia: Discodorididae). Jour- 
nal of the Molluscan Studies 74: 143-151. 

Espinosa, J. and J. Ortea. 1999a. be nuevas especies del 
género Hyalina Schumacher, 1817 (Mollusca: Neogas- 
tropoda: Marginellidae) del Me ur Caribe de Costa Rica y 
costas de Cuba. Avicennia 10/11: 165-169. 

Espinosa, J. and J. Ortea. 1999b. Descripcion de nuevas mar- 
ginelas (Mollusca, Neogastropoda: Marginellidae) de 
Cuba y del Caribe de Costa Rica y Panama. Avicennia 
10/11: 165-179. 

Espinosa, J. and J. Ortea. 2000. Descripci6n de un género y 
once especies nuevas de Cystiscidae y Mz wreinellidae ( Mol- 
lusca: Neogastropoda) de | Caribe de Costa Rica. Avicennia 
12/13: 95-114. 

Espinosa, J. 2001. Descripci6n de una nueva especie de 
ee (s.L.) Blainville, 1S28. Avicennia Suplemento 4: 
21- 

E repinosa, i J. Ortea, and J. Magana. 2001. Deseripcién de una 
nueva especie de Me ‘lanella Bowdich, 1822 y redescripcién 
de Melanella arcuata (c. B. Adams, 1850), Avicennia 
Suplemento 14, pp. 120-124. 

Espinosa, J. and J. Ortea. 2002. Nuevas especies de margineli- 
formes de Cuba, Bahamas y el Mar Caribe de Costa Rica. 
Avicennia Suple mento 15: 101-128. 

Espinosa, ]. and J. Ortea. 2004, Nuevas especies de moluscos 
gasteropodos marinos Mollusca: Gastropoda de las Baha- 
mas, Cuba y el Mar Caribe de Costa Rica. Revista de la 
Academia Canarias de Ciencias. 15: 207-216. 

Gosliner, T., J. Ortea, and A. Valdés. 2004. New data on tropi- 
cal Eastern Pacific Chromodorididae (Nudibranchia: Do- 
ridina) with description of a new species of Mexichromis 
Bertsch, 1977. Proceedings of the California Academy of 
Sciences 55; 588-597. : 

Hermosillo, A. and A. Valdés, 2007. Five new species of aeolid 
nudibranchs (Mollusca, Opisthobranchia) from the tropi- 
cal eastern Pacific. American Malacology Bulletin 22: 119— 

37 : 

Houart, R and C. M. Hertz. 2006. A review of Typhisopsis 
Jousseaume, L880, and Typhisala Jousseaume, LSS1 (Gas- 
es Muricoidea) of the eastern Pacific, The Nautilus 
120; 52-65, 

Magana, I. 4. Espinosa, and J. Ortea. 2003, Descripcién de dos 
nuevas especies del género Prunum Herrmannsen, 1852 
(Mollusca: Gastropoda: Marginellidae) del Caribe y el 
Pacifico de Costa Rica. Avicennia 16; 121-128. 

Ortea, J.. J. Espinosa, and Y. Camacho. 1999, Especies del 
genero Polycera Cuvier, 1816 (Mollusca: Nudibranchia) 
recolectadas en la epifauna de algas rojas del Caribe de 
Costa Rica y Cuba. Avicennia 10/11: 157-164. 

Ortea, J. and J. Espinosa, 2000. Nueva especie del género Jano- 


F. Villalobos-Rojas et al., 2008 


Page 165 


lus Bergh, 1884 (Mollusca: Nudibranchia) de Cuba y 
Costa Rica. Avicennia 12/13: 79-83. 

Ortea, J. 2001a. El género Doto Oken, 1815 en el Mar Caribe: 
Historia natural y descripcion de nuevas especies. Avicen- 
nia Suplemento 3: 1-46. 

Ortea, J. 2001b. Descripcién de una nueva especie de Ancula 
Loven, 1S46, Avicennia Suplemento 4: 49. 

Ortea, J. and J. Espinosa. 2001a. Descripci6n de una nueva 
especie del género Rissoella Gray, 1S47. Avicennia Suple- 
mento 4: 36. 

Ortea, J. and J. Espinosa. 2001b. Descripcién de una nueva 
especie de Philinopsis Pease, 1560, Avicennia Suplemento 
4: 41. 

Ortea, J. and J. Espinosa. 2001c. Descripci6n de una nueva 
especie de Ercolania Trinchese, 1872. Avicennia Suple- 
mento 4: 45-47. 

Ortea, J. and J. Espinosa. 2001d. Descripcién de una nueva 
especie de Dendrodoris Ehrenberg, 1831. Avicennia 
Suplemento 4: 52-53. 

Ortea, J. and J]. Espinosa. 2001le. Descripcién de una nueva 
especie de Adrana H. y A. Adams, 1858. Avicennia Suple- 
mento 4: 61-64. 

Ortea, J., J. Espinosa, and L. Moro. 2001. pen bike de una 
nueva especie de Philine Ascanius, 1772. Avicennia Suple- 
mento 4: 38-40. 

Ortea, J. and M. Caballer. 2002. Nuevos datos sobre el género 
Eubranchus Forbes, 1838 (Mollusca: Nudibranchia) en 
aguas templadas del Atlantico Oeste. Avicennia Suple- 
mento 15: 77-90. 


Ortea, J. and J. Espinosa. 2002. Nuevas Especies del género 
Elysia Risso, 1S1S (Mollusca: Sacoglossa) con caracteres 
singulares. Avicennia, Suplemento 15: 129-140. 

Ortea, J., M. Caballer, y J. Espinosa. 2003. Nuevos Aeolidaceos 
(Mollusca: Gastropoda) de Costa Rica. Avicennia 16; 129- 
142. 

Ortea, J., M. Caballer, and L. Moro. 2004. Dos aeolidaceos con 
ceratas rojos de la region macaronesica y el Mar Caribe 
(Mollusca: Nudibranchia). Vieracea 32: 83-96. 

Richling, 1. 2001. New species of Helicinidae from Costa Rica 
(Mollusca: Neritopsina). Schriften zur Malakozoologie 17; 
1-8. 

Richling, I. 2004. Classification of the Helicinidae: Review of 
morphological characteristics based on a revision of the 
Costa Rican species and application to the arrangement of 
the Central American mainland taxa (Mollusca: Gas- 
tropoda: Neritopsina). Malacologia 45: 195-450, 

Taylor, D. W. 2003. Introduction to Physidae (Gastropoda, Hy- 
grophila); biogeography, classification, morphology. Re- 
vista de Biologia Tropical 51: 1-299. 

Valdés, A. and Y. Camacho-Garcia. 2000. A new species of 
Cyerce Bergh, 1871 (Mollusca, Sacoglossa, Polybranchi- 
idae) from the Pacific Coast of Costa Rica. Bulletin of 
Marine Science 66; 445—456. 

Valdés, A. and Y, Camacho-Garcia. 2004. “Cephalaspidean™ 
Heterobranchs (Gastropoda) from the Pacific Coast of 
Costa Rica. Proceedings of the California Academy of Sci- 
ences 55: 459-497. 


THE NAUTILUS 122(3):166-170, 2008 


Page 166 


Dilemma japonicum new species (Bivalvia: Anomalodesmata: 
Poromyidae): A new record of the genus from the 


Northwest Pacific 


Takenori Sasaki 

The University Museum, 
The University of Tokyo 
7-3-1 Hongo, Bunkyo-ku, 
Tokyo 113-0033 

JAPAN 
sasaki@um.u-tokyo.ac.jp 


José H. Leal 


P. O. Box 1580 


The Bailey-Matthews Shell Museum 


Sanibel, FL 33957 USA 
jleal@shellmuseum.org 


ABSTRACT 


The fourth species of the bivalve genus Dilemma Leal, 2008, is 
described from disarticulated valves collected off central Japan. 
The discovery of these specimens represents a significant range 
extension for the genus to the Northwest Parihie, The new 
species is distinguished from the other three known species by 
its surface sculpture, shape of escutcheon, and hinge. The pro- 
dissoconch, indicative of lecithotrophic dev elopment, and shell 
microstructure, with outer homogeneous and inner nacreous 
layers, are illustrated for the first time for a species of the 
genus. 


Additional Keywords: Septibranchia, Poromyoidea, Japan, 
nacreous layer, prodissoconch 


INTRODUCTION 


Dilemma is a recently established, unusual genus of 


Anomalodesmata (Leal, 2008). The first described spe- 
cies of the genus, D. inexpectatum (Crozier, 1966), was 
dredged Goan bathyal depths off northern New Zealand 
and originally assigned to Corculum, family Cardiidae 
(Crozier, 1966) due to general similarities in ‘shell shape. 
However, its original familial allocation was questionable 
from habitat and depth alone, because Corculum species 
inhabit shallow lagoons of tropical waters and utilize pho- 
tosynthesis-derive ad energy by symbiotic dinoflage ‘lke S 
(Farmer et al., 2001). Subse quent discovery of obviously 
related species with preserve -d soft parts ‘allowed Leal 
(2008), using details of the macroanatomy, to demon- 
strate (1) that Corculum inexpectatum and the other, 
then newly found species are related and deserved 
grouping under a new genus, (2) that the new genus 
should be included in the Poromyidae, and (3) that spe- 
cies of Dilemma are carnivores. In fact, carnivory is a 
feeding habit that is common in the Anomalodesmata, a 
group of bivalves mostly present in the deep sea. 
Three species were originally (Leal, 2008) assigned to 
the genus: (1) Dilemma frumarkernorum Leal, 2008, 


from off Key West Florida, 229 m, (2) D. spectralis Leal, 
2008, from off Vanuatu, 950-961 m, and (3) D. inexpec- 
tatum (Crozier, 1966) from off Three Kings Islands, New 
Zealand, $05 m. Their geographic distribution is disjunct, 
but shell morphology is surprisingly similar between the 
species of Florida and Southwest Pacific. 

During a research cruise off central Japan, in 2000, the 
senior author collected small unpaired valves of Corculum- 
like bivalves, but its locality (240-273 m) was obviously 
too deep for Corculum. In es the inner shell layer 
is iridescent, indicating presence of a nacreous layer. 
This conchological feature alone precluded inclusion of 
this new species in the Cardiidae. However, lack of soft 
parts preve ted a better resolved familial or supra- 

familial placement of the new taxon. The publication by 

Leal (2008) allowed allocation of the unknown bivalve in 
that newly established genus of the Poromyidae. In this 
paper, we describe this most interesting new species and 
report a new record of the genus from the Northwest 
Pacific. 


MATERIALS AND METHODS 


Four unpaired valves were sorted from sediments 
dredged trom southeast of Kamogawa, Chiba Prefecture, 
Japan, with a biological dredge (R/V TANSEI-MARU, cruise 
KT-00-05, station 1, Aen m, 34°59.963' N, 
140°27.159' E— 35°00.020' 140°28.427' E, May 17, 
2000). All specimens were so and disarticulated. The 
shells were coated with platinum and vanadium and pho- 
tographed with a scanning electron microscope (Hitachi 
S-2250N), following a standard method (see Geiger et 
al., 2007). The holotype and two paratypes are deposited 
in the Department of Historical Geology and Paleontol- 
ogy, The University Museum, The University of Tokyo 
(U MUT), and one paratype in The Bailey-Matthews 
Shell Museum, Sanibel, Florida (BMSM), 


SSS SSS SSS SSS SSD 


T. Sasaki and J.H. Leal, 2008 


Figures 1+. 


valve. 1. Holotype, UMUT RM 29689. 2. Paratype 


1, UMUT RM29690, 3. Paratype 2, UMUT RM29691, 4. Paratype 


Dilemma japonicum new a saa 1. Lateral view. 2. Posterior view. 3-4. Interior view. 1-3. Left valve. 4. Right 


3, BMSM 


17983. Abbreviations: et = cardinal-like ico e = escutcheon; k = keel; It = lateral tooth; p = projection in hinge; s = socket. 


SYSTEMATICS 


Superfamily Poromyoidea Dall, 1886 
Family Poromyidae Dall, 1886 
Genus Dilemma Leal, 2008 


Type Species: Dilemma frumarkernorum Leal, 2008, 
by original designation. 


Diagnosis: Shell compressed anteroposteriorly and 
expanded laterally; lateral outline of articulated valves 
cardioid; umbones projecting dorsally and located ante- 
riorly; sharp oblique carina dividing anterior and poste- 
rior regions; maximum growth axis having ca. 30° against 
anteroposterior axis; hinge axis short with cardinal-like 
tooth and socket in each valve; lateral tooth reduced, 
present only in right valve; ligament external, double- 
layered: lithodesma absent; shell interior lined with 
sheet-like nacreous layer. See Leal (2008) for anatomical 
characters. 


Dilemma japonicum new species 
Figures 1-12 


Diagnosis: Posterior region of shell sculptured by 
rough, lamellate, thin, commarginal ribs; surface irregu- 
larly punctate on anterior region but puncta radially ar- 
ranged on posterior region; escutcheon not distinct de- 
marcated; prominent projection present in posterior re- 
gion of hinge. 


Description: Shell thin, fragile, compressed antero- 
posteriorly; shell height larger than shell length. Anterior 
and posterior regions clearly demarcated by sharp keel 
Figure 1, k): anterior region narrower and less inflated 


than posterior. Surface of anterior region smooth, flat- 
tened in mz acroscopic view, but microscopici ally punctate 
(Figure 5). Surface of posterior region sculptured by sev- 
eral weak radial ribs (Figure 3), aad more prominently so 
by dense commarginal ribs, also punctate in enlarged 
view, but puncta arranged neatly in radial direction un- 
like on anterior region (Figure 6). Umbo angulated by 
keel and rib along posterior margin (Figure 7); umbonal 
cavity deep: umbo involute (Figure 3). Prodissoconch 
simple, flattened, shield-like, ca. 200 2m in length, with- 
out prominent sculpture or division into prodissoconch I 
and II (Figure 8). Hinge provided with single cardinal- 
like tooth and socket on each valve (Figures 3-4, 7, et, s); 
cardinal-like tooth located anterior to socket on left valve 
(Figure 3), their position reversed on right valve (Figure 
4). Lateral tooth small, present only on right valve ( (Fig- 
ure 4, It). Single sharp projection ( (Figures : 3-4, 7, p) 
prominent on posterior side of hinge in similar postion 
and size in right and left valves. Ligament and outlines of 
muscle scars not observed in disarticulated valves. Shell 
margin consisting of two layers, outer homogeneous 
structure and inner nacreous structure (Figure 9-10, 
OL, IL). In outer layer elongate granules arranged ver- 
tically but lacking clear boundary. In inner layer suboval 
tablets fusing in growth region (Figure 11) 
sheet-like layers forming nacre (Figure 12). 


numerous 


Type Material: Holotype, 4.7 mm (height: SH) « 2.0 
mm (length: SL), UMUT RM29689 (Figure 1); Paratype 
1,5.6 mm (SH) « 2.4 mm (SL), UMUT RM29690 (Fig- 
ure 2); Paratype 2, 6.0 mm (SH) x 3.0 mm (width) 
UMUT RM29691 (Figure 3); Paratype 3, 9.5 mm (SH) 
4.0 mm (SL), BMSM 17983 (Figure 4). 


Page 168 


THE NAUTILUS, Vol. 122, No. 3 


Figures 5-8. ; 
hinge. 8. Prodissoconch. 5-6. Holotype, UMUT RM29689. 
tooth; p = projection in hinge; pd = prodissoconch; s = socket 


Southeast off Kamogawa, Chiba Pre- 
34°59,963'-35°00.020" N, 


Type Locality: 
fecture, Japan. 240-273 m, 
140°27.159'—140°28.427' E. 


Distribution: Known only from the type locality. 


Etymology: The species epithet recognizes the coun- 
try of the type loc: lity. The epithet japonicum agrees in 
vender with the name Dilemma, a late-Latin neuter noun 
derived from the Greek. Of the previously named species 
in the genus, inexpectatum is a neuter gender epithet, 
frumarkernorum has a genitive ending that is not to be 
influenced by the gender of the genus, and spectralis is 
an epithet originally used by Leal (2008) as a noun in 
apposition, 


DISCUSSION 


The inclusion of the new species in the genus Dilemma 
is well supported by the presence of the diagnostic shell 
characters of the genus: ante roposte riorly compresse od 
shell, cardioid outline in articulated valves, shi arp carina 
dividing the shell into the anterior and posterior regions, 


Dilemma japonicum new species. 5-6. Detail of surface sculpture. 


Anterior slope. 6. Posterior slope. 7. Detail of 


7-8. Paratype 2, UMUT RM29691. Abbreviations: et = cardinal-like 


cardinal-like tooth and socket in each valve, and nacreous 
shell interior. Obviously there is no other group having a 
combination of these characters in bivalves. 

From the three previously known species of the genus, 
Dilemma inexpectatum (Crozier, 1966) from northern 
New Zealand is most similar to the new species in having 
rough commarginal sculpture in the posterior region, 
However, the two species differ in two characters: the 
rows of pores are absent and the escutcheon is distinct in 
D. inexpectatum. As summarized in Table 1, the four 
species in the genus can be distinguished among them- 
selves by four he Il characters: sculpture in the posterior 
region; the presence or absence of small pits; the dis- 
tinctness of the escutcheon; and the presence or absence 
of the posterior projection of the hinge. The paired pos- 
terior projections are present only in D. japonicum, they 
are not part of the hinge teeth syste m, because they do 
not articulate as a tooth and a socket between the right 
and left valves. Untortunately, the outlines of muscle 
scars were unclear in the spe ales of the new species. 

The original description of risa potas 

“Shell apparently nacreous internally Leal, 2008: : 
but did not offer any detailed dese ios or se 


T. Sasaki and J.H. Leal, 2008 


Page 169 


Figures 9-12. Dilemma japonicum new species. Shell microstructure. 9-10. Observed at small broken part of shell margin, vertical 
section of homogenous structure in outer layer (OL) lined by nacreous structure in inner layer (IL). 11-12. Oblique (11) and 
horizontal (12) views of nacreous layer at inner shell surface. Paratype 2, UMUT RM29691. 


on the shell microstructure of the three originally in- 
cluded species. The description of the inner nacreous 
shell layer in the present study confirms its presence in 
Dilemma. 

A shell consisting of outer homogeneous and inner 
nacreous layers (Figure 9-12) supports inclusion of 
the new species in Anomalodesmata (Taylor et al., 1973; 
Prezant, 1998). The microstructure of the outer layer is 


somewhat similar to a simple prismatic structure in that 
elongate granules are arranged vertically. The outer layer 
is identified as a homogeneous structure, because, unlike 
typical prismatic structure, crystals lack sharp bound- 
aries. It should be observed, however, that environmen- 
tal changes can alter otherwise organized shell micro- 
structures to appear irregular, smoothed, and homoge- 
neous (R. Prezant, pers. comm.). 


Table 1. Shell characters and distribution of the four Dilemma species. 


D. frumarkernorum 


D. spectralis 


D. japonicum 


D. inexpectatum new species 


Sculpture in posterior Dense, coarse growth Smooth 
region lines 

Small pits on exterior Absent Present 
surface 


Escutcheon Not clearly separated 


Projection posterior Absent Absent 
to beak 
Distribution Off Key West Florida, Off Vanuatu, 


USA, 229 m 


Not clearly separated 


Southwestern Pacific, 
950-961 m 


Rough commarginal Rugose. lamellate 
ribs and grooves 


Absent 


commarginal ribs 
Present 
Clearly defined Not clearly separated 
Absent Present 
Off Three Kings Off central Japan, 
Islands, New Zealand, Northwest Pacific. 
$05 m 240-273 m (dead 


Page 170 


THE NAUTILUS, Vol. 122, No. 3 


The prodissoconch of Dilemma japonicum new species 
indicates non-planktotrophic, lecithotrophic develop- 
ment. It measures ca. 200 jm in length (Figure §), a size 
that fits into the predicted size range (135-230 xm) of 
prodissoconchs of bivalves with lecithotrophic develop- 
ment (Ockelmann, 1965; Jablonski and Lutz, 1983). The 
developmental modes of the other species of the genus 
are unknown. 


Although two of the other live-collected species of 


Dilemma were found attached to hard substrata, the mi- 
crohabitat of the new species is unknown. The bottom 
sediment from the type locality contained numerous 
dead shells and other biogenic fragments such as those of 
bryozoans and sponges. Because most bivalves were dead 
and disarticulated, large part of samples in the dredge 
haul was inferred to be transported and soe cunuilered. 
Specifying the actual microhabitat of the species is an 
interesting target for future sampling. 

The description of Dilemma japonicum extends the 
geographic range of the genus to the Northwest Pacific, 
in addition to previous records from off Florida and the 
Southwestern Pacific (off Vanuatu and New Zealand). 
The genus has a broad geographic range, extending 
across the Panama land bridge and Eastern Pacific we 
rier. Future sampling of deep- sea hard substrates may 
yield additional records of Dilemma species from other 
locations in the Pacific and in other oceans. 


ACKNOWLEDGMENTS 


The authors are indebted to the reviewers, Riidiger 
Bieler, Robert S. Prezant, and Richard C. Willan, for 
their comments and suggestions. Richard Willan and Rii- 
diger Bieler helped with etymologies and confirmed the 
gender of the name Dilemma. The samples of the new 
species became available with the kind assistance of Dr. 
Suguru Ohta (formerly Ocean Research Institute, Uni- 


versity of Tokyo), other colleagues on board and crew of 
R/V TANSEI-MARU ) (Japan Agency for Marine-Earth Sci- 
ence and Technology, formerly Ocean Research Insti- 
tute, University of Tokyo) during the cruise KT-00-05. 


LITERATURE CITED 


Crozier, M. A. 1966. New species and records of Mollusca from 
off Three Kings Islands, New Zealand. Transactions of the 
Royal Society of New Zealand, Zoology 8: 39-49. 

Farmer, M. A, W. K. Fitt, and R. K. Trench. 2001. Morphology 
of the symbiosis between Corculum cardissa (Mollusca: 
Bivalvia) and Symbiodinium corculorum (Dinophyceae). 
Biological Bulletin 200: 336-343. 

Geiger, D. L., B. A. Marshall, W. F. Ponder, T. Sasaki, and 
A. Warén. 2007. Techniques for collecting, handling, and 
propane small molluscan specimens. Molluscan Re- 
search 27(1): 1-49 

Jablonski, D. and R. A. Lutz. 1983. Larval ecology of marine 
benthic invertebrates: paleobiological implications. Bio- 
— Reviews 58; 21-89, 

Leal, J. H. 2008. A remarkable new genus of carnivorous, 
A bivalves (Mollusca: Anomalodesniats: Poromyidae) 
with descriptions of two new species. Zootaxa 1764: 1-18. 

Ockelmann, W. K. 1965. Developmental types in marine bi- 
valves and their distribution along the Atlantic coast of 
Europe. In: Cox, L. R. and Peake, J. F. (eds.) Proceedings 
of the First European Malacological Congress, London, 
1962, Conchological Society of Great Britain and Ireland, 
and the Malacological Society of London, London, 
pp- 25-35. ; 

Prezant, R. S. 1998. Subclass Anomalodesmata. In: Beasley, 
P. L., Ross, G. J. B. and Wills, E. (eds.) Mollusca: The 
Southern Synthesis. Fauna of Australia. Vol. 5, Part A. 
SCIRO Publishing, Melboume, pp. 397-249. 

Taylor, J. D., W. D. Kennedy, and A. Hall. 1973. The shell 
structure and mineralogy of the Bivalvia. II. Lucinacea- 
Clavagellacea, Conclusions. Bulletin of the British Mu- 
seum (Natural History), Zoology 22: 255-294, 


THE NAUTILUS 122(3):171-177, 2008 


Page 17] 


On the genus Heteroschismoides Ludbrook, 1960 (Scaphopoda: 
Gadilida: Entalinidae), with descriptions of two new species 


Victor Scarabino 

Département Systématique et E volution 
Muséum national d'Histoire naturelle 
UMS Taxonomie-Collections CP51 

55 rue de Buffon 

75005 Paris, FRANCE 
scarabino_victor@yahoo. fr 


Janeiro 
Av. Pasteur, 455 


Carlos Henrique Soares Caetano 


Departamento de Zoologia, 
Universidade Federal do Estado do Rio de 


Rio de Janeiro, 22.290-240 BRAZIL 
cheaetano@zipmail.com.br 


ABSTRACT 


Heteroschismoides is a deep-sea genus characterized by shells 
being between 10 and 20 mm as adults, sculptured by 9 to 10 
prominent primary ribs and a unique deep irregular apical fis- 
sure on dorsal side, considered until now to include a single 
species: Dentalium subterfissum Jeffreys, 1877. During revision 
of material from several expeditions carried out by the Museum 
National d'Histoire Naturelle, Paris, two new species were 
identified and are here described: H. meridionalis new species 
and H. antipodes new species In addition, designation of the 
lectotype of H. subterfissus is proposed, as well as new records 
for this species in the northeastern Atlantic ocean are given. 
Heteroschismoides meridionalis new species is closely related 
to H. subterfissus, but the first has a smaller maximum diameter 
of shell and smaller apical aperture diameter. Heteroschis- 
moides antipodes new species from Solornon Islands is smaller 
than other two species for both, shell length and fissure exten- 
sion. The distance of point of maximum curvature from the 
apex in H. antipodes new species is located nearer to the apex 
than in H. meridionalis new species and H. subterfissus. The 
results here obtained considerably enlarge the geographical 
distribution of the genus and suggest a worldwide bathyal and 
abyssal distribution for this genus. 


Additional Keywords: Heteroschismoides subterfissus, Het- 
eroschismoidinae, tusk shell, lectotype, new species, Brazil, 
Solomon Islands, deep-sea 


INTRODUCTION 


The genus Heteroschismoides was proposed by Lud- 
brook in 1960 to include the deep sea species De ntalium 
subterfissum Jettreys, 1877 (for further supraspecific in- 
formation see Steiner and Kabat, 2001). The most re 
markable character of the genus is a unique, deep and 
irregular apical fissure at the dazed! side of shell not present 
among other Scaphopoda (Chistikov, 1982; Scarabino, 
1995). Since Chistikov (1982), this genus has been con- 
sidered as monospecific and restricted to the Atlantic 
Ocean. During the revision of material from several ex- 
peditions carried out by the Muséum national d'Histoire 


naturelle, Paris, in the northeastern and southwestern 
Atlantic Ocean (Brazil) as well as off Solomon Islands in 
the Pacific, a number of specimens corresponding to the 
genus were identified. In a preliminary sorting, Brazilian 
specimens were placed under H. subterfissus, but the 
unexpected finding of representatives in the tropical Pa- 
cific, decided us to undertake the present revision. Here, 
we propose the lectotype of Heteroschismoides subter- 


fissus and describe two new species, extending consider- 


ably the geographical distribution range of the genus. 


MATERIALS AND METHODS 


The material was collected during several expeditions car- 
ried out by the MNHIN and IrREMER, known as Brocas I to 
XI (1972-1981), THALASSA (1970-73) and INCAL (1973): 
in the Gulf of Gascony; ABYPLAINE (1982): off Portugal 
and Spain; NORATLANTE (1969): large coverage of the 
North Atlantic; BIACORES (1971): off Azores; BIOVEMA 
(1977): VeMA Trench; MD55 (1987): Southeastern Bra- 
zil: and SALOMON 2 (2004): off Solomon islands. 

In view to better define the species, a morphometric 
analysis was attempted based on selected undamaged 
shells of the three species. Shell measurements were 
taken according to Shimek (1989), Steiner (1999) and 
Steiner and Linse (2000), including length (L); maximum 
diameter (Max), that is in this case equal to the anterior 
aperture diameter; maximum curvature (Arc); distance 
of point of maximum curvature from the apex (Larc); 
apical aperture diameter (Apd). Besides those measure- 
ments, we also estimated the extension of the apical fis- 
sure (Fiss) located at dorsal side of shell. To assure in- 
dependence among variables, we performed a prelimi- 
nary correlation analysis among all variables with strongly 
correlated variables (r > 0.70) be sing excluded. The non- 
parametric Kruskal-Wallis test was employed to examine 
for differences of each morphometric parameter (un- 
transformed data) between species. Dunn’s multiple 
comparisons test was used a posteriori to assess signifi 
cant differences between species. A multivariate approach 


aaa: 179 
Page 172 


THE NAUTILUS, Vol. 122, No. 3 


was carried out utilizing Discriminant Function (DF) 
Analysis to integrate all morphometric data in a single 
analysis. To perform this analysis, we standardized the 
morphometric data following Romesburg (1984). 

Institutional abbreviations used in the text are; BMNH: 
The Natural History Museum, London; IFREMER: Insti- 
tut Francais de Recherche pour lExploitation de la Mer; 
MNR]J: Museu Nacional, Rio de Janeiro, Brazil; MNHN: 
Muséum national d’Histoire Naturelle, Paris; USNM: 
National Museum of Natural History, Smithsonian Insti- 
tution, Washington, DC, USA. Other abbreviations are: 
CP: beam trawl; DC: “Charcot” dredge; stn: station; lv: 
live-collected; dd: shell only. 


SYSTEMATICS 


Order Gadilida Starobogatov, 1982 

Suborder Entalimorpha Steiner, 1992 

Family Entalinidae Chistikov, 1979 

Subfamily Heteroschismoidinae Chistikov, 1982 
Genus Heteroschismoides Ludbrook, 1960 


Type Species: Dentalium subterfissum Jeffreys, 1877 
(by original designation). Recent, northeastern Atlantic 
Ocean. 


Description (Modified from Scarabino, 1995: 302): 
Shell up to 20 mm length, slightly curved, regularly ta- 
pering. Translucent grey when the animal is alive, 
opaque to polished dhe or cream when empty. Longi- 
tudinally sculptured by 9-10 primary ribs that reach a 
anterior aperture or vanish towards it. Intercostals spaces 
convex or straight, smooth or finely striated longitudi- 
nally by 8 to 14 lines. Secondary ribs, predominantly a 
single one, can be present in each intercostal space. Apex 
with a long and wide irregular fissure on dorsal side early 
observed in late embryonic stages and juvenile shells. 
Section polyg gonal, more notorious at the central zone of 
shell, fading or not towards the slightly laterally com- 
pressed fan oral aperture in adult specimens. Shell ee 
tures and repairs are frequently observed, as well < 
twisting of shell. 

RapuLa: Rachidian tooth with anterior margin 
rounded and lateral margins thick; laterals with sharp 
pointe ral prime ary a and 4 importe int denticles; mar- 
ginal slightly curved with conspicuous lateral processes, 
better observed in light microscope. 


Distribution: Recent, Atlantic Ocean and tropical Pa- 
cific Ocean, bathy al- -aby ssal. 


Remarks: Scarabino (1995: 302) mentioned by mis- 
take “10-12 primary ribs”; in the light of new information 
this number is here corrected as 9 to 10 primary ribs. 


Molecular data of H. subterfissus has been published by 
Steiner and Dreyer (2003). 


Heteroschismoides subterfissus (Jeffreys, 1877) 
(Figures 1-6, 15) 


Dentalium subterfissum Jettreys, 1877: 154; 1883: 660, pl. 49, 
fig. 3; Watson, 1879: 516; 1886: 10, pl. 1, fig. 10; Pilsbry 
and Sharp, 1897: 61, pl. 7, figs. 15-19; Warén 1980: 53. 

Dentalium (Heteroschisma) sibie wrfissum: Henderson, 1920; 58. 

Dentalium (Dentalium) subterfissum: Nickles 1979: 47, fig. 5a b. 

Heteroschismoides subterfissus: Chistikov, 1982: 675, figs. 3, 5; 
Steiner 1998: 78: Steiner and Kabat, 2001: 446; 2004: 651. 

Description: Shell up to 9.2 mm length, slightly 

curved, regularly tapering. Translucent shiny grey, pres- 

enting agtuies and repairs. Longitudinal sculpture of 
nine primary notorious ribs and ¢ a single secondary one in 
between, more notorious in the dorsal half, all reaching 
the thin oral aperture. Intercostals spaces conv on finely 

striated longitudinally by eight lines. Apex with a 2.3 mm 

long, wide, irregular fissure on dorsal side. Transversal 

section poly gonal, slightly compressed laterally, less ap- 
parent at apex. 


Type Material: — Lectotype, here selected (see remarks), 
the largest (9.2 mm) of the three specimens of lot USNM 
175018, Ireland, 54°19’ N,11°50'’ W, 1180 fims [2158 m] 
(Porcupine 1869 stn 16); other paralectotypes: USNM 
175017, Off West coast of Ireland, (Porcupine 1869 stn 
19a), 2 specimens; USNM 175019, Greenland, 56°11'N 
37°41'W, 1450 fms [2646 m], Valorous stn 12, 1 speci- 
men: USNM 175020, 55°40'N 12°46’W, 1476 fms [2694 
m], Porcupine L569 stn 21, 3 specimens; off Azores, 
Challenger Expedition stn 78, 37°26! 95°13’ W, 
Azores, 1000 fathoms [1825 m], 7 specimens and 3 frag- 
ments, BMNH_ 1887.2.9.36-40. Two other paralecto- 
types from nae gee expedition are deposited at 
BMNH under code 1885.11.5.1393—4, with a mention 
“further syntypes in USNM 175017-020", 

The fact of the species has been partially described on 
the base of a young specimen (see Remarks) and that the 
specimen illustrate od was not found in the collections re- 
vised, lead us to designate the lectotype based on an 
adult shell. 


Measurements of Lectotype (mm): Length 9.20, 
oral aperture 1.00, apex 0.12, are 0.42 at 1.9 from apex. 


Type Locality: West of Ireland, 54°19'N, 11°50'W, 
1489 m, Porcupine 1869, stn 16 (here selected). 


Other Material Examined: Off Gabon (Af 
S, 8°18’ E, stn DS20, 2514 m (1 dd); Gulf of Guinea, 
4°40°N, 5°39’2E): stn DS30, 3109 m (S dd) (both WaLbDA 
Expe eee CNEXO, Nicklés, 1979), all MNHN. 


Yica), 2°32’ 


Figures 1-14. Heteroschismoides species. 1-6. H. subterfissus (Jeffreys, 1977). 1. Lectotype USNM 175018, 9.2 mm. 2. Para- 


lectotype. 3. Detail of the apical fissure. 4. Specimens showing embryonic shell. 
10. H. meridionalis new species. 


microsculpture on the outer surface of intercostals space, 7- 


5. Detail of longitudinal ribs. 6. Detail of 
7. Holotype, MNHN, 14 mm. 8. 


Paratype, detail of apex. 9. Detail of sculpture. 10. Detail of microsculpture on the outer surface of intercostals space. 11-14. 
Heteroschismoides antipodes new species. 11. holotype, MNHN, 9.95 mm. 12. Detail of apex. 13. Detail of sculpture. 14. Detail of 


microsculpture on the outer surface of intercostal space, 


2008 


V. Searabino and C. H. S. Caetano, 


Page 174 THE NAUTILUS, Vol. 122, No. 3 


1haem 


Figures 15-17. Radulae of Heteroschismoides species. 15. H. subterfissus, rachidian and lateral teeth, intemal view. 16. H 
antipodes new species, external view of lateral tooth (left), rachidian, lateral, and marginal teeth (right), in internal view. 17. H 
meridionalis new species: rachidian and lateral teeth, internal view. 


New Records: Central Atlantic Ocean: Azores Islands: 11°06’ W, 2466 m, 3 lv, 1 dd: stn CP 04, 56°33’ N, 11°11! 
BIACORES 1971: stn DS 54, 38°12’ N, 28°15’ W, 1S10 m, W, 2438-2513 m, 7 lv, 7 dd: stn CP 05, 56°00! N, 12°29’ 
2 dd, stn DS 165, 37°33’ N, 25°58’ W, 2085-2050 m, 1 W, 2884 m, 1 dd: stn CP 06, 55°02.3' é. 12°40’ W, 
dd, DS 173, 37°57' N, 26°08’ W, 3225 m, 1 dd: Norar- 2888-2893 m, 3 lv, 1 dd: stn CP 07, 55°03’ N, 12°46’ 

LANTE 1969: stn 02, 53°55’ N, 17°52’ W, 2456 — 9420, 3 2895-2897 m, 1 dd; stn CP 09, 50°15’ N, 13°16’ 
lv; stn 84, 36°21' N, 08°43’ W, 2871-2875 m, 3 lv, 3 dd: 2659-2691 m, 2 lv; stn WS OL, 50°19" N, 13°08’ . 
stn 85, 36°25’ N, 08°48’, 257: 3-2580 m, 2 ie 3. dd; 2550-2539 m, 5 lv; stn WS 02, 50°19’ N, 12°55' W, 
THALASSA ee Sth X 336, 44°11’ N,O5°10' W, 1850— pees m, | lv: stn WS 03, 48°19’ N, 15°23’ W, 4829 
2050 m, 1 Iv, 1 dd; Eastern Atlantic Ose an: INCAL 1976: m, 2 dd: stn WS 09, 47°28.8 N-47°27.9’ N, 09° 234! W, 
stn. DS *" 57°59’ N, 10°40’ W, 2091 m, 56 lv, 7 dd: stn. 4277 m, 1 dd,; stn OS 01, 54°14’ N, 13°11’ W, 2634 m, 
DS 02, 57°59’ N, 10°49’ W, 2081 m, 51 lv, 3 dd: stn. DS L lv; Brocas I: stn DS 32, 47° 32" N. 08°05’ W, 2138 m, 
05, ee N, 11°12’ W, 2503 m, 24 lv, 7 dd; stn. DS 07, 3 lv, 3 dd; Bocas HI: stn DS 37, 47°32’ N, 08°35’ W, 
55°01" 12°31’ W, 2884 m, 4 lv, 1 dd; stn. DS 07, 2110 m, 2 dd; Brocas IV: stn DS 52. 44°06' N, 04°22' W, 
56°27’ : 11°11’ W, 2884 m, 5 lv, 1 dd: stn. DS 08, 2006 m, 2 dd: stn DS 51, 44°11’ N, 04°15’ W, 2430 m, 
55°02’ N, 12°35’ W, 2891-2884 m, 3 lv, 2 dd: stn. DS 09, L lv; stn DS 58, 47°34" N, 09°08’ a 2775 m, 3 dd: stn 
55°08’ N, 12°53’ W, 2897 m, 2 lv, 2 dd; stn. CP O1, DS 62, 47°33’ N, 08°40' W, 2175m, 1 dd: stn DS 64, 
57°58’ N, 10°55’ W, 2040-2068 m, 12 lv; stn. CP 02, 47°29’ 'N, 08°35' W, 2156m, 4 lv, 1 dd; BIOGAS V: stn 
57°58’ N, 10°42’ W, 2091 m, 2 dd: stn. CP 03, 56°38' N, CP 07, 44°09.8' N, 04°16.4' W, 2170 m, 5 dd; Brocas VI 
stn DS 61, 47°34.7' N, 08°38.8" W, 2250 m, 1 lv, 1 dd; 
6 ee ee stn DS 62, 47°33’ N, 08°40' W, 2175 m, 1 dd; stn DS 63, 
47°33' N, 08°35’ W, 2126 m, 1 lv; stn CP 08, 47°03’ N, 
>| | 08°39' W, 2177 m, 2 lv, 1 dd; stn CP 09, 47°33’ N, 08°44’ 
a} ° © Hsubterfissus (Jetheys, 1877) | W, 2171 m, 1 lv: stn CP 1 44°0S' N, 04°16’ W, 1995 m, 
| 3 dd; stn DS 71, 47°34’ N, 08°34’ W, 2194 m, 3 lv, 5 dd: 
3 a? stn DS 86, 44°04.8' N, 04°18.7' W, 1950 m, 1 dd; stn DS 
2 S7, 44°05’ N, 04°19’ W, 1913 m_3 lv: Blocas VIL: stn CP 
x | Ps 26, 47°33" N, 08°34" W, 2115 m, 1 lv; Biocas VII: stn 
e . 5 * KG 145, 47°33’ N, 08° ea W, 2170 m, 1 dd: stn KG 147, 
a| *% , eo. oe 17°33.40' N, “08°40.70" W, 2190 m, Iv, dd; stn KG. 149. 
: 6 ° 3° : 17°33’ N, 08°39' W, 2165 m, 1 dd: stn KG 151, 47°34’ 
a 4° " : N, 08°39' W, 2205 m, 1 lv; BloGas IX: stn CP 34, 47°32' 
2 a® N, 08°25’ W, 1970 m, 1 dd; BioGas XI: stn CP 37, 47°34’ 

° N, 08°41' W, 2175 m, 2 Wy. 

4 3 2 1 0 1 2 3 4 5 Distribution: — Eastern Atlantic Ocean: Greenland, off 
DF1 West Ireland, Gulf of Gascony, off Spain and Portugal to 
Figure 18. Discriminant function analysis of shell morpho Gulf of Guinea, West Africa (Je ffreys, 1877; Watson, 


netric parameters of Heteroschismoides species 1S79. ISS6. Loc: ard. LS9S: Nickles. 1979) _ Alive in 940 


V. Searabino and C. H. S. Caetano, 2008 


2987 m, shells down to 4529 m (present paper). Richest 
depth tor live material: 2040-2503 m (present paper). 


Remarks: _ Jeffreys partially described the - cies based 
on a juvenile specimen, since he mentioned ~ 
end bulbous” and illustrated this morphology in plate 49, 
fig. 3. However, no juvenile specimen was detected 
among the type material. In addition, he wrote that the 
slit is in the “under or ventral side”, which in fact corre- 
sponds to the dorsal side in normal position of the ani- 
mal. Jeffreys also mentioned the presence of 12 to 16 
ribs, but we did not find any specimens with that number 
of ribs. However, it is possible that secondary ribs and 
intercostal striae might have been, in part, counted in- 
appropriately for the original description. To clarify, in 
the description we enumerated only the primary ribs 
because the number of secondary ribs varies with age 
and among specimens but the number primary ribs is 
constant among specimens and through the life of the 
specimen. Specime ns with embryonic shell are often ob- 
served. 

Heteroschismoides subterfissus was also mentioned in 
the CHALLENGER expedition from “stn 120 - 8°37’ S 
34228’ W, Pernambuco, South America 675 fathoms 
{1232 m] red mud” (Watson, 1886). The lot in which this 
reference was based was not located in BMNH collec- 
tions by the senior author. In the distribution paragraph, 
Watson (1886) wrote “Habitat.- Davis Strait. 1785 fath- 
oms. North Atlantic; various Stations off the West Coast 
of Ireland. 1180 to 1476 fathoms (Jeff.).” He did not 
mention Pernambuco. 

The material from Pernambuco may belong to the new 
species described below or, as sugge sted by Henderson 
(1920: 58). could also correspond to specimens of Per- 
tusiconcha callithrix (Dall, 1889), since young specimens 
of this species, specially those with a heoken apex, can 
easily be misidentified as species of Heteroschismoides 
(Scarabino, pers. observ). 


Heteroschismoides meridionalis new species 
(Figures 7-10, 17) 


Description: Shell 14 mm length, slightly curved, 
regularly tapering, opaque white. Longitudinal sculpture 
composed of nine rounded-edge primary ribs, all reach- 
ing but fading towards the oral aperture. No secondary 
ribs. Intercostal spaces concave in posterior three quar- 
ters and straight to convex at the anterior fifth, present- 
ing 12 very faint, fine striae throughout. Apex with a 2 
mm long, ‘wide irregular fissure on dorsal side. Cross- 
section polyg gonal, less anaes at ape x ‘and at the slightly 


Ricasue cients of the Holotype (mm): Length 14.0, 
oral aperture 1.0, apex 0.1, arc 0.8 at 6.7 from apex. 


Holotype (lv) MNHN 20902 and 14 


Type Material: ' 
13 MNHN 20903-20906, 1 MNRJ 


paratypes (dd), 
12707. 


Type Locality: Off Espirito Santo, Brazil, 1$°59.1' S 
37°47.8' W, 1540-1550 m {MD 55 stn DC 70]. 


. poste rior 


Material Examined: Southwestern ne Ocean: 
MD 55 stn CP 68, 18°55.6' S,37°49.1' W, 1200-1500 m, 
1 dd (paratype MNHN 20903); stn OG 70, 18°59.1’ S, 
37°47'8 W, 1540-1550 m, 1 lv (holotype MNHN 20902), 
26 dd (7 paeype MNHN 20904, 1 paratype MNRJ 
12707): stn DC 72, 19°00.4' S, 37°48.8S' W, 950-1050 m, 

4 dd ( (paratypes MINHN 20905); stn CB 77, 19°40.6" S, 
ce W, 790-940 m, 9 dd (2 patie: MNHN 
20906): stn. CB 79, 19°O1.S' S, 37°47.8' W, 1500-1575 
m, 10 dd: sth CB 106, 23°54.2' S, 42°10.5' W, 830 m, 15 
dd; stn. CB 107, 24°00.3' S, 42°14.4' W, 1020 m, 11 dd. 


Etymology: Meridionalis: Southern. Name refers to 
the taxon’s distribution in the Southern Hemisphere. 


Distribution: Brazil: Espirito Santo and Sao Paulo. 
Alive in 1540 m, shells in 790-1540 m depth. 


Heteroschismoides antipodes new species 
(Figures 11-14, 16) 


Description: Shell to 9.95 mm length, slightly curved 
at the apex, re gularly tapering. Opaque white. I songitu- 
dinal sculpture ron ten primary ribs, all reaching the “onl 
aperture. Intercostal spaces concave and smooth. Apex 
with a wide irregular fissure 1.7 mm long on dorsal side. 
Cross section polyg gonal, less obvious at apex, slightly lat- 
erally compressed at the anterior aperture. 


Measurements of Holotype (mm): Length 9.95, oral 
aperture 0.9, apex, 0.1, are 0.4 at 2.76 from apex. 


Type Material: Holotype (dd) MNHN 20907 and 10 
paratypes (7 lv, 3 dd) MNHN 20908-20910 


Type Locality: Solomon Islands, 07°49.3" S, 
157°41.2' E, 1045-1118 m [SALOMON 2 Stn, CP2217]. 


Material Examined: Solomon Islands. SALOMON 2 
stn CP2182, 08°47.0' S, 157°37.9’ E, 762-1060 m, 16 dad. 
stn CP2197, 08°24.4' S, 159°22.5' E, 897-1057 m, 2 lv, 
2 dd (3 paratypes MNHN 20909); stn CP2217, 07°49.3' 
S, 157°41.2' E, 1045-1118 m, 1 ly, 3 dd (holotype 
MNHN 20907 and 3 paratypes MNHN 20908), stn 
CP2218, 07°56.3' S, 157°34.6' E, 582-864 m, 4 lv 
(paratypes MNHN 20910) 14 dd: stn CP2253, 7°26.5' S, 
156°15.0’ E. 1200-1218 m. 10 dd. 


nia. From Greek antipodes, meaning those 
living on diametrally opposed places on Earth. 


Distribution: Solomon Islands, alive in 1200 m:; shells 
762-1200 m depth. 


MORPHOMETRIC RESULTS AND DISCUSSION 


Table 1 lists the mean, standard deviation, minimum, 
and maximum values for each morphometric parameter 
for the three species. Significant differences in all mor- 
phometric parameters were observed between the spe- 
cies (Kruskal-Wallis test; p<0.01), except for the maxi- 
mum curvature (Arc). This univariate comparison 
showed that H. meridionalis new species and H. subter- 


Page 176 


THE NAUTILUS, Vol. 122, No. 3 


Table 1. Shell morphometric parameters with minimum (min), maximum (max), mean values and standard deviation (+SD) for the 
three new species of Heteroschismoides. L: length; Max: maximum diameter (= anterior aperture diameter); Arc: maximum curvature; 
Lare: distance of point of maximum curvature from the apex: Apd, apical aperture diameter; Fiss: fissure length. KW: Kruskal-Wallis 
test); ns: not significant; *p<0.001. The horizontal lines at Dunn’s test represents absence of statistical differences. 


H. meridionalis (n=12) H. antipodes (n=8) H. subterfissus (n=23) 
mean (+SD) min—max mean (+SD) min-max mean (+SD) min—max KW Dunn's test 
L 13.3 (1.6) 11.1-16.5 8.9 (0.8) 7.9-10.7 14.2 (2.6) 10.0-19.0 19.19° st sb an 
Max 1.0 (0.05) 0.9-1.1 1.0 (0.07) 1,0-1.2 4 (0.16) 1.0-1.7 29.76° st an sh 
Arc 0.5 (0.4) 0.1-1.2 0.5 (0.1) 0.4-0.7 0.5 (0.1) 0.2-1.0 0.82 ns st sb an 
Lare 6.1 (0.8) 4.8-7.2 3.5 (0.6) 2.14.0 5.4 (2.5) 2.2-11.4 14.98° st sb an 
Apd 0.1 (0.05) 0.1-0.2 0.2 (0.00) 0.2-0.2 0.2 (0.05) 0.2-0.3 20:10" st an os 
Fiss 2.1 (0.5) 1.5-3.1 1.4 (0.2) 11-15 1.9 (0.5) 1.2-3.0 13.32° st ban 


fissus are more similar, differing significantly only in 
maximum diameter (Max) and apical aperture diameter 
(Apd) (Table 1). Heteroschismoides meridionalis new 
species is also less shiny and the intercostal spaces are 
more densely striated: 12-15 striae (observed on stained 
or coated specimens due to the difficulty to be observed 
without any treatment) against S—10 in H. subte -rfissus 
(well observed under inert these data was obtained from 
counting 10 specimens of each species). 

Individuals of Heteroschismoides antipodes new spe- 
cies are significantly smaller than individuals of the other 
two species for both shell le ngth and fissure extension. 
The distance of point of maximum curvature from apex 
(Lare) in H. antipodes new species specimens is located 
nearer to apex than in specimens of both H. meridionalis 
new species and H. subterfissus (Table 1). In addition, 
H. antipodes specimens are opaque, due to their coarsest 
surface microsculpture. Specimens of this species have 
smooth intercostal spaces without secondary ribs and 
striae, and the primary ribs are more apparent. 

The multivariate discriminant function analysis was 
able to distinguish three groups among the specimens we 
examined that correspond to three species (Wilks’ 
Lambda = 0.12; F676 = 24.27; p < 0.0000). This analysis 
classifies about 97% of the cases correctly (only 1 out of 
43 specimens were incorrectly classified) (Figure 3). The 
model was constructed with 3 parameters (Max, L : Max, 
Lare). The variable L, which was highly correlated with 
Lare (r = 0.80) and Fiss (r = 0.74), was excluded. The 
Discriminant Functions (DF) based on the raw coeffi- 
cients of canonical variables are shown below: 


DF] = 2.5439Max + 1.0378L : Max — 0.7366Lare 
DF2 = 0.7249Max + 2.1404L : Max — 0.832S8Lare 


The radulae of the three species, here illustrated, are 
quite similar and further studies would be necessary to 
test the existence of interspecific differences, as it is the 


case in most Scaphopoda. Finally, the findings of Het- 
eroschismoides in the Solomon Islands show the genus to 
have worldwide distributed instead of being confined to 
the eastern Atlantic Ocean. However, better sampling of 
areas of the Indian Ocean (one of the ie »ss known tropical 
areas for scaphopod diversity) could eventually confirm 
whether the genus is also represented in that ocean. 


ACKNOWLEDGMENTS 


The authors want acknowledge the support and help re- 
ceived from Philippe Bouchet, Virginie Héros, and Phil- 
ippe Maestrati (MNHN), Ellen Strong (USNM), Kathie 
Way (BMNH), and Anders Warén (Swedish Museum of 
Natural History, Stockholm). Ronald Shimek is also 
gratefully acknowledged for his constructive comments 
on the manuscript. 


LITERATURE CITED 


Chistikov, S. D. 1952 The modem Entalinidae (Scaphopoda, 
Gadilida), 1. Subfamily Heteroschismoidinae - 1 [in Rus- 
sian]. Zoologicheskii Zhurnal 61 (5): 671-682. 

Dall, W. H. 1889. Reports on the results of dredging, under the 
supervision of Alexander Agassiz, in the Gulf of Mexico 
(1877-78) and in the Caribbean Sea (1879-S0), by the 
U.S. Coast Survey Steamer “Blake”, Lieut.-Commander 
C.D. Sigsbee, U.S.N., and Commander J.R. Bartlett, 
U.S.N., commanding. XXIX. Report on the Mollusca. Part 
2, Gastropoda and Scaphopoda. Bulletin of the Museum 
of Comparative Zodlogy at Harvard College 1S: 1-492, pls. 
1O—40. 

Jeffreys, J. G. 1877. New and peculiar Mollusca of the order 
Solenoconchia procured in the “Valorous” expedition. An- 
nals and Magazine of Natural History (4) 19: 153-158. 

Jeffreys, J. G. 1883. ie the Mollusca procured during the 
“Lighting” and “Porcupine * expeditions, 1868-70. Pro- 
ceedings of the Zoological Society of London 1882: 656— 
687, pl. 49-50 


V. Searabino and C. H. S. Caetano, 2008 


Page 177 


Henderson, J. B. 1920. A monograph of the East American 
scaphopod mollusks. United States National Museum Bul- 
letin 111: 1-177, 20 pls. 

Locard, A. 1898S. Mollusques Testacés. In: Expéditions Scien- 
tifiques du Travailleur et du Talisman pendent les années 
LSSO-1SS3. Masson, Paris, 515 pp., 1S pls. 

Nicklés, M. 1979. Scaphopodes de /Ouest-Africain (Mollusca, 
Scaphopoda). Bulletin du’ Muséum National d’ Histoire 
Naturelle, Section A. Zoologie, Biologie et écologie Ani- 
males (séries 4) 1(1): 41-77, 1 pl. 

Pilsbry, H. A. and B. Sharp. ee Scaphopoda, 1-280, in: 
Manual of C Jonchology, 17: 1-280, Conchological Section, 
Academy of Natural Sciences, Philade Iphia. 1 S97: 1-144, 
pls. 1- 26; 1S9S: i-xxxii + 145-280, pls. 27-37]. 

Romesburg, H. C. 1984. Cluster Analysis for Researchers. Life- 
time Le sarning Publications, Be ‘mont, 334 pp. 

Scarabino, V. 1995. Scaphopoda of the tropical Pacific and 
Indian Oceans, with descriptions of 3 new genera and 42 
new species. In: Bouchet, P. (ed.), Résultats des Cam- 
pagnes MUSORSTOM, vol. 14. Mé se s du Muséum na- 
tional d'Histoire naturelle 167: 189-37 

Shimek, R. L. 1989. Shells eee ae aa systematics: a 
review ot the slender, shallow-water Cadulus of the North- 

eastern Pacific (Scaphopoda: Gadilida). The Veliger, 32, 
233-246, 

Steiner, G. 1997. Scaphopoda from the Spanish coasts. Iberus 
15: 95-111. 

Steiner, G. 1998. Phylogeny of Scaphopoda (Mollusca) in the 
light of new anatomical data on the Gadilinidae and some 


eae matica, and a reply to Reynolds. Zoologica Scripta 
97: 73-82. 

Steiner, CG. 1999. A new genus and species of the family An- 
ulidentaliidae (Sez iphopoda: Dentaliida) and its systematic 
ae Journal of Molluscan Studies 65: 151-161. 

Steiner, G. and K. Linse. 2000. Systematics and distribution of 
the ey (Mollusca) in the Beagle Chanel (Chile). 
Mitteilungen aus dem ee Zoologischen Mu- 
— und Institut 97; 13—3¢ 

Steiner, G. and H. Dreyer. ae Molecular phylogeny of 
Scaphopoda ( (Mollusea) inferred from 18S rDNA se- 
quences: support for a Scaphopoda-Cephalopoda clade. 
ZA yes Scripta 32: 343-356, 

Steiner, G. and A. R. Kabat. 2001. Catalogue of Se ei 
taxa of Scaphopoda (Mollusca). Zoosystema 23: 433-460 

Steiner, G. and A. R. Kabat. 2004. Catz alogue of spe cies-group 
names of Recent and fossil Scaphopoda ( (Mollusca). Zoo- 
systema 26; 549-726 

Warén, A. 1980. Marine Mollusca described by John Gwyn 
Jeffreys, with the location of the type material. Concho- 
logical Society of Gre . Britain and Ireland, Special Pub- 
lesion 1. 1-60, pls. 1-8. 

Watson, R. B. 1879. vialinecs of the H.M.S. CHALLENGER Ex- 
pedition. H. The Solenoconchia, comprising the genera 
Dentalium, Siphodentalium, and Cadulus. Journal of the 
Linnean Society of London 14 (78): 506-529. 

Watson, R. B. 1886. Report on the Scaphopoda and Gaster- 
opoda collected by H.M.S. Challenger during the years 
1873-76. Challenger Report, Zoology 15, pt. 42, 756 pp. 
50 pls. 


THE NAUTILUS 122(3):178-181, 2008 


Page 178 


Parasitism of Monogamus minibulla (Olsson and McGinty, 


1958) 


(Gastropoda: Eulimidae) on the red sea-urchin Echinometra 
lucunter (Linnaeus, 1758) (Echinodermata: Echinometridae) on 


the Caribbean coast of Mexico 


Norma Emilia Gonzalez-Vallejo 
Depto. Ecologia Acuatica 

El Colegio de la Frontera Sur 

Apdo. Postal 424, 77014 

Chetumal, Quintana Roo, MEXICO 
negonzale@ecosur.mx 


ABSTRACT 


Eulimids are gastropod mollusks parasites of echinoderms. In- 
tertidal red sea-urchins, Echinometra lucunter, collected along 
the Mexican Caribbean shores were examined for the presence 
of those ectoparasitic gastropods. The analysis of 206 sea- 
urchins resulted in 56 having eulimids on them (27% preva- 
lence), mostly living in pairs, with about 7 eulimid pairs per 
sea-urchin. The eulimid ¢ gastropods found are Monogamus 
minibulla (Olsson and MacGinty, 1958), desonbed from 
Panama. Its host was unknown. This article consists of the first 


record of this host-parasite relationship. A short description of 


the shell structure and some details of the biological relation- 
ship are given. The females are much larger than the males, 
and the latter are usually attached to the female bodies. Most 
female eulimids live inside the hypertrophied foot, which this 
allows for partial oe of the entire body. A few females, 
however, were found living fixed directly on the echinoderm 
integument. The position of the g gastropod on the sea-urchin is 
fixe d via boring through the skeleton. Anchoring is achieved via 
the proboscis. Some ‘females had egg capsules ( (0.9-1.5 mm) 
fixed to their bodies; each is a spherical structure attached to 
the body by a short stalk. Egg capsules contained 14-15 em- 
bryos in different stages of development. The only previously 
publishec d account of this type of relationship inv olves a similar 
species een on Echinometra mathaei (Blainville, 1825) from 
the Red Sea. 


Additional Keywords: Rosenia, Echinometra mathaei, Monoga- 
mus interspinea 


INTRODUCTION 


The red sea-urchin Echinometra lucunter (Linnaeus, 
1758) is frequently found on shallow water rocky- 
bottoms along the Me xican Caribbean shores, and occa- 
sionally in sea-grass beds. In that region, some specimens 
living in the reef lagoons host one or more tiny whitish 
e alorud gastropods firmly attached to their bodie 3S. 
Ecological associations including eulimids gastropods 


have been recognized (Warén, 1983). The pee de- 
scription of the eulimid Rosenia minibulla Olsson and 
McGinty, 1958, was made from an empty shell found on 
the beach of Isla Bocas, Panama; however at that time, its 
host was unknown. 

Liitzen (1976) studied the sea urchins Echinometra 
mathaei (Blainville, 1825) from the Red Sea and the 
Western Pacific Ocean and found eulimid parasites on 
them. He named a new genus as Monogamus because he 
found these eulimids always living in pairs; he included 
two different species, one (M. entopodia Liitzen, 1976) 
living inside the hypertrophied tube feet of E. matthaei 
from the Red Sea, and another one (M. interspinea 
Liitzen, 1976) living on the integument of E. matthaei 
from several localities in the Western Pacific Ocean. 

Warén (1983) established Rosenia Schepman, 1914, as 
a junior synonym of Pelseneeria Koehler and Vaney, 
1908. However, the transfer of Rosenia minibulla to Mo- 
nogamus was informally proposed by Warén and 
Mooleal yeek (1989), perhaps because both share internal 
growth lines in the body whorl. In addition, the former 
euthor had found individuals of the species living on E. 
lucunter. 

The faunal associations of the red sea-urchin were 
a in the southern Caribbean Sea by Schoppe and 

Werding (1996), but these authors did not find eulimids 
ad with the red sea-urchin in the study localities. 
In this contribution, my objectives are to report the para- 
sitism of the eulimid Monogamus minibulla (Olsson and 
McGinty, 1958), on the red sea-urchin Echinometra lu- 
cunter (Linnaeus, 1758), found along the Caribbean 
coast of Mexico, and to describe some aspects of the 
relationship between the two species. 


MATERIALS AND METHODS 


The material was collected along several places in the 
Caribbean coast of Quintana Roo, Mexico: Isla Contoy 


N. E. Gonzalez-Vallejo, 2008 


Page 179 


(21°30'S.4” N, 86°47'45.3” 28/11/01); Playa Paraiso, 
C ne (20°35'21.5" N, Revie W, 24/111/01); Playa 
Ana and José (19°54'22.4" N, 87°26'14.5" W, 2 kin S 
from the junction to Boca Paila, 18/II/01); Majagual 
(18°43'/28” N, 87°42'05” W, 22/11/00): Punta Herradura 
(18°32'23” N, 87°44'32"” W, 27/VI/00): and Buenavista 
(18°30'42” N, 87°45'30" W, 15/V/99). 

Sea-urchins were eaunted and those with eulimids 
fixed, sorted and kept individually, and identified accord- 
ing to Hendler et al. (1995). The number of eulimids per 
sea-urchin was counted to determine the prevalence: the 
relationship between the number of infested hosts (sea- 
urchins) and the total number of hosts; and the intensity: 
number of parasites (eulimids) present in infected host 
(Margolis et al. 1952). Voucher specimens were depos- 
ited in the Swedish Museum of Natural History, Stock- 
holm, Sweden, SMNH 27858, and in the Reference Col- 
lection of El Colegio de la Frontera Sur (ECOSUR), 
Chetumal, Mexico). 


RESULTS 


Monogamus minibulla (Olsson and McGinty, 1958) 


Description: Mean shell length: 1.85 mm (range 1.5- 
2.5), width 1.15 mm (range 1.0-1.5). Females are easily 
distinguished due to their body-size, being twice as large 
as males. The shell is tiny, spherical, fragile, and trans- 
parent. The spire is low, the nuclear apex is long and 
erect like a stump (mucro). The shell color is white with 
glossy texture, with a very fine suture line. There are two 
post- -nuclear whorls: the first one is small and convex 

with a small shoulder close to the suture, which is be step r 
seen on the aperture side. The body whorl is large, and 
inflated, polished, with some delicate growth lines. The 
aperture is wide, the outer lip thin, expanded along its 
median part. There is no operculum. When the animal is 
alive, the mantle and visceral mass are deep-red. There 
are two distinct black eyes. Tentacles were not observed. 


Parasitism: From a total of 206 sea-urchins, 56 were 
infested with eulimids. The prevalence was of 0.27 (pres- 
ence of the parasite in 27% of the specimens). The in- 
tensity was of 7.03: there was an average of 7.03 eulimid 
pairs per sea-urchin (including only hose that were in- 
fested by at least one eulimid [T TT able 1]). 


Attachment Locations of Parasites on Hosts: Many 
parasites have a preference for the oral side of the host, 


although some were located on the dorsal side 
(periproct); in only two cases parasites were found on the 
peristomial membrane (buccal tube foot). Most eulimids 
invade the tube feet, becoming inserted in its integu- 
ment. This invasion can produce a hypertrophy of that 
tube foot. Often, the deformation is such that when the 
eulimid grows into adulthood, the tube feet are modified 
beyond recognition. The eulimids, in sal ae juveniles 
with a single teleoconch whorl, use the skin of the tube 
as a protective hood; later, as the animal grows, its own 
mantle (the pseudopallium) covers the fragile shell, ex- 
cept for the apical part, which is left uncovered (Fig- 
ure 1). 

Another attachement site is the interambulacral zone. 
Eulimids that are not covered by the pseudopallium are 
attached by the snout (plus pharyngeal anchor). Few oth- 
ers may be found adhering to the integument at the base 
of spines, where according to stage of shell growth, it 
may be collapsed, broken, and later disappear, leaving an 
open space where the “eulimid-scar” becomes evident. 

Eulimids may bore the sea-urchin test: this may be 
seen clearly when detached specimens were placed on a 
host. They may also enlarge the pore associated with the 
foot tube: a typical feeding pattern of ectoparasitic eu- 
limids involves insertion of the proboscis through the 
tube pore followed by suction of coelomic contents cat the 
sea-urchin. 

The gastropods were found mostly in pairs, but, less 
frequently, two or three males were found associated 
with a single larger female. These pairs look like irregu- 
larly shaped protrusions of the body wall, each with one 
or two circular openings through which the eulimid shell 
tips may project. Egg capsules were observed packed in 
globular sacks (3-5), attached for a short stalk to the body 
of the female, and partially or completely covered by the 
pseudopallium. Each capsule (size 0.09-1.5 mm in 
length) contained 14-15 embryos. Egg capsules con- 
tained all stages from early egg stage to small juveniles 
ready to hatch, 


DISCUSSION 


Olsson and McGinty (1958) described Monogamus 
minibulla based on an empty shell. In the present study 
details of the shell morphology are presented based on 
several alive spe cimens an ‘those fixed on the sea- 
urchins. Warén in Warén and Moolenbeek (1989) briefly 
pointed out the presence of M. minibulla associated to 


Table 1. Intensity and prevalence of Monogamus minibulla on Echinometra lucunter on the Caribbean coast of Mexico. 


Localities Sea urchins With parasites 


Eulimid females Prevalence 


Intensity 


Isla C Contoy 27 13 
P. Paraiso 3 ] 
P. Ana y José 51 14 
Majagual 120 23 
P. Herradura 3 3 

2 2 


Buenavista 


268 20.61 0.45 
] 1.00 0.30 
114 8.14 0.27 
9 0.39 0.19 
2 0.66 1.00 
] 5.50 1.00 


Page 150 


THE NAUTILUS, Vol. 122, No. 3 


Figures 1-4. 
3. Male (left, 


Echinometra lucunter. However, they only recorded the 
dimensions of the shell (1.5 mm of longitude and a larval 
shell of 1.0 mm), because their contribution was focused 


mainly in the relationship of the eulimid Trochostilifer 


eucidaricola (Warén and Moolenbeek, 1989) associated 
with the pencil sea-urchin Eucidaris tribuloides (La- 


marck, 1816). This study constitutes the first report of 


the relationship of Monogamus minibulla with Echi- 
nometra lucunter for the Mexican Caribbean region. 
Liitzen (1976) reported the presence of Monogamus 
entopodia in the tube feet of the Red Sea sea-urchin. In 
the Mexican Caribbean, M. minibulla infests the tube 
feet, but was also observed attached to the epidermis 
host, using its snout and tissues of the host but without 
being covered or protected with the pseudopallium. The 
two different patterns were rarely observed on the same 
sea-urchin. Liitzen (1976) had found the same pattems 
but on sea urchins collected in cifferent localities and 
regarded them as belonging to two different species 


Further, Echinometra from the Indian and western Pa- 


Vonogamus minibulla. 1. On Echinometra lucunter (host test = 16.54 min diameter 


1.5 mm) and female (right, 2.5 mm). 4. Female, anterior and lateral views (1.5 mim height 


8.72 mm height); 2. In situ. 
1.4 mm width.) 


cific oceans have been found to belong to three different 
morphospecies that might be re productive ‘ly isolated 
from each other as we I ( Arakaki and Uehara, 1999). 
Thus, it seems that Caribbean sea-urchins have not di- 
verged as much as those present in the Indo-Pacific 
Oceans. 

Few studies have detailed the prevalence and intensity 
of eulimids parasites, Salazar and Reyes (1995) studie d 
the relationship of Thyca sn Berry, 1959, and the 
starfish Phataria unifascialis (Gray, 1840), finding that 
the association was more Foes in two sites of the Gulf 
of California where human influence was more accentu- 
ated. In this case, only one of the localities (Playa Ana 
and José) is he avily influenced by development, and is a 
protected, semi-enclosed embayment, with many resi- 
dences and tourists. That locality presented two sea- 
urchins that had unusually high infestations of about 100 
eulimids each. Most of the sea-urchins were adjusted 
close to the average of about 7 eulimid pairs per sea- 


urchin. 


N. E. Gonzalez-Vallejo, 2008 


Page 18] 


Warén (1983) pointed out that many tropical eulimids 
that attach their egg capsules to their shells to or to their 
hosts, have almost invariably been found with spawn, an 
indication that spawning might be continuous. On the 
other hand, Liitzen (1976) suggested for M. entopodia 
that the presence of different developmental stages 
found simultaneously in samples of egg capsules Sidi: 
cates continual production of eggs, which is similar to 
what happens in M. minibulla. 

A thorough study of the association is required to elu- 
cidate the mechanisms of feeding, including histology. 


There is need to clarify whether the proboscis sucks of 


part of the liquid fraction of the host’s coelom or, as 
Liitzen (1976) pointed out for M. entopodia, the eulimid 
feeds exclusively of the area adjacent to attachment on 
the sea-urchins. 

To determine the specificity of this association it is 
necessary to do experiments and observations in the col- 
lection place, to determine whether, as Warén (1983) 
pointed out, there is lack of specificity of the parasite on 
a single species guest. Although I had opportunity to 
examine some specimens of Echinometra viridis Agassiz, 
1863, and I did not observe any attached eulimids, it 
could be said that if some preference exists in this case, 
but also this would have to corroborate it. Further, in 
2001 T had examined about 50 specimens belonging to E. 
lucunter from Guana Island, but none had any associated 
eulimid parasite. 


ACKNOWLEDGMENTS 


Many thanks to Anders Warén (Swedish Museum of 


Natural History), who always answered my questions 
about eulimids and is always ready to help. To Sergio 
Salazar-Vallejo and Maria Ana Tovar, who reviewed pre- 
vious drafts of the manuscript and to colleagues at 
ECOSUR for collecting the sea urchins for the present 
study. The careful reviews by Anders Warén and an 


anonymous referee enhanced the clarity of this contri- 
bution. Humberto Bahena helped improve the quality of 
the photographs. 


LITERATURE CITED 


Arakaki, Y. and T. Uehara. 1999. Morphological comparison of 
black Echinometra individuals among those in the Indo- 
Ney Pacific. Zoological Science 16: 551-558. 

Hendler, G., J. E. Miller, D. L. Pawson, and P. M. Kier. 1995. 
Sea vee Sea-urchins and Allies: Echinoderms of Florida 
and the Caribbean. Smithsonian Institution Press, Wash- 
ington, 390 pp. 

Liitzen, J. 1976. On a new genus and two new species of proso- 
branchia (Mollusca) parasitic on the tropical sea urchin 
Echinometra mathaci. Israel Journal of Zoology 25: 38-51. 

Margolis, L., G. W. Esch, J. C. Holmes, A. M. Kuris, and A. 
Schad. 1982. The use of ‘ecological terms in parasitology 
(report of an ad hoc Committee of the American Society of 
Parasitologists). Journal Parasitology 68: 131-133. 

Olsson, A. A. and T. L. McGinty. 1958S. Recent marine mollusks 
from the Caribbean coast of Panama with the description 
of some new genera and species. Bulletins of American 
Paleontology 39(177): 1-58. 

Salazar, A. and H. Reyes. 1998. Parasitismo de Thyca callista 
(Gastropoda: Capulidae) ) sobre Phataria unifascialis (As- 
teroidea: Ophidiasteridae) en el Golfo de California, 
México. Revista de Biologia Tropical 46(3): 1-4. 

Schoppe, A. and B. Werding. 1996. The boreholes of the sea 
urchin genus Echinometra (Echinodermata: Echinoidea: 
Echinometridae) as a microhabitat in tropical South 
America. Publicazioni della Stazione Zoologica di Napoli I, 
Marine oe 17: 181-186, 

Warén, A. 1983. A generic revision of the family Eulimidae 
(Gastropoda:Prosobranchia) The Journal of Molluscan 
Studies, Supplement 13, 96 pp. 

Warén, A. and R. Moolenbeek. 1989. A new eulimid gastropod, 
Trochostilifer eucidaricola, parasitic on the pencil urchin 
Eucidaris tribuloides trom the southern Caribbean. Pro- 
ceedings of the Biological Society of Washington 102: 
169-175. 


THE NAUTILUS 122(3):182, 2008 Page 182 


Erratum 


Due to an editorial lapse, the article by Simone and Cunha (2008) in the most recent issue of The Nautilus incorrectly listed “FMNH” 
instead of the correct form “UF” as the collection acronym for the Florida Museum of Natural History. 


LITERATURE CITED 


Simone, L. R. L. and C. M. Cunha. 2008. Revision of the genus Spinosipella (Bivalvia: Verticordiidae), with descriptions of two new 
species from Brazil. The Nautilus 122: 57-78. 


THE NAUTILUS 122(3):183-184, 2008 Page 153 


Notices 


Subscriptions rates for The Nautilus will increase slightly for 2009. Individual and institutional subscriptions will change respectively 
to $50 and $85. Shipping costs will remain the same. These adjustments are necessary given increases in production, materials, and 
mailing costs. 

The Nautilus staff has been making a concerted effort to increase the quality and quantity of articles published in the journal. 
The graph below shows the number of pages published in the last seven volumes (the number of pages for current volume 
122 is projected). The increase in number of pages has been steady, and we intend to keep this positive rate for the next 


volumes. 


250 The Nautilus - Pages/Volume 


200 


150 


No. Pages 


100 


50 


116 117 118 119 120 121 122 


Volume Number 


Given the increase in number of pages, the cost of one page to subscribers has actually declined in the past seven years, from 
$0.25 in 2002 [volume 116] to $0.19 in 2008 [volume 122], to individual subscribers, and from $0.40 to $0.33 to institutional 


subscribers. 


The graph below shows the price per page to individual and institutional subscribers and relevant trends for the past seven years 
based on subscription costs and number of pages. Please consider these figures and trends when renewing your subscription! 


05 The Nautilus - Cost per Page to Subscriber 


Cost in US Dollars 


—e— Individual 
—B— Institutional 
——— Trend (Individual) 


—-—- Trend (Institutional) 


116 117 118 119 120 121 122 


Volume Number 


Page 154 THE NAUTILUS, Vol. 122, No. 3 


The Nautilus has signed an agreement with the Biodiversity Heritage Library (BHL) Project. Ten major natural history museum 
libraries, botanical libraries, and research institutions have joined to form BHL. The agreement entails digitizing back issues of The 
Nautilus for open electronic access to the entire journal except for the last three volumes. One of the participating members of BHL, 
the National Museum of Natural History (Smithsonian Institution) Library in Washington, DC, started digitizing the journal. 


The Nautilus will be posted in the BHL portal (www.biodiversitylibrary.org) and in The Bailey-Matthews Shell Museum web site 
(shellmuseum.org/nautilus). Accordingly, BHL will receive 29 volumes of the journal previously scanned by friends of the journal. 
The Nautilus and its editor want to acknowledge the time and effort contributed by David Campbell, Wayne Harland, Fabio 


Moretzsohn, Richard E. Petit, and Colin Redfern in that first phase of the electronic access project. 


Sponsored in part by the State of Florida, Department 
of State, Division of Cultural Affairs, the Florida Arts 
Council and the National Endowment for the Arts 


yp PREP EG 
8 & ® 
Qe a x ee 
=) Cc} 
gS me NATIONAL 
p x ENDOWMENT 
a) a FOR THE ARTS 


OF cyrswe® 


si 


INSTRUCTIONS TO AUTHORS 


THE NAUTILUS publishes articles on all aspects of the 
biology, palgontolony and systematics of mollusks. 
Manuscripts describing original, unpublished research 
and review articles will be considered. Brief articles, not 
exceeding 1000 words, will be published as notes and do 
not require an abstract. Notices of interest to the mala- 
cological community will appear in a notices section. 


Manuscripts: Each original manuscript and accompa- 
nying illustrations should be submitted to the editor pref- 
erably via e-mail or as hardcopy in triplicate. 

Text must conform to the dimensions of 8% x 11-inch 
paper, double-spaced, and single-column throughout (in- 
cluding literature cited, tables, and figure captions). Au- 
thors should follow the general recommendations of Sci- 
entific Style and Format—The CSE Manual for Authors, 
Editors, and Publishers, available from the Council of 
Science Editors at www.councilscienceeditors.org. The 
first mention of a scientific name in the text should be 
accompanied by the taxonomic authority, including year. 
Latinized names and other words to be printed in italics 
must be underlined; leave other formatting indications to 
the editor. Metric, not English, units are to be used. Geo- 
chronologic modifiers should be capitalized only when 
units are formally recognized: for instance, use Late Cre- 
taceous but early Miocene. Likewise, only modifiers of 
formally recognized chronostratigraphic units are capi- 
talized: use Lower Jurassic but upper Oligocene. 

The sequence of sections should be title page, ab- 
stract, introduction, materials and methods, results, dis- 
cussion, acknowledgments, literature cited, tables, figure 
captions, figures. The title page should include the title, 
author’s name(s) and address(es). If corresponding au- 
thor is not the senior author, please indicate. The ab- 
stract should summarize in 250 words or less the scope, 
main results, and conclusions of the article. Abstracts 
should be followed by a list of additional key words. All 
references cited in the text must appear in the Literature 
Cited section and vice-versa. Please follow a recent issue 
of THE NAUTILUS for bibliographic style, noting that 
journal titles must be unabbreviated. Information on 
plates and figures should be cited only if not included 
within the pagination of cited work. Tables must be num- 
bered and each placed on a separate page. If in doubt. 
please follow a recent issue of the journal for sequence of 
sections and other style requirements. 


Illustrations: Illustrations are rendered either at full- 
page width (maximum width 17 cm) or column width 
(maximum width 8.2 cm). Please take these dimensions 
into consideration when preparing illustrations. Page- 
width illustrations ideally sould span the entire width of 
printed page (17 cm). “Tall” page-width illustrations 
should be avoided, square or “landscape” formats work 
better. Please design plates accordingly, such that there 
will be enough space left at the bottom of printed page 
for plate caption. (Digital technology has made this task 
much easier.) 


All line drawings must be in black, clearly detailed, 
and completely labeled. Abbreviation definitions must be 
included in the caption. Line drawings must be high 
resolution files at least 600 dpi (dots per inch) resolution 
at actual size. Standard digital formats for line drawings 
include .tif, .bmp, .psd, .eps, and .pdf. 

Photographs may be submitted in black-and-white or 
color, preferably in RGB mode if in color. Standard digi- 
tal formats for photographs include .tif, .psd, .jpg, or .pdf. 
Photographs must be high resolution files at least 300 dpi 
resolution at actual size. 

If more than one figure is included in an illustration, 
all figures are to be consecutively numbered (Figures 1, 
OT Bee ihe NOT Figures 1A, 1B, 1C,..., NOR Plate 1, 
Figure 1, .. .). In illustrations with more than one figure, 
make sure that blank areas between figures is kept to a 
minimum, thereby allowing for more area for each indi- 
vidual figure. 

Compressed files (e.g., .jpg) may be used to facilitate 
transmission of files during original submission, but may 
not be acceptable at final submission (see below). 


Voucher Specimens: Deposition of the holotype in a 
recognized institutional, public collection is a require- 
ment for publication of articles in which new species- 
level taxa are described. Deposition of paratypes in in- 
stitutional collections is strongly encouraged, as is the 
deposition of representative voucher specimens for all 
other types of research work. 


The Editorial Process: Upon receipt, all manuscripts are 
assined a number and acknowledged. The editor reserves 
the right to return manuscripts that are substandard or 
not appropriate in scope for THE NAUTILUS. Manu- 
scripts deemed appropriate for the journal will be sent 
for critical review to at least two reviewers. The review- 
ers’ recommendations will serve as basis for rejection or 
continuation of the editorial process. Reviewed manu- 
scripts will be sent back to authors for consideration of 
the reviewers’ comments. The revised version of the 
manuscript may at this point be considered accepted for 
publication by the journal. 


Final Submission: Authors of accepted manuscripts are 
required to submit a final version via e-mail to the editor 
at jleal@shellmuseum.org. Please do not send low-resolu- 
tion or compressed illustration files at this stage. Send any 
files larger than 20 Mb on a CD or DVD to the editor. 


Proofs: After typesetting, proofs will be sent to the au- 
thor. Author should read proofs carefully and send cor- 
rections to the editor within 48 hours. Changes other than 
typesetting errors will be charged to the author at cost. 


Offprints: An order form for offprints will accompany 
the proofs. Offprints will be ordered through the as ae 
Authors with institutional, grant, or other research sup- 
port will be asked to pay for page charges at the rate of 
$60 per page. 


@ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper) 


i 
i) 


Hl 


os 


| in 

ua y ay 

i a be iy : i 

a . nae nie on a ¥ 
sy ha i: 


i) : \ i i : : pitts ey eal uN 7 
mf At Fa) WE: hy i) au i i, . y} i wal { 
aa en Ni os ) M ie sap ae My ‘ aay 


14 
1 


mth 


a 


ino 
ny ia 


f Ay iy 
4 i oH ois Ie¥ 
ie a A 7 f Me i i we 4, @ ahi . 
; i ihe 1 oA Ya a) re fy At) Nh, im ii an hit ay, .. in 
‘ As Pad - 
tha | Donn Lara x, oP ‘ rt 4 ne a ; Tila! 7 ie 7 are Si) ik Mf 
th . yh f a te ty: 4 7 a , . i ui Mt vA ee ‘ id i f a ie ie i, i” Lyf VA] a Et, Ve 
; q NETS) on is) i f ) tnt ‘ is) Mi: vt Bi | nie i ai ty 1 by! iy ’ a) f Ht f v | H gi oad : ity i I My to aya 
ALEGRE Min Rin RATT ni ony CE MR a an ea ns 
| 7 an : uy ' " Ve 7 pie) u alte Mi ; 7 iy i i t . hi Oia . a ‘pi I" ie Re (i (yt wan ) ‘ fl ; 
| 7 i ey me iar wt ‘ a | ‘ a y AL : ie ‘ WP ? re it ‘ us oF 
ee Ma ee iy?’ ery, 1; nn meek wea f Tear ara 
| i pe om ‘ j Ai ; v | A eae Oh bal Ny ae ) yi ih a f i ty dt ; 
PLAC any , a 7 a i/ "OAL ' vip : | iy ran abe Ae ry up t Mi i r i 
aime vi an i yi Anes Hit ite Ta Pee rae iN it ly c er 
+ 4 if) ; M iM chi i 7 i A Oty i | rir 4 hi he i yl if f hau A he a 
- q my rn My ‘ - a ye = Pp in : ke pot | hn A 
Ah ri 't A a iu +, . 7 nt | ay . pM : i 4 i ) : Bi 
: ys e Fj i ) ) ! t oe - j] | ry r ‘i + i, P 
ol ie OC! aed it Fe : PG a hall i . 4) : 
‘| f yv' H , Vi } my | li ; ' = NT 7 ohh 
’ | } i =i an) ' : ‘ ii AU ' 
7 y he - Py) a - 5 fal | t . : : aa) - 
. ay tet M, PM a, 4a | 
. ny ‘ ; ; ai yee, ‘| ne Ane i: i yt | i j ij Ur iy ' i 
1 | 1S ’ | ah) - . An My 1 re 
Fe eee - ‘ a | ; ij : 7 ln rm i < aly Wh cit 
Line 1" pene tye gold aAaNt j.! ; 
it ny ’ yal : if ‘ an hoa ; 
i - i ~ i , _ z ve ix ; spend ‘ 
7 : "hy ry) iw 1} | i, } i y! Hy) 
; 7 iv ti i oi : 1 a ' : 
! oi J 1 
; ; Aly, vit y ri hy i) 
: lk 4 J mel ell Ly 4 < 
1 i j U 16 ; ' it 
; i ; 1 L Lat i ! ' . 
ns) ; ! 1 ! n M 1, i 1A 
1 ; { ap) i i A 
ny on , i ; 4 i ; i {3 Wi ,! 
\ yy! 7 pe ft I. 
v | J _ 7 { lio yl i ‘4 nf 
| ; i P, , 
t | ! i 
ht ip f 
7 f t 7 gt ' : rey 
: \- yy! i 
. . 4 ! uy 4 ee 
oY oy oe it i a 
( - : - i cea ‘ cal a v4 7 
‘ : ; nic . Mi ve ee A 
; . totae) | foe de ee ie ity 
‘ ; i , 
vy) i to ! t!. ne Th, ‘ 
{ : | 
ite . . | a, i " i 
‘ ‘ ° : - A . ™ A Ti A 
i ir ts by " 
1 t 1 7 j - : \ F) Ht \s ms . 
. j i i a a i ~ ' sit 
| ! i i 7 
i ’ ws) - vy 
| | yal a eae 
1 i | ti : f | , 
H Ly J i tet 
| ‘ - i i : i 
! vi i 7 aM 
| { v : ih i : Ro 
; 7 d f ; i : \ 
i | i fi] Fi ! 1 tf / ny 
I q ti 7 ' : 
| | Y ! { ] a a 
: 7 | ' v 
- i ‘ 
; : ; ; vt, | ne 
: Pipe ae) | ' 
, Sse aie eid 
} j i! t op tor i ee 1° | 


THE NAUTILUS 


Volume 122, Number 4 
December 22, 2008 
ISSN 0028-1344 


A quarterly devoted 
to malacology. 


EDITOR-IN-CHIEF 

Dr. José H. Leal 

The Bailey-Matthews Shell Museum 
3075 Sanibel-Captiva Road 
Sanibel, FL 33957 


BUSINESS MANAGER 


Mary Jo Bunnell 

The Bailey-Matthews Shell Museum 
3075 Sanibel-Captiva Road 
Sanibel, FL 33957 


EDITOR EMERITUS 


Dr. M. G. Harasewych 
Department of Invertebrate Zoology 
National Museum of 

Natural History 
Smithsonian Institution 


Washington, DC 20560 


CONSULTING EDITORS 
Dr. Riidiger Bieler 
Department of Invertebrates 
Field Museum of 


Natural History 
Chicago, IL 60605 


Dr. Arthur E. Bogan 

North Carolina State Museum of 
Natural Sciences 

Raleigh, NC 27626 


Dr. Philippe Bouchet 

Laboratoire de Biologie des 
Invertébrés Marins et Malacologie 

Muséum National d'Histoire Naturelle 

55, rue Buffon 

Paris, 75005 France 


Dr. Robert H. Cowie 

Center for Conservation Research 
and Training 

University of Hawaii 

3050 Maile Way, Gilmore 409 

Honolulu, HI 96822 


Dr. Robert T. Dillon, Jr. 
Department of Biology 
College of Charleston 
Charleston, SC 29424 


Dr. Eileen H. Jokinen 
§234 IE. North Shore Road 
Sault Ste. Marie, MI 49783 


Dr. Douglas S. Jones 

Florida Museum of Natural History 
University of Florida 

Gainesville, FL 32611-2035 


Dr. Harry G. Lee 
4132 Ortega Forest Drive 
Jacksonville, FL 32210 


Dr. Charles Lydeard 

Biodiversity and Systematics 
Department of Biological Sciences 
University of Alabama 

Tuscaloosa, AL 35487 


Bruce A. Marshall 
Museum of New Zealand 
Te Papa Tongarewa 
P.O. Box 467 
Wellington, NEW ZEALAND 


Dr. James H. McLean 
Department of Malacology 
Natural History Museum 
of Los Angeles County 
900 Exposition Boulevard 
Los Angeles, CA 90007 


Dr. Paula M. Mikkelsen 

Paleontological Research 
Institution 

1259 Trumansburg Road 

Ithaca, NY 14850 


Dr. Diarmaid O Foighil 

Museum of Zoology and Department 
of Biology 

University of Michigan 

Ann Arbor, MI 48109-1079 


Dr. Gustav Paulay 

Florida Museum of Natural History 
University of Florida 

Gainesville, FL 32611-2035 


Mr. Richard E. Petit 
P.O. Box 30 
North Myrtle Beach, SC 29582 


Dr. Gary Rosenberg 

Department of Mollusks 

The Academy of Natural Sciences 
1900 Benjamin Franklin Parkway 
Philadelphia, PA 19103 


Dr. Angel Valdés 
Department of Malacology 
Natural History Museum 
of Los Angeles County 
900 Exposition Boulevard 
Los Angeles, CA 90007 


Dr. Geerat J. Vermeij 
Department of Geology 
University of California at Davis 
Davis, CA 95616 


Dr. G. Thomas Watters 
Aquatic Ecology Laboratory 
1314 Kinnear Road 
Columbus, OH 43212-1194 


SUBSCRIPTION INFORMATION 


The subscription rate for volume 
123 (2009) is US $50.00 for 
individuals, US $85.00 for 
institutions. Postage outside the 
United States is an additional US 
$10.00 for regular mail and US 
$25.00 for air delivery. All orders 
should be accompanied by payment 
and sent to: THE NAUTILUS, P.O. 
Box 1580, Sanibel, FL 33957, USA, 
(239) 395-2233. 


Change of address: Please inform 
the publisher of your new address at 
least 6 weeks in advance. All 
communications should include both 
old and new addresses (with zip 
codes) and state the effective date. 


THE NAUTILUS (ISSN 0028-1344) 
is published quarterly by The Bailey- 
Matthews Shell Museum, 3075 

Sanibel-Captiva Road, Sanibel, FL 


33975. 


Periodicals postage paid at Sanibel, 
FL, and additional mailing offices. 


POSTMASTER: Send address 
changes to: THE NAUTILUS 
P.O. Box 1580 

Sanibel, FL 33957 


CONTENTS 


NAUTILUS 


Volume 122, Number 4 
December 22, 2008 
ISSN 0028-1544 


Daniel L. Geiger 


Sven N. Nielsen 
Claudio Valdovinos 


Juliana M. Harding 
Stefanie M. Gera 
Roger Mann 


Maria Cristina Dreher Mansur 


Claus Meier-Brook 
Cristian Ituarte 


Maria Eugenia Segade 


Guido Pastorino 


Alexandre Dias Pimenta 
Diogo Ribeiro do Couto 
Paulo Marcio Santos Costa 


Luiz Ricardo L. Simone 


New species of scissurellids from the Austral Islands, French Polynesia, and the 
Indo-Malayan eee (Gastropoda: Vetigastropoda: Scissurellidae, 


Anatomidae, Larocheidae) ..........cccecceecesecesecescescecacecscecscesenerecseensceneeeseensecsseeaee 185 
Early Pleistocene mollusks of the Tabul Formation, South-Central Chile......... 201 


Radula morphology in veined rapa whelks, Rapana venosa (Valenciennes, 1846) 
(Gastropoda: Muricidae) from C eee ake Bay, USA wis. .ictisesecutasesindantatnrnts 217 


A new species of Sphaerium Scopoli, 1777, from southern Brazil (Bivalvia: 


Some remarks on the gross anatomy of Adelomelon ferussacii 
(Donovan, 1824) (Gastropoda: Volutidae) from the coast of Pate iwonia, 


A new species and a new record of Muricidae (Gastropoda) from Brazil: genera 
Pierynotus-and Leptotrophoticicteyicdescsnaistiutivaiioe anitie ctieinecvdedelibedeceeees ss 244 


A new species of Chlamydoconcha Dall, 1884, from southeastern Brazil (Bivalvia: 


Ghilamiydocori chidae).o.<.asasssseves sos csvereion azensstatvsarnessariesenensioayessis apetsesasveedd nutans ots DO 
Research Note 
Jeremy S. Tiemann Sinistral Campeloma decisum (Say, 1817) (Gastropoda: Viviparidae) from the 
Kevin S. Cummings Box iver DM O18is.24 see cesrckesissnusicciedennen Bresbws eieaetaunsaeapcbeuieidebangccsw ake Waser Nakstedsaeasseuste ODO 
UB YoC0) Sed a 11 25 RRC ce OR 261, 263 
TAU 2 aes eek ccs Re ce tsa sat ae ected te ch aateece ea tcc p at voik ou daise touachetadettatasvatairsceichasnnds@aias cuanser beets satis vaadtedetieeiteaysgsatteeseuters 264 
IN UNC Safes -ac lee tgh tn nas ta, Apachettsusetveatics tien toaa baa banteabete sb cste toast tut etch Paes abe vtsiaGe chad bi cnesucoute ttaeies Sedecenates eenense ce tae lapaae eeu 264 
Ath Ore Vr 6 asics seetendev sac heg Shee levdheciead Sea bd canada atone ond eae enc asisilacatosetdeca cguawien Peatengatanietee ceentetee ee eesands dddpueads foe sa tes 265 


MBLWHOI Library 


JAN 9 7 2008 


WOORS HOLE 
Massachusetts 02543 


DOT BR WDE 


ODA 


STATEMENT OF OWNERSHIP, MANAGEMENT, AND CIRCULATION 


Publication Title, THE NAUTILUS. 
Publication No., 0028-1344. 

Filing Date, September 24, 2008. 

Issue Frequency, Quarterly. 

No. of Issues Published Annually, Four. 
Annual Subscription Price, US $80.00. 


Complete Mailing Address of Known Office of Publication, 3075 Sanibel-Captiva Road, Sanibel, FL 33957 USA. 


Complete Mailing Address of Headquarters, same as 7. 
Full Names and Complete Mailing Addresses of Publisher, The Bailey-Matthews Shell Museum, 3075 Sanibel-Captiva Road, 
Sanibel, FL 33957 USA. 

Editor, Dr. José H. Leal, address as above. 

Managing Editor, Mary Jo Bunnell, address as above. 


Owner, Shell Museum and Educational Foundation, Inc., address as above. 


Known Bondholders, Mortgagees, and Other Security Holders Owning or Holding 1 Percent or More of Total Amount of 
Bonds, Mortgages, or Other Securities, None. 
The purpose, function, and nonprofit status of this organization and the tax exempt status for federal income tax purposes has 
not changed during the preceding 12 months. 

Publication Name, THE NAUTILUS. 

Issue Date for Circulation Data Below, September 26, 2008. 


Extent and Nature of Circulation 


Average 
12 months 


A. 
B. 


Total Number of ¢ Jopies 
Paid Circulation 
1. Paid/Requested Outside-County Mail Subscriptions 
2. Paid In-¢ sounty Subscriptions 
3. Sales Through Dealers and Carriers, Street Vendors, 
Counter Sales, and Other Non-USPS Paid Distribution 
4. Other Classes Mailed Through the USPS 
Total Paid and/or Requested Circulation 
Free Distribution by Mail 
1. Outside-County 
2. In-County 
3. Other Classes Mailed Through the USPS 
Free Distribution Outside the Mail 
Total Free Distribution 
Total Distribution 
Copies not Distributed 
Total 
Percent Paid and/or Requested Circulation 


420 


309 
0 


0) 
16 
325 


12 

0 

0) 

0) 

12 
337 
$3 
420 
97% 


Single 
Issue 


420 


319 
0 


0 
15 
334 


12 

0) 

0 

0 

12 
346 
74 
420 
97% 


THE NAUTILUS 122(4):185-200, 2008 


Page 185 


New species of scissurellids from the Austral Islands, French 
Polynesia, and the Indo-Malayan Archipelago (Gastropoda: 
\ ‘etigastropoda: Scissurellidae, Anatomidae, Larocheidae) 


Daniel L. Geiger 

Santa Barbara Museum of Natural History 
Invertebrate Zoology 

2559 Puesta del Sol Road 

Santa Barbara, CA 93105 USA 
geiger@vetigastropoda.com 


ABSTRACT 


Four new species of Scissurellidae, Anatomidae, and Larochei- 
dae are described from the Austral Archipelago, French 
Polynesia and the Indo-Malayan Archipelago: Sinezona 
danieldreieri new species, S. wiley new species (both Scissur- 
ellidae), Anatoma rapaensis new species (Anatomidae) and 


Trogloconcha lozoueti new species (Larocheidae). One of 


the species is currently only known from the Austral Islands 
T. lozoueti), while the others seem to show a_ broad 
Indo-Malayan Archipelago to western Pacific distribution. 
Shells of all species and the radulae of S. danieldreieri, 
T. lozoueti, and A.rapaensis are illustrated with scanning elec- 
tron micrographs. 


Additional Keywords: Mollusca, Rapa Island, radula 


INTRODUCTION 


The present contribution describes some microgas- 
tropods in the families Scissurellidae, Anatomidae, and 
Larocheidae known from the Austral Islands and be- 
yond. The impetus stems from a French expe dition to 
the southeastern-most Austral island, Rapa (see Lozouet 
et al.. 2004; 2005 for details); for the species found 
at Rapa, additional material from other localities located 
in various institutional collections has also been inclu- 
ded. This article is part of a series describing the bio- 
diversity of this remote archipelago (e.¢., Schwabe and 
Lozouet, 2006). 

Scissurellidae sensu lato comprises ean of small 
basal marine snails in Vetigastropoda (Geiger et al., 
2008). The phylogenetic position a peer is cur- 
rently unsettled. Whereas Geiger and Thacker (2005; 
unpublishe -d data) resolved a Lepetodrilus (Lepetodrili- 
dae) + Scissurella, Sinezona, Sukashitrochus (Scissurel- 
lidae sensu stricto) clade with more distantly related 


Anatoma (Anatomidae), Kano (2008) using the same 


three genes (Histone 3, COI, 18S) recovered in some of 
his analyses a clade uniting all three of these lineages 
(Lepe todrilus, Sinezona, Anatoma). Scissurellidae and 
Anatomidae are characterized by a slit or foramen. in 
the shell (lacking in Larocheidae), a rhipidoglossate rad- 
ula, and a lack of nacre. Approxim itely 140 species are 
currently described with an additional 90 remaining to 
be formally recognized (Geiger, 2003; 2008). To date, no 
species in ‘these families ie ever been recorded from 
the Austral Islands. The species described herein are all 
new to science, one with its range restricted to the 
Austral Islands, and three others known from several 
localities in the broad Indo-Pacific. 


MATERIALS AND METHODS 


Standard methods for scanning electron microscopy 
(SEM) were employed as de taile d in Geiger (2006a, 
b, c) and Geiger et al. (2007). Terminology for shell 
tae and details on method of whorl count have 
been given elsewhere (Geiger, 2003; Geiger and Si isaki, 
2008). Specimens cited are dry lots; *: comple te” indi- 
cates wet-preserved lots with animals. All de »pth indica- 
tions refer to bottom depth. 

Institutional abbreviations used in the text are: BRC: 
Bret Raines Collection, Kansas City, USA; DLG: Daniel 
L. Geiger Collection, Los Angeles, USA; JTC: Jean 
Trondlé Collection, Paris, France; LACM: Los Angeles 
County Museum of Natural History, Los Angeles, oe 
MNHN: Muséum national d’Histoire naturelle, Paris, 
France; SBMNH: Santa Barbara Museum of Natural 
History, Santa Coabare. USA: USNM: United States Na- 
tional Museum of Natural History, Washington (DC) 
USA; ZMA: Zoological Museum, Amsterdam, The Neth- 
erlands. 

Other abbreviations used in text are: M: Monotypy 
OD: Original designation; SEM: Scanning electron 
microscope/microscopy/micrograph 


Page 186 


THE NAUTILUS, Vol. 122, No. 4 


SYSTEMATICS 


Scissurellidae Gray, 1547 
Sinezona Finlay, 1926 


Type species: Schismope brevis Hedley, 1904 (OD). 


Remarks: The genus has recently been treated by 
Marshall (2002) and Geiger (2003). 


Sinezona danieldreieri new species 
(Figures 1-9) 


Description: Shell small (to 0.77 mm), trochiform. Pro- 
toconch of | whorl, with strong axial sculpture not 
reaching apical suture, apertural varix not connected to 
smooth embryonic cap, apertural margin straight. Tele- 
oconch I of 0.875 whorls, suture at ponnlien cee 
mately 9-15 strong, raised axial cords; first spirals after 
0.5 whorls. Tleoconch 1 Il of 0.6 whorls, approximately 
four fine spiral cords at onset of selenizone between 
selenizone and deep suture below periphery, descending 
noticeably on last 0.25 whorls; shoulder slightly concave, 
approximately 13-20 axial cords, raised near suture, as 


high as wide towards selenizone; approximately 10 spir- 
als, distinct cords near suture, diminishing to very fine 
cords near selenizone; interstices with fine irregular 
growth lamellae. Base with constriction below ete ni- 
zone, approximately 15 elevated spiral cords, crossed 
and run over by approximately 16 fine spiral lines. Um- 
bilicus cone- shape sd, moderately wide, wall smooth, at 
distinct angle to base. Aperture De: shaped, roof overhan- 
ging. Selenizene above periphery, keels quite strong, 
moderately elevated, distinct linules, elongated feranten 
closed anterior ly. 


OpercuLuM: Thin, flexible, multispiral, with central 
nucleus. 


Raputa (Ficures 8-9): Rachidian tooth triangular cusp 
with approximately five equal sized deatclee Lateral 
teeth 1-3 similar; outer edge of cusp with 4-5 denticles. 
Lateral tooth 4 reduced in size, hook-shaped. Lateral 
tooth 5 enlarged by broadening, approximately six, large 
denticles on inner edge, 1-2 small ones on outer edge. 
Central denticle of inner marginal teeth large, approxi- 
mately three denticles on inner edge, four on outer 


Figure I. 


Holotype of Sinezona danicldreicri new species. 
Guinea, 5.283° S, 150.131° E, 0-5 m. Seale bar, shell 


(SBMNET $3540). Lumu Reef, Kimbe Bay, New Britain, Papua New 
500 jim. Seale bar, protoconch = 100 jum, 


D. L. Geiger, 2008 


Page 187 


edge; outer marginal teeth spoon-shaped, bilaterally 
symmetrical with approximately a dozen fine denticles 
on each side. Radular interlock moderate. 


Type Material: Holotype: SBMNH $3540: Figure 1. 24 
Paratvpes: 3-4 m, Little Santa Cruz Island, Mindanao, 
6.854° N, 122.04° E (USNM 812453, 1: Figure 2). 
1.5 m, Okinawa, Oku, Japan, 26.847° N, 128.287 
E (LACM 77-61, 3: one illustrated Figure 3). 20 m, Off 


Figures 2-4. Paratypes of Sinezona danieldreicri new species 
122.04° E, 3-4 m (USNM 812453, ] 
Island. Pamilacan Island. Philippines, 9.5° N, 123.917° E, 20 m (AMS 406342, 1). Scale bars, shell = 500 jum. Scale bars, protoconch 
= 100 um 


3. Okinawa, Oku Japan, 26.847° N, 128.287° E, 


Bohol Is, Pamilacan Island, 9.5° N, 123.917° E (AMS 
106342, 1: Figure 4). 


Type Locality: 0-5 m, Lumu Reef, Kimbe Bay, New 
Britain, Papua New Guinea, 5.283° S, 150.151° E. Col. 
D. L. Geiger Oct. 2005. 


Etymology: The name honors Daniel Dreier, son of 
Douglas and Hanna Dreier of Santa Barbara, who have 


Little Santa Cruz Island, Mindanao, Philippines, 6.584° N 
1.5 m (LACM 77-61, 3). 4. Off Bohol 


Page 188 


THE NAUTILUS, Vol. 122, No. 4 


made generous contributions to the Santa Barbara 
Museum of Natural History. 


Distribution: Indo-Malayan Archipelago to southern 
Polynesia; shallow reef environments. 


Other Material Examined: FRENCH POLYNESIA: 0 m, 
Rapa Iti Island, Rapa, Austral Islands, 27.622°  S, 
144.302° W (MNHN, 1). 0 m, Kotuaie Point, Tupuaki 
Bay, Rapa, Austral Islands, 27.577° S, 144.343° W 
(MNHN, 1). 2 m, North of Pukitarava, Rapa, Austral 


Islands, 27.597° S, 144.227° W (MNHN, 4: two illu- 
strated Figures 5-6). 2-4 m, N of Aturapa Island, Rapa, 
Austral Islands, 27.572° S, 144.350° W (MNHN, 4). 2m, 
N of Rapa Iti Island, Rapa, Austral Islands, 27.620° S, 
144.303° W (MNHN, 1). 2 m, N of Anatakuri Bay, Rapa, 
Austral Islands, 27.623° S, 144.307° W (MNHN, 15). 
3-24 m, Hiri Bay, Rapa, Austral Islands, 27.622° S, 
144.370° W (MNHIN, 4). 5 m, Anarua Bay, Rapa, Austral 
Islands, 27.605° S, 144.378° W (MNHN, 7: one 
illustrated Figures 7, 8-9 [radula]). 6 m, Mei Point, 


Figures 5-7. Sinezona danieldreicri new species. 5-6. North of Pukitarava, Rapa, Austral Islands, French Polynesia, 27.597° S, 
144.227° W. 2m (MNIIN). 7. Anarua Bay, Rapa, Austral Islands, French Polynesia, 27.605° S, 144.378° W,5 m (MNHN). Radula of 


pecimen is shown in Figures S8—9. Seale bar. shell = 500 tum Seale bar, protoconch = 100 ttm. 


D. L. Geiger, 2008 


Page 189 


Figures 8-9. 


Radula of Sinezona danieldreieri new species, from specimen shown in Figure 7. Anarua Bay, Rapa, Austral Islands, 


27.605° S, 144.378° W, 5 m. 8. Lateral teeth 4 (L4) and 5 (L5) plus marginal teeth. 9. Half width of radula with central field. Scale 


bars = 10 um. 


Rapa, Austral Islands, 27.637° S$, 144.303° W (MNHN, 1; 
MNHN, 1). 6 m, SW of Gotenaonao Point, Rapa, 
Austral Islands, 27.645° S, 144.320° W (MNHN, 3). 
15-20 m, Vavai, Rapa, Austral Islands, 27.590° S, 
144.381° W (MNHN, 1; MNHN, 1). 16-20 m, Mei 
Point, Rapa, Austral Islands, 27.637° S, 144.303° W 
(MNHN, 2). 1S m, Rarapai Island, Rapa, Austral 
Islands, 27.572° S, 144.368° W (MNHN, 1). 27 m, 
Kauira Point, Rapa, Austral Islands, 27.592° S$, 144.347° 

(MNHN, 1). 30 m, NW of Tauna Island, Rapa, 
Austral Islands, 27.605° S, 144.303° W (MNHN, 17). 
36 m, around Rukuaga Point, Rapa, Austral Islands, 
27.568° S. 144.368° Ww (MNHN, 8). 

PuiLipPINEs: Canipo Island, Cuyo Islands, Palawan 
Province, 11° N, 120.948° E (USNM 808195, 1). 

Guam: 0.7 m, Ajayan Bay, S. Guam, 13.245° N, 
144.717° E (LACM 77-19, 2). 


Remarks: The protoconch sculpture shows some in- 
traspecific variability in the material examined. More 
specifically, the strong axial cords are sometimes discon- 
tinuous. This character is not restricted to the present 
species, hence, it does not seem to be of any systematic 
value. Those axial cords are also somewhat variable in 
width, the thickenings occasionally seeming to form 
some faint spiral slemente. Given the intra- and inter- 
specific variability of protoconch sculpture at this level 
of detail, the presence/absence of faint spiral elements is 
considered irrelevant for systematic purposes. Discrete 
differences can be observed between strong and weak 
axial cords as utilized previously in a systematic context 
(Geiger, 2003). 

Sinezona plicata (Hedley, 1899) from the Indo-Pacific 
has an overall wider and larger shell (2.3 mm; Geiger 
and Jansen, 2004b), more pronounced but fewer raised 
axial cords, and a eg aga with spiral lines. Sinezona 
ferriezi (Crosse, 1867) (? = S. hoernesi Semper, 1865: 
whereabouts of ee types unknown, original illu- 
stration lacking detail) with broad Indo- Malayan 


Archipelago distribution, is about twice as large, has a 
smooth protoconch with a distinct apertural varix, and 
>1 teleoconch I whorls. Sinezona globosa Geiger, 2006, 
from New Caledonia and Wallis Island is more rounded 
in overall appearance, lacks the elevated spiral cords, is 
larger (3.1 mm; Geiger, 2006a), and has spiral sculpture 
on the protoconch. Sinezona_ macleani Geiger, 2006, 
with broad Indo-Malayan distribution is more globular 
in overall outline, larger (2.3 mm; Geiger, 2006a), has a 
protoconch with flocculent sculpture, lacks the raised 
axial lines and has a wider umbilicus. All the above 
species grow to much larger size and would be imma- 
ture with open slit at the maximum size of S. daniel- 
dreieri. 


Sinezona wileyi new species 
(Figures 10-16) 


Description: Shell medium size (up to 1.08 mm) tro- 
chiform turreted. Protoconch slightly sunken in, with 1 

whorl, fine irregular axials, no apertural varix, apertural 
margin straight. Teleoconch I with 0.6-0.7 whorls, su- 
ture at periphery, usually with approximately 14-22 faint 
to distinct fine axial cords, usually without spiral lines 
(occasionally faintest spiral threads only visible by SEM 
[Figure 11]), interstices with irregular thickenings. Tele- 
oconch IT with up to 1.25 wots, suture below periph- 
ery. Shoulder with approximately 32-44 _ barely 

perceptible to distinct fine commarginal axial cords, 
stronger near suture becoming less distinct towards sele- 
nizone; approximately eight baiely perceptible to very 
indistinct spiral cords, concentrated on middle of shou 
der; occasional specimens with predominant spiral 
sculpture (Figure 11). Base distinctly constricted below 
selenizone, axial cords of same strength and density as on 
shoulder, starting in lower portion of constriction be low 
selenizone; approximately 17 fine, irregularly spaced spi 
ral cords, starting in lower portion of constriction with 
onset of axial lines. Umbilicus wide, walls straigh 


1 


Page 190 


THE NAUTILUS, Vol. 122, No. 4 


Figure 10. Holotype of Sinezona wileyi new species. Lumu Reef, Kimbe Bay, New Britain, Papua New Guinea, 5.283° S, 
150.131° BE, 0-5 m, (SBMNH 83541). Scale bar, shell = 500 jim. Scale bar, protoconch = 100 jun, 


smooth, periphery at distinct angle to base, bordered by 
carina. Aperture subquadratic, D-shaped, roof overhan- 
ging. Selenizone above periphery, keels moderately 
strong, quite elevated, lunules clistinct at recular interval: 
foramen narrow, elongated, margins converging and 
touching, without raphe. Animal unknown. 


Type Material: Holotype: SBMNH 83541 (Figure 10). 
35 Paratypes: 0-5 m, Lumu Reef, Kimbe Bay, New Brit- 
ain, Papua New Guinea, 5.283° S$, 150.131° E (DLG 
639, 19: one illustrated in Figure 15). Laisse de plage 
extérieure, Mururoa, Tuamotu, French Polynesia, 
21.842° S$, 138.895° W (JTC, 2: one illustrated in Figure 
12). 20 m, Herald Pass, W of Ndravuni Island, NW 
end of Great Astrolabe Reef, Fiji, 18.767° S, 178.467° E 
(LACM §5-135, 2: one illustrated in Figure 14). 20 m, 
Herald Pass, W of Ndravuni Island, NW end of Great 
Astrolabe Reef Fiji, 18.767° S, 178.467° E (LACM 85- 


Or 


35, 2). 10-20 m, Off Francis Island in Beqa Lagoon, 


just off of Beqa Island (south of Viti Levu), 18.300° S, 
178.067° E (DLG 279, 10: one illustrated in Figure 16). 


Type Locality: 0-5 m, Lumu Reef, Kimbe Bay, New 
Britain, Papua New Guinea, 5.283° $, 150.131° E. 


Etymology: The name honors the collector of the first 
known specimens of the species Tony Wiley of River- 
side, California, USA. 

Distribution: — Tropical Indo-Malayan Archipelago; 
shallow shelf, 


Other Material Examined: =F Rencu PoLyNesia: 36 m, 
Around Rukuaga Bay, Rapa, Austral Islands, 27.568" S, 
144.568° W (MNHN, 20: Figures 11, 13). 2-4 m, N of 
Aturapa Island, Rapa, Austral Islands, 27.572° S, 144.350° 
W (MNHIN, 1). 5 m, Anarua Bay, Rapa, Austral Islands, 
27.605° S, 144.375° W (MNEIIN, 4). 45 m, Haurei Bay, 
Rapa, Austral Islands, 27.613° S, 144.305" W (MNITIN, 1). 


D. L. Geiger, 2008 


Page 19] 


Figures 11-13. Sinezona wileyi new species. 11-12. Around Rukuaga Bay, Rapa, Austral Islands, French Polynesia, 27.568° S, 
144.368° W, 36 m (MNHN, 20). 13. Laisse de plage extérieure, Mururoa, Tuamotu, French Polynesia, 21.842" S$, 138.895° W 
Paratype: JTC. 2). Scale bars. shell = 500 pm. Scale bars, protoconch = 100 fm 


NeW CaLeponia: 25-30 m, Santal Bay, Lifou, 20.822 
S. 167.173° E (MNHN, 1). 55-57 m, Koumac Sector, 
20.672° S, 164.195° E (MNHN, 1). 5—25 m, Touho Sec- 
tor. 20.878° S, 165.325° E (MNHN, 1). 8 m, Touho 
Sector, 20.742° S, 165.265° E (MNHN, 1 

Fiji: 10-20 m, Off Francis Island in Bega Lagoon, just 
ott of Bega Island (south of Viti Levu), 18.300° S$ 
178.067° E (DLG 279. 10 


Papua New Guinea: 70 m, off South Ema Reef, Kimbe 
Bay, New Britain, 5.202° S, 150.152° E (DLG 768, 3 


Remarks: The organization of the foramen in this 
species seems to be intermediary between the typical 
open slit in Scissurella and the closed foramen in Sine- 
zona Although the margins of the slit are converging 
and fused anteriorly, a raphe is not evident The facts 


Page 192 


THE NAUTILUS, Vol. 122, No. 4 


that there is a marked downward deflection of the last 
quarter whorl and that the roof of the peristome 
attaches below the periphery of the previous whorl dem- 
onstrate that those specimens are fully grown and share 
more similarities with Sinezona than ‘with Scissurella 
with an open slit. Accordingly, S$. wileyi is placed in 
Sinezona. 


The species seems to have variable strengths of sculp- 
ture. Most oo have predominant axial sculpture 
(Figures 10, 14-15), while in occasional specimens 
(Figure 11) rea ‘asia sculpture is barely perceptible and 
the “spiral structures, particularly on the adumbilical part 
of base, become most prominent. As there are interme- 
diate specimens (Figure 13) with distinct adumbilical 


Figures 14-16. 


Sinezona wileyi new species, paratypes. 14. Herald Pass, W of Ndravuni Island, NW end of Great Astrolabe Reef, 


Fiji, 18.767° S, 178.467 a 20 m (LACM $5-135, 2). 15. Lumu Reef, Kimbe Bay, New Britain Papua New Guinea, 5.253° S, 
150.131° E, 0-5 m (DLG 639, 19). 16. Off Franc is Island in Beqa Lagoon just olf of Beqa Island ( south of Viti Levu), Fiji, 18.3° S, 


178.067° E, 10-20 m agen 10). Seale bars, shell 


500 pum. Scale bars, protoconch 


LOO tun 


D. L. Geiger, 2008 


Page 193 


spiral lines and relatively weak axial lines, combined 
with identical condition of the protoconch and very 
small size, these sculptural differences are conside red 
to constitute intraspe cific variability. 

There are no similar ae s in the Indo-Pacific. Sine- 
sona ferriesi (Crosse, 1867) with broad Indo-Malayan 
Archipelago distribution is most similar, but is about 
twice the size of S. wileyi, has a smooth protoconch with 
distinct apertural varix, a teleoconch I with more than 
whorl, is more rounded overall, and a distinct raphe 
anterior to the closed foramen. Sinezona garciai Geiger, 
2006, from the Caribbean, shares the protoconch sculp- 
ture, the absence of an apertural varix on the proto- 
conch, the relatively short teleoconch I with 0.6 whorls, 
and the narrow umbilicus. However, in addition to its 
occurrence in a separate ocean, S. garciai has a more 
inflated shell with an oval aperture, which connects to 
the previous whorl barely below the periphery, has only 
about 0.6 teleoconch TH whorls, and bears a distinct 
raphe anterior to the foramen. 


Anatomidae McLean, 1989 
Anatoma Woodward, 1859 


Type Species: Scissurella crispata Fleming, 1828 (M) 


Remarks: The genus has recently been treated by 
Marshall (2002), Gei iger (2003), and Geiger and Jansen 
(20042). 


Anatoma rapaensis new species 
(Figures 17-23) 


Description: Shell of medium size (to 1.82 mm), tro- 
chiform biconical. Protoconch with 0.75 whorls, floceulent 
sculpture, no apertural varix, apertural margin slightly si- 
nusoid. Teleoconch I with 0.3-0.4 whorls, approximate- 
ly 9-12 axial cords, interstices with fine flocculent 
sculpture, occasionally (holotype) flocculent sculpture 
somewhat concentrated in position of selenizone. Teleo- 
conch IE with up to 2.25 whorls, suture immediately below 
selenizone in early growth, offset by width of selenizone 


Figure 17. 


Holotype of Anatoma rapaensis new species. St. 43, Haurei Bay, Rapa, Austral Islands, French Polynesia, 27.613° S 
144.305° W, 45 m. (MNHN 20823). Scale bar shell = 1 mm. Scale bar protoconch = 100 tm. 


Page 194 


THE NAUTILUS, Vol. 122, No. 4 


Figures 18-20. Anatoma rapaensis new species, paratypes (MNHN). 18-19. St. 43, Haurei Bay, Rapa, Austral Islands, French 
Polynesia, 27.613° S, 144.305° W, 45 m (MNHN 20824). 20. St. 48, around Rukuaga Point, Rapa, Austral Islands, French Polynesia, 


in fully grown specimens. Shoulder convex, approximately 
53-66 (n = 4) axial cords on first teleoconch H whorl, same 
density on remaining whorls; first fine spiral cords after 
0.125 whorls, approximately 7-10 after one teleoconch I 
whorl, approximately 15-22 at apertural margin of fully 
grown specimen, becoming less distinct and more un- 


evenly spaced towards apertural margin: intersections of 


27.568° S, 144.368° W, 36 m (MNHN, 30). Scale bars, shell = 1 mm. Scale bars, protoconch = LOO jum 


axial and spiral cords forming minute points. Base contin- 
uously sloping with narrow umbilicus; axial cords of same 
density and strength as on shoulder, approximately 19 
spiral cords, fine spiral cords below selenizone turning 
into low steps from mid-base onwards; intersection of 
spiral and axial cords with fine points. Aperture ovoid 
D-shaped, roof overhanging, basal adumbilical portion 


D. L. Geiger, 2008 


Page 195 


flared. Selenizone at periphery, keel moderately strong, 
moderately elevated, slit open, margins converging 
towards apertural margin. 


OpercuLum (Ficure 23): As large as aperture, thin, cor- 
neous, nucleus central, multispiral. 


Raputa (Ficures 21-22): Rachidian tooth trapezoid, 
central denticle with 3—4 denticles on each side. Lateral 
teeth 1-3 similar, development of cusp reduced periph- 
erally, terminal denticle largest, 4-2 denticles on outer 
edge. Lateral tooth 4 reduced in size, hook-shaped, with 
one minute point on each side. Lateral tooth 5 enlarged 
with four strong denticles on inner margin. Inner mar- 
ginal teeth elongated, terminal denticle largest, 3-4 
smaller denticles on inner margin, three larger denticles 
on outer margin. Outer marginal teeth with cup-shaped 
cusp with many small denticles. 


Type Material: Holotype: MNHN 20823 (Figure 17). 
60 Paratypes: 45 m, RAPA St. 43, Haurei Bay, Rapa, 
Austral Islands, French Polynesia, 27.613° S, 144.305° 
W (MNHN 20824, 29; two illustrated in Figures 18-19). 
36 m, RAPA St4S, Around Rukuaga Point, Rapa, 
Austral Islands, 27.568° S, 144.368° W (MNHN, 30; 
one illustrated in Figure 20). 145 m, MUSORSTOM 9 
St. CP 1159, Eiao Island, 7.972° S, 140.728° E (MNHN, 
1: Figures 21-23 [radula, operculum)). 


Type Locality: 45 m, RAPA St. 48, Haurei Bay, Rapa, 
Austral Islands, French Polynesia, 27.613° S, 144.305° W. 


Etymology: The species is named alter its type locali- 
ty, Rapa Island, French Polynesia. 


Distribution: Tropical Indo-Malayan Archipelago and 
Western Pacific; shallow shelf to upper slope. 


Other Material Examined: FRencu Potynesia: LO-15 m, 
Pake Bay, Rapa, Austral Islands, 27.617° S, 144.310° 
W (MNHIN, 2). 30 m, NW of Tauna Island, Rapa, Austral 
Islands, 27.605° S, 144.303° W (MNHN, 50). 33 m, 
Haurei Bay, Rapa, Austral Islands, 27.612° S, 144.318° W 
(MNHN, 10). 52-57 m, SE of Tauna Island, Rapa, 
Austral Islands, 27.608° S, 144.295° W (MNHN, 50: 
MNHN, 20). 100 m, North of Raivavae, Austral Islands, 
23.828° S, 147.693° W (MNHN, 2). 

PHILIPPINES: 92-97 m, 12.517° N 
(MNHN, 8). 

New Careponia: 105-110 m, Poindimie Sector, 
20.817° S, 165.317° E (MNHN, 9). 250-350 m, south- 
ern New Caledonia, 22.500° S, 166.400° E (MNHN, 1). 
250-350 m, southern New Caledonia, 22.500°  S, 
166.400° E (MNHN, 1). 495 m, southern New Caledo- 
nia, 22.367° S, 166.233° E (MNHN, 3). 600-616 m, 
northern New Caledonia, 18.817° S, 163.250° E 
(MNHN, 3). 


. 120.650° E 


Figures 21-23. 


Aadula and operculum of Anatoma rapaensis new species. Eiao Island, Marquesas Islands, 7.972° S, 140.725° E, 


145 m (MNHN, 1). 21-22. Radula. 23. operculum. Scale bar, radula Figure 21 = 50 wm. Scale bar, radula Figure 22 = 20 mm. Scale 


bar, opeculum = 500 um. 


Page 196 


THE NAUTILUS, Vol. 122, No. 4 


Marouesas Isutanps: 145 m, Eiao Island, 7.972° S, 
140.728° E (MNHN, I: paratype), 200-220 m, Ua Huka 
Island, 8.900° S, 139.633° W (MNHN, 6). 352-358 m, 
Hiva Oa Island, 9.850° S$, 139.150° W (MNHN, 1). 

Fit: 149-168 m, S of Viti Levu, 18.207° S, 178.5 
(MNHN, 1). 260-305 m, S of Viti Levu, 18.308° S, 
178.097° E (MNHN, 6). 275-430 m, S. of Viti Levu, 
18.297° S, 177.907° E (MNHN, 1). 441-443 m, S. of Viti 
Levu, 18.320° S, 177.862° E (MNHN, 1). 


Remarks. Anatoma rapaensis is characterized by the 
rather flattened overall shape, the short teleoconch I 
with less than 0.5 whorls, and the fine reticular sculpture 
that is axial-dominated on the first half teleoconch I 
whorl, subsequently becoming spiral-dominated. The 
most similar species is Anatoma (sensu lato) exquisita 
Schepman, 1908, from the Indo-Malayan Archipelago; 
the comparison is based on SEM imaging of the holo- 
type (ZMA 3.08.101; Figure 24). The shell is slightly 
taller, the spiral and axial lines are denser, forming more 
prominent points at their intersection, and the axial lines 
are slightly stronger than the spiral lines even in larger 


specimens (specimens 1.S—3.5 mm examined): protoconch 
and teleoconch I are eroded beyond recognition. Anatoma 
paucispiralia Bandel, 1998, from Satonda, Indonesia, has a 
smooth protoconch, a shorter teleoconch I (0.125 vs. 0.3— 
0.4 whorls), and a slightly undulating shoulder profile. All 
other Indo-Pacific species are either more turreted or have 
noticeably different sculpture on shoulder and base. The 
only other documented case of a species that shows a 
change of sculpture on the teleoconch I is Anatoma jane- 
tae Geiger, 2006, known from 2,500 m off the west coast of 
North America (Geiger, 2006c). 


Larocheidae Fleming, 1927 
Trogloconcha Kase and Kano, 2002 


Type Species: Trogloconcha ohashii Kase and Kano, 
2002 (OD). 


Remarks: The genus was recently treated by Geiger 
(2003). 


Trogloconcha lozoueti new species 
(Figures 25-30) 


Figure 24. Holotype of Anatoma exquisita Schepman, 1908 (ZMA 3.08.101). Siboga Station 95, 5°43.5' N, 119°40' E, 522 m 
between Sabah, Malaysia, and southeastern Tawitawi Group, Philippines]. Seale bar, shell = 1 mm. Scale bar, protoconch = 100 jum 


D. L. Geiger, 2008 Page 197 


27. Trogloconcha lozoueti new species. 25. Holotype Vavai, Rapa Austral Islands, French Polynesia, 27.590° S 
144.381° W, 15-20 m (MNHN 20825). 26-27. Radula from specimen shown in Figure 30. Hiri Bay, Rapa, Austral Islands, French 
525 


7 > 95_< 
Figures 25-2 


Polynesia, 27.622° S_ 144.37° W, 3-24 m (MNHN). Scale bar, shell = 1 mm. Scale bar, protoconch = 100 tum. Scale bar, radulae = 10 pum 


Page 195 THE NAUTILUS, Vol. 122, No. 4 


fy 


Figures 28-30. Trogloconcha lozoucti new species. Paratypes. 28-29. Vavai, Rapa, Austral Islands, French Polynesia, 27. 
144.351° W, 15-20 m (MNHN 20826). 30. Hiri Bay, Rapa, Austral Islands, French Polynesia, 27.622° S, 144.37° W, 3-24 m 


MNHN, 20). Scale bar, shell 28 = 1 mm. Seale bars, shell 29-30 = 200 tum. Seale bars, protoconch = 100 jun 

Description: Shell medium size (to 1.17 mm), trochi- 31 spiral cords, from suture to mid-base as fine spiral 
form globular, with rapidly increasing whorls. Proto- cords, transitioning to low spiral steps in 5-6 adumbilical 
conch with 0.75 whorls, flocculent sculpture somewhat spirals: approximately 70 fine axial cords on last whorl, 
spirally arranged, weak apertural varix apertural margin forming distinct points at intersections; overall appear- 
convex. Teleoconch with up to two whorls, approximate- ance of shell fine reticulate and spiky. Suture bordered 
ly 15 axial cords on first 0.3-0.5 whorls with no spiral by strong irregularly lamellate thickening on shoulder. 
sculpture, interstices with fine flocculent sculpture; on- Base anomphalus with weak callus in umbilical region. 
set of spiral cords alter 0.5 whorls; approximately \perture oval, roof overhanging. 


D. L. Geiger, 2008 


Page 199 


OpercuLuM: Corneous, round, multispiral, with central 
nucleus, covering only approximately 1/3 of aperture. 


Raputa (Ficures 26-27): Rachidian tooth triangular, 
cusp with central denticle largest, three denticles on 
each side, arranged in convex curve. Lateral teeth 1-4 
similar, L-shaped, cusp with apical denticle largest, 3-4 
denticles on outer margin, 1-3 denticles on inner mar- 
gin; lateral tooth 5 enlarged by half, apical denticle larg- 
est, 3-4 denticles on each side. Immer marginal teeth 
with elongated shaft central denticle largest, 3-4 denti- 
cles on inner margin, 5-6 denticles on outer margin; 
outer marginal teeth with cup shape cusp, with many 
fine denticles on each side. Radular interlock of central 
field strong. 


Type Material: Holotype: MNHN 20825. 26 Para- 
types: 15-20 m, RAPA St. 32, Vavai, Rapa, Austral 
Islands, 27.590° S, 144.381° W (MNHWN 20826, 6: two 
illustrated in Figure 28-29). 3-24 m, RAPA St. 9, Hiri 
Bay, Rapa, Austral Islands, 27.622° S, 144.370° W 
(MNHN, 20: one illustrated in Figure 30). 


Type Locality: 15-20 m, RAPA St. 32, Vavai, Rapa, 
Austral Islands, French Polynesia, 27.590° S, 144.381° W. 


ee Named in honor of Pierre Lozouet 
(MNHN) for his accomplishments particularly in the field 
of fossil mollusks including Scissurellidae sensu lato, and 
his work in connection with the Rapa expedition. 


Other Material Examined: FRENCH PoLyNesIA: 36 m, 
Around Rukuaga Point, Rapa, Austral Islands, 27.568° S, 
144.368° W (MNHN, 100). 33 m, Haurei Bay, Rapa, 
Austral Islands, 27.612° S, 144.318° W (MNHN, 6). 52- 
57 m, SE of Tauna Island, Rapa, Austral Islands, 27.608° 
S, 144.295° W (MNHN, 20). 30 m, NW of Tauna Island, 
Rapa, Austral Islands, 27.605° S, 144.303° W (MNHN, 
3). 8 m, S of Tarakoi Island, Rapa, Austral Islands, 
27.093° S, 144.308° W (MNHN, 1). 52-57 
Tauna Island, Rapa, Austral Islands, 27.608° S, 144.295° 
W (MNHN, 9) 


Remarks: = Trogloconcha ohashii from southem Japan 
has fewer spiral and axial elements that also form elevated 
points and a wide umbilicus. Trogloconcha tesselata Kase 
and Kano, 2002, from the Indo-Malayan Archipelago has 
a smooth protoconch and lacks the fine points at the 
intersection of axials and spirals. Trogloconcha christinae 
Geiger, 2003, from Western Australia lacks spiral sculp- 
ture and has flattened-flocculent protoconch sculpture. 
The new species is currently only known from the 
Austral Islands. The radula represents the general veti- 
gastropod pattern (rhipidoglossate) and : shores the typi- 
cal larocheid arrangement of similar lateral teeth 1-4, 
without reduced, hook-shaped lateral tooth 4; the radula 
confirms the placement of the species in Larochaeidae. 
The lack brood pouch seen in Larochaea and 
Larocheopsis places the species in Trogloconcha. 
Juveniles have a proportionally wider appearance 
with the suture approximately at the periphery of the 


SE of 


previous whorls. Fully grown specimens have a more 
elevated appearance with the suture connecting well 
below the periphery of the previous whorl. 

penpaen I 


DISCUSSION 


The new species all belong to recognized genera. They 
are diagnosed by particular character combinations not 
known from any described species. The two Sinezona 
species are among the smaller species in the genus, 
while the Anatoma and Trogloconcha species are within 
the usual size range for their respective genera. All spe- 
cies show characters of shell and radular morphology 
that are known from other species; no new character 
states were found. The combination of particular fea- 
tures and their particular strength of development, how- 
ever, are unique for each of the new species. 

The distribution of three species reaches beyond ve 
Austral Islands towards the equator (Sinezona wile yi, S 
danieldreieri, and Anatoma rapaensis), while one spe- 
cies (Trogloconcha lozoueti) is currently only known 
from Rapa Island. This pattern suggests a closer faunal 
affinity of the Austral archipelago pene the broad Indo- 
Malayan Archipelago, rather ae with the temperate 
Southern Ocean. The two more widely distributed spe- 
cies S. danieldrieri and S. wileyi are also those that have 
been found in deeper water, confirming the idea that 
deep-water species in general show a wider distribution 
than those restricted to shallow water. 


ACKNOWLEDGMENTS 


I thank on behalf of P. Lozouet all colleagues who 
participated in “RAPA 2002” and especially the mollusk 
group: R. von Cosel, V. Héros, A. Le Goff , P. Maestrati, 
J.-L., Menou, S. Schiaparelli and _ J. Trondlé. Many 
thanks also to Claude Payri (University of Papeete, 
Tahiti), who organized the logistics with the French 
Navy and the mayor and local "cone of Rapa. Pierre 
Lozouet prov ided helpful criticism on a draft version of 
the present contribution. Diego Zelaya and an anon- 
ymous reviewer improved the manuscript with helpful 
pointers. Total Foundation funded the expedition within 
its “Coral Reef Biodiversity Programme.” The SEM 
investigation was supported by NSF MRI-0420706 
erant to Henry Chaney, Michi vel Caterino, and Daniel 
L. Geiger. Visiting curatorships from AMS and MNHN 
to DI c are kindly acknowledged. 


LITERATURE CITED 


Geiger, D. L. 2003. Phylogenetic assessment of characters 
proposed for the generic classification of Recent Scissur- 
ellidae (Gastropoda: Vetigastropoda) with a description of 
one new genus and six new species from Easter Island 
and Australia. Molluscan Research 23: 21-83. 

Geiger, D. L. 2006a. Eight new species of Scissurellidae and 
Anatomidae (Mollusca: Gastropoda: Vetigastropoda) from 


Page 200 


THE NAUTILUS, Vol. 122, No. 4 


around the world, with discussion of two new senior syno- 
nyms. Zootaxa 1128: 1-33. 

Geiger, D. L. 2006b. Sasakiconcha elegantissima new genus 
and species (Gastropoda: Vetigastropoda: Anatomidae?) 
with disjointly coiled base. The Nautilus 120: 45-51. 

Geiger, D. L. 2006c. A new blind Anatoma species from the 
bathyal of the northeastern Pacific (Vetigastropoda: Ana- 
tomidae). Molluscan Research 26: 108-122. 

Geiger, D. L. 2008. Monographing micromolluscs: A case 
study on Scissurellidae s.]. (Vetigastropoda). Zoosymposia 
1: 133-145. 

Geiger, D. L. and P. Jansen 2004a. Revision of the Australian 
species of Anatomidae (Mollusca: Gastropoda: Vetigastro- 
poda). Zootaxa 414; 1-35. 

Geiger, D. L. and P. Jansen. 2004b. New species of Australian 
Scissurellidae (Mollusca: Gastropoda: Vetigastropoda) 
with remarks on Australian and Indo-Malayan species. 
Zootaxa 714; 1-72. 

Geiger, D. L., B. A. Marshall, W. F. Ponder, T. Sasaki, and 
A. Warén. 2007. Techniques for collecting, handling, 
and pa small molluscan specimens. Molluscan 
Research 27: 1-50. 

Geiger, D. L., A Niitzel, and T. Sasaki. 2008. Vetigastropoda. 
In: Phylogeny and Evolution of the Mollusca. W. F. Pon- 
der and D. R. Lindberg (eds). University of California 
Press, Berkeley, pp. 297-330. 

Geiger, D. L. and T. Sasaki. 2008. Four new species of Anato- 
midae (Mollusca: Vetigastropoda) from the Indian Ocean 


(Reunion, Mayotte) and Australia, with notes on a novel 
radular type for the family. Zoosymposia 1: 247-264. 
Geiger, D. L. and C. E. Thacker. 2005. Molecular phylogeny of 
Vetigastropoda reveals non-monophyletic Scissurellidae, 
Trochoidea, and Fissurelloidea. Molluscan Research 25: 

47-55. 

Kano, Y. 2008. Vetigastropod phylogeny and a new concept of 
Seguenzioidea: independent evolution of copulatory 
organs in the deep-sea habitats. Zoologica Scripta 37: 
1-21. 

Lozouet, P. R. von Cosel, V. Héros, A. Le Goff, P. Maestrati, 
J.-L. Menou, S. Schiaparelli, and J. Tréndlé. 2004. 
LAtelier Rapa 2002 (Polynésie frangaise). Xenophora 
107: 17-28. 

Lozouet, P., R.von Cosel, V. Héros, A. Le Goff, P. Maestrati, 
J.-L. Menou, S. Schiaparelli, and J. Tréndlé. 2005. Biodi- 
versity gradient in the Pacific: first results of RAPA 2002 
(French Polynesia). In: G. Richard (ed.) Les mollusques 
dans la recherche actuelle. Actes du Iéme Congres 
International des Sociétés Européennes de Malacologie 
(La Rochelle 24-27 juin 2003): 93-99. 

Marshall, B. A. 2002. Some Recent scissurellids form the New 
Zealand region, and remarks on some scissurellid genus 
group names (Mollusca: Gastropoda). Molluscan Re- 
search 22; 165-181. 

Schwabe, E. and P. Lozouet. 2006. Chitons (Mollusca, Poly- 
placophora) from Rapa, the southernmost island of 
Polynesia. Zoosystema 28: 617-633. 


THE NAUTILUS 122(4):201-216, 2008 


Page 20] 


Early Pleistocene mollusks of the Tubul Formation, 


South-Central Chile 


Sven N. Nielsen! 
GeoForschungsZentrum Potsdam 
Sektion 3.1 

Telegrafenberg 

14473 Potsdam GERMANY 
nielsen@gpi.uni-kiel.de 


Claudio Valdovinos 

Center of Environmental Sciences, EULA-Chile 
Universidad de Concepeidn! 

Casilla 160-C 

Concepcion, CHILE 

and 

Patagonian Ecosystems Research Center (CIEP) 
Coyhaique, CHILE 

evaldovi@udec.cl 


SES 


ABSTRACT 


Early Pleistocene mollusks of the Tubul Formation are re- 
viewed and 23 species are recognized. These are twelve 
bivalves, ten gastropods, and one scaphopod. With two excep- 
tions, all of them are known from the Recent. The fauna is 
compared with Miocene, Pliocene-Pleistocene, and Recent 
faunas of the Chilean coast. The composition of the Tubul 
fauna resembles the Recent one of the Magellan Region, sug- 
gesting cooler water temperatures for south-central Chile dur- 
ing the early Pleistocene than at present. 


Additional Keywords: Mollusca, eastern Pacific, climate 


INTRODUCTION 


The marine late Pliocene to early Pleistocene of Chile 
comprises localities distributed along almost the whole 
length of the coast of Chile (Figure 1). The northern 
localities attributed to the Pliocene and Pleistocene are 
exclusively nearshore environment outcrops between 
Peninsula Mejillones and Coquimbo that have been de- 
scribed in detail by Herm (1969) and Le Roux et al. 
(2004, 2005, 2006). Herm (1969) described the faunas 
from these localities and also commented on the fauna 
of the La Cueva Formation, which overlies the Miocene 
Navidad Formation of central Chile. Pliocene faunas 
from southern Chile have been described from the 
islands Guafo (Frassinetti, 1997, 2000) and Guamblin 
Frassinetti and Covacevich, 1995). Another 
late Miocene or early Pliocene age from central Chile is 
known from Lo Abarca near San Antonio (Covacevich 
and Frassinetti, 1990: DeVries, 2003). 


’ Current address: Institut fiir Geowissenschaften, Christian- 
Albrechts-Universitat zu Kiel, Ludewig-Meyn-Str. 10, 24118 
Kiel, GERMANY. 


fauna of 


The Tubul Formation (Figure 2) was first described by 
Feruglio (1949), with its type area located on the south- 
ern side of the Gulf of Arauco (37°14! S, 73°26’ W). 
Onshore outcrops are limited to the east by the Cor- 
dillera de Nahuelbuta and to the west by the Pacific 
Ocean. Feruglio (1949) described the sediments of the 
Tubul Formation as light-gray soft tuffaceous sandy 
mudstones with fossil remains, which can reach a 
thickness of up to 100 m (Garcia, 1968). Using macro- 
fossils, Briiggen (1950) showed that these sediments 
can be correlated with those of Coquimbo. From their 
study of microfossils, Martinez and Osorio (1968) con- 
cluded that the basal part of the Tubul Formation 
corresponds to the middle Pliocene. However, later 
studies by Martinez (1976) in Caleta Hueton led to 
the conclusion that the base of the Tubul Formation 
corresponds to the late Pliocene. Subsequently, Bird- 
Bagoczky (1979) recognized a lower and an upper 
member: the lower member has a thickness of approx- 
imately 50 m and is formed by very fine, slightly 
calcareous, dark-gray sandstones, including abundant 
fossil marine fauna; the upper member a6 has a 
thickness of approximately 50 m and is formed by 
light-gray to olive fine sandstones and mudstones with 
rare Foals: 

According to Pineda (1983, 1986), deposition of the 
sediments of the Tubul Formation was rapid, in a quiet 
zone close to the coast, maybe lagoonal. These sedi- 
ments should correspond to the Plio-Pleistocene-bound- 
ary (Pineda, 1983, 1986) and occur at the following 
localities: Punta Pichicui, Estero Chupalla, Villa Alegre, 
Los Alamos, Quebrada Raquilco, El Tique, Estero 
Licauquen, and Minas Trihueco. According to Pineda, 
the Tubul Formation represents a transgressive se- 
quence. The presence of quartz-sand and a conglomer- 
ate at the base of the formation has been interpreted as 
a former coastal line. Above this level, very fine clayey 
sandstones have been deposited, which should re present 


Page 202 THE NAUTILUS, Vol. 122, No. 4 


| 18°S 


20°S 


(22°S 


30°S 


32°S 


34°S 


36°S 


38°S 


Elevation (m) 
40°S 7000 


42°S 


= 


44°S 


46°S 


48°S 


50°S 


52°S 


6 5 
16 
4 9 


1 54°S 


2 


1 3.5 78101214 1719 
13 18 ™ Beg kilometers 


M.9L 
M.bL 

c 

0 

8 
M.99 


Figure 1. Extant ranges of bivalve and gastropod species occurring at Las Penas. 1. Ennucula grayi. 2. Tindariopsis sulculata. 3. 
g 8 I I 8 gray 
Malletia chilensis. 4. Zygochlamys patagonica. 5. Cyclocardia velutinus. 6. Macoma inornata. 7. Darina solenoides. 8. Ensis macha. 
yg Y i y 
9. Retrotapes exalbidus. 10. Pandora cistula. 11. Epitonium magellanicus. 12. Fusitriton magellanicus. 13. Trophon geversianus. 14. 
“Xymenopsis” dispar. 15. Chorus giganteus. 16. Nassarius taeniolatus. 17. Adelomelon ancilla. 18. Bela paesleri. 19. Scaphander 
interruptus. Stars indicate Plio—Pleistocene fossil localities. 


S. N. Nielsen and C. Valdovinos, 2008 


Page 203 


5880 km 


ARAUCO 
GULF 


Punta 
las Penas 


5878 


Figure 2. 


a shallow-marine environment. Radic et al. (2005) cited 
a thickness of 100 to 500 m for the Tubul Formation 
based on well data and seismic lines. 

The molluscan fauna decribed in this paper comes 
from the coastal bluffs near Las Pefias (Figure 2), which 
represent the uppermost part of the Tubul Formation. 


MATERIALS AND METHODS 


The mollusks described in this study belong to the col- 
lection of the late Professor Lajos Bir6 and are housed in 
the Departamento de Ciencia de la Tierra, Universidad 
de Concepcion, ple The material comes from near 
Punta Las Penias (37°14/32” S, 73°26'02” W, Figure 2), 
south of the village Tubul. and is designated by ihe let- 
ter ‘T° in the collections. The number of specimens 
inspected for each species is indicated by a number in 
parentheses if the number is greater than one. Species 
are not formally described because all except two are 
well known, Modern geographic ranges of species are 
modified from Valdovinos (1999). 


SYSTEMATICS 


Class Bivalvia Linnaeus, 1758 
Family Nuculidae Gray, 1824 
Genus Ennucula Iredale, 1931 


Ennucula grayi (dOrbigny, 1846) 
Figure 3) 


Nucula obliqua Sowerby, 1833: 5, pl. 16, fig. 21. 

Nucula grayi 7 Orbigny, 1846: vol. 5, 625 (nomen novum 
for N. obliqua Somerby, non Lamarck, 1819). 

Ennucula grayi (d’Orbigny).—Soot-Ryen, 1959: 13, ph. 1 
fig. 8: Villarroel and Stuardo, 1998: 13 3, figs. 8, 70, 71, 

3, LOT—-109. 

iephaceate araucana (Philippi) —Villarroel and Stuardo, 

1998: 165 (partim), figs. 142-143. 


Area of the Tubul Formation and sample site at Las Penas. Height of coastal bluff about 18 m. 


Ennucula valdiviana (Philippi) —Villarroel and Stuardo, 
1998: 165 (partim), figs. 134-137 

Ennucula lebuensis (Philippi).—Villarroel and Stuardo, 
1998: 165 (partim), figs. 138-139. 


Material Examined: 47 specimens: T/4 (2), T/5 (6), 
T/7 (5), T/11, T/17, T/204, T/237, T/386, oie T/1276 
(3), Daal , T/1394—-1397, T/1537, T/1749-1754, T/1968 
(7), T/1969 (2), T/1989-1990, T/2048, T/2090. 


Measurements: Width 19.2 mm, length 16.7 mm 


(T/237) 


Occurrence: Recent: Coquimbo to Magellan Region; 
Plio-Pleistocene: Arauco to ?Isla Guafo, 


Remarks: = Nucula barrosi Philippi, 1887 (=N. arau- 
cana Philippi, 1587), N. lebuensis Philippi, 1887, and 
N. valdiviana Philippi, 1887, are Miocene species from 
the Navidad, Ranquil, and Sto. Domingo formations of 
central and southern Chile. However, their generic 
placement is not known and therefore possible relation- 
ships to E. grayi remain unclear. The specimens men- 
tioned and figured by Villarroel and Stuardo (199%) 
belong in E. grayi and not in one of these Miocene 
species. The taxonomic positions of these older species 
must await a revision of the Chilean Miocene nuculoids. 


Genus Tindariopsis Verrill and Bush, 1897 


Tindariopsis sulculata (Gould, 1852) 
(Figure 4) . 


Nucula striata King and Broderip, 1832: 337 (non 
Lamarck, 1805). 

Nucula sulculata Gould, 1852: 12, 434, pl. 37, figs. 539 
a-e (Couthouy MS). 

Nucula elegans Hupé, 1854: 305, Conquiliologia pl. 5 
fig. 7; Philippi, 187: 189, pl. 31, fig. 6. 

Tindariopsis sulculata (Gould).—Dell, 1964: 149; Villar- 
roel and Stuardo, 1998: 144-145, figs. 39-41, SO-S2 
157-159. 


THE NAUTILUS, Vol. 122, No. 4 


S. N. Nielsen and C. Valdovinos, 2008 


age 205 
Page 205 


Nuculana sulculata (Gould)—Dell, 1971: 167, pl. 1, figs. 
14, pl. 2, fig. 9. 

Nuculana elegans (Hupé).—Frassinetti and Covacevich, 
1995: 51, text-fig. 3a, pl. 1, figs. 1-5; Frassinetti, 
1997: 60, pl. 1, figs. 5—7. 

Tindaria sulculata (Gould).—Griffin and Nielsen, 2008: 
16, pl. 5, figs. 3-5. 


Material Examined: ca. 1S50 specimens: T/1 (92), T/S 
(24), T/9 (89), T/12 (17), T/13 (36), T/14 (17), T/15 (48), 
T/16 (9), T/1S (198), T/19 ges T/32 i T/53-54, 
ne (4), T/149 (9), T/164 (2), T/168 (8), T/206, T/207 

), T/236 (4), T/296 (3), T/542, T/97 4 (2), T/982-984 (ca, 
ea T/1083 (7), T/1182-1187 (ca. 90), T/1235 (6), 
T/1270 (293), T/1308 (ca. 30), T/1379 (92), T/1518 (11), 
T/1536 (21), T/1624, T/1681 (ca.45), T/1682 (6), T/1802— 
1804 (ca. 90), T/1911 (23), T/1912 (ca. 90), T/1913, 
T/1970 (15), T/2001 (5), T/2009 (23), T/2042 (7). 


Measurements: Width up to 16 mm (1/1682), 
Occurrence: Recent: Talcahuano to Strait of Magel- 


lan; Plio-Pleistocene: Arauco to Isla Guamblin. 


Remarks: As for many other species, there are different 
names for Recent and fossil records of the same species. 
Villarroel and Stuardo (1998, p. 145), in their review of the 
Recent and fossil species of Chilean Protobranchia, noted 
that “The ornamentation of the shell of T. sulculata is very 
similar to that of the fossil species 7. elegans (...). It differs 
from that by its shorther length and by the truncation of its 
extreme posterior.” However, we do not consider these 
differences as sufficient for specific separation and regard 
them as intraspecific variation instead. 

Family Malletiidae H. and A. Adams, 1858 

Genus Malletia Moulins, 1832 

Malletia chilensis Moulins, 1832 

(Figure 5) 


Malletia chilensis Moulins, 1832: 85, pl. 1, figs. 1, 8. 
Malletia inequalis Dall, 1908: 219, 383. 
?Malletia sp. Frassinetti and Covacevich, 1995: 50. 


Material Examined: Nine specimens: T/544(2), 
T/1091-1093, T/1917, T/2176-2178. 


Measurements: Figured specimen width 26 mm 
(T/1092). 
Occurrence: Recent: Coquimbo to Magellan Region; 


Plio-Pleistocene: Arauco to Isla Guamblin. 


Remarks: According to Villarroel and Stuardo (1998), 
there are four or five species of Malletia known from 
Chile, all of which are rather similar to each other. Apart 
from M. chilensis, they recorded M. patagonica Mabille 


and Rochebrune, 1889, M. inequalis Dall, 1908, and 
M. magellanica (Smith, 1875). Malletia hyadesi Mabille 
and Rochebrune, 1889 was listed as a probable synonym 
of M. patagonica. Malletia volckmanni ( (Philippi, 1887) 
most probably is of Miocene age and has not been 
revised and its placement should be regarded with caution. 


Family Mytilidae Rafinesque, 1815 
Genus ?Mytilus Linnaeus, 1758 


Mytilus sp. 
(Figure 6) 


Material Examined: One specimen: T/106. 


Measurements: Length 58 mm (T/106). 


Remarks: According to Valdovinos (1999) there are 
nine species of Mytilidae known from Chile. Based on 
the main characters of the our shell fragment, such as 
the pointed morphology of the umbo and fine external 
growth striae, our specimen may correspond either to 
he genus Mytilus or Choromytilus. In southern Chile 
the species Choromytilus chorus (Molina, 1782) and 
Mytilus edulis chilensis Hupé, 1854, coexist, but it is 
not possible to identify our specimen to species level. 
Mytilidae of uncertain generic and specific status are 
known from Miocene through Recent and are in need 
of a detailed revision. 


Family Pectinidae Rafinesque, 1815 
Genus Zygochlamys thering, 1907 


Zygochlamys patagonica (King and Broderip, 1532) 
(Figure 7) 


9907 


Pecten patagonicus King and Broderip, 1832, 5: 337. 

Pecten tenuicostatus Hupé, 1854: 291, pl. 5, fig. 4; Phi- 
lippi 1887: 203, pl. 47, fig. 1; Méricke, 1896: 580, 
pl. 12, figs. 13-16. 

Chlamys patagonica patagonica (King and Broderip). 
—Beu, 1985: 1-11, pl. 1, figs. 1-4. 

Zygochlamys Paes (King and Broderip).—Waller, 
1991: 28- 30, pl. 2, figs. 13, 14; Griffin and Nielsen, 
2008: 37, pl. 16, f igs. 5, 6. 

Chlamys a ee (Hupé).—Frassinetti and Covace- 
vich, 1995: 52, text-fig. 3b, pl. 1, figs. 6-9. 


Material Examined: 305 specimens: 1/63, T/79, 
T/135, T/147, T/166, T/175, T/180, T/222-223, T/243, 
T/255, T/256, T/293, T/305, T/307—309, T/311—-312, 
T/329-334, T/361-380, T/387—399, T/432-434., 
T/491-504, T/577-578, T/725-726, T/72S8—-772, T/775-— 
777, T/SO6—-820, T/1006—1017, T/1039, T/L074—1077, 
T/1080-1082,  T/1156-1157, T/1160—-1166, T/1269, 


Figures 3-13. 


Bivalvia. 3. Ennucula grayi (dOrbigny, 1846) [length 19.2 mm]. 4. Tindariopsis sulculata (Gould, 1852) [length 


16.0 mm]. 5. Malle‘tia chilensis Moulins. 1832 [length 16.0 mm]. 6. Mytilus sp. [length 58.0 mm]. 7. Zygochlamys patagonica (King 


and Broderip, 1832 


1844) [length 16.0 sin] 10. Darina solenoides (King and Broderip, 1832 


[length 75 mm]. 8. Cyclocardia velutinus (E. A. Smith, 1881) [length 18.5 mm]. 9. Macoma inornata (Hanley, 


) Hength 44.0 mm] 11. Ensis macha (Molina, 1782) {length 


97.2 mm]. 12. Retrotapes exalbidus (Dillwyn, 1817) [length 62.0 mm]. 13. Pandora cistula Gould, 1850 {length 23.0 mm]. 


Page 206 


T/1280-1281, T/1284, T/1309-1310, T/1358, T/1371- 
1378, T/1380-1386,  T/1389-1392,  T/1415-1424, 
T/1514-1516,  T/15538-1559,  T/1579-1586, = T/1632, 
T/1677-1680, T/1714-1725, T/1827, T/1863-1875, 
T/1964-1967, T/2003-2008, T/2030-2031, — T/2033, 
T/2037-2038, T/2049, T/2052, T/2055-2056, T/2059- 


2065, T/2154—2167, T/2181-2192, T/2225-2239. 
Measurements: Height 75 mm, width 73 mm (T/223). 


Occurrence: 
Magellan; Plio-Pleistocene: Arauco to Isla Guamblin. 


Remarks: Because revision of Chilean pectinids is not 
within the scope of this work, relationships of Z. patago- 
nicus to the northern species described by Herm (1969) 
are not discussed here. Zygochlamys patagonicus is 
the type species of Psychrochlamys Jonkers, 2003. 
However, like Griffin and Nielsen (2008) and Dijkstra 
and Marshall (2008), we consider Psychrochlamys a 
synonym of Zygochlamys. 


Family Carditidae Fleming, 1528 
Genus Cyclocardia Conrad, 1867 


Cyclocardia velutinus (E.A. Smith, 1881) 

(Figure §) 

Cardita (Actinobolus) velutinus E.A. Smith, 1881: 42, 
pL 5, fig. 8. 

Corda acloenaini Philippi, 1SS7: 167, pl. 37, fig. 4. 

Cardiocardita volckmanni (Philippi). _Frassinetti and 
Covacevich, 1995: 53, text-fig. 3c, pl. 1, figs. 10-17; 
Frassinetti, 1997: 65, pl. 1, figs. 17-20. 

Material Examined: 19 specimens: T/507-510, T/ 

639, T/786-787, T/1546-1552, T/162S—1630, T/1756, 

T/2256. 


Measurements: 
(T/508). 


Width 18.5 mm, length 17 mm 


Occurrence: Recent: Arica to Beagle Canal; Plio- 
Pleistocene: Arauco to Isla Guamblin. 


Remarks: A number of different Cyclocardia species has 
been reported from southern Chile (Forcelli, 2000). 


Family Tellinidae Blainville, 1814 
Genus Macoma Leach, 1819 


Macoma inornata (Hanley, 1S44) 
(Figure 9) 


Tellina inornata Hanley, 1844, 1847: 315, pl. 59, fig. 127. 

Tellina tubulensis Philippi, 1887: 134, pl. 32, fig. 7. 

Macoma (Psammacoma) inornata (Hanley).—Soot-Ryen, 
1959: 62, pl. 4, figs, 34-36. 

Macoma_ tubulensis 
pl. 2, fig. 6 


Frassinetti, 1997: 72, 


(Philippi).- 


Material Examined: Eight specimens: T/64, T/721-— 
724, T/789, T/1236, T/2067. 


Measurements: Width figured specimen (T/64) 


26 mm 


Recent: Island of Chiloé to Strait of 


THE NAUTILUS, Vol. 122, No. 4 


Occurrence: Recent: Atacama to Gulf of Ancud; Plio- 
Pleistocene: Arauco to Isla Guamblin. 


Remarks: Herm (1969, p. 119) described a specimen 
of Ardeamya sp. as the only tellinid from the Pliocene 
south of San Antonio. Since Herm knew the Philippi 
collection, it must be assumed that it is not conspecific 
with the species reported here. 


Family Mactridae Lamarck, 1809 
Genus Darina Gray, 1853 


Darina solenoides (King and Broderip, 1832) 
(Figure 10) 


Erycina solenoides King and Broderip, 1832: 335. 
Darina solenoides (King and Broderip)—Smith, 1905: 
337, 


Mactra sp.—Frassinetti and Covacevich, 1995: 54. 
Material Examined: One specimen: T/1238. 


Measurements: Width figured specimen (T/1238) 


44 mm. 


Occurrence: Recent: Strait of Magellan and Tierra 
del Fuego; Plio-Pleistocene: Arauco to Isla Guamblin. 


Remarks: Herm (1969) discussed the problems of 
related forms appearing in the Pliocene of Coquimbo 
to San Antonio which he placed in the genus Mulinia 
Gray, 1837. 


Family Pharidae H. and A. Adams, 1858 
Genus Ensis Schumacher, 1817 


Ensis macha (Molina, 1782) 
(Figure 11) 


Solen macha Molina, 1782: 203; Hupé, 1854: 369, pl. 8, 
fig. 6. 

Solen oladiolus Sowerby, 1839: 153, pl. 43, fig. 4. 

Solen ‘gladiolus? Sowerby. —Philippi, 1887: 169, pl. 34 
fig, 9. 

Ensis macha (Molina). 

Ensis sp.—Frassinetti, 1997: 67. pl. 2 


—Carcelles, 1944: 292. 
, figs. 4-5. 


Material Examined: 38 specimens: T/192, T/579-580, 
T/793—797, T/1084, T/1 101-1102, T/1175—1176, T/1285— 
1290, T/1368—1370, T/1412—1414, T/1694, T/1707—-1712, 
T/1923, T/1972, T/2000, T/2068, T/2226-2227. 


Measurements: Up to 97.2 min (T/1707). 


Occurrence: Recent: Caldera to Magellan Region; 
Plio-Pleistocene: Coquimbo to Isla Guafo. 


Remarks: —Ensis macha is one of the few species which 
has a very wide geographical range in Pliocene and Re- 
cent faunas of Chile. During the Pleistocene it also 
reached Mejillones (Herm, 1969; own observations). 


Family Veneridae Rafine sque, 1815 
Genus Retrotapes del Rio, 1997 


tetrotapes exalbidus (Dillwyn, 1817) 
(Figure 12) 


S. N. Nielsen and C. Valdovinos, 2008 


Page 207 


Venus exalbida Chemnitz, 1795: 225, pl. 202, fig. 2 (non 
binom.); Dillwyn, 1S17: 170. 

Venus aerea Hupeé, 1854: 338 

Venus subalbicans Hupé, 1854: 339. 

Venus araucana Philippi, LSS7: 117, pl. 17, fig. 6; Tavera 
and Veyl, 1958: 170, pl. 4, fig. ite 

Samarangia exalbida (Dilhwyn).—Carcelles, 1944: 287, 
pl. 12, figs. 93, 94. 

Eurhomalea araucana (Philippi).—Frassinetti, 1974: 47, 
figs. 1-2; Frassinetti and Covacevich, 1995: 54, 
text-fig. 3c, pl. 1, fig. 1S; Frassinetti, 1997: 74, pl. 2, 
fig. 6. 

Retrotapes exalbida (Dillwyn).—del Rio, 1997: 80-82, 
figs. 22, 23, 41. 

Retrotapes meAlbius (Dillwyn).—Reid and Osorio, 2000: 
139, fig. 5]; Griffin and Nielsen, 2008: 7, 35, pl. 1 
figs. 2-4, pl. 16, figs. 1-3. 


Material Examined: 669 specimens: T/21—25, T/44— 
49, T/56, T/5S—62, T/65-73, T/75, T/78, T/S4-109, 
T/136-140, T/151, T/153-163, T/173-176, T/1S1—-191, 
T/19S8—203, T/20S—213, T/216-221, T/226-235, T/241- 
942. T/249-253, T/257-263, T/266-272, T/277-282, 
T/319-328, T/400-421, T/449-490, T/516-539, T/545— 
557, T/607-614, T/653-671, T/681—703, T/710-719, 
T/S02, T/S29-SS1, T/901-910, T/919-968,  T/97S, 
T/98S7—-989, T/1027—-1037, T/1109-1114, T/1128-1134, 


T/1142-1154,) T/1158-1159,  T/1194-1197, T/1212- 
1217, = T/1228-1229, = T/1237-1268, = T/1298-1306, 
T/1316-1324,  T/1333-1356, T/1425-1428,  T/1511— 


1518, 1/1534, T/153S-1544, T/1645-1676, T/17S1- 
1796, =T/1805-1809, = T/1817-1826, — T/1835-1862, 
T/1949-1963, T/2011-2014,  T/2035-2036, T/2039— 
2041, T/2046, T/2058, T/2081, T/2093-2097, T/2140- 
2147, T/2230-2249. 


Measurements: Up to width 62 mm, length 54.5 mm 
(T/967). 
Occurrence: Recent: Chiloé to Strait of Magellan: 


Plio-Pleistocene: Arauco to Isla Guamblin. 


Remarks: A number of different Eurhomalea species, 
most probably now belonging in Retrotapes, has been 
reported from northern Chile (Herm, 1969). 


Family Pandoridae Rafinesque, 1815 
Genus Pandora Bruguiere, 1797 


Pandora cistula Gould, 1850 
(Figure 13) 


Pandora cistula Gould, 1850: 217. 
Kennerlyia patagonica Dall, 1915: 450. 


Pandora sp.—Frassinetti and Covacevich, 1995: 55, pl. 1, 
fig. 19. 


Material Examined: T/141—146. 1/345, 1T/353-354, 
T/615-634, T/781. T/783, T/784 (3), T/972-973, T/986 
(2), T/1273 T/1291-1292, T/1311-1312, T/1405-1411, 
T/1535 (8), T/1755, T/2175, T/2225. 


Measurements: Width 23 mm (T/22235). 


Occurrence: Recent: Valparaiso to Magellan Region; 
Plio-Pleistocene: Arauco to Isla Guamblin. 


Remarks: This species is the only living species of 
Pandora recorded in Chile (see Forcelli, 2000). 


Class Gastropoda Cuvier, 1795 
Family Epitoniide ie Berry, 1910 
Genus Epitonium Roding, 1798 
Subgenus Nitidiscala de Boury, 1909 


Epitonium (Nitidiscala) magellanicus (Philippi, 1845) 
(Figure 14) 


Scalaria magellanica Philippi, 1845: 46. 

Scalaria (Opalia) magellanica Strebel, 1905: 656, pl. 23 
figs. 44a—f. 

Cirsotrema (Coroniscala) magellanicum (Philippi). — 
Wenz, 1938: 798, fig. 2322. 

Epitonium (Nitidiscala) magellanica(Philippi).— 
1987: 70, fig. $2. 

siaee biroi - rassinetti and Covacevich, 1995: 56, 
pl. 2, figs. 1-2; Frassinetti, 1997: 135, pl. 1, figs. 4-5. 


Rael Examined: One specimen: T/2071. 
Height 20.5 mm. 


Occurrence: Recent: Strait of Magellan; Plio-Pleisto- 
cene: Arauco to ?Isla Guamblin. 


Ramirez, 


Measurements: 


Remarks: A number of different epitoniid species has 
been reported from southern Chile (Forcelli, 2000). The 
type material of Epitonium biroi Frassinetti and Cova- 
cevich, 1995 from Guamblin was not revised, so it 
remains unclear if it represents a different species or a 
synonym. 


Family Naticidae Guilding, 1834 
Genus Euspira Agassiz, 1838 


Euspira guamblinensis Frassinetti and Covacevich, 1995 
(Figure 15) 


Polinices (Euspira) guambline nsis Frassinetti and Cova- 
cevich, 1995: - text-fig. dae, pl. 2, figs. 3-5; Fras- 
sinetti, 1997; 135, pl. 1, “fips, 6-9. 


Material Examined: 37 specimens: 1/30, T/152, 
T/215, T/335-337, T/778, T/779, T/976, T/1177-1181, T/ 
1272, T/1279, T/1627, T/1684, T/1696, T/1704, T/1705, 
T/1777, T/1899, T/1901, T/1902, T/1909, T/1940-1942, 
T/1991, T/2053, T/2068, T/2151, T/2210-2212, T/2255. 


Measurements: Height up to 22.5 mm, figured speci- 


men (T/1909) height 14 mm. 


Occurrence: Recent: Extinct; Plio-Pleistocene: Ara- 
uco to Isla Guamblin. 

Remarks: None of the extant naticid species rev- 
iewed by Pastorino (2005a) matches Euspira guambli- 
nensis in its callus characters and we regard it as a valid 
species. 

Family Ranellidae Gray, 1854 

Genus Fusitriton Cossmann, 1903 


THE NAUTILUS, Vol. 122, No. 4 


Page 208 


S. N. Nielsen and C. Valdovinos, 2008 


Page 209 


Fusitriton magellanicus (Réding, 1798) 
(Figure 16) 


Murex magellanicus Chemnitz, 1788: 275, pl. 164, 
fig. 1570 (non binom.). 

Neptunea magellanica Réding, 1798: 116. 

Triton cancellatum Lamarck, 1816: 4 

Priene (Fusitriton) cancellatus (Lamarck).—Cossmann, 
1903: LOY, fig. S. 

Argobuccinum  (Fusitriton) cancellatus | (Lamarck).— 
Wenz, 1938: 1058, fig, 3022. 

Argobuccinum (Argobuccinum) magellanicum  (Réd- 
ing).—Carcelles, 1944: 247, pl. 2, fig. 23. 

Argobuccinum (Fusitriton) magellanicum (Réding). — 
Carcelles and Williamson, 1951: 286. ; 

Fusitriton cancellatus (Lamarck).-Smith, 1970: 475, pl. 
42, figs. 4-10. 

Fusitriton magellanicus (R6ding).—Cernohorsky, 1977: 
107, fig. 3. 

Material Examined: 20 specimens: T/150, T/171, 

T/178, T/214, T/315, T/340, T/511, T/512, T/572, T/971, 

T/1100, T/1282, T/1360, T/1575, T/1731, T/2026, 

T/2039, T/2169, T/2214, T/2252. 


Measurements: 
73 mm. 


Figured specimen (T/2026) height 


Occurrence: Recent: Los Vilos to Magellan Region 
and Juan Fermdandez Archipelago; Plio-Pleistocene: 
Arauco. 


Remarks: This is a well known species that clearly 
does not present any taxonomic problems. 

Genus Sassia Bellardi, 1872 

Sassia leucostomoides (Sowerby, 1846) 

(Figure 17) 


Triton leucostomoides Sowerby, 1846: 240, pl. 4, fig. 64. 

Sassia leucostomoides (Sowerby).—Frassinetti, 1997: 136, 
pl. 1. figs. 11-14; Griffin and Nielsen, 2008: 49, 
pl. 21, figs. 9-14. 


Material Examined: Six specimens: T/310, T/574; 
T/2032, T/2153, T/2196, T/2215. 


Measurements: Figured specimen (T/574) height 
44.4 mm. 
Occurrence: Recent: Extinct; —Plio-Pleistocene: 


Arauco to Isla Guafo. 


Remarks: This species was originally described from 
Guato (Sowerby, 1846) and this fecond is the first from 
another locality. 


Family Muricidae Rafinesque, 1S15 
Genus Trophon Monttort, 1810 


Trophon geversianus (Pallas, 1774) 
(Figure 1S) 


Buccinum geversianus Pallas, 1774: 33 

Murex magellanicus Gmelin, 1791: 3548. 

Fusus geversianus (Pallas)—Hupé, 1854; 167. 

Trophon geversianus (Pallas) —Gould, 1852: 227, pl. 6, 
fig. 297 


, pl. 3, figs. 1, 2. 


22/. 


Material Examined: Nine specimens: T/513, 1/569, 
T/575, T/640, T/798, T/1361, T/1574, T/2029, T/2253. 


Measurements: Height 62 mm (1/640), figured spec- 
imen (T/1361) 39 mm. 


Occurrence: — Recent: Chiloé to Magellan Region; 
Plio-Pleistocene: Arauco., 


Remarks: The Chilean species of the genus Trophon 
present a high intraspecific variability, with many species 
described as synonyms of T. geversianus. A detailed re- 
vision of the genus Trophon, including T. geversianus, 
was recently provided by Pastorino (2005b). 


Genus “Xymenopsis” Powell, 1951 


“Xymenopsis” cf. “X.” dispar (Rochebrune and Mabille, 
1889) 


(Figure 19) 


Fusus dispar Rochebrune and Mabille, 1589: H57, pl. 2 
fig. 3. 

Trophon dispar (Rochebrune and Mabille).— 
and Williamson, 1951: 289, 

Xymenopsis dispar (Rochebrune and Mabille)—Powell, 
1951: 159. 


Carcelles 


Material Examined: 16 specimens: T/352, T/586 (5), 
T/588 (7), T/1357, T/2219, T/2245. 


Measurements: Figured specimen (1/352) 22 mm. 


Occurrence: Recent: Tierra del Fuego; — Plio- 


Pleistocene: Arauco. 

Remarks: — Pastorino and Harasewych (2000) stated 
that this species does not belong in Xymenopsis but did 
not provide a more appropriate generic-level placement 
in the Muricidae. 


Genus Chorus Gray, 1847 


Chorus giganteus (Lesson, 1830) 
(Figure 20) 


Monoceros giganteus Lesson, 1830; 405, N° 165, pl. 11, 
fig. 4. 


Figures 14-25. 


Gastropoda and Scaphopoda. 14. Epitonium (Nitidiscala) magellanicus (Philippi, 1845) [height 25.0 mm]. 15. Euspira 


guamblinensis Frassinetti and Covacevich, 1995 [height 22.5 mm]. 16. Fusitriton magellanicus (Réding, 1798) [height 73.0 mm]. 17. 
Sassia leucostomoides (Sowerby, 1846) [height 44.4 mm]. 18. Trophon geversianus (Pallas, 1774) [height 62.0 mm]. 19. “Xymenopsis” ct 
X.” dispar (Rochebrune and Mabille, 1889) [height 22.0 mm], 20. Chorus giganteus (Lesson, 1830) [height 95.0 mm]. 21. Nassarius 


taeniolatus (Philippi, 1845) [height 11.0 mm], 22. Adelomelon ancilla | 


Ligthfoot, 1786) [height 142.0 mm], 23. Bela paesleri Strebel 


1905 [height 16.0 mm]. 24. Scaphander interruptus Dall, 1889 [height 18.5 mm], 25. Dentalium sp. [length 72.0 mm| 


Page 210 


THE NAUTILUS, Vol. 122, No. 4 


Monoceros giganteus? Lesson.—MGricke, 1896: 564. 

Rapana ( (Chorus) gigantea (Lesson). —Wenz, 1941: 1038, 
fig. 3080. 

Chorus giganteus (Lesson).—Carcelles, 1954: 271, pl. 5, 
figs. 1-11; Herm, 1969: 135, pl. 15, figs. e 4b; 
DeVries, 1997: 132, pl. 1, figs. 1-4, pl. 4, fig. 4. 

Chorus grandis (Philippi). -~Herm, 1969: pl. 15, figs. 


Material Examined: 17 specimens: T/576, T/985, 
T/1174, T/1193, T/1904—1905, T/1907—-1908, T/1910, 
T/1984, T/2170-2172, T/2208, T/2213, T/2272, one 


specimen without number. 


Measurements: Height up to 95 mm (T/9S5), figured 
specimen (T/1905) 50. 5 mm. 


Occurrence: Recent: Papudo to Calbuco; Plio-Pleis- 
tocene: southern Peru to Arauco. 


Remarks: The species of Chorus have been reviewed 
by Herm (1969) and DeVries (1997). DeVries (1997) 
also gave new diagnoses and descriptions. Regarding 
the stratigraphic range of Chorus giganteus cited by 
DeVries (1997), its presence in the Tubul Formation 
implies a maximum age of very late Pliocene to early 
Pleistocene for these deposits. 


Family Nassariidae Iredale, 1916 
Genus Nassarius Duméril, 1806 


Nassarius taeniolatus (Philippi, 1845) 
(Figure 21) 


Buccinum taeniolatus Philippi, 1845: 69; Hupé, 1854: 
207, pl. 4, fig. 9. 

Nassarius taeniolatus (Philippi)—Keen, 1971: 609, fig. 
1313 


Material Examined: Three specimens: T/586 (2), 
T/1687. 
Measurements: 11 mm (T/1687). 


Occurrence: Recent: Papudo to Chonos Archipelago; 
Plio-Pleistocene: Arauco. 


Remarks: This species is the only representative of 
Nassarius in southern Chile, while two more species are 
found living off central and northern Chile (N. gayi 


(Kiener, 1835) and N. dentifer (Powys, 1835); see Mar- 
incovich, 1973) 


Family Volutidae Rafinesque, 1815 
Genus Adelomelon Dall, 1906 


Adelomelon ancilla (Ligthfoot, 1786) 


(Figure 22) 


Voluta ancilla Lightfoot, 1786: 84. 

Voluta gracilis Wood, 1528: 59, pl. 3, fig. 2. 

Cimbiola ancilla (Lightfoot)—Pace, 1902: 28, pl. 7, 
figs. 1-16 

Adelomelon (Adelomelon) ancilla 
1938: 1347, fig. 3816. 

Adelomelon ancilla (Lightfoot).—Dall, 1906; 143. 


(Lightfoot).—Wenz, 


Material Examined: One specimen: T/1227 


Measurements: Figured specimen (T/1227) 142 mm. 


Occurrence: Recent: Chiloé to Magellan Region; 
Plio-Pleistocene: Arauco to ?Isla Guafo. 


Remarks: It is unclear if A. reconditus pecan 
1997, is a different species or a juvenile A. ancilla. 
large specimen of presumably A. reconditus was found 
on ‘Cua (coll. Nielsen), but preservation does not allow 
assignment to a species. The Chilean fossil species of 
Adclomelan were recently revised by Nielsen and Fras- 
sinetti (2007a). 


Family Turridae H. and A. Adams, 1853 

Genus Bela Leach in Gray, 1847 

Bela paesleri Strebel, 1905 

(Figure 23) 

Bela paessleri Strebel, 1905: 588, pl. 22, figs. 35, 35a—b. 
ae lia (Bela) paessleri (Strebel).—C sarcelles. 1950: 67. 


Material Examined: 
T/590 (7), T/1626. 


Measurements: 


14 specimens: T/589 (6, in part), 


Height 16.7 mm (T/1626). 

Occurrence: Recent: Valdivia to Magellan Region; 

Plio-Pleistocene; Arauco, 

Remarks: This species is in need of revision, but be- 

cause the protoconch is not preserved in these speci- 

mens, it is not possible to verify its generic position. 
Family Scaphandridae Montfort, 1510 

Genus Scaphander Montfort, 1510 


Scaphander interruptus Dall, 1889 

(Figure 24) 

Scaphander interruptus Dall, 1889, 12; 297, pl 12, 
fig. 12. 

Material Examined: Six specimens: T/244, T/583, 

T/584 (2), T/975, one specimen without number. 


Measurements: 
18.5 mm. 


Figured specimen without number, 


Occurrence: Recent: Panama to Magellan Region; 
Plio-Pleistocene: Arauco. 


Remarks: The Recent and fossil Chilean Cephalaspi- 
dea are in need of revision. Scaphander cosmophilus 
(Sowerby, 1546) is known from the Pliocene of Isla 
Guafo (Frassinetti, 2000) and possibly from Isla Guam- 
blin (Frassinetti and Covacevich, 1995). Scaphander cos- 
mophilus was placed in the Genus Kaitoa by Griffin and 
Nielsen (2008). It is not clear if the two species are 
closely Aen 


Class Scaphopoda Bronn, 1562 
Family Dentaliidae Gray, 1847 
Genus Dentalium Linnaeus, 1758 


Dentalium sp. 
(Figure 25) 


S. N. Nielsen and C. Valdovinos, 2008 


Material Examined: Two specimens: T/1937-1938. 


Measurements: 
2mm. 


Figured specimen (T/1937) length 


Remarks: Dentaliidae are well represented in Chile at 
least since the Miocene (see e.g., Philippi, 1SS87; Herm, 
1969). A review of the group has never been under- 
taken. 


ASSOCIATED MACROFAUNA 


Several taxa outside the mollusks are known from the 
Tubul Formation. The Bir6 collection contains about 
225 specimens of barnacles, remains of two decapod 
crabs, bryozoans, about 260 specimens of the brachio- 
pod Mage llania venosa Solander, 1786, two different 
echinoids and some shark teeth. 

Some of the fauna represented in the collection does 
not come from the Tubul Formation but from the Mio- 
cene Ranquil Formation, most probably from the nearby 
locality Punta El Fraile (see Nielsen et al., 2004; Finger 
et al., 2007). Miocene representatives in the Tubul col- 
lection include the nautiloid cephalopod Aturia cubaen- 
sis (Lea, 1S41) and the olivid gastropod Lamprodomina 
dimidiata (Sowerby, 1846). 


COMPARISON WITH OTHER FAUNAS 
AND CONCLUSIONS 


U — the Tubul Formation is the late Miocene to 
early Pliocene Ranquil Formation (Garcia, 1968; Pineda, 
1986: Finger et al., 2007), the fauna of which is similar 
to that of the Navidad Formation south of Valparaiso 
(see Philippi, 187; Groves and Nielsen, 2003; Nielsen, 
2004, 2005; Nielsen and Frassinetti, 2003; Nielsen et al., 
2004). This fauna has been reworked and is of early to 
middle Miocene age (DeVries and Frassinetti, 2003; 
Finger et al., 2007) and contains many tropical to sub- 
tropical gastropod genera, such as Nerita, Strombus, 
Xenophora, Distorsio, Echinophoria, Ficus, Terebra, 
and Architectonica (Philippi. 1887; Covacevich and 
Frassinetti, 1980: Nielsen, 2005; Nielsen and DeVries, 
2002; Nielsen and Frassinetti, 2007b), that disappeared 
from Chile during the late Miocene climate cooling. 
A number of the gastropod genera described herein 
(Epitonium, Trophon, Chorus, Nassarius) were already 
present in Chile during the Miocene, but were repre- 
sented by different species. If and how these are related 
to the living taxa has yet to be investigated. 

The Pliocene to Pleistocene faunas from northem Chile 
have a different composition than that of the Tubul For- 
mation, containing abundant Trochoidea, Fissurella, Tur 
ritella, Crucibulum, Trochita, Crepidula, different species 
of Naticidae (including Sinum cymba), Argobuccinum, 
Nassarius, Acanthina, Chorus, and Oliva peruviana 
(Herm, 1969: DeVries, 1997, 2003; own data), represent- 
ing today’s coastal fauna of northem to central Chile. 
However, these are nearshore faunas with many taxa from 


Page 21] 


rocky coasts, while that of Tubul is a soft-bottom fauna 
from slightly deeper water, missing typical rocky-shore 
elements such as Fissurella, hosnihiana: and Concholepas, 
which are present in the Plio-Pleistocene deposits of 
northem Chile and in the modern fauna along the whole 
Chilean coast (McLean, 1984; DeVries, 1995, 2003). 

The Tubul fauna has previously been compared to 
faunas coming from Guamblin and Guafo islands farther 
south (Frassinetti, 1997, 2000: Frassinetti and Covace- 
vich, 1995; Table 1). The fauna from Guamblin is close 
to the Tubul fauna and Frassinetti and Covacevich 
(1995) and Frassinetti (2000), comparing both faunas 
directly, cited new Tubul species (Epitonium biroi, Tro- 
phon covacevichi, and Hindsiclava ignorata), and sug- 
gested that Epitonium biroi may be conspecific with 
E. (Nitidiscala) magellanicus, Trophon covacevichi: may 
be conspecific with Trophon geversianus, and Hindsi- 
clava ignorata may be conspecific with Bela paesleri. 
However, to confirm these synonymies more material 
should be directly compared. 

The fauna described by Frassinetti (1997, 2000) from 
Guatfo contains a mixture of species also present at 
Guamblin island and Tubul and species known from 
older deposits, e.g., Panopea chiloensis, Incatella chilen- 
sis, Echinophoria sp. (cited as Semicassis sp.), Chorus 
doliaris, Penion spp., and Gemmula subaequalis. It seems 
likely that these taxa represent a fauna intermediate 
in age between the older faunas and the Tubul-Guam- 
blin fauna but it is possible that specimens of the 
older fauna have been reworked from older beds and 
incorporated into younger sediments as has been 
demonstrated for the Navidad, Ranquil and Lacui for- 
mations of Mio-Pliocene age (Finger et al., 2007). The 
presence of Chorus doliaris and Incatella chilensis 
suggests a latest Miocene age for the Guafo fauna 
(DeVries, 1997, 2007). 

The first appeareance of Chorus giganteus in south- 
erm Peru during the latest Pliocene (DeVries, 1997) 
makes this a likely maximum age for the Tubul fauna. 
Indeed, this agrees well with strontium isotope stratigra- 
phy data from calcitic shells of Zygochlamys patagonica 
coming from the same locality, which confirm an early 
Pleistocene age (unpublished data). The two apparently 
extinct species present, Euspira guamblinensis and Sas- 
sia leucostomoides, may well be found living off Chile in 
the future. Comparison with living ranges of the species 
found in the Tubul Formation ( Figure 1) permits one to 
identify two regions with 16 species in common, The 
fauna of the Tubul Formation resembles most closely 
the present-day faunas of northern Chiloé island and the 
Magellan Region, whereas northern faunas of Pliocene 
to Pleistocene age have a composition similar to that of 
modern northern and central Chile. Water temperatures 
for the Arauco area are accordingly interpreted to have 
been colder than today. It is pre ssently not pe ‘ the 
more northern species (Figure 1, numbers 6, 15, 16 
previously tolerated cooler waters or if the cabo ‘mm spe- 
cies (Figure 1, numbers 7, 11, 14) tolerated warmer 
waters. However, it is possible that an additional factor 


THE NAUTILUS, Vol. 122, No. 4 


Table 1. Comparison of mollusk species found at Tubul, 


Guafo and Guamblin. 


Tubul Guafo Guamblin 
Bivalvia 
Ennucula grayi x ? (barrosi) 
Tindariopsis x x 
sulculata 
Malletia chilensis x ? (sp.) 
Mytilus sp. x 
Zygochlamys x ? (cf. hupeanus)  (tenuicostatus ) 
patagonica 
Lucinoma sp. x 
Cyclocardia x x (volckmanni) (volckmanni) 
velutinus 
Macoma inornata (tubulensis) 
Darina solenoides x (Mactra sp.) 
Ensis macha x x (sp.) 
Retrotapes x x (araucana) x (araucana) 
exalbidus 
Pandora cistula (sp.) 
Panopea x 
chiloensis 
Gastropoda 
Epitonium x ? (biroi) ? (biroi) 
magellanicus 
Turritella 
chilensis 
Euspira x x ; 
guamblinensis 
Echinophoria sp. x (Semicassis) 
Fusitriton x 
magellanicus 
Sassia x 
leucostomoides 
Cymatium 
remotum 
Trophon x ? (covacevichi) — ? (sp.) 
geversianus 
Trophon parcus x 
ee nopsis” cf. x ? (huilliche) 


X.” dispar 
Xymenopsis hero 
Chorus giganteus 
Chorus doliaris 
Nassarius 
taeniolatus 
Nassarius gayi 
Phos chilensis 
Penion spp 
Adelomelon : ? (reconditus ) 
ancilla 
Bela paesleri 
Hindsiclava 
ignorata 
Cryptogemma 
senex 
Gemmula 
subaequalis 
Scaphander x ? (cosmophilus ) 
interruptus 


Scaphopoda 
Dentalium sp : ? (sp 


? (cosmophilus) 


independent of water temperature is responsible 
for these widely separated species triplets today in con- 
trast to their common occurrence during the early 
Pleistocene. 


ACKNOWLEDGMENTS 


First of all we express our gratitude to Profesor Lajos 
Biro-Bagoczky (1929-1993) who, together with others, 
assembled this collection during more than 15 years. 
Arturo Quinzio (Departamento Ciencias de la Tierra, 
Universidad de Concepcion, Chile) made this work 
possible by giving access to this collection. Klaus Bandel 
(Geolc gisch-P Paliontologisches Institut, Universitit 
Hamburg, Germany) is Hhanlked for discussion on the 
Tubul Formation and fauna. Daniel Melnick (GFZ 
Potsdam) helped with the DEM map for the distribution 
figure. Comments and suggestions by Tom DeVries 
(Burton, USA) improved the manuscript. Work of SNN 
was supported financially by Deutsche Forschungsge- 
meinschaft grant Ni 699/4-1 and 4-2 


LITERATURE CITED 


Adams, H. and A. Adams. 1853-1858. The Genera of Recent 
Mollusca; Arranged Becoraine to Their Organization. 
Vol. 1, 484 p.; Vol. 2, 661 p.; Vol. 3, 138 pls., John 
van Voorst, oa 

Bellardi, L. 1872. I molluschi dei terreni terziari del Piemonte 
e della Liguria. Part 1: Cephalopoda, Pteropoda, Hetero- 
poda, Gasteropoda (Muricidae et Tritoniidae). Loescher, 
Torino, 264 pp., 15 pls. 

Berry, S. S. 1910. Report on a collection of shells from Peru, 
with a summary of littoral marine Mollusca of the Peru- 
vian zoological province. By William Healy Dall, 1909, 
Proceedings of the USNM_ 37, pp. 147-294, pls. 20-28. 
The Naudhus 2 23: 130-132. 

Beu, A. G. 1955. Pleistocene Chlamys patagonica delicatula 
(Bivalvia: Pectinidae) off southeastern Tasmania and his- 
tory of its species gr oup in the Southern Ocean. Depart- 
ment of Mines and Energy, South Australia, Special 
Publication 5: 1-11. 

Blainville, H. M. D. de 1814. Sur la classification méethodique 
des animaux mollusques, et établissment dune nouvelle 
considération pour y parvenir. Bulletin des sciences, par 
la Société philomatiques de Paris 2: 175-180. 

Biro-Bagéezky, L. 1979. Contribucion al conocimiento de la 
Formacion Tubul, Plioceno Superior, Provincia de Arauco 
(37° 14’ Lat. Sur). Congreso Geologico Chileno, No. 2, 
Actas 3: H33—H44. 

Bronn, H. G. 1862-1866. Die Klassen und Ordnungen der 
Weichtiere (Malacozoa), wissenschaftlich dargestellt in 
Wort und Bild. Band 3, Abt. 1: Kopflose Weichtiere 
(Malacozoa Acephalo). Band 3, Abt. 2: Kopftragende 
Weichtiere (Malacozoa Cephalophora), Winter, Leipzig, 
1499 pp., 136 pls. 

Briiggen, J. 1950. Fundamentos de la geologia de Chile. Insti- 
tuto geografico militar de Chile, Santiago, 374 pp. 

Bruguiére, J. G. 1791-1816. Tableau E neyelopé ‘dique et Méth- 
odique des trois régnes de la Nature... Vers, ‘Coauilles et 
Polypiers. Paris, 3 vols. 


S. N. Nielsen and C. Valdovinos, 2008 


Page 213 


Carcelles, A. R. 1944. Catalogo de los moluscos marinos de 
Puerto Quequén. Revista del Museo de La Plata. Nueva 
Serie, Zoologia 3: 233-309. 

Carcelles, A. R. 1950. Catalogo de los moluscos marinos de la 
Patagonia. Anales del Museo Nahuel Huapi 2: 41-100. 

Carcelles, A. R. 1954. Observaciones sobre los géneros 
Nucella, Chorus y Concholepas. Comunicaciones del 
Instituto Nacional de Investigaciones de las Ciencias Nat- 
urales, Museo Argentino de Ciencias Naturales “Bernar- 
dino Rivadavia” (Ciencias Zoologicas) 2: 255-275. 

Carcelles, A. R. and S. I. Williamson. 1951. Catalogo de 
los moluscos marinos de la Provincia Magallanica. Revista 
del Instituto Nacional de Investigaciones de las Ciencias 
Naturales, Museo Argentino de Ciencias Naturales 
“Bernardino Rivadavia” (Ciencias Zoologicas) 2: 225-383. 

Chemnitz, J. H. 1788. Neues systematisches Conchylien-Cabi- 
net, fortgesetzt durch J. H. Chemnitz. ...Vol. 10, Ras- 
pische Buchhandhing, Niirnberg, 376 pp., pls. 137-173. 

Chemnitz, J. H. 1795. Neues systematisches oo Cabi- 
net, fortgesetzt durch J. H. Chemnitz. ...Vol. 11, Ras- 
pische Buchhandlung, Niirmberg, 310 pp., pls. eas 

Cemohorsky, W. O. 1977. The taxonomy of some Southern 
Ocean Mollusca (Gastropoda) ) mainly Antarctic and Sub- 
antarctic. Records of the Auckland Institute and Museum 
14: 105-119. 

Conrad, T. A. 1867. Description of a new genus of Astartidae. 
American Journal of Conchology 3: 191. 

Cossmann, M. 1903. Essais de paléoconchologie comparee, 
vol. 5. M. Cossmann, Paris, 215 pp. 

Covacevich, V. and D. Frassinetti. 1980. El género Ficus en el 
Mioceno de Chile Central con descripcion de F. gayana 
sp. nov. Gastropoda: Ficidae. Boletin del Museo Nacional 
de Historia Natural, Chile 37: 281-294. 

Covacevich, V. and D. Frassinetti. 1990. La fauna de Lo 
Abarca: hito bioestratigrafico y paleoclimatico en el 
Terciario Superior marino de Chile Central. Se soundo 
Simposio sobre el Terciario de Chile, Actas: 51-71. 

Cuvier, G. 1795. Second mémoire sur lorganisation et les rap- 
ports des animaux a sang blanc, dans lequel on traite de la 
structure des Mollusques et de leur division en ordres, lu 
ala Société dhistoire naturelle de Paris, le 11 Prairial, an 
Il. Magazin Encyclopédique, ou Journal des Sciences, 
des Lettres et des Arts 2: 433-449. 

Dijkstra. H. H. and B. A. Marshall. 2008. The Recent Pecti- 
noidea of the New Zealand region (Mollusca: Bivalvia: 
Propeamussiidae, Pectinidae and Spondylidae). Mollus- 
can Research 28: 1-SS. 

Dall, W. H. 1889. Preliminary report on the collection of Mol- 
lusca and Brachiopoda obtained in 1887-88. Scientific 
results of explorations by the U.S. Fish Commission 
Steamer “Albatross”, no. 7. Proceedings of the United 
States National Museum 12: 219-362. 

Dall. W. H. 1906. Notes on some names in the Volutidae. The 
Nautilus 19: 143-144. 

Dall, W. H. 1908. The Mollusca and the Brachiopoda. Bulletin 
of the Museum of Comparative Zoology Harvard 43(6): 
205-487. 

Dall. W. H. 1915. A review of some bivalve shells of the group 
Anatinacea from the west coast of America. Proceedings 
of the United States National Museum 49: 441-456. 

deBoury. E. A. 1909. Catalogue des sous-genres des Scaladea 
Journal de Conchyliologie 57: 256-258. 

del Rio, C. J. 1997. Cenozoic biogeographic history of the 
eurythermal genus Retrotapes, new genus (subfamily 


Tapetinae) from southern South America and Antarctica. 
oe Nautilus 110: 77-93. 

Dell, R. K. 1964. Antarctic and subantarctic Mollusca: Amphi- 
neura, nae eee and Bivalvia. Discovery Reports 33: 
93-250. 

Dell, R. K. 1971. The marine Mollusca of the Royal Society 
Expedition to Southern Chile, 1955-1959. Records of the 
Dominion Museum 7: 155-233. 

DeVries, T. J. 1995. Concholepas Lamarck, 1801 (Neogastro- 
poda: Muricoidea): a Neogene genus native to “South 
America. The Veliger 38: 284-297, 

DeVries, T. J. 1997. A review of the genus Chorus Gray, 
1847 (Gastropoda: Muricidae) from western South Amer- 
ica. Tulane Studies in Geology and Paleontology 30: 
125-145. 

DeVries, T. J. 2003. Acanthina Fischer von Waldheim, 1807 
(Gastropoda: Muricidae), an ocenebrine genus endemic 
to South America. The Veliger 46; 332-350. 

DeVries, T. J. 2007, Cenozoic Turritellidae (C Vanegas ) from 
southern Peru. Journal of Paleontology 81: 331-351. 
DeVries, T. J. and D. Frassinetti. 2003. Range extensions and 
biogeographic implications of Chilean Neogene mollusks 
found in Peru. Boletin del Museo Nacional de Historia 

Natural, Chile 52; 119-135. 

Dillwyn, L. W. 1817. A Descriptive Catalogue of Recent Shells 
2. J. & A. Arch, London. 

a Orbigny, A. D. 1831-1847. Voyage dans TAmeérique Méridio- 
nale. Vol. 5, part 3, Mollusques. Chez P. Bertrand, Paris, 
758 pp. + Atlas $5 pls. 

Dumeril, A. M. C. 1806. Zoologie Analytique, ou méthode 
eee de classification des animaux. Paris, 344 pp. 
Feruglio, 1949. Descripeion geologica de la Patagonia. 
Direccién General de Yacimientos Petroliferos Fiseales 

(Buenos Aires) 2: 549 pp. 

Finger, Kk. L., S. N. Nielsen, T. J. DeVries, A. Encinas, and 
D. E. Peterson. 2007. Paleontologic evidence for sedi- 
mentary displacement in Neogene foreare basins of cen- 
tral Chile. Palaios 22: 3-16. 

Fleming, J. 1828. A History of British Animals, exhibiting the 
descriptive characters and systematic arrangement etc. 
Bell and Bradfute, Edinburgh, 565 pp. 

Forcelli, D. 2000. Moluscos magall4nicos, Guia de moluscos 
de Patagonia y Sur de Chile. Vasquez Mazzini Editores, 
Buenos Aires; 200 PP: 

Frassinetti, D. 1974. El género Venus en la coleccion de fosiles 
terciarios y cuartarios de R. A. Philippi (1887). Boletin del 
Museo Nacional de Historia Natural, Chile 33; 43-51. 

Frassinetti, D. 1997. Molluscos del Plioceno Superior de Isla 
Guato, sur de Chile, Parte I. Bivalvia. Boletin del Museo 
Nacional de Historia Natural, Chile 46: 55-79. 

Frassinetti, D. 2000. Moluscos del Plioceno Superior marino de 
Isla Guafo, sur de Chile. Parte Il. Gastropoda. Boletin del 
Museo Nacional de Historia Natural, Chile 49: 131-161 

Frassinetti, D. and V. Covacevich. 1995. Moluscos del Plioceno 
Superior marino de Isla Guamblin, Archipielago de los 
Chonos, sur de Chile. Revista Geol6gica de Chile 22: 47-73 

Garcia, F. 1968. Estratigrafia del Terciario de Chile Central 
In: Cecioni C. G. (Ed.) El Terciario de Chile, Zona cen- 
tral. Editorial Andres Bello, Santiago pp. 25-58. 

Gmelin, J. F. 1791. Caroli a Linné Systema Naturae per Regna 
Tria Naturae, Secundum Classes, Ordines, Genera, Species, 
cum characteribus, differentiis, synonymis, locis. Editio 
decima tertia, aucta, reformata, cura J. F. Gmelin. Tomus I, 
Pars 6, Vermes testacea. G.E. Beer, Leipzig, pp. 30214120 


Page 214 


THE NAUTILUS, Vol. 122, No. 4 


Gould, A. A. 1850. Descriptions of shells from the United 
States Exploring Expedition. Proceedings of the Boston 
Society of Natural History 3: 343-348. 

Gould, A. A. 1852. Mollusea and Shells. United States Explor- 
ing Expedition during the Years 1835-1542 Under the 
Command of Charles Wilkes, U. S. N. Vol. 12. Gould 
and Lincoln, Boston, xv + 510 pp., 52 pls. 

Gray, J. E. 1824. Shells. Supplement to Appendix, Parry’s Voy- 
age for the discovery of north-west passage in the years 
1819-1820. Appendix 10, Zoology: 240-246. 

Gray, J. E. 1837. A synoptical catalogue of the species of cer- 
tain tribes or genera of shells contained in the collection 
of the British Museum and the author's cabinet; with 
descriptions of new species. The Magazine of Natural 
History, and Journal of Zoology, Botany, Geology and 
Mineralogy, new series 1; 370-376. 

Gray, J. E. 1847. A list of the genera of recent Mollusca, their 
synonyms and types. Proceedings of the Zoological 
Society of London 15: 129-219. 

Gray, J. E. 1853. A revision of the genera of some of the 
families of Conchifera or bivalve shells. Annals and Maga- 
zine of Natural History, ser. 2, 11: 33-4. 

Gray, J. E. 1854. On the division of ctenobranchous gasteropo- 
dous Mollusca into larger groups and families. Proceed- 
ings of the Zoological Society of London 21: 32-4. 

Griffin, M. and S. N. Nielsen. 2008. A revision of the type 
specimens of Tertiary clears from Chile and Argentina 
described by d Orbigny 1842), Sowerby (1846), and 
Hupe (1854), Journal of Systematic Pale ontology 6: 
251-316. 

Groves, L. T. and S. N. Nielsen. 2003. A new Late Miocene 
Zonaria (Gastropoda: Cypraeidae) from Central Chile. 
The Veliger 46: 351-354. 

Guilding, L. 1834. Observations on Naticina and Dentalium, 
two genera of molluscous animals. Transactions of the 
Linnean Society of London 17; 29-35, pl. 3. 


Hanley, S. 1842- 1856. An illustrated and descriptive catalogue of 


recent bivalve shells. Pp. 1-392, suppl. pls. 9-24, pp. 1-24. 

Herm, D. 1969. Marines Plioziin und Pleistoziin in Nord- und 
Mittel-Chile unter besonderer Beriicksichtigung der 
Entwicklung der Mollusken-Faunen. Zitteliana 2: 1-159. 

Hupé, H. 1854. Malacologfa y conquiliologia. In: Gay, C. (ed), 
Historia fisica y politic i de Chile, Zoologia 8, p. 5-355, 
Atlas 2, Conquiliologia: 1-6, Malacologia: 1-8, E. Thunot 
& Co., Lesauvage, Paris. 

Ihering, H. von 1907. Les mollusques fossiles du Tertiaire et 
du Crétacé Supérieur de l’Argentine. Anales del Museo 
Nacional de Buenos Aires, 3(7): 611 pp. 

fredale, T. 1916. On two editions of Dumeril’s Zoologie Analy- 


tique. Proceedings of the Malacological Society of 
| g o . 


London 12: 79-84. 

Iredale, T. 1931. Australian molhiscan notes. No. 1. Records of 
the Australian Museum 18: 210-235. 

Jonkers, H. A, 2003. Late Cenozoic-Recent Pectinidae (Mol- 
lusca: Bivalvia) of the Southern Ocean and neighbouring 
regions. Monographs of marine Mollusca 5: 1-125, 

Keen, A. M. 1971. Sea Shells of Tropical West America. Stan- 
ford University Press, Stanford, 1064 pp 

Kiener, L. C. 1835. Spécies Geneéral et Iconographie des 
Coquilles Vivantes Comprenant la Collection du Museum 
d'Histoire Naturelle de Paris, la Collection de Lamarck, 
celle de Massena (Appartenant Maintenant M. le Baron 
3. Delessert), et les Découvertes Recentes des Voyageurs. 
Genre Buccinum. Rousseau Libraire, Paris, 5] pp. 


King, P. P. and W. J. Broderip. 1832. Description of the Cirri- 
pedia, Conchifera and Mollusca in a collection formed by 
the officers of HMS Adventure and Beagle employed 
between the years 1826 and 1830 in surveying the south- 
ern coast of South America including the Strait of Magel- 
lans and the coast of Tierra del Fuego. Zoological Journal 
5: 332-349. : 

Lamarck, J. B. P. A. 1805. Mémoires sur les fossiles des envi- 
rons de Paris, comprenant la détermination des espéces 
qui appartiennent aux animaux marins sans vertebres et 
dont la plupart sont figurés dans la collection des vélins 
du Muséum. Extraits des Annales du Muséum National 
(Histoire Naturelle. Volume VI (1805): 23e mémoire, 
Genres Pinna, Mytilus, Mediola, Nucula, 117-126. 

Lamarck, J. B. P. A. de M. 1809. Philosophie zoologique, ou, 
Exposition des considerations relative a histoire naturelle 
des animaux. Dentu and the author, Paris, 309 pp. 

Lamarck, J. B. P. A. de M. 1811. Suite de la détermination des 
especes de Mollusques Testacés, Voluta. Annales du Mu- 
séum National d'Histoire Naturelle 17; 54—S0. 

Lamarck, J. B. P. A. de M. 1816. Mollusques et polypes divers. 
Part 23, oe pls. 3891-488. In J. G. Bruguiere, J. P. B. 
A. Lamarck, and G. P. Deshayes, Encyclopédie Meéthodi- 
que. Paris. 

Lamarck, J. B. P. A. de M. 1819. Histoire naturelle des animaux 
sans vertebres, presentant les charactéres geneéraux ete) 
Volume 6. Chez Pauteur, au Jardin du Roi, Paris, 343 pp. 

Le Roux, J. P., C. Gomez, J. Fenner, and H. Middleton. 2004. 
Sedimentological processes in a scarp-controlled rocky 
shoreline to upper continental slope environment, as 
revealed by unusual sedimentary fe Bee in the Neogene 
Cc oquimbo Formation, north-central Chile. Sedimentary 
Geology 165: 67-92. 

Le Roux, J. P., C. Gomez, C. Venegas, J. Fenner, H. Middle- 
ton, M. Marchant, B. Buchbinder, D. Frassinetti, C. Mar- 
quardt, kK. M, Gregory-Wodzicki, and A. Lavenu. 2005. 
Neogene-Quaternary coastal and offshore sedimentation 
in north-central Chile: record of sea-level changes and 
implications for Andean tectonism. Journal of South 
American Earth Sciences 19: 83-98 

Le Roux, J. P., D. M. Olivares, S. N. Nielsen, N. D. Smith, 
H. Middleton, J. Fenner, and S. E. Ishman. 2006. Mio- 
cene-Pliocene bay sedimentation as controlled by regional 
crustal behaviour, local tectonics and eustatic sea-level 
changes in the Coquimbo Formation at the Bay of Ton- 
goy, conal't Jhile. Sedimentary Geology 184: 133-153. 

Lea, I. 1841. Notice of the Oolitic Formation in America, with 
descriptions of some of its organic remains. Transactions 
of the American Philosophical Society new series 7: 
251-260. 

Leach, W. E. 1819. A list of invertebrate animals discovered by 
His Majestys Ship Isabella, in a voyage to the Arctic 
Regions. In Ross, J., Voyage of discovery . exploring 
Baffin’s B: wy, Appendix 2: 61-64. 

Lesson, R. P. 1830. Mollusques. In Duperrey, M. L. 1. (ed.) 
Voyage autour du Monde, Exécuté par Ordre du oe sur 
la Corvette de Sa Majesté, La Coquille, pendant les 
Années 1822, 1823, 1824 et 1825. Zoologie 2(1): 239- 
448 (1830): Atlas: pls. 1-16, (1831). Paris. 

Lightfoot, J. R. 1786. A catalogue of the Portland Museum, 
lately the property of the Duchess Dowager of Portland, 
deceased; which will be sold by auction, by Mr. Skinner 
and co. on Monday the 24th of April, 1786. Skinner & 

o., London, viii + 194 pp. 


S. N. Nielsen and C. Valdovinos, 2008 


Page 215 


Linnaeus, C. 175S. Systema naturae per regna tria naturae, 
secundum classes, ordines, genera, species, cum charac- 
teribus, differentiis, synonymis, locis. Tomus I. Editio 
decima, reformata. Holmiae, Impensis Direct. Laurentii 
Salvii, $24 pp. 

Mabille, J. and A. T. Rochebrune. 1SS9. aaa ie Mission 
Scientifique du Cap Horn, 1882-3, Zoologie 2: 1-143. 

Marincovich, L. Jr. 1973. Intertidal mollusks of f Iquique, Chile. 
Bulletin of the Natural History Museum of Los Angeles 
County 16: 149. 

Martinez, R. 1976. Hallazgo de Sphaeroidinella dehiscens 
dehiscens (Parker and Jones) en el Plioceno de Arauco: 
su significado para la reinterpretacion del Ne6geno Supe- 
rior en Chile. Congreso Geologico Chileno, No.1, Actas 1: 
C125-C142. 

Martinez, R. and R. Osorio. 1968. Foraminiteros pliocénicos 
de Chile Central, I: edad y paleoecologia de la Forma- 
cidn Tubul. In: Cecioni, C. G. (ed.), El Terciario de Chile, 
Zona central. Editorial Andres Bello, Santiago: 155-165. 

McLean, J. H. 1984. Systematics of Fissurella in the peruvian 
and magellanic faunal provinces (Gastropoda: Prosobran- 
chia). Natural History Museum of Los Angeles County 
Contributions in Science 354: 70 pp. 

Molina, G. I. 1782. Saggio sulla Storia Naturale del Chili. 
T. d Aquino, Bologna, 367 pp. 

Montfort, P. D. D. 1810. Conchyliologie systématique, et clas- 
sification méthodique des coquilles; offrant leurs figures, 
leur arrangement generique, leurs descriptions caracteris- 
tiques, leurs noms; ainsi que leur synonymie en plusieurs 
langues. F. Schoell, Paris, 2: 676 pp. 

Moulins, M. Des 1832. Description d'un nouveau genre de 
coquille vivante, bivalve, des mers du Chili. Actes de la 
Société Linnéenne de Bordeaux 5: 83-92. 

Moricke, W. 1896. Versteinerungen der Tertiiirformation von 
Chile. Neues Jahrbuch fiir Mineralogie, Geologie und 
Palaontologie, Beilage Band 10: 548-612. 

Nielsen. S. N. 2004. The genus Olivancillaria (Gastropoda, 
Olividae) in the Miocene of Chile: rediscovery of a senior 
synonym and description of a new species. The Nautilus 
118: $S—92. 

Nielsen, S. N. 2005. ea Strombidae, Aporrhaidae, and 
Struthiolariidae (Gastropoda, Stromboidea) from Chile: 
their significance to biogeography of faunas and climate 
of the south-east Pacific. Journal of Paleontology 79: 
1120-1130. 

Nielsen, S. N. and T. J. DeVries. 2002. Tertiary Xenophoridae 
(Gastropoda) of western South America. The Nautilus 
116: 71-78 

Nielsen. S. N. and D. Frassinetti. 2003. New and little known 
species of Pseudolividae (Gastropoda) from the Tertiary 
of Chile. The Nautilus 117: 91-96. 

Nielsen S. N. and D. Frassinetti. 2007a. The Neogene Voluti- 
dae (Gastropoda, Psi Soa ) from the Pacific coast 
of Chile. Journal of Paleontology $1: 82-102. 

Nielsen S. N. and D. Frassinetti. 2007b. The Miocene Archi- 
tectonicidae (Gastropoda) of Chile. Pal&éontologische 
Zeitschrift 81: 291-303. 

Nielsen. S. N., D. Frassinetti, and K. Bandel. 2004. Miocene 
Vetigastropoda and Neritimorpha (Mollusca, Gastropoda) 
of Central Chile. Journal of South American Earth 
Sciences 17: 73-88. 

Pace. S. 1902. On the anatomy and relationships of Voluta 
musica, Linn.; with notes upon certain other supposed 


members of the Volutidae. Proceedings of the Malacolog- 
ical Society of London 5: 21-31. 

Pallas, P. S. 1774. Spicilegia Zoologica quibus novae imprimis et 
obscurae animalium species iconibus, descriptionibus atque 
commentariis illustrantur. Berolini, 1 part 10, 41 pp., 44 pls. 

Pastorino, G, 2005a. Recent Naticidae (Mollusca: Gastropoda) 
from the Patagonian Coast. The Veliger 47; 225-258. 

Pastorino, G. 2005b. A revision of the genus Trophon Mon- 
tfort, SLO (Gastropoda: Muricidae) from southern South 
America. The Nautilus 119: 55-82. 

Pastorino, G. and M. G. Harasewych. 2000. A revision of the 
Patagonian genus Xymenopsis Powell, 1951 (Gastropoda: 
Muricidae). The Nautilus 114; 38-58. 

Philippi, R. A. 1845. Diagnosen einiger neuer Conchylien. 
Archiv fiir Naturgeschichte 11; 50-71. 

Philippi, R. A. 1887, Die Tertiéren und Quartiiren Versteiner- 
ungen Chiles. Brockhaus, Leipzig, 266 pp. 

Pineda, V. 1983. Evolucion paleogeogratica de la peninsula de 
Arauco durante el Cretacico superior Terciario. Memoria 
de Titulo (Unpublished), Universidad de Chile, Departa- 
mento de Geologia, Santiago, 268 pp. 

Pineda, V. 1986. Evolucion paleogeogratica de la cuenca sedi- 
mentaria Cretacica - Terciaria de ‘eon In: Frutos, J., 
R. Oyarztin and M. Pincheira (eds.), Geologia y Recursos 
naturales de Chile. Editorial de la Universidad de Con- 
cepcion, Concepcion, Vol. 1: 376-390. 

Powell, A. W. B. 1951. Antarctic and subantarctic Mollusca: 
Pelecypoda and Gastropoda. Discovery Report 26: 47-79. 

Powys, W. L. 1835. Characters and observations on new 
genera and species of Mollusca and Conchifera collected 
by Mr. Cuming. Proceedings of the Zoological Society 
London 1835: 93-96. 

Radic, J. P., P. Alvarez, and L. Rojas. 2005. Tectonostratigraphic 
evolution of the Arauco-Itata foreare basin, Central Chile. 
6th International Symposium on Andean Dynamics (ISAG 
2005, Barcelona), Extended Abstracts: 586-587. 

Rafinesque, C. S. 1815. Analyse de la Nature ou Tableau de 
lUnivers et des Corps Organisés. Palermo, 224 pp. 

Ramirez, |. 1987. Moluscos de Chile. Imprenta Museo Nacio- 
nal de Historia Natural, Chile. Vol. 2 (1987) Mesogastro- 
poda. 

Reid, D. G. and C. Osorio. 2000. The shallow-water marine 
Mollusca of the Estero Elefantes and Laguna San Rafael, 
southern Chile. Bulletin of the natural History Museum 
London (Zoology) 66; 109-146. 

Rochebrune, A. T. de and J. Mabille. 1889. Mollusques. In: 
Mission Scientifique du Cap Horn (1852-1583), Zoologie 
6(2): 1-126. 

Roding, P. F. 1798. Museum Boltenianum. Pars secunda con- 
tinens conchylia. J.C. Trappii, Hamburg, 199 pp. 

Schumacher, C. F. 1817. Essais d'un nouveau systeme des 
habitations des vers testaces. Schultz, Copenhagen, 
287 pp. 

Smith, E, A. 1875, Descriptions of some new shells from Ker- 
guelen’s Island. The Annals and Magazine of Natural His- 
tory, series 4, 14: 67-73. 

Smith, E. A. 1881. Mollusca and molluscoidea of the zoological 
collections made during the survey of HMS “Alert” in 
the Strait of Magellan and on the coast of Patagonia 
Proceedings of the Zoological Society of London: 22—44 

Smith, E. A. 1905. On a small collection of Mollusca from 
Tierra del Fuego. Proceedings of the Malacological 
Society of London 6: 333-339 


Page 216 


THE NAUTILUS, Vol. 122, No. 4 


Smith, J. T. 1970. Taxonomy, distribution and phylo- 
geny of the cymatiid gastropods Argobuccinum, Mediargo 
and Priene. Bulletins of American “Paleontology 56(254): 
441-573. 

Solander, D. C. 1786. A Catalogue of the Portland Museum. 
London, 194 pp. 

Soot-Ryen, T. 1959. Reports of the Lund University Chile 
Expedition. 1948-49. 35. Pelecypoda. Lunds Universitets 
Arsskrift. N.F. Avd. 2, 55(6):; 1-86. 

Sowerby, G. B. 1833. Characters of new species of Mollusca 


and Conchifera collected by Mr. Cuming. Proceedings of 


the gore cts Society of London 2(25): 194-202. 

Sowerby, G. B. 1839. Molluscous animals and their shells, by 
J. E. GC ray, continued by G. B. Sowerby. In: Beechey, 
F. W. (ed.) The zoology of C apt. Beechey’ 5 voyage to the 
Pacific and Behring’s Straits, London: 103-155. 

Sowe rby, G » B.;. I. 1846, Appendix. Descriptions of Tertiary 
fossil shells from South America. In: C, Darwin, Geologi- 
cal observations on South America. Smith, Elder & Co., 

London: 249-268. 

Sowerby, J. 1S12-1846. The mineral conchology of Great Brit- 
ain; or coloured figures and descriptions of those remains 
of testaceous smimnals or shells, which have been pre- 
served at various times and depths in the earth. Volumes 
1-7. Sowerby, London. 

Strebel, H. 1905. Beitriige zur Kenntnis der Molluskenfauna 
der Magalhaen-Provinz. No. 3. Zoologische Jahrbiicher. 


Abteilung fiir Systematik, Geographie und Biologie der 
Tiere 22: 575-666. 

Tavera, J. and C. Veyl. 1958. Reconocimiento geolégico de la 
Isla Mocha. Anales de la Facultad de Ciencias Fisicas y 
Matemiaticas, 14-15:157-186. i 

Jaldovinos, C. 1999. Biodiversidad de moluscos Chilenos: 
base de datos taxonémica y distribucional. Gayana 63: 
111-164. 

Verrill, A. E. and K. J. Bush 1897. Revision of the genera of 
Ledidae and Nuculidae of the Atlantic Coast of the 
United States. American Journal of Science, serie 4, 3 
(153): 51-63. 

Villarroel, M. and J. Stuardo, 1998. Protobranchia (Mollusca: 
Bivalvia) chilenos recientes y algunos fosiles. Malacologia 
40; 113-229, 

Waller, T. R. 1991. Evolutionary relationships among commercial 
scallops (Mollusca: Bivé via: Pectinidae). In: Shumway, S.E. 
(ed.), Scallops: Biology, Ecology and Aquaculture. Elsevier, 
Amsterdam, 1-73. 

Wenz, W. 1935-1944. Gastropoda, Teil I. In: Schindewolf, 
O. H. (ed.), Handbuch der Paliiozoologie 6. Borntraeger, 
Berlin, 1639 pp. ; ; 

Wood, W. 1828. Index Testaceologicus: or a Catalogue of Shells: 
with the Latin and English names, references to figures 
and places where found, ... Supplement, ... (References 
from Lamarck’s “Animaux sans Vertebres,” adapted to the 
figures in the ... “Index”, ...). London. iv + 59 pp. 


THE NAUTILUS 122(4):217-227, 2008 


Radula morphology in veined rapa whelks, Rapana venosa 
(Valenciennes, 1846) (Gastropoda: Muricidae) from 


Chesapeake Bay, USA 


Juliana M. Harding" 

Stefanie M. Gera 

Roger Mann 

Department of Fisheries Science 
Virginia Institute of Marine Science 
Gloucester Point, Vi irginia 23062 USA 
jharding@vims.edu 


ABSTRACT 


Radula length, width, number of transverse rows of teeth, and 
rachidian tooth dimensions (central cusp height, central cusp 
base width, and rachidian tooth base width) were examined in 
relation to leined rapa whelk shell length. Radula length and 


width increase linearly with whelk shell length. The number of 


transverse rows of radular teeth increase with whelk shell 
length. Within an individual, central cusp height of the rachi- 
dian tooth increases with increasing distance from the anterior 
of the radula. Central cusp height of the rachidian tooth, an 
indicator of tooth wear or use, was least for teeth in rows 1 and 
11. Teeth in radular row 21 appear to be in a transition zone 
from high to low wear or use. Within a radula and within a size 
class. the ratio of central cusp base width to rachidian tooth 
base width does not change. Central cusp base width, central 
cusp height and rachidian tooth base width were significantly 
smaller in females than in males indicating sexual dimorphism 
in rachidian tooth shape for rapa whelks. Patterns of wear as 
indicated by central cusp base width to central cusp height ratio 
values were not significantly different between sexes and may 
serve as an indication that feeding strategies and/or prey may 
be similar between animals of different sex but similar size. 


Additional Keywords: Neogastropoda, rachidian teeth, allometry, 
ontogeny 


INTRODUCTION 


The radula is a chitinous ribbon-like series of nearly 
colorless transverse tooth rows resting atop the radula 
membrane (Wu, 1965; Radwin and Wells, 1968). Muri- 
cid gastropods use the anterior teeth when drilling holes 
in bivalve prey (Carriker, 1961, 1981; Fujioka, 1985). 

As anterior teeth are worn down, they are replaced by 
younger teeth that are formed in the radular sac and 
gradually moved forward along the radula (Isarankura 


; : 
* Author for correspondence 


and Runham, 1968; Carriker, 1981). Muricid radulae 
have between 100 and 500 transverse rows of teeth (e.g., 

Carriker, 1961; Radwin and Wells, 1968; Fujioka, 1985). 
Each transverse row of teeth consists of a central rachi- 
dian (R) tooth and two slender marginal teeth (M) in the 
tooth formation M + R + M (Carriker, 1969). The cen- 
tral rachidian tooth in each transverse row is responsible 
for most of the rasping and physical shell removal during 
drilling while the marginal teeth synchronously tear fle sh 
from prey (Carriker, 1969: Carriker et al., 1974; Krutak, 
1977). Thus, the rachidian teeth show more wear, or 
reduction in size with use, than marginal teeth found in 
the same transverse rows (Carriker et al., 1974). This 
trend is particularly evident at the anterior end of the 
radula where the rachidian cusps in the most anterior 
row(s) may be completely removed by use (Carriker, 
1969, 1974; Fujioka, 1985). 

Veined rapa whelks (Rapana venosa, Valenciennes 
1846, Muricidae) are predatory marine gastropods that. 
while originally native to Japanese and Korean waters 
(Tsi et ale 1983), have successfully invaded marine and 
estuarine habitats in the Black, Adriatic, Aegean, Medi- 
terranean (Mann et al., 2004), and North Seas (Vink 
et al., 2005) as well as the Rio de la Plata (Pastorino 
et al., 2000) and Chesapeake Bay, USA (Harding and 
Mann, 1999). At the present time, the Ches sapeake Bay 
rapa whelk population is the only known population of 
bs whelks in North America. 

Rapa whelks provide an unusual opportunity to 
investigate allometric changes in radula morphology 
across a wide size range of individuals mee they 
reach terminal shell lengths in excess of 170 mm (Wu, 
1988: Harding and Mann, 2005). Like other muricids 
(Paine, 1966), rapa whelks experience ontogenetic 
shifts in diet (Harding and Mann, 2001) as well as 
predation strategy and resulting predation signatures 
in prey valves ( Harding et al.. 2007). Small ( 
shell length) rapa whelks drill their prey (Harding and 
Mann, 2001; Harding et al., 2007) including barnacles 


35 mm 


Page 215 


THE NAUTILUS, Vol. 122, No. 4 


(Balanus sp., Chthamalus sp.), mussels (Mytilus sp., 
Geukensia demissa), soft shell clams (Mya arenaria), and 
oysters (Crassostrea virginica). At shell lengths above 
35 mm, rapa whelks eat larger bivalves (Harding and 
Mann, 2001) including oy sters and northern quahogs 
(Mercenaria mercenaria) and typically either edge bore 
their prey or leave no signatures (Morton, 1994: Harding 
et al., 2007). 

We quantitatively describe radula and rachidian tooth 
morphology for a size range of rapa whelks from Chesa- 
peake Bay, USA. Rachidian teeth in Rapana have a large 
central cusp flanked by two smaller cusps (Avalawa: 
1964: Wu, 1965). Shell length, the maximum dimension 
from the tip of the spire to the bottom of the siphonal 
canal, is used as the metric of whelk size. Shell length 
does not fluctuate with season or other factors. Relation- 
ships between shell length and radula dimensions are 
quantitatively désoribed: for male and female rapa 
whelks. Within each radula, rachidian tooth morphology 
is described along the length of the radula by measuring 
ratios of rachidian tooth central cusp base width to cen- 
tral cusp height and central cusp base width to rachidian 
tooth base width. The resulting ratios are compared 
between teeth along the length of an individual radula 
and across radulae from male and female whelks as well 
as from a size range of Chesapeake Bay rapa whelks. 


MATERIALS AND METHODS 


Rapa whelks with shell lengths (SL) in excess of 70 mm 
were obtained from the lower Chesapeake Bay, USA as 
donations to the Virginia Institute of Marine Science 
(VIMS) rapa whelk bounty program. Rapa whelks less 
than 70 mm SL were cultured at VIMS, Gloucester 
Point, Virginia to supplement the lower SL range of rapa 
whelks because individuals less than 70 mm SL were not 
available through the bounty program. 

At the time of whelk collection, SL was measured in 
mm and whelks were assigned to shell length classes 
that were established to evenly categorize the potential 
SL range (1-180 mm SL). Petite, small, medium, and 
large classifications corresponded to whelk SL ranges 
: less than 45 mm, 45.1-90 mm, 90.1-135 mm, and 

5.1-180 mm, respectively. 

“Whe Iks were frozen after collection and thawed to 
facilitate dissection and removal of radulae. Whelks 
were sexed during dissection and distinguished 7 male 
or female on the basis of penis length and gonad color 
alter Mann et al. (2006). For the purposes of discussion 
herein, true females (penis length = 0 mm, bright yellow 
gonad) and imposex females (penis length - -20) mm, 
bright ye low gonad) are groupe dl toge shes per Mann 
et al. (2006) 
the whelk probosc is. However, two control radulae were 
removed by soaking the proboscis for 24 hours in 10% 
sodium hydroxide to ensure that dissection removed the 
entire radula intact. Only intact radulae were used 
this study 


Typic: ally, radulae were dissected out of 


After removal from whelks, images of the complete 
radula were taken using a digital camera mounted on a 
dissecting microscope for measurement of total radula 
length, total anterior to posterior distance (mm) and rad- 
ula width, the maximum lateral distance across the bases 
of the marginal and rachidian teeth at the first transverse 
row of teeth (Figure 1). The odontophore was removed 
and then the rachidian teeth were systematically 
removed from every 10th transverse tooth row (e.g., 
Row 1, 11, 21 etc.) ) along the radulae moving from ante- 
rior to posterior (Figures 2 and 3). Tooth rows were 
removed with a size 10 scalpel blade for larger indi- 
viduals, and with sharpened needles for cultured 
individuals less than 66 mm SL. 

Digital images were taken of each individual rachidian 
tooth after removal with the tooth positioned convex 
side down. Typical magnification of individual teeth 
used for digital images ranged from 50 for whelks with 
SL greater than 147 mm to 90 for whelks less than 
45-50 mm SL. Measurements (mm, Figure 4) of the 
rachidian tooth central cusp base width (L1), maximum 
central cusp height (L2) measured from the tip of the 
central cusp to the midpoint of L1, and the maximum 
rachidian tooth base width (13) were made on the 
resulting images. The terminology used to describe 
tooth morphology follows that of oak! 1993). 


Data Analyses: Significance levels for all statistical 
tests were set at alpha = 0.05 a priori. Fishers multiple 
comparison tests were used for post hoc comparisons 
when appropriate. 


Radula Allometry and Gross Morphology: — Linear 
and power regression models were used to describe 
relationships within sexes between rapa whelk shell 
length and radula length, radula width, and number of 
rows of transverse teeth per radula. The relationship 
between radula length and both radula width and the 
number of transverse rows of teeth within radulae from 
males and females were also examined with both linear 
and power regressions. The slopes of all morphological 
regressions \were compared between sexes with t tests 
(per Zar, 1996) on raw data or on logarithm transformed 
data if the power model was deemed more appropriate 
than the linear model. 


Rachidian Tooth Dimensions: The relationship bet- 
ween central cusp base width (L1, Figure 4) and size 
class of the whelk from which it came was evaluated with 
a three-way ANOVA (whelk size class x tooth row x 
sex) with the response being the maximum base width of 
the rachidian tooth. These data satisfied assumptions 
of homogeneity of variance after the logarithm. trans- 
formation was applie sd but not normality. 

The reli BHOUSHED be uh en rachidian tooth central cusp 
height (LL2, Figure 4), the size class of the whelk from 
which it came, and the tooth row was evaluated with a 
three factor ANOVA (hell size class tooth row 
sex). These data satisfied neither the assumption of 
normality nor homogeneity of variance regardless of the 


J. M. Harding et al., 2008 


Page 219 


Figures 1-3. Radula of veined rapa whelk. 1. Radula from an 84.2 mm shell length (SL) whelk showing general morphological 
features and orientation. 2. A rachidian tooth from row | of the radula from a 134 mm SL whelk showing wear. 3. An unworn 
rachidian tooth from row 41 from a 122 mm SL whelk. Scale bar = 1 mm. Abbreviations: r = row number; R = rachidian; Od = 


odontophore; A = anterior; P = posterior. 


transformation (logarithm, natural logarithm, square 
root, arcsine). 

The relationship between rachidian tooth base width 
L3, Figure 4), row number, sex, and size class of the 
whelk from which it came was evaluated with a three- 
way ANOVA (whelk size class x tooth row x sex). These 
data satisfied the assumption of homogeneity of variance 
without transformation but did not satisfy the assump- 
tion of normality regardless of the transformation (loga- 
rithm, natural logarithm, square root, arcsine) and were 
analyzed without transformation. 

The ratio of rachidian tooth central cusp base width 
L1) to central cusp tooth height (L2) was calculated for 


each rachidian tooth. Using a ratio that compares base 
width to tooth height is appropriate in a structure where 
both the base width and tooth height change along the 
length of the structure with ontogeny. Not only does the 
ratio allow for scaling when comparing individual teeth 
along the radula length, but it can also be used as an 
index of wear because width does not change with use 
In this case a central cusp base width: central cusp tooth 
height ratio >1 is indicative of wear. The first 81 rows 
were chosen for analysis because every radula dissected 
had at least 81 rows. These data satisfied neither the 
assumption of homogeneity of variance nor normality 
regardless of the transformation (logarithm, natural 


Page 220 


THE NAUTILUS, Vol. 122, No. 4 


Figure 4. Veined rapa whelk rachidian tooth with the mea- 
surements made in this study identified: rachidian tooth central 
cusp base width (L1), rachidian tooth central cusp height (L2), 
and rachidian tooth base width (L3). Scale bar = 0.25 mm. 


logarithm, square root, arcsine). A three-way ANOVA 
(whelk size class x tooth row number x sex) was used 
to evaluate the ratio of central cusp width to central cusp 
height (L1:L2). 

The ratio of central cusp base width (L1) to rachidian 
tooth base width (L3) for each rachidian tooth was eval- 
uated with a 3 way ANOVA (size class x row x sex) to 
describe potential changes in tooth shape with ontogeny. 
Data satisfied assumptions of homogeneity of variance 
without transformation but not nor staliy (either with or 
without transformation, e.g., logarithm, natural loga- 
rithm, square root, arcsine). acces in the L1:L3 ratio 
across whelk size classes reflect ontogenetic changes in 
tooth pam that may be related to sexual dimor- 
phism (Fujioka, 1982, 1984) and which may act to dis- 
perse relatively greater strike force during f feeding in 
larger whelks. 


RESULTS 


Only radulae from whelks collected when water tem- 
peratures were above 11-12°C and feeding were used 


(Harding, unpublished data). All radulae examined were 
intact. Descriptive morphological data were collected 
from 39 rapa whelk radulae. These radulae were from 
rapa whelks with shell lengths between 20.2 mm and 
174 mm (Table 1, Figure 5). Rapa whelk radula lengths 
ranged from 4.33 to 51.05 mm with corresponding 
radula widths of 0.23 to 2.67 mm and total number of 
transverse rows of teeth of 89 to 210, respectively. 
Radula length was an average of 21.4% (standard error 
= 0.61%) of shell length. 


Radula Allometry and Gross Morphology: Regres- 
sion coefficients for the fitted linear and power regres- 
sion models used to describe relationships between rapa 
whelk shell length and radula morphology and between 
rapa whelk radula measurements are given in Table 2A. 

The linear model is suggested as a better oe of 
the relationship between rapa whelk shell length (SL) 
and radula length (RL; Figure 6A) for both sexes since 
the coefficients of determination from both models are 
identical (Table 2A) and the linear model provides the 
simplest description of the data. The slope for the SL— 
RL relationship in males is significantly higher than that 
for females (t-test, Table 2B). 

The power model more accurately described the rela- 
tionships between rapa whelk SL and radula width (RW; 
Figure 6B) by predicting a radula width equal to 0 at a 
shell length equal to 0. The coefficient of determination 
for the line sar model describing the relationship between 
shell length and the number of rows of radular teeth was 
higher (females = 0.63, males = 0.76, Table 2A) than that 
of the corresponding power model (females = 0.58, 
males = 0.69, Table 2A) for both sexes and the linear 
model predicted a positive number of rows of teeth at 
shell lengths of 0 mm (Figure 6C). 

The relationship between radula length (RL) and rad- 
ula width (RW) was described with a power model for 
both sexes (Table 2A, Figure 6D) which predicted a 
radula width of 0 at a radula length of 0 and had a higher 
coefficient of determination than the corresponding lin- 
ear model (Table 2A). The linear model describing the 
relationship between radula length (RL) and number 
of transverse rows of teeth has a higher coefficient of 
determination than the corresponding power model 


Table 1. . se id of ey whelks used in this study with basic statistics on radulae. Abbreviations used below are as follows: 
F = female, M = male, Avg = average for female and male whelks combined, SL = shell length, mm, SEM = standard error of the 
mean in miner ses, RL = radula le meth, mm, RW = radula width, mm; NRT = number of rows of teeth. 
# of Whelks Avg SL Ave RL Ave RW Avg NRT Avg RLSL % 
Whelk size class (F/M) (SEM) (SEM) (SEM) (SEM) (SEM) 
Petite (<45 mm SL) F=5 25.64 (2.64) 9.28 (0.56) 0.48 (0.18) 106 (6.07) 20.59 (0.58) 
M=5 34.72 (3.07) 7.62 (0.95) 0.71 (0.31) 122.20 (4.78) 21.72 (0.81) 
Small (45,1—-90 mm SL) F=5 70.38 (5.47) 15.93 (2.21) 0.97 (0.15) 133.80 (5.05) 92.38 (1.65) 
M=6 69.13 (6.21) 17.63 (1.92) 1.22 (0.13) 126.67 (3.86) 25.56 (1.79) 
Medium (90.1-135 mm SL) F = 4 104.75 (5.07) 22.32 (2.52) 1.47 (0.03) 137 (12.71) 21.37 (2.53) 
M=5 120.2 (5.3) 29.85 (1.58) 2.04 (0.10) 149.2 (7.19) 25 (1.51) 
Large (135.1-150 mm SL) F=2 153 (10.0) 35,7 (3,2) 1.99 (0.24) 177.5 (16.5) 23.3 (0.57) 
M=7 153.71 (4.47) 12 (2,23) 2.22 (0.10) 182.29 (6.38) 26.76 (1.21) 


J. M. Harding et al., 2008 


Page 221 


H@ Females 
© Males 


w 


Number of whelks 


Midpoint of rapa whelk shell length size class (mm) 


Figure 5. Shell length (mm) frequency distribution for the 
39 veined rapa whelks whose radulae were examined in this 
study. 


(females = 0.78 vs. 0.71, males = 0.81 vs. 0.72, Table 2) 
and predicts a positive number of tooth rows at radula 
lengths of 0 mm (Figure 6E). 


Rachidian Tooth Dimensions: Rachidian tooth cen- 
tral cusp base width (Ll, mm; Figure 4) increased 
significantly with increasing whelk size class (Table 3, 
Figures 7A and 7D). There were no significant differ- 
ences in central cusp base width observed between rows 
within a size class for the 81 rows of teeth that were 
examined. Central cusp base width was. significantly 
larger in males than in females (Fisher's test, Table 3). 
Differences between male and female central cusp base 
width were particularly evident in the medium and large 
size classes (Figure 7A and 7D). 

The height of the rachidian tooth central cusp (L2, 
mm) varied significantly with size class, row number, 
and sex (ANOVA, Table 3, Figures 7B and 7E). In gen- 
eral, larger whelks have larger central cusp heights in 
rows 31 through 81 than whelks of other size classes. 
Within all size classes and both sexes, central cusp 
heights from rows | and 11] are significantly less than in 
rows 31 through 81 (ANOVA, Table 3, Figures 7B 
and TE). Central cusp heights from rows 1, 11 and 21 in 
large whelks are significantly different from central cusp 
heights in petite whelks but similar to cusp heights ob- 
served in rows | and 21 for medium whelks and row 11 
for small whelks (ANOVA, Table 3, Figures 7B and 7E). 
Female whelks have significantly lower L2 values than 
male whelks (Fisher's test, Table 3, Figures 7B and 7E). 

Rachidian tooth base width (L3, mm) increases signi- 
ficantly with increasing whelk size class (ANOVA, 


z= 
o 
© 
) 
a 
= 
a 
+ 
“a 
* 
iss} 


4 
am 
= 
=r 
Be 
ae 


iary of regression statistics used to describe rapa whelk radula morphology and wear patterns. The linear model equation was y 


Table 2A. 


used for the power model was y = a X 


radula 


? = standard ert 


coefficient; SE 


Abbreviations used below are as follows: coef 


RW = radula width, mm: NRT 


number of rows of teeth. 


man: 


h, 


Males 


Coef b (SE) 


Females 


Coet b (SE) 


p Value Regression 


F Statistic 


) 


4 


Coeft a (SE 


R? 


p Value Regression 


F Statistic 


= 
Y 


5 


Ro 


Model 


Relationship 


Ooranno rst 
AawmMe tae as 
OARDMHSOD al 
GKeKEoOCKOCtTOSKS 
an 


1.08 


(0.08) 


40.35 
0.43 


69 


76 


ANDnRoOnorse 
ORANAANDFAMS 
GHMHHBOrKS 
ADGONAMMNMV 
[oon oe | 


.29) 


P 


we) 
aA 
cD 
o> 
Oman amon oO 
DAADOOD OM on 


= 
2 : cS = ac 
| s i _— aml 
~ © F me ZF, 
iam jam xX al 
vx gv x g Ee 
- - - - m 
Sp) Sp) NY aa aa 


Power 


Page 222 THE NAUTILUS, Vol. 122, No. 4 


60 A —e—Females 
* -no- Males 
50 5 a 
o ra 
= 40 a ® 
s a 
= 30 ae 
© op 
o-" o 
3 oe : 
& 20 oo" ; 
~ “Ya 
ou 6 ; 
10 i) 


0 50 100 150 200 


Radula width (mm) 
Radula width (mm) 


0 50 100 150 200 0 10 20 30 40 50 60 
290 Shell length (mm) 236 Radula length (mm) 
200 
= = 
e g 180 
iE ; 160 
” wn 
5 © 140 
3 2 
E E 420 
Zz Zz 
100 {’ 
80 
0 50 100 150 200 0 10 20 30 40 50 60 


Shell length (mm) Radula length (mm) 


J. M. Harding et al., 2008 


Page 223 


Table 2B. Summary of t-tests comparing regression equa- 
tions for female and male whelks given in Table 2A that are 
recommended for descriptions of these re lationships. T-tests 
were performed for the power model using logarithm 
transformed data. Abbreviations are the same as those used 
in Table 2B above. 


Relationship Comparison Model p value 
SL vs. RL Female vs. Male Linear <0.05- 
SL vs. RW Female vs. Male Power >0.05 
SL vs. NRT Female vs. Male Linear >0.05 
RL vs. RW Female vs. Male Power >0.05 
RL vs. NRT Female vs. Male Linear >0.05 


Table 3, Figures 7C and 7F). However, no differences 
were observed in rachidian tooth base widths between 
radular tooth rows within a whelk size class and within a 
sex (ANOVA, Table 3, Figures 7C and 7F). Rachidian 
tooth base width was significantly larger in male whelks 
than in female whelks and this trend is particularly evi- 
dent in the medium and large size classes (Table 3, 
Figures 7C and 7F). 

Large whelks had significantly higher ratios of rachi- 

dian central cusp base width (Li) to central cusp height 

(L2) than all other whelk size classes (ANOVA, Table. 3: 
Figure 8). The first row of teeth in the radulae had 
significantly higher L1:L2 ratios than all other rows 
(ANOVA, Table 3 . Figure 8). The eleventh row of teeth 
also had an L1:L2 ratio ie was significantly higher than 
that observed in rows 21-81 (ANOVA, Table 3 Figure 8). 
Since central cusp base to height (L1:L2) ratios >1 are 
indicative of tooth wear, the anterior 1-11] rows of teeth 
are more worn than newer teeth occurring in rows 2] 
and higher. Within each size class and sex, a wide range 
of L1:L2 values was observed for row 1 and/or row 11 
(Figure §). This variability was the result of one or two 
individual whelks per size class having very low central 
cusp heights ( extreme wear) in row 1 or row 11, the rows 
of teeth that are activ ely used in feeding. Patterns of wear 
as indicated by L1:L2 ratio values were not significantly 
different between sexes (Table 3, Figure 8). 

The ratio of central cusp base width to rachidian base 
width (L1:L3) was significantly affected by size class and 
sex (ANOVA, Table 3, Figure 9). Within a size class and 
within a sex, the ratio of central cusp base width to 
rachidian tooth width did not change significantly with 
row number. Male whelks had greater L1:L3 ratios than 
female whelks (Fisher's test, Table 3, Figure 9). 


DISCUSSION 


Radula length 
teeth in the radula increase with increasing rapa whelk 


floridana): 


radula width, and number of rows of 


shell length. Ontogenetic increases in radula length and 
the number of rows of teeth with shell length have 
also been documented for other muricid species (e.g. 
Stramonita floridana, Radwin and Wells, 1968 (as Thais 
Cronia margariticola and Morula musiva, 
Fujioka, 1954; Thais bronni and T. clavigera, Fujioka, 
1985; Nucella lapillus, Kool, 1993). 

The relationship between rachidian tooth base width 
and central cusp base width also changes with ontogeny 
but does not change in relation to the anterior-posterior 
location on the radula. That is, within an individual and 
within a size class, rachidian teeth examined from rows | 
through 81 display similar scaling of central cusp base 
width to rachidian tooth base swith, Rachidian teeth in 
female rapa whelks tend to have smaller central cusp 
base width as well as tooth base widths when compared 
to male whelks within the same size class. Fujioka (1982, 
1984) describes similar ontogenetic changes in rachidian 
tooth shape including an increase in earl cusp base 
width for Cronia margariticola, Morula musiva, and Dru- 
pella sp. in relation to sexual dimorphism. The observed 
ontogenetic changes in rachidian tooth shape may reflect 
morphological shifts designed to accommodate greater 
rachidian tooth strike force re sulting from the sealing of 
the buccal complex at increased w he Ik sizes. Pre sumably 
there is an ontogenetic scaling relationship in effect to 
optimize the force provided by the buccal mass muscula- 
ture and minimize the damage to rachidian teeth through 
use that is reflected in the shape of the tooth. 

Rachidian tooth wear, as indicated by the ratio of 
central cusp width (L1) to central cusp height (L2), 
decreases with increasing distance from the anterior 
(oldest) end of the radula. Rachidian teeth in the first 
11 rows of the radula have central cusp heights that are 
less than central cusp heights in rows 21 through $1 in 
all size classes. Carriker et al. (1974) describe rachidian 
cusps that have been worn off leaving only the tooth 
base in the anterior rows of rachidian "teeth of Urosal- 
pinx cinerea folleyensis. In laboratory studies with Thais 
bronni and T. clavigera, Fujioka (1985) observed that 
rows of teeth worn by feeding yin 5-15% of the total 
number of radular rows. The whelk Acanthina spirata 
uses approximately 8 to 20 teeth in each rasping stroke 
as these whelks feed on mussels (Hemingway, 1975). 
These data are consistent with our observations for rapa 
whelks, where at least the Ist and 11th rows of the radula 
were used, and the 21st row acted as a transition between 
the part of the radula the whelk was actively using to feed 
and the more posterior section that was unused. 

The observed changes in rachidian tooth wear may 
reflect ontogenetic change ’s in predation strategy, diet, 
or possibly both. Differences in predation strategy 
are potentially reflected in the observed changes in 


Figure 6. 


Relationships for female (n = 16) and male (n = 23) 


veined rapa whelks ranging from 20.2 to 174 mm SL between shell 


length (SL) and radula length (RL, A), SL and radula width (B), SL and the number of rows of teeth (C), RL and radula width (D 


and RL and number of rows of teeth (E) with fitted regressions 


(female = solid, male = dashed) that were used to describe the 


relationships. Linear regression models are plotted for panels A, C, and E. Power models are presented in panels B and D 


Regression equations and descriptive statistics are given in Table 2 


Page 224 


THE NAUTILUS, Vol. 122, No. 4 


(L1, mm; SEM) 


Average central cusp height Average central cusp base width 
(L2, mm; SEM) 


Average rachidian tooth base width 
(L3, mm; SEM) 


Figure 7. 


the mean, SEM) for { 
of the mean, SEM 


females (¢ 


@ Petite female (n=5) Medium female (n = 4) 9 4 0 Petite males (n= 5) » Medium males (n = 5) 


fF A. a Small female (n= 5) am Large female (n = 2) D. 4 Smalimales (n= 6) g Large males (n = 7) 
op ee ae ahah te tat 

0.3 
é foPoE gE Pe # 

0.25 


0.2 


0.15 


0.1 


0 20 


40 60 80 (0) 20 40 60 


Rachidian tooth row number 


80 
Rachidian tooth row number 


Graphs of rachidian tooth row number in relation to average central cusp width (L1) with error bars (standard error of 
les (A) and males (D) from all size classes, average central cusp height (L2) with error bars (standard error 
males (B) and males (E) from all size classes, and average rachidian tooth base width (L3, SEM) for 


and males from all size classes 


J. M. Harding et al., 2008 


Page 225 


Table 3.) Summary of ANOVA results comparing tooth 
morphology across whelk size classes and rows within the 
radulae. Asterisks indicate statistical significance at an alpha 
value of 0.05. Abbreviations used below are as follows: 1= petite 
size class, 2 = small size class, 3 = medium size class, 4 = large 
size class. NA = Not applicable. Refer to Figure 4 for a 
description of L1, L2, and L3. 


Fisher's test 


Test Response Factors — p value results 
ANOVA. Central cusp Size class <0.01) 4>3>2>1 
base Row 0.80 NA 
width (L1) Sex <0.01° Male > Female 
ANOVA. Central cusp Size class <0.01° 4,3>2>1 
height (L2) — Row <0.01° 1,11 < 31-8] 
Sex <0.01° Male > Female 
ANOVA. Rachidian Size class <0.01) 4>3>2>1 
tooth base Row 0.99 NA 
width (L3) Sex <0.01° Male > Female 
ANOVA Wear Size class 0.02" 4>3,2,4=1 


(ratio L1/L2) Row <0.01" 1 > 21-81, 
11> 31-81 
Sex 0.16 NA 
ANOVA. Shape (ratio Size class <0.01° 4>2,3>1 
LI/L3) Row 0.90 NA 
Sex <0.01° Male > Female 


rachidian tooth wear with ontogeny because the rachi- 
dian teeth are actively used during shell drilling. Meth- 
ods of feeding w hich require penetration of prey valve 
shells with the radula (e. g., drilling) will leave more wear 
on the rachidian teeth than non- drilling methods of at- 
tack. Therefore, examination of pathadian tooth wear 
along the radula and differences in wear depending on 
size “class may give an indication as to transitions in 
feeding strategies : different size classes of rapa whelks 
(Figure 8). High | evels of rachidian tooth wear (L1:L2 
ratio >1) in rows 1 and 11 were associated with the 
petite and large size classes (Figure 8A). The smallest 
whelks (<45 mm SL) typically leave drill holes (i.e., wall 
bores) in the valves of their prey (Harding et al., 2007). 
Although large (>135 mm SL) rapa whelks do not 
always leave ae signatures in prey valves, edge 
bore signatures are left instead of drill holes (Harding, 
Kingsley-Smith, Mann, unpublished data) when signa- 
tures are present. The observed L1:L2 values for rows 

1 and 11 in the large size class are driven by one male 
ee SB). It is possible that this individual had been 
using its radula to penetrate prey shells and that the other 
large whelks had not. Since the large whelks used herein 
were wild caught and had unknown feeding histories, we 
cannot say - with certainty. Relatively less wear (L1:L2 
ratio <1.2 ) in row 11 was observed in rapa whelks with 
shell . of 45 to 135 mm (small and medium size 
classes) that do not typically drill their prey (Harding 
et al., 2007, Figure 8). Similar wear patterns (L1:L2 ratio) 
were observ edi for males and females and may serve as an 
indication that feeding strategies and/or prey may be simi- 
lar between animals of different sex but similar size. 


A. Females Petite ir = . 
n = 


Average [central cusp base width(L1)/central cusp height (L2)], SEM 


6 
o Petite (n=5 
B.|Males a Small i = a} 
° Medium (n= 5) 
Q Large (n=7) 
5 
4 
3 
2 


& 6 
we 
[3] 
oe 
ee 
Une 


0 20 40 60 80 
Rachidian tooth row number 


Figure 8. Rachidian tooth wear as indicated by the average 
ratio of rachidian tooth central cusp base width (L1) to central 
cusp height (L2) in relation to rachidian tooth row number for 
females (A) and males (B) from all size classes. 


If a whelk is using the radula to penetrate the shell, 
the chemical composition of the prey shell may affect 
the level of wear observed on the rachidian teeth. Oyster 
and mussel shells have more calcite and are relatively 
softer than other bivalves with predominantly aragonite 
shells (Carter, 1980). Drilling through aragonite shells 
has the potential to cause more wear on eachidk in teeth 
than shell Es tration of calcite shells. Examination of 
the radulae from rapa whelks using drilling to consume a 
species-specific diet is a topic for future research that 
would provide data to address changes in radula wear 
with regard to prey shell hardness. 


‘ DIG 
Page 226 


THE NAUTILUS, Vol. 122, No. 4 


0.35 e Petite (n = 5) 
. a Small (n = 5) 
A. Females » Mediunntn=4y 
= Large (n = 2) 
0.3 7 
0.25 
7 
Pye! 
0.2 
0.15 
0.35, 
|B./Males 


oO 
wo 
T 
rre—t 
HH 
HH 
HoH 


att fd t | 


Average [central cusp base width(L1)/rachidian tooth base width (L3)], SEM 


02 o Petite (n = 5) 

: 4 Small (n =6) 
° Medium (n = 5) 

5 Large (n= 7) 


0 20 40 60 80 
Rachidian tooth row number 


Figure 9. Average ratio of central cusp base width (L1) to 
rachidian tooth base width (L3, with standard error of the 
mean, SEM) by rachidian tooth row for females (A) and males 
(B) from all size classes. 


ACKNOWLEDGMENTS 


This manuscript is dedicated to the late Dr. Melbourne 
R. Carriker, whose passion for and detailed work with 
Muricids serves as both a standard of excellence and 
an inspiration. Thanks are extended to all local fisher- 


men that have participated in the Virginia Institute of 


Marine Science rapa whelk bounty program since its 
inception in 1998. Melissa Southworth, Ethan Jestel, 
Catherine Ware, Erica Westcott, Steven Goodbred, 
David Kerstetter, Peter Kingsley-Smith, Amy Bohannon, 
Meredith Fagan, Meghan Harris, Rhonda Howlett, 
Rae Marie Johnson, Courtney Harris, Karen Capossella, 
John Hansen, and Matthew Robinson assisted with 


whelk collections from local fishermen. Drs. Greg 

Capelli, Randy Chambers, John Kraeuter, Ms. Melissa 
Southworth, and an anonymous reviewer provided valu- 
able comments on earlier versions of this manuscript. 
This work was completed in partial fulfillment of the 
requirements for an undergraduate Honors degree 
(SMG) from the Department of Biology, College of 
William and Mary, Williamsburg, Virginia. This is contri- 
bution number 2928 from the Virginia Institute of 
Marine Science, Gloucester Point, Virginia. 


LITERATURE CITED 


Arakawa, k. 1964. A study on the radulae of the Japanese 
Muricidae (2): The genera Vexilla, Nassa, Rapana, Murex, 
Chicoreus, and Homalocantha. Venus 22: 355-364. 

Carriker, M. R. 1961. Comparative functional morphology of 
boring mechanisms in gastropods. American Zoologist 
1: 263-266. 

Carriker, M. R. 1969. Excavation of boreholes by the gastro- 
pod, Urosalpinx: an analysis by light and scanning elec- 
tron microscopy. American Zoologist 9: 917-933 

Carriker, M. R., J. G. Schaadt, and V. Peters. 1974. Analysis by 
slow-motion picture photopograph and scanning electron 
microscopy of radular function in Cl Irosalpinx cinerea 
follyensis (Muricidae, Gastropoda) ) during shell penetra- 
tion. Marine Biology 25; 63-76, 

Carriker, M. R. 1981. Shell penetration and feeding by natica- 
cean and muricacean predatory gastropods: a synthesis. 
Malacologia 20; 403-422. 

Carter, J. 1980. Environmental and biological controls of bi- 

valve shell mineralogy and microstructure. In: Rhoads, D 
and R. Lutz (Eds.) Skeletal Growth of Aquatic Organisms. 
Plenum Press, New York, pp. 69-113. 

Fujioka, Y. 1952. On the secondary sexual characters found in 
the dimorphic radula of Drupella (Gastropoda: Muricidae) 
with reference to taxonomic revision. Venus 40; 203-223. 

Fujioka, Y. 1984. Sexually dimorphic radula in Cronia mar- 
gariticola and Morula musiva (Gastropoda: Muricidae). 
Venus 43: 315-330. 

Fujioka, Y. 1955. Seasonal aberrant radular formation in Thais 
bronni (Dunker) and T. clavigera (Kuster) (Gastropoda: 
Muricidae). Journal of Experimental Marine Biology and 
Ecology 90; 43-54. 

Harding, J. M. and R. Mann. 1999. Observations on the bio- 
logy of the veined rapa whelk, Rapana venosa (Valenci- 
ennes, 1546) in the Chesapeake Bay. Journal of Shellish 
Re ees 18: 9-17. 

Harding, J. M. and R. Mann. 2001, Growth rates of larval and 
juvenile veined rapa whelks Rapana venosa, from Chesa- 
peake Bay, USA, from hatch through age 1. International 
Conference on Marine Bioinvasions, 9-11 April 2001, 
New Orleans, LA, USA (Abstract). 

Harding, J. M. and R. Mann. 2005. Veined rapa whelk (Rapana 
venosd) range extensions in the Virginia waters of 
Chesapeake Bay, USA. Journal of Shellfish Research 24: 
3581-355. 

Harding, J. M., P. Kingsley-Smith, D. Savini, and R. Mann. In 
press. Comparison of predation signatures let by Atlantic 
oyster drills (Urosalpinx cinerea Say, Muricidae) and 
veined rapa whelks (Rapana venosa Valenciennes, Muri- 
cidae) in bivalve prey. Journal of Experimental Marine 
Biology and Ecology. 


SSS SSS SSS 


J. M. Harding et al., 2008 


Page 227 


Hemingway, G. T. 1975. Functional morphology of feeding 
in the predatory whelk, Acanthina spirata, Gastropoda: 
Prosobranchia). Bulletin of the American Malacological 
Union, Inc. pp. 64-65 (Abstract). 

Isarankura, K. and N. W. Runham. 1968. Studies on the re- 
placement of the gastropod radula. Malacologia 1: 71-91. 

Kool, S. P. 1993. The systematic position of the genus Nucella 
(Prosobranchia: Muricidae: Ocenebrinae). The Nautilus 
107: 43-57. 

Krutak, P. R. 1977. Gastropod radulae: their potential in the 


fossil record. Transactions of the Gulf Coast Association of 


Geological Societies 27: 314-322. 

Mann, R., A. Occhipinti, and J. M. Harding. 2004. Current 
status of global invasions by the marine gastropod Rapana 
venosa. First special report of the International Council 
for Exploration of the Seas on the status of introduced 
species. International Council for Exploration of the 
Seas/ICES. 24th Annual meeting of ICES (WGITMO), 
Goteberg, Sweden, March 20-21, 2002. 

Mann, R., J. M. Harding, and E, Westcott. 2006. Occurrence 
of imposex and seasonal patterns of gametogenesis in the 
invading veined rapa whelk Rapana venosa from Chesa- 
peake Bay, USA. Marine Ecology Progress Series 310: 
129-138. 

Morton, B. 1994. Prey preferences and method of attack 
by Rapana_ bezoar (Gastropoda: Muricidae) from 
Hong kong. pp. 309-325. In: B. Morton (Ed.). The 


malacofauna of Hong Kong and Southern China III. 
Hong Kong University Press, Hong Kong. 

Paine, R. 1966. Function of labial spines, composition of diet, 
and size of certain marine gastropods. The Veliger 9: 17-24. 

Pastorino, G., P. E. Penchaszadeh, L. Schejter, and C. Bremec. 
2000. Rapana venosa (Valenciennes, 1846) (Mollusca: 
Muricidae): A new gastropod in south Atlantic waters. 
Journal of Shellfish Research 19: 897-899. 

Radwin, G. E. and H. W. Wells. 1968. Comparative radular 
morphology and feeding habits of Muricid Gastropods 
from the Gulf of Mexico. Bulletin of Marine Science 18: 
72-85. 

Tsi, C. Y., X. T. Ma, Z. K. Lou, and F. S. Zhang. 1983. Ilustra- 
tions of the fauna of China (Mollusca), Vol 2. Science 
Press, Beijing. 

Vink, R., D. Nieweg, and H. Post. 2005, Rapana venosa (Valen- 
ciennes, 1846): a new invasive species for the Netherlands 
(and England?). Spirula 345: 152-155 (In Dutch with 
English abstract). 

Wu, S. kK. 1965. Studies of the radulae of Taiwan muricid 
gastropods. Bulletin of the Institute of Zoology, Academia 
Sinica 4: 95-106. 

Wu, Y. 1988. Distribution and shell-height relation of Rapana 
venosa Valenciennes in the Laizhou Bay. Mar. Sci./ 
Haiyang Kexue 6: 39-40. 

Zar, J. 1996. Biostatistical analysis. 3rd edition. Prentice Hall, 
New Jersey. 


THE NAUTILUS 122(4):228-235, 2008 


Page 228 


A new species of Sphaerium Scopoli, 1777, from southern Brazil 


(Bivalvia: Sphaeriidae) 


Maria Cristina Dreher Mansur 
Av. Arlindo Pasqualini, 410 
91760-140 Porto Alegre, BRAZIL 
mcmansur@terra.com.br 


Sommergasse 10 


GERMANY 


Claus Meier-Brook 


D-72119 Ammerbuch-Reusten 


Cristian Ituarte 

Museo Argentino de Ciencias Naturales 

Ay. Angel Gallardo 470, C1405DJR, Buenos 
Aires 

ARGENTINA 


ituarte@mail.retina.ar 


ABSTRACT 


Sphaerium cambaraense new species is described based on 
samples collected in the beginning of the summer near the 
headwaters of Taquari River (Jacui River Basin) on the basaltic 
plateau of southern Brazil. This is the first record for the genus 
Sphaerium in Brazil and in South America outside the Andes. 
Sphaerium cambaraense is characterized by a relatively large 
and solid shell, a high triangular shell outline and a solid hinge 
plate. Compared with Sphaerium forbesi (Philippi, 1869), 
S. cambaraense has a more strongly triangular shell outline 
and beaks not prominent. Sphaerium lauricochae (Philippi, 
1869), another similar species from Bolivia, Chile, and Peru, 
has a more rounded shell outline. 


Additional Keywords: Freshwater, Rio Grande do Sul, South 
America 


INTRODUCTION 


According to Dreher-Mansur and Meier-Brook (2000), 
the family iene is represented by two subfamilies: 
Euperinae, including the genera Eupera Bourguignat, 
1854, and Byssanodonta rt Orbigny, 1846; and Sph- 
aeriinae with three genera, Sphae rium Scopoli, 1777, 
Musculium Link, 1807, and Pisidium Pfeiffer, 1821. 
Four species of the genus Sphaerium are known from 
South America: Sphaerium aequatoriale Clessin, 1879, 
from Ecuador (Kuiper and Hinz, 1984); S. forbesi (Phi- 
lippi, 1869) recorded from Peru and Bolivia (Haas, 
1949), from Bolivia (Haas, 1955), from Colombia, Peru, 
Bolivian Andes (Kuiper and Hinz, 1984), and Chile 
(Ituarte, 1995); S. lawricochae (Philippi, 1869), from 
Peru, Bolivia and Chile (Kuiper and Hinz, 1984), and 
Chile (Ituarte, 1995): S. titicacense (Pilsbry, 1924), from 
Peru and Bolivia (Kuiper and Hinz, 1984). According to 
these authors, they are concentrated in the Central 
Andes from Ecuador to North Chile including high- 
altitude Bolivia, at 2000 to 4700 m altitude; varying in 


size between 3 and 12.5 mm. Haas (1949) reports 


Sphaerium (S.) boliviense (Sturany, 1900), from the 
highlands of Bolivia and from a lake in Junin, Peru. He 
examined also material of this species collected by Sioli 
in the regions of the rivers Maué-Agi and Tapajos, 
ibutaes of the Amazon River, which constitutes the 
first record of Sphaerium for Brazil. According to Kui- 
ae and Hinz (1984), S. boliviense is a junior synonym of 
S. forbesi. 

The record of Sphaerium observationis by Mansur 
et al. (1991) for Mirim Lagoon in southern Brazil is a 
misidentification, as that species is not a Sphaerium. 
According to the revision by Ituarte (1995), Pisidium 
abionbatonss Pilsbry, 1911, only occurs in the southern 
Argentina, not in Brazil. 

Sphaerium cambaraense new species is the fifth 
Sphaerium species from South America and the first spe- 
cies of the genus described from southem Brazil, in the 
highlands near the Atlantic Ocean, a location geographi- 
cally very distant from the Andes and Amazon River. 


MATERIALS AND METHODS 


Specimens were collected with a plastic sieve with mesh 
size of about 0.8 mm. Specimens were sorted from the 
sediment with fine feather tweezers; anesthetized in 
small vials containing water with menthol crystals, fixed 
in a5% formalin solution for 24 hours, rinsed for 24 hours 
in tap water and preserved in 70% ethanol. Soft parts of 
specimens for scanning electron microscopy (SEM) were 
removed with tweezers, shells cleaned with a soft and 
fine brush and rinsed several times in distilled water. 
Dried shells were glued on stubs with light-silver glue 
(Porolon Equipment, Herts) or metallic adhesive tape 
(TO66 Silver tape 9 mm, Hert-Scotch), coated with gold, 
and observed either in a Cambridge Stereoscan 250 Mk2 
or Philips scanning electron microscopes. Shell micro- 
structure was studied by fracturing shells at the middle 
of the height, parallel to commarginal ridges. Terminolo- 
gy for shell microstructure follows Dyduch- Falniowska 
(1983) and Dreher-Mansur and Meier-Brook (2000). 


M. C. D. Mansur et al., 2008 


Page 229 


Stomach nomenclature follows Purchon (1958, 1960). 
The shape indices, height index [I = H/L] and convexity 
index [Ci = W/H], were calculated according to the cri- 
teria followed by Ituarte (1996). 

Abbreviations use in the text are: MACN, Museo 
Argentino de Ciencias Naturales, Buenos Aires; MCN, 
Museu de Ciéncias Naturais, Fundagao Zoobotanica do 
Rio Grande do Sul, Porto Alegre; MCP, Museu de Cién- 
cias e Tecnologia da Pontificia Universidade Catélica, 
Rio Grande do Sul, Porto Alegre, Brazil. 


Genus Sphaerium Scopoli, 1777 


Sphaerium cambaraense new species 
(Figures 1-20) 


Diagnosis: Distinguished by the relatively large and 
shes shell with trapezoidal tending to tric angui: ar shell 
outline, low and wide beaks, subce sntrally locate a without 
marked nepionic cap, and broad and solid hinge plate. 


Description: SHELL: Solid, relatively large (maximum 
observed L: 11.22 mm), slightly convex (Ci = 5345). 
Shell outline high 85+2), trapezoidal tending to 
triangular. Dorsal margin has pronounced curve. Anteri- 
or and posterior margins gradually descending and 


Figures 1-5. 


Sphaerium cambaraense new species. 1, 2. 


outer shell surface. 4, 5. Inner shell surface, detail of pores. Scale bars: 1, 


Holotype MCP Mol. Outer view of left and ri 


gently curved below middle of total height, without 
marked ange Posterior end slightly truncated, oblique 
(Figures 1, 2). Ventral margin long and evenly curved. 
Shell surface silky, glossy, with very fine irregularly 
distributed radial lines: weaker on beaks and irregular 
commarginal, sometimes coarse, striae (12 or more per 
0.5 mm in the middle of the shell (Figure 3). Outer shell 
surface light yellowish brown; pale ae at beaks, more 
ventrally gr ayish brown with complete or incomplete 
ye sllow concentric bands running ae anterior to poste- 
rior margins; a large yellow band near ventral m: irgin, 
Beaks without marked embryonic cap. Inner shell sur- 
face white, grayish at muscle scars. Beaks slightly proso- 
gyrous, low and wide, slightly see above dorsal 
margin, subcentrally located | Figures 1, 2, 6, 7). 

Hinge plate strikingly curved solid, bioad, reaching 
0.5 mm width in middie region in specimens of 1] mm 
length, slightly narrower at jeval of cardinal teeth. Hinge 
Cae. arched, pe articul: irly below cardinal teeth (Figures 6-9). 

Cardinal teeth strong, close to dorsal margin. Right 
cardinal tooth, C3, short, strongly curved, posterior end 
enlarged in a grooved cup (Figure 10). Left cardinal teeth: 
outer cardinal tooth, C4, thin, strikingly oblique, located 
immediately behind C2, anterior end slightly overlapping 


ght valves. 3. Detail of 
2=4mm; 3 = 400 um; 4 = 4 um; 5 = 40 pm 


AUTILUS, Vol. 122, No. 4 


Figures 6-11. Sphaerium cambaraense new species. Paratype (MACN-In 37063). 6. Inner view of right valve. 7. Inner view of 


right valve. 8. Hinge of right valve. 9. Hinge of left valve. 10. 


5 


cardinal teeth (C2 and C4) and ligament. Scale bars: 6, 7 = 2 mm: 


2; inner cardinal tooth, C2, short high columnar, deey 


arched into a \ shape Figure 11). Right lateral teeth 
somewhat short, strong, with distal cusps | Figure 8): left 
lateral teeth relatively long, strong, and high (Figure 9). 
Ligament internal but exteriorly visible, slightly protrud- 


ed in larger specimens Figures LO, 11) 


Yetail of right cardinal tooth (C3) and ligament. 11. Detail of left 
S,9 =1 mm; 10, 11 = 500 um 


Suett Microsrructrure: Imer shell surface perforated 
wey 2 a1 2) (Pig 

by numerous pores (55/400 tim> to 13/160 tum) (Figures 

f, 5) representing the opening of tubuli that cross entire 

calcareous part of shell (shown in part in Figure 12 


Openings of pores on inner surface surrounded by a 


funnel-shaped depression and in mouth circled by arim 


M. C. D. Mansur et al., 2008 Page 23 


Figure 12. Sphaerium cambaraense new species. Shell microstructure from a: periostracum (top) to f: the endostracum (bottom 


Inserts a-g are details of: a, periostracum; b, granular layer; ec, diagonal layer forming a composite prismatic structure; d, diagonal 


layer. crossed structure: e, diagonal layer forming a pseudo crossed lamellar structure; f, palisade structure; g, diagonal layer, and 
y 


internal surface of the endostracum (arrow). Scale bars: 12 = 50 tm; inserts: a-d = 4 tm; e = 20 tum; f = 10 pm; g = 2 Lm 


Page 232 THE NAUTILUS, Vol. 122, No. 4 


M.C. D. Mansur et al., 2008 


Pz ge 2 9' 3h 3 


(Figure 4). Shell relatively thick reaching 200 ptm in 


central area (Figure 12). Shell structure consists of 


five layers: periostracum, granular layer, diagonal layer, 
palisade, diagonal layer, and endostracum. Below the 
2 um thick periostracum (Figure 12, a), a granular 10 pm 
thick layer appears (Figure 12, b); grains concentrated 
in some points irregularly and sparsely distributed, look- 
ing like cone-shaped bars arranged perpendicularly to 
shell surface. These change into irregularly and com- 
pactly arranged grains of different sizes. More internally, 
shell structure changes gradually into a diagonal layer 
(Figure 12, e, d, e), w here three different patterns may 
be distinguished: first (about 50 jim thick), composite 
prismatic structure, showing vertical feather-like pattern 
(Figure 12, ¢); second (about 40 tm thick), occupies 
the central part of shell, gradually changing into a 
cross-lamellar structure ( Figure 12, d): the third (approx- 
imately 60 [m), resembles a pseudo crossed-lamellar 
structure (Figure 12, e). Close to the inner shell surface 
there are one or two palisade layers, narrow (each about 
1 um thick) (Figure 12, f). The palisade structure is 
followed by a diagonal layer with lamellae oriented 
in one direction (Figure 12, g); below, a very narrow 
layer, the endostracum (1.0 to 1.5 pm thick) ( (Figure 12, 
arrow). i 
Anatomy (Ficures 13-20): Inner and outer demi- 
branchs well-developed, outer smaller demibranch 
reaching half of height of inner demibranch (Figure 13). 
Brood sacs occupy anterodorsal part of inner demi- 
branchs, embryos contained in sacs showed different 
developmental stages, denoting sequential brooding. 
Up to six dev eloping embryos found in largest examined 
specimen (>11 mm length). Largest, tertiary, brood sac 
contained two embryos (>1 mm length) (Figure 13); 
secondary sac located under lar gest one, contained three 
or four small embryos. Primary sac attached lower on 
inner demibranch. (Specimens for this study were col- 
lected at the beginning of the Southern Hemisphere 
summer (January, 1994) and many of the specimens 
larger than 9 mm showed brood sacs.) Anal and branchi- 
al openings extended in diverging short siphons, nearly 
equal in size. Anal siphon wader at base and more stre- 
tched out (Figure 13). Two labial palps (Figure 13, 14) 
on each side of mouth, triangular-shaped in lateral view: 
opposite contacting walls with 12 small folds, tapering 
toward distal ends. Mantle musculature (Figure 15) with 
relatively short siphonal retractors; inner radial mantle 
muscles arranged in eight to nine bundles (Figure 15). 
Fan-shaped stomach (Figures 17, 18) bent to right 
side, laterally covered by “digestive gland. Dorsal hood 
relatively short and left duct well dev pile Stomach 
internally (Figure 17) shows very simplified structures 


on dorsal hood and right side such as short gastric shield 
under dorsal hood; minor typhlosole and rejecting tract 
beginning at right side after descending from dorsal 
hod: ne an dley ation slightly wrinkled bel tween intesti- 
nal groove and rejection tract; anterior fold absent. In- 
testine opening associated to style sac in center of floor; 
major typhlosole arched in front with two expansions 
that end respectively at left and right duct openings, not 
penetrating in ducts, which allocates the stomach to 
Type IV; left and right duct openings well developed 
and ramified in ieee: secondary ducts. 

Intestine short and simple (Figure 18), anterior part 
associated to style sac, broad, descending straight to 
floor of visceral mass; mid-intestine strikingly stretched 
forming single loop; hindgut straight, ascending and 
bending toward rectum that ends into an anal papilla. 
Nephridia of closed type; dorsal and outer lobes fused, 
impossible to distinguish in dorsal view (Figures 19, 20); 
external wall of outer lobe ornamented with many small, 
rounded extrusions; nephridia in young specimens usu- 
ally of open type and with separ ated lobes. Funnel rela- 
tively long and wide; proximal loop with smooth walls 
forming three ascending rings then run backward up to 
posterior adductor miasele. turning laterally in direction 
to lateral loop. Lateral loop straight, elongated, running 
along outer side and partially covered laterally by outer 
lobe. Excretory sac sub-triangular, rounded in front and 
relatively small. No valve at the entering of the distal 
loop into excretory sac. 


Type Locality: Lajeado da Margarida (50°15.75' W, 
29°0.87' S: 870-880 m altitude) on Camisa River, Antas 
River Basin, considered to be the headwaters of Taquari 
River, the main tributary of Jacui River, South Atlantic 
Brazilian Basin. 


Type Material: Holotype MCN 38821; Paratypes: 
MCN 33919 (33 specimens), MCP sa (6 specimens), 
MACN-In 37063 (two specimens), 12 Jan. 1994. 


Etymology: The specific epithet refers to the City and 
Municipality of Cambara, close to the type locality. 


Distribution and Habitat: Known only from the type 
locality. The Municipality of Cambara is situated in 
northeastern Rio Grande do Sul State, Brazil, in the 
highest part named Planalto Riograndense. From the 
phy siographic point of view, this region is characterized 
by a basaltic shield covered by ion grass steppes and 
Araucaria Forest mixed with the seiithieastern limits 
of the Atlantic Forest. The altitude varies from 850 to 
1050 m, and in the winter, temperatures fall below 
freezing and there is the occasional snow. The rivers that 
cross the region, flanked by a low gallery forest, have 


Figures 13-20. 


Sphaerium cambaraense new species. Schematic drawings of soft anatomy. 13. Gross anatomy (left mantle lobe 


removed). 14. Folded surfaces of inner and outer left labial palps. 15. Outer view of left mantle lobe showing the inner radial rmantle 

muscles. 16. Dorsal view of the stomach and digestive gland. 17. Floor of the stomach after removing the roof. 18. Inner view of the 
organs in the visceral mass. 19. Dorsal view of nephridia, posterior adductor muscle and posterior foot retractors. 20. Lateral view of 
left nephridium. Scale bars: 13, 15, 16, 18 = 2 cm; 14, 17, 19, 20 = 2 mm 


Page 234 


THE NAUTILUS, Vol. 122, No. 4 


hard bottoms formed by flattened basaltic stones, and 
currents are strong. The collecting sites were small 
ponds along the river course, where currents were low- 
ex, allowing the accumulation of decayed leaves and very 
soft, dark, and fine sand deposits, not deeper than 1 m, 
where specimens settle. Together with the Sphaerium 
samples, many specimens of one species of Pisidium sp. 
(MCN 33918), and one of Diplodon sp. (MCN 33920) 
not yet identified, were found. 


DISCUSSION 


Sphaerium cambaraense is similar to Sphaerium forbesi 
(Philippi, 1869) (from Bolivia, Chile, and Peru). However, 
S. cambaraense has a more decidedly triangular shell 
outline, beaks not full with not marked ne pionic shell: 
in addition, S. cambaraense is larger than S. forbesi. 
Sphaerium lauricochae Philippi, 1869), also reported from 
Bolivia, Chile, and Peru, differs from S$. cambaraense 
by its more rounded shell outline. In relation to shell 
thickness and microstructure, §. cambaraense is similar to 
the European species Sphaerium rivicola (Lamarck, 
1818) and Sphaerium corneum (Linnaeus, 1758): 
Dyduch-Falniowska (1983) reported for these species (as 
well as for Musculium lacustre (Miiller, 1774) and several 
Pisidium species) six different layers, (1) lari (2) 
homogeneous-granular layer, (3) granular layer; (4) diago- 
nal layer (composite prismatic structure), (5) palisade 
structure, and (6) endostracum. Only S. corneum and 
S. rivicola showed a different structure for the diagonal 
layer, referred to as “crossed-lamellar structure” (Dy ee 
Falniowska, 1983). This structure was also found i 
S. cambaraense, however, the diagonal pattern of ie 
plates has a different arrangement: in S. rivicola the 
oblique plates show a horizontal herringbone pattern and 
in S. cambaraense some rows of plates are oblique not 
forming a horizontal pattern. These latter resemble in part 
the pseudo crossed-lamellar structure found in the Corbi- 
culidae. The periostracum layer in S. cambaraense is thin- 
ner than in S. corneum and the homogeneous granular 
layer is lacking. The simplified eondinen: of posterior and 
right side of stomach is quite similar to that found in 
the Euperinae (Dreher-Mansur and Meier-Brook, 2000). 
The nephridium is relatively similar compared to 
S. corneum (Dreher-Mansur and Meier-Brook, 2000), 
but the excretory sac and proximal loop are shorter and 
the valve at the insertion of the distal loop into excretory 
sac is lacking in S. cambaraensis; the multilobed surface 
of lateral lobe is observed for the first time in Sphaerium. 
Based on the similar morphology of the ne heat 
Korniushin (1998) propose d that South American spe- 
cies traditionally assigne d to Sphae rium ac tue lly be- 
long to Musculium. According to Park and O Foighil 
2000), the usefulness of the fine anatomy of ne phridi- 
um in the fi umily Sphaeriidae is relative, due to its high 
il pl. isticity. Coole oY and O F ‘oighil (2000), 
tochondrial 16S r DNA gene se quences 


morphol NOL 


based T1 n 


] 


observed that th Sphacrium/Musculium clade exhibit 
moderate | low levels of genetic dive rgences and 


the same asynchronous or sequential brooding pattern 
(L.e., brooding sacs contain more than one “develop- 
ing generations of embryos). Nevertheless, Cooley and 
O “Foighil (2000) recognized Musculium as a monophy- 
letic group. We allocate the new species in Sphaerium 
until more evidence is available to help with this ques- 
tion. More recently, a phylogenetic analysis of the 
Sphaeriinae (Lee and O Foighil, 2003) based on a molec- 
ular study of nuclear (ITS- 1) and mitochondrial (16S) 
gene sequences of 15 species from North and South 
America, Europe, Asia, and Australia, recovered a 
strongly supported monophyletic group of sequential 
brooders (Musculium and Sphaerium). However, the 
analyses indicate that Musculium and Sphaerium sensu 
lato are not natural groups, proposing a new classification 
system comprising five subgenera within Sphaerium: 
Sphae rium sensu stricto Scopoli, 1777; Musculium 
Link, 1807; Amesoda Rafinesque, 1520; Sphaerinova 
Iredale, 1943, and Herringtonium Clarke, 1973. As these 
subgeneric groupings, however well supported by 
molecular data, are not defined from the morphological 
point of view, it is not possible at this point to place 
S. cambaraense within the new scheme. 


ACKNOWLEDGMENTS 


We are grateful to the curators of MCN mollusk 
collection: Ingrid Heydrich and Silvia D. Hahn for 
providing the loan of specimens; to Prof. Dr. Cecilia 
Volkmer Ribeiro for the help in sampling and to 
Prof. Dr. Wolfgang Maier from Tiibingen University, 
Germany, for SEM sessions; Fabian Tricarico from the 
MACN SEM unit for his fine work with SEM photography. 
C.L. is a researcher from the Consejo Nacional de Investi- 
gaciones Cientificas y Técnicas (CONICET), Argentina. 


LITERATURE CITED 


Cooley, L. R. and D. O Foighil. 2000. Phylogenetic analysis 
of the Sphaeriidae (Mollusca: Bivalvia) based on partial 
mitochondrial 16S rDNA gene sequences. Invertebrate 
Biology 119: 299-308. 

Dyduch- Falniowska, A. 1983. Shell Microstructure and Sys- 
tematics of Sphaeriidae (Bivalvia, Eulamellibranchiata), 
Acta Zoologica Cracoviensia 16: 251-296. 

Haas, F. 1949. Land und Siisswassermollusken aus dem Amazo- 
nas Gebiete. Archiv fiir Molluskenkunde 78(4/6); 149-156. 

Haas, F. 1955. On some small collections of inland shells from 
ms America. Fieldiana, Zoology 34(35): 361-387. 

Ituarte, C. 1995. Nuevos registros de 1 Sat oe Pfe es 1821 y 

Spha verium Scopoli, 1777 (Bivalvia: Sphaeriidae) en C shile, 
Bolivia y Noroeste argentino, Neotropica 41( 0S. 106): 
31-41. 

Ituarte, C. F. 1996, Argentine species of Pisidium Pfeiffer, 
1821, and Musculium Link, 1807 (Bivalvia: Sphaeriidae) 
The Veliger 39 189-203. 

Korniushin, A. 1998a. Notes on the anatomy of some species 
of Sphaerium s. (Mollusca, Bivalvia) from the tropical 
region with revision of their taxonomic status. Vestnik 


319 


3): 3-12. 


99 
zoologii OAs 


SS SS ES 


M. C. D. Mansur et al., 2008 


Page 23: 


Korniushin, A. 199Sb. A comparative investigation on nephri- 
dia in fingernail and pill clams. Malacological Review, 
Supplement 7: 53-63. 

Kuiper, J. G. J. 1962. Notes sur le systématique des pisides. 
Journal de Conchyliologie 102: 53-57. 

Kuiper, J. G. J. and W. Hinz. 1984. Zur Fauna der Klein- 
muscheln in den Anden (Bivalvia: Sphaeriidae). Archiv 
fiir Molluskenkunde 114(4/6): 137-156. 

Lee, T. and D. O Foighil. 2003. Phylogenetic structure of the 
Sphaeriinae, a global clade of freshwater bivalve molluscs, 
interred from nuclear (ITS-1) and mitochondrial (16S) 
ribosomal gene sequences. Zoological Journal of the Lin- 
nean Society 137: 245-260. 


Mansur, M. C. D. and C. Meier-Brook. 2000. Morphology of 


Eupera Bourguignat 1854, and Byssanodonta Orbigny 
1846 with contributions to the Phylogenetic Systematics 


of Sphaeriidae and Corbiculidae (Bivalvia: Veneroida). 
Archiv fiir Molluskenkunde 128(1/2): 1-59. 

Mansur, M. C. D., C. Schulz, M. G. O. Silva, and N. M. R. 
Campos-Velho. 1991. Moluscos bivalves limnicos da Esta- 
cao Ecologica do Taim e areas adjacentes, Rio Grande do 
Sul, Brasil. Iheringia, Série Zoologia, 71: 43-58. 

Park, J. Kk. and D. O Foighil. 2000. Sphaeriid and corbiculid 
clams represent separate heterodont bivalve radiations 
into freshwater environments. Molecular Phylogenetics 
and Evolution 14: 75-88. 

Purchon, R. D. 1958. The stomach in the Eulamellibranchia: 
Stomach type IV. Proceedings of the Zoological Society of 
London 131; 487-525. 

Purchon, R. D. 1960. The stomach in the Eulamellibranchia: 
stomach types IV and V. Proceedings of the Zoological 
Society of London 135: 431-489, ; 


THE NAUTILUS 122(4):236-243, 2008 


Page 236 


Some remarks on the gross anatomy of Adelomelon ferussacii 
(Donovan, 1824) (Gastropoda: Volutidae) from the coast 


of Patagonia, Argentina 


Maria Eugenia Segade 

Guido Pastorino! 

Museo Argentino de Ciencias Naturales 
Ay. Angel Gallardo 470 3° piso lab. 57 
C1405D]R Buenos Aires, ARGENTINA 


ABSTRACT 


Specimens of the volutid Adelomelon ferussacii (Donovan, 
1824) were dissected. These were collected during an extraordi- 
nary low tide on February, 2006, at Playa La Mina, Puerto San 
Julian, Santa Cruz Province, Argentina oy 09’ S, 67°37’ W). 
The gross anatomy, along with radulae and shell ultrastructure, 
are described for the first time. Analysis of these features sug- 
gests a close relationship with the other species of the genus 
Adelomelon Dall, 1906, and raises doubts about the validity 
of Pachycymbiola Thering, 1907, a genus in which it has been 
included by recent authors. 


Additional Keywords: Southwestern Atlantic, Mollusca, 


Patagonia 


INTRODUCTION 


The family Volutidae includes the most conspicuous spe- 
cies among all endemic mollusks of the Southwestern 
Atlantic. This is so not only because of their large size, 
but also because of their interesting reproductive 
(Penchaszadeh, 1976, 1999: Luzzatto, 2006), 
Volutes live in quite diverse environments (i.e., subtidal 


biol Py 


to deep water; soft, muddy, sandy, or mixed bottoms; 
cold, temperate, or warm waters). Several species are 
extremely rare, while others are quite common. At least 


two species, particularly in Uruguay, are the subject of 


Zidona 
dufresnet (Donovan, 1823) and Adelomelon brasiliana 
(Lamarck, 1811). Other spec ies are likely to be « xploite od 
in the near future (Giménez and Penchaszadeh, 2002). 


extensive commercial fisherie 3+ The se are 


In addition, volutids usually aré » top predators with great 


ecological importance in the marine realm. Species of 


volutes studied thus far generally have been found to be 
pre dators on other mollusks and/or scavengers, 


} 
Author for correspondence gpastorino@macn,gov.ar 


About 30 names have been applied to Volutidae from 
the western Atlantic, from Venezuela to Argentina. Of 
these, 16 are living in Argentine waters. Most are poorly 
known, with dain. deived only from their original de- 
scription, which generally included very little biological 
information. In addition: ‘the taxonomic status of several 
names awaits clarification. 

Adelomelon ferussacii (Donovan, 1824), a locally com- 
mon volute similar to A. brasiliana, is known only from 
its shell. Recent findings of egg capsules suggest a compl- 
etely different re productive Diology ( (Pe aehasrade h and 
Segade, submitted). 

In this paper we provide new anatomical and ultra- 
structural data for Adelomelon ferussacii derived from a 
large ontogenetic series of live specimens, and compare 
them with other species of the same genus. 


MATERIALS AND METHODS 


Specimens used in this study were hand-collected dur- 
ing an extraordinary low tide on February 2006 at Playa 
La Mina, Puerto San Julian, Santa Cruz Province 
(49°09' S, 67°37’ W) (Figure The surtace water 
temperature was 15°C. Animals were relaxed in freez- 
ing sea water, preserved in ethanol 70% and dissected 
under a stereoscopic microscope. Radulae were pre- 
pared according to the method described by Solem 
(1972). Photographs were taken using digital cameras 
and images were also digitally processed. Shell ultra- 
structure data were procured from freshly fractured 
colabral sections taken from the central portion of 
the lip on the last whorl of several individuals. MACN: 
Museo Argentino de Ciencias Naturales. 


SYSTEMATICS 


Family Volutidae Rafinesque, 1S15 
Subfamily Zidoninae H. Adams and A. Adams, 1853 
Genus Adelomelon Dall, 1906 


Adelomelon ferussacii (Donovan, 1824) 


enn nnn TT EEE 


M. E. Segade and G. Pastorino, 2008 


Page 237 


T T “ATS yh is 
Buenos Aires = : 


\ Golfo 
San Matias 
> 


Figure I. 
Santa Cruz, Argentina. 


Voluta ferussacii Donovan, 1824, 2, pl. 67; Sowerby, 1546 
1: 203, pl. 46, fig. 7; Reeve, 1549 6, pl. 10, fig. 23, 
spec. 39; Dall, 1907: 362. 

Voluta rudis Gray in Griffith and Pidgeon, 1834: 601, pl. 
30, fig. 1. 

Voluta oviformis Lahille, 1895: 20, pl. 1 
figs. 53-56, pl. 7, figs. 121-138, pl. 1 
Voluta oviformis typica Lahille, 1895: 20. 

Voluta oviformis longiuscula Lahille, 1895: 20. 

Voluta oviformis fratercula Lahille, 1895: 20. 

Voluta (Cymbiola) ferussacii Donovan.—Strebel, 1906: 
100, pl. 9. figs. 46, 46a, 4549. 


figs. 1-2, pl. 2, 
0. figs. 4-9. 


Map showing the study area of Playa La Mina, 


Adelomelon (2?) ferussacit Donovan.—Clench — and 
Turner, 1964: 157, PI. 98, figs. 1-3. 

Adelomedon (sic) ferussacii (Donovan).—Castellanos, 
1970: 110, pl. §, fig. 7 

Adelomedon (sic) ferrusacii |sic]|—Castellanos, 1970b: 1, 
figs. 6, 9. 

Adelomelon (Pachycymbiola) ferussacii (Donovan, 1824). 
—Weaver and DuPont, 1970: 108, pl. 45C, 45D. 
Adelomelon ferussacii (Donovan, 1524).—Castellanos 

and Landoni, 1992: 12, Pl. 1, fig. 8. 
Pachycymbiola ferussacii (Donovan, 1824) —Poppe and 
Goto, 1992: 116, pl. 38, figs. 3-5. 


Description: Shell medium in size, up to 122 mm, 
solid, fusiform; color grayish-brown. Aperture semi- 
circular, dark-brown oe Protoconch of 144 smooth 
whorls; teleoconch of up to 4 slightly convex whorls; 
spire low, sometimes somewhat upturned; spire angle of 
80°, suture well defined. Columella curved, orange, with 
three to six folds set obl raed to siphonal Geciale, regu- 
larly distributed except for the anteriormost one, w nels 
is separate from the others. Columellar callus usually 

weak, but sometimes thick. Siphonal canal fairly broad 
and shallow. Growth lines span the surface, sometimes 

producing irregular costae. Shell ultrastructure showing 
three layers: a crossed! lamellar aragonite layer in w hich 
the cry stal planes are arranged perpendicul: uly to the 
growing edge of the shell, cand also perpe sndicular to 
the adel e layer (25% shell thickness); a middle layer 
(50% shell thickness) of colabrally aligned crossed- 
lamellar aragonite, and an outer layer (25% shell thick- 
ness) of amorphous calcite (Figure 1S). The innermost 
layer varies in thickness according to the sector of the 
lip: along the most curved sector it appears thicker, 
while it is reduced or absent towards the ends of 
the lip. a arrangement is quite similar to that found 
in shells of A. brasiliana (Figure 19) and A. beckii 
Broderio: 1836). 

Embryonic shells very thin (at hatching stage), whitish 
in the first whorls and dark brown in the ‘|: ist protoconch 
whorl. Surface covered by 8-10 regularly spaced spiral 
threads in the last whorl, no ple iits visible. Calcarella 
reduced, weakly pronounced. 

Foot, head, and siphon are finely mottled purple 
in color. The contact surface of the foot is whitish. Foot 
and shell length are similar. Operculum absent. Head 
broad and flattened, with two short tubular tentacles 
that separate the lateral lappets from the central one. 
Eyes are very small and located near the base of the 
tentacles. The penis emerges directly behind the right 
cephalic lappet (Figure 21). The siphon, also well pig- 
mented, is muscular with paired and symmetrical si aa 
nal appendages emerging from the base of the eee 
and spanning half its length. 

The bipectinate osphradium has aproximately 100 
equal leaflets. The ctenidium is 14 times as long as thi 
osphradium. The hypobranchial gland is thin. The ie 
boscis shows the same color pattern as the foot, head 
and siphon. Mouth opening is triangular 


THE NAUTILUS, Vol. 122, No. 4 


M. E. Segade and G. Pastorino, 2008 


Page 239 


Figures 18-19. 


Radular ribbon narrow, up to 17.4 mm in length (n = 13; 
x = 1L.7; DS = 1.97), with 49-74 rows (proportional to shell 
length) with one tricuspid central tooth per row (Figure 22). 
The radulae increase the number of teeth with age (Fig- 
ure 32). Central tooth thin and long, anterior profile con- 
cave with a ventral-posterior thickening (Figures 23 and 
24). Lateral cusps of the rachidian tooth are similar in size 
to the central one. Each cusp with a dorsal shallow indenta- 
tion or groove present, where the corresponding previous 
cusp imbricates (Figures 24-25). Embryonic radulae 
showing about 15 rows of teeth where lateral cusps are 
shorter than central cusps (Figure 26; Table 1). 

Salivary glands (racemose glands) large and irregu- 
larly shaped. Accessory salivary glands (tubular glands) 
very long and extremely convoluted. distally expanded, 
ov erlying dorsal surface of saliv ary glands. The tubular 
glands can easily be separ ated from the racemose 
glands. Ducts of the accessory salivary glands descend 
laterally to oesophagus and are sub kernal with respect 
to the mouth. They never fused and end separately. 
Ducts of salivary glands become embedded in the 


Shell ultrastructure. 18. Adelomelon ferussacii (Donovan, 1824). 19. Adelomelon brasiliana (Lamarck, 1811). 
Scale bars = 500 tm. 


oesophagus anterior to small valve of Leiblein. The an- 
terior esophagus runs behind the proboscis, and passes 
through the nerve-ring where an externally inconspicu- 
ous valve of Leiblein is located. The gland of Leiblein is 
very long and extremely conv oluted (Figure 20). The 
posterior oesophagus continues to the stomach, 


Material Examined: (D = specimen was dead when 
sampled: A = alive) Punta Cavendish, Puerto Deseado, 
Santa Cruz province, 2 D, collected in 5 m depth 
(MACN-In 31354); Patagonian coast, 1 D (MACN-In 
11385); Playa Cabo San "Pablo, Tierra del Fuego, 4 D 
(MACN-In 12532); Punta Sinaia, Tierra del Fuego, 
Expedicion Facultad, 5 D (MACN-In 12531); Ushi, 
Tierra del Fuego, 1 D (MACN-In 9441); Punta Marta, 
Rio Grande, Tierra del Fuego, 9 D (MACN-In 35113): 
Bahia San Sebastian, Tierra del Fuego, 1 D (MACN-In 
35393): San Sebastian south, Tierra’ del Fuego, 4 D, 
Exp. Facultad de Ciencias (MACN-In 12530); Play a del 
Rio Grande, Santa Cruz province, 1 D, (MACN-In 
6647); 52°20! S, 6S°18! W, Station 28, Cabo Virgenes. 


Table 1. Dimensions (mm) of some adults of Adelomelon ferussacii from MACN collection, 

Specimen 1 2 3 4 5 6 ¢ 8 9 10 11 12 13 
Sex 3 3 + 3 = 3 3 - Y 2 + 2 : 
Shell length (mm) (SL) 76.5 75 58 91 78 85 75 72 76 74.5 76 114 75 
Aperture length (AL) 60 57 46 69 60 64 59 56 58 58 56 S9 56.5 
AL/SL 0.78 0.76 0.79 0.76 0.77 0.75 0.79 0.78 0.76 0.78 0.74 0.78 0.75 
Spire angle (°) 80 78 77 83 82 S4 59 SO 83 78 Sl $5 82 
Radula length (RL) 10.5 11.3 9.5 12.2 12.] 13.1 11.4 10.4 9.6 11S 12.1 17.4 11.7 
Number of radular teeth 56 50 49 53 61 58 56 52 50 60 58 74 57 


Figures 2-17. 


Shells of A. ferussacii and A. brasiliana, 2-13. Adelomelon ferussacii (Donovan, 1824). 2-4. MACN-In 37015 


Puerto San Julian, Santa Cruz. 5-6. Holotype of Voluta rudis Griffith and Pidgeon, 1834, NHM 19920177. 7-10. Embryo shells (at 


hatching stage) MACN-In 3 


Buenos Aires. Scale bars = 10 mm. 


7014. Puerto San Julian, Santa Cruz. 11-13. Juvenile shells MACN-In 37012, 3 kin North of Puerto San 
Sebastian. Tierra del Fuego. 14-17. Adelomelon brasiliana (Lamarck, 1811), 


embryo shells MACN-In 37015, off Mar del Plata, 


Page 240 


THE NAUTILUS, Vol. 122, No. 4 


Santa Cruz province, A.R.A. Bahia Blanca, 1 D, in 11 m 
(MACN-In 24080): Bahia Laura, Puerto Deseado, Santa 
Cruz province, 1 D (MACN-In 9199-16); Bahia San 
Sebastian, Tierra del Fuego, 1 D (MACN-In 21154); 
Estancia “Viamonte”, Rio Grande, Tierra del Fuego, 
2 D (MACN-In 27219): Isla Quiroga, Puerto Deseado, 
Santa Cruz province, 1 D (MACN-In 26199); Punta 
Arenas, Chile, 1 D (MACN-In 9040-27); Rio Grande, 
Tierra del Fuego, 24 D (MACN-In 12529); Playa 
La Mina, Puerto San Julian, Santa Cruz Province (67° 
37’ W, 49° 09’ S), 10 A in low tide (MACN-In 37487). 


Distribution: Adelomelon ferussacii is a typical com- 
ponent of the Magellanic province, ranging from the 
province of Santa Cruz to Southern Chile. More north- 
ern citations of the species (e.g., Clench and Turmer, 
1964; Castellanos and Landoni, ve are here referred 
to A. brasiliana. No specimens of A. ferussacii from re- 
vised collections (MLP and MACN) were found outside 
of this range. 


Distribution According to Previous Records in 
the Literature: Puerto Gallegos, Punta Arenas, (Stre- 
bel, 1906); Santa Cruz coast, (Lahille, 1895); Southern 
Patagonia and Magellanic region (Carcelles and William- 
son, 1951); Golfo San Matias to Straits of Magellan 
(Clench ane Turner, 1964), however they only examined 
specimens from the acre of Puerto Deseado, Cabo 
Buen Tiempo, Rio Gallegos, Bahia de la Posesion and 
San Gregorio in Chile; Golfo San Matias to Magellan 
Straits (Weaver and DuPont, 1970); Santa Cruz (Castel- 
lanos, 1970b); Southern coast of Buenos Aires province 
to Magellan Straits (Castellanos and Landoni, 1992); 
Southeastern coast of Argentina, south to the Magellan 
Straits (Poppe and Goto, 1992). 


DISCUSSION 


Clench and Turner (1964) suggested the inclusion of 


Voluta ferussacii in the genus Adelomelon Dall, 1906, 
because of shell similarity with A. brasiliana. However, 
at that time they examined no complete specimens to 
confirm such generic placement. Weaver and Du Pont 
(1970) mentioned that no live specimens were collected. 
Later, Castellanos (1970) illustrated the radula confirm- 
ing the generic placement suggested by Clench and 
Turner (1964). However, data on the gross anatomy had 
not previously been reported. 

The last comprehe nsive taxonomic revision of the 
family Volutidae from the southwestern Atlantic was 
prepared by Clench and Turner (1964, 1970). They de- 
scribed the new subfamily Odontocymbiolinae and the 
new genus Odontocymbiola, and finally resolved the 
confusion of previous authors (e.¢., Pilsbry and Olsson, 
1954) between Adelomelon ancilla (Lightfoot, 1786) 
and Odontocymbiola magellanica (Gmelin, 1791). These 
authors also described two new species: O pescalia anc 
A. riosi (Clench and Turner, 1964). The latter was in- 


cluded in the new beenus Weaveria. After their work, 


21 


S 


Figures 20-21. Adelomelon ferussacii (Donovan, 1824). 20. 
Anterior alimentary system. 21. Dorsal view of head, siphon, 
and penis of a male specimen. ae, anterior esophagus; asg, 
accesory salivary gland; el, cephalic lappet; dasg, duct acce- 
sory salivary gl: ind; e, eye: gl, gland of Leiblein; nr, nerve ring; 
Ps penis; pd, penial duct: pe, posterior esophagus; rs, radular 
sac; s, siphon; sg, salivary gland; t, tentacle. 


several additional new species were described, in partic- 
ular from Brazil (e.g., Leal and Bouchet, 1989; Leal and 
Aios, 1990). 

Most of the literature dealing with southwestern At- 
lantic volutids is primarily taxonomic. Anatomical data 
are scarce and usually drawn from one or relatively few 
spe clmens, sometimes incomple te. Exce ptions are the 
papers by Novelli and Novelli (1982) and Aycgaguer 

(2002), pi articularly the latter, in sehich the authors de- 
scribe ( in some detail the an: itomy of Zidona dufresinei. 

Clench and Turner (1964) and Aycaguer (2002) men- 
tioned that a beckii, A. ancilla, A. brasiliana, 
and Zidona dufresnei all have characteristic racemose 
salivary glands loosely intertwined in the tubular 


M. E. Segade and G. Pastorino, 2008 Page 24] 


Figures 22-31. Radulae of Adelomelon species. 22-26. Adelomelon ferussacii (Donovan, 1824). 22. Frontal view. Scale bai 
250 um. 23. Lateral view. Scale bar = 200 um. 24-25. Rachidian teeth. Scale bar = 200 um. 26. Radula of an embryo. Scale bai 
50 um. 27-31. Adelomelon brasiliana (Lamarck, 1811). 27. Radula of an embryo. Scale bar = 50 um. 28-29. Rachidian teeth. Scale 


bar = 200 um. 30. Lateral view. Scale bar = 200 um. 31. Frontal view. Scale bar = 250 tm 


age IAI 
Page 242 


THE NAUTILUS, Vol. 122, No. 4 


Number of teeth 


0 20 40 60 80 100 120 
Shell length (mm) 


Figure 32. Relationship between radular and shell length 


(R = 0.9178). 


accessory glands. The radula in these species is also 
rachiglossate with a unique central tricuspid tooth. 
These characters, together with several shell similarities 
placed A. ferussacii in the Zidoninae subfamily. 

Radulae of the species A. ferussacii and A. brasiliana 
are quite similar (Figures 22 and 31). However, A. 
brasiliana has a wider and more convex base of the 
rachidian teeth than A. ferussacii (Figures 24 and 
28). Lateral cusps of the rachidian teeth are shorter 
and wider in A. brasiliana than in A. ferussacii. 
These differences are probably specific characters 
and are also present in the embryos of both species 
(Figures 26 and 27). 

The name Pachycymbiola was proposed by ther- 
ing (1907: 209) as a subgenus of Ade lomelon for 

A beatiand. which is aeuially the type species. Pilsbry 
ee Olsson (1954) and later Searabino et al. (2004) ) pro- 
moted Pachycymbiola to generic rank, and mentioned as 
main characters an ovate shell with a short spire, a free 
oval egg capsule, and a protoconch without calcarella. 
Del Rio and Martinez (2006) also treated Pachycym- 
biola at the generic rank following the latter authors. 
They described five Tertiary species, three new, under- 
this genus, pointing out as main differences from Adelo- 
melon the ovate shape of the shell, the short spire and 
the protoconch without calcarella, in agreement with 
Searabino et al. (2004). Adelomelon brasiliana has all 
these features and a large free ovoid ovicapsule, with 
9 to 33 embryos per capsule (Penchaszadeh and 
de Mahieu, 1976; Luzzatto, 2006). On the other hand, 
A. beckii and A. ancilla, which belong in the subgenus 
Adelomelon, have smaller egg capsules, always att: ache d 
to hard substrates (Penchaszadeh et al., 1999). Adelome- 
lon ferussacti presents egg capsules similar to those of 
\. beckii, which are globose hemispherical and flexible, 
with a white opaque color and a leathery texture, 
attached to a hard substrate such as stones or rocky 
substr. te Penchaszadeh and Segade, in preparation). 
The inclusion of A. ferussacii in the subgenus Pachy- 
suggested based on some external 


Srnbios vas first 


similarity to A. brasiliana. However the ege capsules do 


not match those of A. ferussacii. In fact they look more 
similar to those of other species of Adelome lon. Also, the 
gross anatomy revised here does not show conclusive 
fennie: to include A. ferussacii in a different group as 
suggested previously. The only character that could 
siacid as a difference are those of shell morphology. 
Pilsby and Olsson (1954), Scarabino et al. (2004) and 
Del Rio and Martinez (2006) concede that shell shape 
place A. ferussacii close to Pachycymbiola. They do not 
mention that a calearella is reduced, present in the 
young of A, fe russacti (Figures 7-13), but absent in 
those of A. brasiliana. At a closer look even the shell 
shape is substantially different, since A. brasiliana has a 
more flattened spire and pronounced shoulder tubercles 
more similar to A. becki than A. ferussacii. Further de- 
tailed anatomical studies will confirm the relationships 
of these Southwestern Atlantic endemic volutids. 


ACKNOWLEDGMENTS 
We thank M. Griffin (UNLPam) and F. Searabino for 


the suggestions that considerably improved the manu- 
script. Bronwyn Gillanders helped to improve the 
English. Two anonymous reviewers made useful sugges- 
tions on an early draft of the manuscript. This contribution 
was supported by Project PICT No. 03-14419 from the 
National Agency for Scientific and Technical Promotion, 
Argentina. We acknowledge funding by the Consejo 
Nacional de Investigaciones Cientificas y Técnicas 
(CONICET) of Argentina, to which GP belongs. 


LITERATURE CITED 


Aycaguer, C. sts Anatomia de Volutas del Atlaintico sudocci- 
dental, I: Anatomia general y del sistema reproductor de 
Zidona ‘een Donovan, 1823) (Neogastropoda: Volu- 
tidae), Comunicaciones de la Sociedad Malacologica del 

Uruguay 8(76-77): 159-180. 

Carcelles, A. and S. Williamson. 1951. Catalogo de los moluscos 

sce de la provincia magallanica. Revista del Instituto 

Nacional de Investigacién de las Ciencias Naturales 2, 

Ciencias Zoologicas: 225-383. 

Castellanos, Z. J. A. de. 1970. Reubicacion de algunas especies 

de Volutidae del Mar Argentino. Neotropica 16: 1-4. 

Castellanos, Z. J. A. de and N. Landoni. 1992. Catalogo 
descriptivo de la malacofauna marina magallanica LO. Neo- 
gastropoda, Volutidae, Volutomitridae, Cancellariidae, 
Olividae y Marginellidae. Comision de Investigaciones 


Cientificas de la Provincia de Buenos Aires, La Plata, 37 pp. 

Clench, W. J. and R. D. Turner, 1964. The subfamilies Voluti- 
nae, Zidoninae, Odontocymbiolinae on Calliotectinae in 
the Western caer Jobnsonia, 4(43): 129-180. 

Clench, W. J. and R. D. Turner. 1970. a family Volutidae in 
the Mah m eee Johnsonia 4: 369-372. 

Dall, W. H. 1907. A review of the American Volutidae. Smith- 
tie Miscellaneous Collections 48(3): 341-3735 

del Rio, C. J. and S. A. Martinez. 2006 The family Volutidae 
(Mollusca-Gastropoda) in the Tertiary of Patagonia 
(Argentina). Journal of Paleontology SO: 919-945 

Donovan, E. 1824. The naturalist’s repository or miscel- 
lany of exotic natural history, exhibiting specimens of 


M. E. Segade and G. Pastorino, 2008 


Page 243 


foreign birds, insects, shells. Sheet EE, FF, pp. 214-216, 
Plate 67. 

Giménez, J. and P. E. Penchaszadeh. 2002. Reproductive cycle 
of Zidona dufresnei (Caenogastropoda: Volutidae) from 
the southwestern Atlantic Ocean. Marine Biology 140: 
755-761. 

Griffith, E. and E. Pidgeon. 1834. The animal kingdon arranged 
in conformity with its organization by the baron Cuvier. The 
Mollusca and Radiata. London, Whittaker & Co., 601 pp: 

Ihering, H. von. 1907. Les Mollusques fossiles du Tertiaire et 
du Cretace superieur de large sntine. Anales del Museo 
Nacional de Buenos Aires (3)7: 1-611. 

Lahille, F. 1895. Contribucion al eens de las volutas argen- 
tinas I. Morfologia externa. Revista del Museo de La Plata 
6: 293-333. 

Leal, J. H. and P. Bouchet. 1989. New deep-water Volutidae 
from off southeastern Brazil (Mollusca: Gastropoda). The 
Nautilus 103: 1-12. 

Leal, J. H. and E. de C. Rios. 1990. Nanomelon vossi, a new 
deep-water Zidoninae from off southern Brazil (Gastro- 
poda: Volutidae). The Veliger 33: 317-320. 

Luzzatto. D. C. 2006. The biology and ecology of the giant free 
egg capsules of Adelomelon brasiliana Lamarck, 1811 
(Gastropoda: Volutidae). Malacologia 49: 107-119. 

Novelli, R. and A. U. G. Novelli. 1982. Aleumas consideragdes 
sobre a subfamilia Zidoninae e notas sobre a anatomia 
de Adelomelon brasiliana (Lamarck, 1811), Mollusca, 
Gastropoda, Volutidae. Atlantica 5: 23-34. 

Penchaszadeh, P. and G. C. De Mahieu. 1976. Reproduccion 
de gaster6podos Prosobranquios del oe surocciden- 
tal. Volutidae. Physis A, 35(91): 145-15 


Penchaszadeh, P. E., P. Miloslavich, M. Lasta, and P. M. S. 
Costa. 1999. Egg capsules in the genus Adelomelon 
(Caenogastropoda: Volutidae) from the Atlantic coast of 
South America. The Nautilus, 113(2): 56-63. 

Pilsbry, H. A. and A. A. Olsson. 1954. Systems of the Volutidae. 
Bulletins of American Paleontology, 35(152): 5-36. 

Poppe, G. T. and Y. Goto. 1992. Volutes. Mostra Mondiale 
Malacologia Cupra Marittima (AP-Italy). L’Informatore 
Piceno Ed., Ancona, 348 pp: 

Reeve, L. 1849. Conchologia Iconica: or, illustrations of the 
shells of moluscous animals. London. 

Scarabino, F., S. Martinez, C. J. del Rio, A. E. Oleinik, H. H. 
Camacho, and W. J. Zinsmeister. 2004. Two new species 
of Adelomelon Dall, 1906 (Gastropoda: Volutidae) from 
the Tertiary of Patagonia (Argentina). Journal of Paleon- 
tology 78: 914-919. 

Sowerby, G. B. II. 1846. Descriptions of Tertiary fossil shells 
from South America. In: Darwin, C. Geological Observa- 
tions on South America. Smith Elder Co., London, 
249-267. 

Solem, A. 1972. Malacological application of Scanning Elec- 
tron Microscopy, H. Radular structure and functioning. 
The Veliger, 14: 327-336, 

Strebel, H. 1906. Beitriége zur Kenntnis der Molluskenfauna 
der Magalhaen-Provinz. 4. Zoologischen Jahrbuchern. 
Abteilung fur Systematik, Geographie und Biologie der 
Tiere, oA 91-174. 

Weaver, C. S. and J. E. Dupont. 1970. Living volutes. A mono- 
graph of the Recent Volutidae of the world. Delaware 
museum of Natural History. Monograph Series No. 1, 
Delaware, xv + 375 pp. 


THE NAUTILUS 122(4):244-251, 2008 


Page 244 


A new species and a new record of Muricidae (Gastropoda) 
from Brazil: genera Pterynotus and Leptotrophon 


Alexandre Dias Pimenta 
Diogo Ribeiro do Couto 
Paulo Marcio Santos Costa 


Departamento de Invertebrados 

Museu Nacional, Universidade Federal do Rio de Janeiro 
20940-040 Rio de Janeiro, BRAZIL 
adpimenta@yahoo.com.br 


ABSTRACT 


Leptotrophon atlanticus, a new species of Leptotrophon 
Houart, 1995, a genus of Trophoninae previously known only 
from the Indo-Pacific Ocean, is described from the northeast- 
ern coast of Brazil, from depths of 240-260 m. Leptotrophon 
atlanticus is very similar to Leptotrophon kastoroae Houart, 
2001, but has shorter spines, which are brown colored. The 
shell microsculpture of L. atlanticus was studied using scan- 
ning electron microscopy, which revealed a protoconch slightly 
granulated on the last quarter of the whorl and a teleoconch 
with growth striae crossed by regularly spaced zigzag spiral 
lines and axial microstriae. The 1 cadule of L. ailantio us fits the 
pattern described for the genus, but the lateral/marginal cusps 
of the rachidian teeth are not as independent of each other, 
being somewhat fused into a common base. This study reports 
the first record from the South Atlantic Ocean of the Murici- 
nae Pterynotus havanensis Vokes, 1970, previously known only 
from the Caribbean region. 


Additional keywords: Muricinae, Trophoninae, South Atlantic, 
biodiversity, tasonomy 


INTRODUCTION 


The family Muricidae is represented in Brazil by 49 
species from six subfamilies, of which members of the 
subfamily Muricinae are most numerous, with 24 spe- 
cies in nine genera (Rios, 1994). Other genera of Mur- 
icinae recorded from the western Atlantic have never 
been recorded from Brazil until now. This includes the 
genus Prerynotus, which is found in deeper-water habi- 
tats (Vokes, 1970), with 
Florida and the 


eight species reported from 
Caribbean (Rosenberg, 2005), in addi- 
tion to mention records from other regions of the world, 
e.g., Ponder (1972) from Australia, Bouchet and Warén 


(1985) from the northeast Atlantic, and Houart (2001) 
from New Caledonia. 

Phe family Trophoninae includes 18 genera 
Vaught, 1989). It is represented in the western Atlantic 


by five genera, of which only Trophon Montfort, 1810, 
occurs in Brazil (Rios, 1994). More recently, Houart 
(1995) described the genus Leptotrophon, whose 26 spe- 
cies are restricted to the Indo-Pacific region, in New 
Caledonia (Houart, 1995, 2001) and eastern Indonesia 
(Houart, 1997). 

This report presents the first record of the genus 
Pterynotus Swainson, 1833 from Brazil, based on Ptery- 
notus havanensis Vokes, 1970, as well as the first record 
of the genus Leptotrophon from the Atlantic Ocean, 
based on a new species described from northeast Brazil. 


MATERIALS AND METHODS 


This report is based on a sample collected on the Cano- 
pus Bank, State of Ceara, in 2005, from a biogenic sub- 
stratum, in 240-260 m depth. Identification of the 
species was based on original and subsequent descrip- 
tions. In cases where fragme nts of soft parts were avail- 
able, an attempt was made to record some anatomical 
characteristics using camera lucida, especially of the 
head-foot and the palli ial cavity. Radulae were prepared 
for SEM according to the methodology described by 
Bandel (1984). 


SYSTEMATICS 


Family Muricidae Rafine asque, LS15 
Subfamily Muricinae Ratine sque, IS15 
Genus Pter ynotus Swainson, 1$35 


Type Species: = Murex pinnatus Swainson, 1$22 by 
subsequent designation. 

Pterynotus havanensis Vokes, 1970 

(Figures 1-12) 

Pterynotus (Pterynotus) havanensis Vokes, 1970 (new 


name for Murex tristichus Dall, 1889, non Murex 
tristichus Beyrich, 1854.) 


\. D. Pimenta et al., 2008 


a a 


Figures 1-8. Pterynotu havanensis Vokes, 1970: 1—4, 7-8. MNR]J 11057 lensth 24.4 mm, width = 16.5 mm. 5-6. MNR] 
11003. 1. Shell in apertural view. 2. Shell in abapertural viev oy Shell in apical vie w. 4. Shell in lateral view. 5. Protoconch. 6. Detail 
f tel I | ». 7. Radulae in dorsal view. 8. Detail of rachidian teeth. Scale bars: 5-6 LOO pum; 7-5 10.0 tum 


Murex (Pteronotus ) tristichus Dall, 1889: 202 pl 15, fig. 2). Pterynotus phan us auct Abbott, 1974: 175, fig. 1856 
Pteropurpura tristic ha.—Dall, 1927: 58 non Dall, 1889 
Mure Pterynotus) tristichu Clench and Farfante Pterynotus phaneus Dall—Radwin and D’Attilio, 1976 


1945: 36, pl. 20, igs. 14 100 (in part 


Page 246 THE NAUTILUS, Vol. 122, No. 4 


Figures 9-12. Pterynotus havanensis Vokes, 1970 (MNRKJ 11057). 9. Head-foot in apertural view, female, 10. Operculum in inner 
view. 11, Operculum in outer view. 12. Palial cavity, female. Scale bars = 1.0 mm. Abbreviations: an, anus; em, columellar muscle; 
en, ctenidium; ct, cephalic tentacle; ey, eye; hg, hypobranchial gland; mb, mantle border; op, operculum; os, osphradium; ov, 
oviduct; si, siphon. 


A. D. Pimenta et al., 2008 


Page 247 


Pterynotus havanensis Vokes—Fair, 1976: 47, pl. 18, 
fig. 164. 

errs (Pterynotus ) phaneus auct.—Harasewych 
and Jensen, 1979: 4, fig. 3 (in part, non Dall, 1889). 


Description: Shell of moderate size (up to 25 mm 
in length), fusiform, thin; color white. Protoconch 
paucispiral, with 1.5 whorls, glassy, slightly bulbous, 
ending in a thin varix. Teleoconch with up to five 
whorls, somewhat convex in outline, with three thin, 
flaring, backward-curved, wing-like varices per whorl, 
with 5-6 digitations along margin of last whorl varices. 
Axial sculpture consisting of fine growth striae, more 
visible in ventral view of varices, where they cross with 
spiral threads, giving it a foliaceous appearance; no 
intervarical nodes present. Spiral sculpture obsolete, 
barely discernible spiral threads, formed by microscopic 
shallow furrows, somewhat undulated and irregularly 
spaced; abapertural view of varices with five-six spiny 
digitations per varix on the last whorl and two digitations 
on spire varices; spine on shoulders of whorls largest. 
Aperture oval, peristome slightly projected, inner lip 
reflected, attached posteriorly, smooth, with small 
undulations corresponding to varical digitations; outer 
lip smooth. Siphon canal rather long, sharply bent aba- 
perturally and to the right side; narrowly open; with 
previous, old imbricate canal termination, recurved to 
the left side. 


RapuLa (FEMALE): Rachiglossate type; radula ribbon 
long and narrow; rachidian teeth with five pointed 
cusps, central cusp the largest, the second largest at the 
margins, each area between central and mar anal cusps 
with an acute tiny cusp; marginal edge rectangular and 
somewhat pointed, base concave: lneral teeth sickle- 
shaped, broader at the base. 


Heap-Foor (FEMALE): Head poorly differentiated, eyes 
located laterally and in middle of somewhat long cephal- 

ic tentacles. Foot large. OPERCULUM horny, ov ate, cover- 
ing entire shell aperture; external surface with terminal 
nucleus and concentric growth lines; imner surface 

attachment area with single horseshoe-shaped scar, not 
positioned centrally, covering less than 50% of total area 
of operculum, with one adventitious layer. 


PaLLIAL Cavity (FEMALE): Mantle border smooth; siphon 
long and narrow, with smooth border, muscle attach- 
ment reaching as far as gill. Osphradium bipectinate, 
unequal, right side about 1.5 times as wide as left, broad 
and long (about half total ctenidium length). Ctenidium 
monopectinate, long and narrow. Hypobranchial gland a 
thin. poorly dev eloped glandular mass covering mantle 
between anterior end on gill and oviduct. Oviduct occu- 
pying about half of pallial cavity length, broad. 

Type Locality: Blake Station 51, off Havana, Cuba, 
between 445 m and $23.5 m. 

Material Examined: MNRJ 11003 (one individual 
and seven empty shells); MNRJ 11067 (one individual); 


MNR] 11057 (one individual); all from Canopus Bank, 
96 miles off Ceara State, 240-260 m depth, from bio- 
genic substratum. 


Geographic Distribution: Off Georgia and Fernan- 
dina (Dall, 1927), Key West (Fair, 1976) Florida, USA; 
Havana, Cuba (type locality); Pleistocene Moin Forma- 
tion, Costa Rica (Vokes, 1992): Golfo of Uraba, Colom- 
bia (Vokes, 1992); Ceara State, Brazil (this study). 


Discussion: Pterynotus havanenis Vokes, 1970, was 
originally described as Murex (Pteronotus) tristichus 
Dall. 1889. The taxon was later included in a long list of 
synonyms of Pterynotus phaneus (Dall, 1889) by Hara- 
sewych and Jensen (1979), based on a wide variation 
found in several shell characteristics such as the axial 
sculpture, intervarical nodes and outer lip. Vokes (1992) 
revalidated P. havanensis, based on the more numerous 
varices on the early teleoconch whorls and the smoother 
shell surface. 

The only available published illustrations of P. hava- 
nensis are those of the holotype (Dall, 1889: pl. 15, fig. 2; 
Clench and Farfante, 1945: pl. 20, figs. 1-4; Vokes, 1970: 
pl. 3, figs. la,b; Abbott, 1974: fig. 1856; Fair, 1976: pl. 13, 
fig. 164. Hasisenyel and Jensen, 1979: fig. 3; Vokes, 
1992: pl 2, fig. 3), in addition to a Pleistocene fossil 
specimen from Costa Rica (Vokes, 1992: pl. 2, fig. 6). 
This study provides the first illustrations of Recent 
specimens (Figures 1-4), except for that of the holotype 
itself. 

The Brazilian specimens herein studied, all from 
Canopus Bank, about 96 miles off the coast of the State 
of Ceara (240-260 m depth), are very similar to the 
holotype. illustrations, ve little sculpturing between 
the varices (Figures 1-2, 4), without the intervarical 
nodes described for P. phaneus, in addition to axial and 
spiral g growth lines, the digitations on the margins of the 
varices and no denteulae outer lip (Figure 1). These 
characteristics, along with the shape sa sculpture of 
the varices, clearly distinguishes P. havanensis from 
P. phaneus. 

Vokes (1992) stated that both P. phaneus and P. hava- 
nensis have denticulations on the inner side of the outer 
lip. However, this detail is not stated in the original or 
subsequent descriptions of this species, as well as in the 
holotype illustrations, which show a smooth outer lip, as 
well as in the specimens from Brazil. Such denticles 
reported by Vokes (1992) probably correspond to the 
undulations by the varical digitations. In addition, Vokes 
(1992) considered that P. hasoneneis bears several vari- 
ces on the early teleoconch whorls, making reference to 
the figure in Harasewych and Jensen (1979). Brazilian 
specimens bear the usual three varices on the first tele- 
oconch whorls. 

The radula herein illustrated ( 
minor differences from the radula of P. phaneus: illu- 
strated in Harasewych and Jensen (1979: 15, fig. 17). 
In P. havanensis, the central and lateral cusps are longer 
in relation to the total length of the tooth; also, the two 
inner cusps are narrower and shorter. 


Figures 7-8) has some 


THE NAUTILUS, Vol. 122, No. 4 


Subfamily Trophoninae Cossmann, 1903 
Genus Leptotrophon Houart, 1995 


Type Species: Leptotrophon caroae Houart, 1995, by 
original designation. Recent, New Caledonia. 


Leptotrophon atlanticus new species 

(Figures 13-26) 

Diagnosis: Shell densely spiny; color cream white with 
brown spines; paucispiral protoconch with slight granu- 
lated microsculpture on last quarter of last whorl. 


Figures 13-20.  Leptotrophon atlanticus new species. Holotype, MNRJ 11004, length 


14. Abapertural vi 


20. Detail of lateral teeth. Scale bars 


Description: Shell up to 8.5 mm in length (holotype), 
biconic, and densely spiny; spire high. Color creamy 
white, with light brown spines in live specimens. Proto- 
conch rounded, paucispiral, with 1.50 to 1.75 whorls, 
with slightly granulated microsculpture, forming faded 
spiral cords on last quarter of last whorl; terminal varix 
low. Teleoconch with up to 4.25 slightly shouldered 
whorls. Suture impressed. Axial sculpture consisting of 
numerous weak, orthocline lamellate growth striae and 
spiny varices, regularly spaced, forming spines at inter- 
ceptions of spiral cords. Spiral sculpture on spire of two 


8.8 mm, width 


{6 mm). 13. \pertural 
15. Apical view. 16. Operculum in inner view. 17. ( Yperculum in outer view. 18. Radulae in dorsal view 


16-17 = 500 tm ; 1S—20 = 10 jum 


\. D. Pimenta et al., 2008 


Figures 21-26. Leptotrophon atlanticus new species. Paratype, MNRJ 11009, length = §.2 mm, width = 4.4 mm. 21. Abapertural 
view. 22. Detail of sculpture on teleoconch whorl. 23. Detail of microsculpture on teleoconch whorl. 24. Protoconch. 25-26. Detail 


ot protoconch sculpture Seale bars = 100 tm 


strong cords, the adapical one located at 1/3 of whorl 
height below the suture; interceptions between axial var- 
ices and spiral cords form two primary spines regularly 
arranged in spiral crowns, each crown with 9-10 spines 
on last whorl of holotype: spines tall, channeled, com- 
monly adapically and backward-curved, slightly larger at 
shoulder: Spaces between adjacent spines have a squa- 
mous appearance, due to crossing of spiral cords with 
axial growth striae: last whorl with four additional spiral 
spiny crowns, regularly arranged at the base, along with 
corresponding spiral cords; secondary squamous small 
spines appear irregularly arranged, interspaced with pri- 
mary spines; holoty pew ith a secc mdary spiral cord appear- 


ing between the two pmmary ones closest to the suture 
on the last half of the last whorl! Microsculpture formed by 
growth striae crossed by regularly spaced ZIPZA spiral 
lines and axial microstriae \perture slightly oval, rounded 


adapically with about 1/3 of total shell height. Columellar 


ip flaring, smooth and adapically adherent. Anal notch 
indistinct. Outer lip smooth, fragile, primary spiral cords 
and growth striae visible within. Siphon canal long, nar- 
‘ow, bent backwards, narrowly open and smooth, with 
seven imbricate old canal terminations 


Raputa (FEMALE): of rachiglossate type; radula ribbon 
ong and narrow; rachidian tooth trapezoidal with five 
pointed cusps; the two lateral cusps adjacent to central 


Cc 


sp somewhat fused into a bifurcated cusp, outermost 
cusps slightly larger; marginal edges rectangular well 


pointed, area between outer cusps and marginal edges 
with two very small folds; base somewhat sinuous; lateral 
teeth sickle-shaped broader at base 


OrercuLuM horny, elliptical, covering entire shell ape 
ture: external surface with terminal nucleus and concen 


tric growth lines: inner surface attachment area with 


single horseshoe-shaped scar, not positioned centrally 


Page 250 


THE NAUTILUS, Vol. 122, No. 4 


covering about 50% of total area of operculum, with one 
adventitious layer. 


Type Material: Holotype: MNRJ 11004 (8.5 mm); 
three paratypes: MNRJ 11009, all from type locality. 


Type Locality: Canopus Bank, 96 miles off Ceara 
State, 240-260 m depth, from biogenic substratum. 


Distribution: Known from type locality only. 


Discussion: The allocation of Leptotrophon in 
Trophoninae follows Houart (1995), who, when describ- 
ing Leptotrophon, stated that he was considering 
Trophoninae in a traditional way, to include typical “Tro- 
phon-like” species, diagnosed. as small, thin spinose 
hells with a flaring columellar lip. In fact, Kool (1993) 
stated that Trophoninae is probably a non-monophyletic 
group and, thus, Houart (1995) considered — that 
future studies would probably show that the genus 
Le ptotrophon would have to be transferred from the 
Trophoninae. 


The new species fits very well in the diagnosis of 


Leptotrophon, and is very similar to several species from 
the Indo-Pacific, including the type species, L. caroae. 

Leptotrophon otlantious bears the characteristic spiny 
sculpture (Figures 13-15), round-ovate aperture, and 
flaring columellar lip (Figure 13). The radula of L. atlan- 
ticus (Figures 15-20) fits the pattern described for Lep- 
totrophon, but the lateral/marginal cusps are not as 
independent of each other, being somewhat fused into a 
common base (Figure 19). In other ae referred to 
Leptotrophon by Houart (1995), the lateral/marginal 
cusps are similar to L. atlanticus [e.g., L. caroae and 
L. acerapex (Houart, 1986). In addition, two small mar- 
ginal denticles may be seen on the outer base of the 
marginal cusps (white arrow in Figure 19), a characteris- 
tic found in other species of Leptotrophon described by 
Houart (1995). 

The protoconch of the species described by Hoaurt 
(1995) shows considerable variation in’ shape, with 
rounded-globose, acuminate, or carinate protoconchs 
present in different species. The type species has a cari- 
nate protoconch, and Leptotrophon atlanticus has a 
rounded-globose protoconch (Figure 24). 

The most similar species are kastoroae Houart, 
1997, and L. perclarus Houart, 2001. Both species are 
larger than L. atlanticus. Leptotrophon perclarus has a 
taller spire, more acuminate apex, and more inflated last 
whorl: mM addition it has more numerous and smaller 
spines. The shell shape of L. kastoroae is almost identi- 
cal to that of L. atlanticus, but the spines are longer and 
are more upward-curved. 

Prey iously to this work. no record of shell microsculp- 
ture for any species o ; Leptotrophon was available. The 
protoconc th is always described as smooth. In fact, the 
protoconch of L atlantic us has an overall smooth ap- 
pearance (Figure 24), but SEM reveals that only its most 
apical region is smooth; the last quarter area, close to the 
varix, bears spiral sets of microscopic granulations (Fig- 


ures 25-26 \ delicate microsculpture also covers the 


entire teleoconch surface, including the spiral cords and 
spines, with axial and spiral microstriae forming a some- 
what reticulate pattern (Figures 22-23). 

The operculum of L. atlanticus is similar in its outer 
surface to those illustrated by Houart (1995) for species 
of Leptotrophon from the New Caledonian region. 
Houart (1995) did not describe the operculum of Lepto- 
trophon internally; the operculum of L. atlanticus has 
one adventitious layer. 

In spite of the several common characteristics in the 
shell and radulae morphology of Leptotrophon atlanti- 
cus and the species from the Indo-Pacific, the generic 
allocation herein used should be considered as provi- 
sional, due to the wide geographic separation between 
the new species and the other species in the genus 
Leptotrophon, most of which bearing protoconchs 
that indicate non-planktotrophic development. All 26 
previously described species of Leptotrophon are re- 
stricted to the Indo-Pacific (Houart, 1995; 1997: 2001); 
Leptotrophon atlanticus new species is the first record 
of this genus outside that region. The bathymetry 
of L. atlanticus, on the other hand, falls within the 
bathymetric range reported for the Indo-Pacific species 
(200-720 m). In ae future, direct comparisons with the 
Indo-Pacific species, especially including characters 
visible only under SEM, would be most helpful to estab- 
lish the degree of similarity among these species. 


ACKNOWLEDGMENTS 


We are grateful to Mr. Roland Houart (Institute Royal 
des Sciences Naturelles de Belgique) for comments on 
the taxonomy of the species. Vinicius Padula (MNRJ) 
and Franklin Noel dos Santos (Universidade Federal do 
Para) provided additional bibliography. Dr. Janet Reid 
revised the English text. Dra. Noemia Rodrigue ’s helped 
with the SEM photos. We also thank two anonymous 
reviewers for their comments and suggestions. Special 
thanks are due to Mr. Antonio Gil Bezerra and Ms. Elisa 
Gradvohl Bezerra, owners of INACE Shipyard (Indtis- 
tria Naval do Ceara) for the loan of the fishing boat and 
to Mrs. José and Marcus Coltro for financial support to 
the collecting efforts. Conselho Nacional de Desenvol- 
vimento Cientifico e Tecnolégico (CNPq) provided 
PROTAX pos-doctoral grant to P.M. Costa. 


LITERATURE CITED 


Abbott, R. T. 1974. American Seashells. 2nd ed. Van Nostrand 
eles Co., New York. 663 p., 24 pls. 

Bandel, kK. 1984. The radulae of Caribbean and other Meso- 
shea da and Neogastropoda. Zoologische Verhandelin- 
gen 214: 1-188 

Bouchet, P. and A. Warén. 1985. Revision of the Northeast 
Atlantic Bathyal and Abyssal Neogastropoda Excluding 
Turridae (Mollusca: Gastropoda). Societa [Italiana di 
Malacologia (Bolletino Malacologico), Supplemento 1: 
123-296 


A. D. Pimenta et al., 2008 


Page 25] 


Clench, W. J. and I. Pérez Farfante. 1945.The genus Murex in 
the Western Atlantic. Johnsonia 1(17) 1-58. 

Dall, W. H. 1889. Reports on the results of dredging, under 
the supervision of Alexander Agassiz, in the Gulf of Mex- 
ico (1877-78) and in the Carribean Sea (1879-1880), by 
the U. S. Coast Survey Steamer “Blake”, Lieutenant- 
Commander C.D. Sigsbee, U.S.N., and Commander J.R. 
Bartlett, U.S.N.. commanding. XXIX. Report on the 
Mollusca. Part I] Gastropoda and Scaphopoda. Bulletin 
of the Museum of Comparative Zoology 1S: 1492. 

Dall, W. H. 1927. Small shells from dredgings off the southeast 
coast of the United States by the United States Fisheries 
steamer “Albatross” in 1885 and 1886, Proceedings of the 
Unieted States National Museum 70(2667): 1-134. [72-85] 

Fair, R. H. 1976. The Murex Book. Published by the author. 
138 pp., 23 pls. 

Harasewych, M. G. and R. H. Jensen. 1979. Review of the 
subgenus Pterynotus (Gastropoda: Muricidae) in the 
Western Atlantic. Nemouria 22: 1-16. 


Houart. R. 1995. The Trophoninae (Gastropoda: Muricidae) of 


the New Caledonian Region. Mémoirs du Muséum 
national d’Histoire naturelle 14(167): 459-498. 

Houart, R. 1997. Mollusca, Gastropoda: The Muricidae Col- 
lected During the Karubar Cruise in Eastern Indonesia. 
Mémoirs du Muséum national d'Histoire naturelle 16 
(172): 287-294. 

Houart, R. 2001. Igensia gen. nov. and eleven new species 
of Muricidae (Gastropoda) from New Caledonia, 
Vanuatu, and Wallis and Futuna Islands. pp. 243-269. In: 
Bouchet. P. and B. A. Marshall. Tropical Deep-Sea 


Benthos vol. 22. Mémoirs du Muséum national d’ Histoire 
naturelle, Paris, 406 pp. 

Kool, S. P. 1993. The systematic position of the genus Nucella 
(Prosobranchia: Muricidae: Ocenebrinae). The Nautilus 
107: 43-57. 

Ponder, W. F. 1972. Notes on some Australian genera 
and species of the family Muricidae (Neogastropoda). 
Journal of the Malacological Society of Australia 2: 
215-248. 

Radwin, G. E. and A. D’Attilio. 1976. Murex Shells of the 
World. Stanford University Press, Stanford, xi + 254 pp., 
32 pls. 

Rios E. de C, 1994. Seashells of Brazil. 2nd ed. Museu Ocea- 
nogratico Prof. E.C, Rios da Fundagao Universidade de 
Rio Grande, Rio Grande. 368 p., 113 pls. 

Rosenberg, G. 2005. Malacolog 4.1.0: A Database of Western 
Atlantic Marine Mollusca. [WWW database (version 
4.1.0)] URL http:/Avww. malacolog.org/. 

Sunderland, K. and L. Sunderland. 1992. Caribbean Murici- 
dae part Il. American Conchologist 20(3): 14-15. 

Vaught, K. C. 1989. A Classification of the Living Mollusca. 
American Malacologists, Inc. Melbourn. 195 p. 

Vokes, E. H. 1970. Cenozoic Muricid of the Western Atlantic 
Region Part V - Pterynotus and Poirieria. Tulane Studies 
in Geology and Paleontology §: 1-50. 

Vokes, E. H. 1992. Cenozoic Muricidae of the Western 
Atlantic region. Part IX - Pterynotus, Poirieria, Aspella, 
Dermomurex, Calotrophon, Acantholabia, and Attiliosa; 
additions and corrections, Tulane Studies in Geology and 
Paleontology 25: 1-108. 


THE NAUTILUS 122(4):252-258, 2008 


Page 252 


A new species of Chlamydoconcha Dall, 1884, from southeastern 
Brazil (Bivalvia: Chlamydoconchidae) 


Luiz Ricardo L. Simone 

Museu de Zoologia da Universidade de Sao Paulo 
Caixa Postal 42494 

04299-970 Sao Paulo, BRAZIL 

Irsimone@usp.br 


ABSTRACT 


The second species in the genus Chlamydoconcha _ is 
described. Chlamydoconcha av alvis new species, occurs off 
the coast of Rio de Janeiro coast, in southeastern Brazil. The 
new species has very reduced valves and a mantle surrounding 


the entire body, two features of the genus. The outer surface of 


the mantle lacks papillae except for a single one located close 


to the excurrent siphon. These are distinctive characters of 


Chlamydoconcha orcutti Dall, 1884, from the eastern Pacific 
coast of North America, the single other known species of the 


genus. Some of the more interesting anatomical characters of 


the new species are: posterior pair of retractor muscles of foot 
free from valves, absence of adductor muscles, gastric style sac 
totally separated from intestine, and the presence of a single 
(excurrent) siphon. 


Additional keywords: Anatomy, western Atlantic, Rio de 
Janeiro 


INTRODUCTION 


The genus Chlamydoconcha Dall, 1884 (type species by 
original designation: C. orcutti Dall, 1884) has been 
known to be monotypic. Chlamydoconcha orcutti occurs 
from California to western Mexico (Carlton, 1979: 
Morton, 1981). The species is Dennen: by reduction 
of the shell, which is restricted to the anterior region of a 
spherical mantle cover; the mantle outer surface has 
many, somewhat equidistantly distributed papillae. After 


the original description, further anatomical studies of 


C. orcutti were done by Bernard (1897) and Morton 
(1981). 

\ sample collected by biologist Vinicius Padula on the 
coast of Rio de Janeiro was sent to the author for study. 


The analysis of the material revealed a new spe cles of 


Chlamydoconcha, formally described herein. This paper 
is also the first discovery of the genus in the Atlantic 
Ocean, representing the second known species in the 
genus. The present description also includes a detailed 
matomy, which is discussed in comparison to C. orcutti 
Bernard, 1897: Morton, 1981) 


The taxonomic allocation of the genus Chlamydo- 
concha has been problematic. It has been included in 
the Galeommatidae (Morton, 1981), but full family 
status has been assigned (Chlamydoconchidae, Bernard, 
1983), as part of the Galeommatoidea. Full superfamily 
status was also considered (Chlamydoconchacea, Keen, 
1969). The Galeommatoidea, are mostly mollusks with 
usual bivalve shells, but may also include highly modi- 
fied, slug-like animals, with internal and reduced shells. 


MATERIALS AND METHODS 


The specimen was delivered preserved in 70% EtOH, 
A photo of the living specimen was taken before pre- 
servation. The dissection of the preserved animal was 
performed by standard techniques, under a_ stereo 
microscope, with the specimen immersed in the alcohol. 
All dissection steps were also photographed (e.g., 
Figures 3-5), Drawings were made with aid of a camera 
luc a, 

Abbreviations used in figures are: an, anus; au, auri- 
cle; by, byssal gland: ee, vill ciliary connection; ce, cere- 
bral ganglion: co, ce webieos -visceral connective; dd, ducts 
to digestive diverticulae; dh, dorsal hood; di, inner 
demibranch; do, outer demibranch; es, esophagus; fg, 
gill food groove; fm, posterior foot retractor muscle; fr, 
anterior foot retractor muscle; ft, foot; ga, genital aper- 
ture; gi, gill; go, gonad; gs, gastric shield; in, intestine; 
ip, inner hemipalp; ki, kidney; mb, mantle border; mo, 
mouth; mt, mantle: ne, nephropore; op, outer hemi- 
palp; pa, pedal aperture of mantle; pe, pericardium; 
pg. pedal ganglia; pl, pallial papilla; pm, pallial muscles: 
pp. palp; rt, rectum; sh, shell; si, excurrent siphon; 
ss, style sac; st, stomach; ty, typhlosolis; ve, ventricle: 
vg, visceral ganglia; vm, visceral mass. 

Institutional abbreviation: MZSP; Museu de Zoologia 
da Universidade de Sao Paulo, Brazil. 


SYSTEMATICS 


Chlamydoconcha avalvis new species 
(Figures 1-20) 


L. R. L. Simone, 2008 Page 253 


Figures 1-9. Chlamydoconcha avalvis Holotype photos. 1-2. Living specimen dorsal and lateral views, photo Vinicius Padula 
3-5. Preserved specimen. 3. Dorsal view. 4. Ventral view. 5. Left view, right mantle lobe partially removed and deflected anterior! 
right gill deflected upM ards. 6. Right valve, outer view (transversal section artificially done). 7. Same, inner view. 8. Left valve, inne: 


view. 9. Right valve, ventral view of its posterior, concave region Scale bars 15 = 2 mm: 6-9 = 0.5 mm 


Page 254 


THE NAUTILUS, Vol. 122, No. 4 


10 


Figures 10-13.) Chlamydoconcha avalvis anatomy. 10. Left view of entire animal, right mantle lobe partially removed and 
deflected anteriorly (right in Figure). 11. Region of right valve, internal surface of mantle removed, showing pallial muscles (pm) 
originating in valve, some adjacent structures also shown, 12. Gill, transversal section in its middle region. 13. Right palp, outer 
hemipalp deflected anteriorly, a short portion if inner demibranch also shown. Scale bars = 1 mm, 


Diagnosis: Species with a single papilla close to excur- 
rent siphon. Anterior pallial gland shallow. Internal shell 
size about 10% of mantle surface; with rounded, almost 

| posterior end. Anterior pair of pedal retractor 
muscles a branch originated from shell. Gastric 


Maly chan l and style Sac Narrow and long. 


Description: = Srevy (Ficures 6-9): Reduced, inequi- 
valve, occupying about 1/10 of mantle, embedded into 
mantle anterior region (Figure 10, sh). Length approxi- 
mately 4 times width. Color white, opaque. Outline softly 
irregular. Both valves asymmetrical; left valve about 4 
shorter than right’ valve (Figure 8) (this may be 


L. R. L. Simone, 2008 


Page 255 


Figures 14-20. Chlamydoconcha avalvis anatomy. 14. Left view of entire animal emphasizing location of digestive tract and 
topology of main muscles, ganglia and glands, animal artificially represented as transparent. 15. Midgut as in situ, right view. 16. 
Same, slit longitudinally to expose inner surface. 17. Renopericardial structures and region, right auricle artificially disconnected 
from gill and deflected upwards, a transversal section of indicated level of right kidney also shown. 18. Cerebral ganglia, posterior- 
slightly right view. topology of esophagus also indicated. 19. Pedal ganglia, right and slightly posterior view. 20. Visceral ganglia, 
right and slightly posterior view. Scale bars = 1 mm. 


abnormal). Shape somewhat deformed and irregular; 
flattened, planar. Prodissoconch rounded, sub-termi- 
nal: located in middle of anterior fifth of valve length; 
shape semispherical, with small dorsal bulging portion; 
0.26 mm long, 0.31 mm height. Outer surface some- 
what irregular with strong commarginal undulations 
and with rounded. concave impressions: with ventral 
edge elevated (Figure 9). Calcareous concretions close 
to periphery on right valve (Figure 8). Periostracum 
extending about 1/3 beyond calcareous portion of each 
valve, wider dorsally; color yellowish, transparent. Hinge 


edentulous. Ligament small, restricted to umbonal 
region, relatively wide (Figures 6-7), pale brown; resili- 
fer absent. Inner surface glossy. Scar of anterior retractor 
muscle of foot occupying about 1/5 of inner surface, 
3 times longer than wide, located just posterior to umbo- 
nal concavity. 


MantTLE (Ficures 1-5, 10): Surrounding body almost 
completely, spherical in contracted condition (Figures 
14). Color pale cream, translucent (living and_preser- 
ved). Outer surface smooth and simple, lacking papillae 


Page 256 


Pedal aperture ventral, longer anteroposteriorly (Fig- 
ures 4, 10); aperture length spank half total mantle length. 
Edges of pedal aperture thick, simple, with winalations: 
thicker anteriorly. Anterior gland as a blind-sac, located in 
anterior, median region, about 1/3 of animal height from 
anterior end of pedal aperture (Figure 4, ag); size equiva- 
lent to 1/30 of mantle outer surface; its aperture central, 
with about 4 of gland size. Excurrent siphon cylindrical, 
small, papilla- like, located about half of animal height 
from posterior end of pedal aperture (Figures 3, 10, si): 
length about 1/20 of animal length; internal surface 
smooth, simple. Single papilla located about 1/5 of animal 
length dorsal to excurrent siphon (Figures 1, 3, pl), on 
median line, solid, size about half of that of siphon. Pair 
of small, low, bulging projections slightly dorsal to anterior 
gland, corre sponding with shell ne Mantle relatively 
thik. mostly hollow, sponge-like. Mantle inner surface 
smooth, simple (F igure 5). 


Main Muscie System (Ficures 11, 14); Adductor muscle 
not seen, possibly immersed in thin layer of visceral 
dorsal muscles. Pair of anterior pedal retractor muscles 
originate about 1/3 from inner surface of valves (scar 
described above), and about 2/3 splayed by antero- 
dorsal region of visceral sac; gradually becoming thicker 
towards ventral, up to anterior half of pedal oneal re- 
gion. Pair of posterior pedal retractor muscles somewhat 
similar to anterior pair; originating in dorsal visceral sac 
side about 4 posterior from that of anterior pair. Thin 
layer of pallial muscles splayed by mantle like a net; 
mi uinly concentrated anteriorly, inserting in anterior pair 
of pedal retractor muscles, in level just anterior to palps. 


Foor anpd Byssus (Ficures 4, 5, 10, 14): Foot narrow, 
longer antero-poste sriorly: leno about half of animal 
le neth; width about 1/5 of animal width; projected ante- 
riorly at about 4 of animal le a Anterior region some- 
what pointed. Byssal gland a narrow furrow located 
subterminally, in posterior region of foot ventral medial 
line; about 1/7 of foot length. Byssal gland thin, hollow, 
chamber depth of about 1/5 of foot length (Figure 14, 
by). No Reis found. 


PaLLIAL Cavity (Ficures 5, 10-13): Surrounding almost 
entire space between mantle and visceral sac, except 
for a dorsal portion correspondent to 1/10 of visceral 
sac surface connected to mantle. Gill eulamellibranch, 
heterorhabdic, occupying about half of pallial cavity, 
mainly in dorsal region (Figure 10), about two times 
longer than wide. Outer demibranch slightly triangular, 
about 2/3 of inner demibranch; anterior region becom- 
ing abruptly narrow, ending about 1/8 of total gill length- 
posterior to inner demibranch anterior end. Inner 
demibranch anterior end slightly rectangular, ending be- 
tween hemipalps. Gills gradu: ully narrowing towé ards pos- 
terior, up to some what pointe ral posterior e nad. About 1/4 
of each gill (their posterior region) Tree from visceral 
mass, connected with each other by cilia. Cilia connect 
outer lamellae of outer demibranch with mantle and 
inner lamellae of inner demibranch with visceral sac 


THE NAUTILUS, Vol. 122, No. 4 


(Figure 12, ee), same ciliary connection between both 
inner demibranchs in their region posterior to visceral 
mass. Connection among gill flame nts by aligned longi- 
tudinal tissue rods equiv alent’ in width to f Slnmnenie: eae 
longitudinal rod separated f rom neighbor rods by dis- 
tance equivalent to 5 filaments. Ve nizal edge of outer 
demibranch simple; filaments very thin tclsont 1/50 of 
gill width), outer connection mostly dorsal. 

Inner demibranch filaments a little shorter than inner 
demibranch itself; ventral edge with food groove. Inner 
gill connection to visceral mass dislocated ventrally, 
separated from remaining dorsal gill connection by dis- 
tance equivalent to half gill width (Figures 5, 12). Palps 
(Figure 15) with size equivalent to 1/10 of that of gill; 
category IT (Stasek, 1963). Hemipalps similar to each 
other; ventral half tall, slightly triangular; dorsal half 
narrow (about “4 of ventral half), emeath, surrounding 
anterior insertion of inner demibranch. Inner surface of 
palp (ventral half) with uniform, transversal folds, about 
20 folds in each hemipalp; more distal folds shorter, 
weakly arched, folds gr adually becoming longer towards 
medial, dorsal region of folds be coming narrower and 
strongly arched, ee ming a folded dorsal furrow in direc- 
tion to mouth: ventral ead of each folds rounded: dorsal 
end weaker; a smooth, narrow area surrounding entire 
edges of hemipalps (Figure 13). Both palps separated 
from each other bya dicaned equivalent to half of lon- 
ger portion of palp length. Mouth surrounded by anteri- 
or and posterior relatively tall lips, inner surface smooth. 


VisceRAL Mass (Ficures 5, 14): Bulging, spherical; sepa- 
ration with foot somewhat distinct. Gonad color cream, 
surrounding most of visceral structures, occupying about 
S0% of outer region. Genital aperture a small slit located 
about 1/20 of visceral height from dorsal edge and from 
nephropore (Figures 14, 17, ga); genital duct not dis- 
cernible. Digestive diverticula restricted to central 
area of anterior region; color pale greenish beige. Reno- 
pericardial structures occupying a bitit 1/10 of visceral 
volume, located in posterior region of dorsal surface. 


CIRCULATORY AND EXCRETORY SysTeMS (FiGures 14, 17): 
Heart of about 1/20 of visceral volume; located anterior 
to kidney; length about 1/8 of total length. Auricles tri- 
angular, insertion with ctenidial veins about “4 of their 
length, located in posterior quarter of gill. Connection to 
ventricle longitudinal, lateral, with about half of ventri- 
cle length. Ventricle occupying about entire pericardial 
length. Kidneys white, extending from pericardium pos- 
terior end to area equivalent to pericardial length toward 
poste rior region, Each kidney about three times longer 
than tall, cosy solid except for inner flattened faman 
running longitudine uly long central region. Each nephro- 
pore a minute slit locate -d just anterior to origin of pair 
of posterior pedal retractor muscles; inside excurrent 
chamber of outer demibranch. 


Dicestive System (Ficures 14-16): Palps described 
above (pallial cavity). Esophagus with about 2/3 of dis- 
tance between pi alps i in width; le neth about 1/5 of that of 


L. R. L. Simone, 2008 


Page 257 


visceral mass; inner surface smooth. Stomach positioned 
transversal, somewhat perpendicular to esophagus, run- 
ning towards right; narrowing gradually (Figure 15); 
estimated volume about 1/20 of that of visceral mass: 
Type IV (Purchon, 1958). Stomach inner surface with 
pair of low, narrow folds located transversally in esopha- 
geal insertion (Figure 16). Dorsal U- shape sd furrow 
ocated just posterior to wala cig insertion (concavity 


stomach, with about 4 of stomach height; its aperture as 
eft end of U-shaped furrow. Ducts of digestive diverti- 


ateral gastric side: left pair slightly longer than right 
pair. Typhlosole very wide on origin of sty Te sac, narrow- 
ing relatively abruptly, running longitudinally in style sac 
eft side as narrow, low fold. Gastric shield with about 
I/S of internal gastric surface; located inside U-shaped 
furrow. Style sac totally separated from intestine; long 
and narrow: width about 70% of that of esophagus; run- 
ning somewhat straight backwards, ending in posterior 
wall of visceral mass. Digestive diverticula described 
above (visceral mass) Intestine originating in right side 
of style sac origin; inner surface aHaorh, simple: initially 
as wide as stomach, gradually becoming narrow up to 1/3 


of its original width after a distance equivalent to that of 


esophagus. Intestine performing tight loops as shown 
in Figure 14; after this, performing “wide, sigmoid loop, 
in such superior branch edges superior surface of viscer- 
al mass, along median line; running towards posterior. 
Anus sessile, simple: located at base of excurrent siphon. 


GenitaL System: Gonad described above (visceral mass). 
Genital pores represented by small slits equivalent in 
size to nephropore (Figures 14, 17, ga), located about 
1/20 of total animal length from nephropore, slightly 
posterior and ventral. No dications on brooding in gills 
was observed. 


CENTRAL Nervous System (Ficures 14, 18-20): Cerebral 
ganglia (Figure 18) located a short distance dorsal to 
mouth; each one with size equivalent to 1.5 esophagus 
diameter. Cerebral commissure narrow, length equiva- 
lent to each ganglion. Pedal ganglia (Figure 19) located 


in middle between cerebral ganglia and posterior end of 


foot: both ganglia completely conhected with each other 


along median line, forming a single, spherical mass of 


equivalent size of each cerebral ganglion. Visceral gang- 
lia (Figure 20) located just ventral to origins of posterior 
pair of pedal retractors; size equivalent to about S0% 
of that of cerebral ganglia, visceral commissure very 
short, ganglia almost touching each other. Cerebro- 
visceral connective very narrow, running through gonad 
Figure 14, co). 


Measurements: Animal length = 15 mm; valve = 3.7 


by 1.2 mm. 


Holotype: MZSP._ 8631S, Vinicius Padula col., 05 


March 2006. 


posterior). Dorsal hood triangular, located at left side of 


cula in two ae each pair located in middle region of 


Type Locality: Brazil, Rio de Janeiro, Cabo Frio, Ilha 
Comprida, 22°51/47” S, 41°56'35" W, about 6 m depth, 
under rocks. 


Distribution: Only known from the type locality. 


Etymology: The specific epithet refers to the apparent 
absence of the shell valves, which are virtually invisible 
in the living animal; a combination of the Latin negative 
prefix a and the noun valvis. 


Comparative Remarks: = Chilamydoconcha avalvis has 
the external surface of the highly developed mantle 
practically lacking papillae ( (Figures 1-4). This is the 
main character differentiating the species from the 
Pacific congener C. orcutti, which has a richness of 
papillae in the outer mantle surface, somewhat equidis- 
tantly disposed (Dall, 1884; Bernard, 1897; Williams, 
1949: Morton, 1981: fig. S). However, a single papilla is 
present in C. avalvis, close to the excurrent siphon; 
C. orcutti also possesses a differentiated papilla in the 
same position (Bernard, 197: fig. 3), which was named 
“defensive papilla” by Morton (1981). 

Anatomically, both Chlamydoconcha show similar 
organization. Mantle enlargement, foot features, posi- 
tion of the valves and main muscles, and internal fea- 
tures of glands and digestive tubes, are similar in the two 
species. The main anatomical differences, beyond the 
above mentioned papillae, are: The shell is proportional- 
ly smaller in C. avalvis (about 1/10 of mantle, Figure 14) 
than that of C. orcutti (about 1/6 of mantle). Although 
the prodissoconch (Figures 6-S) is very similar in both 
species, the posterior end of the shell of C. avalvis is 
more squarish than that of C. orcutti; in which the pos- 
terior end of the shell is pointed (Bernard, 1897: fig. 15; 
Morton, 1981, figs. 4-5). The anterior gland of C. avalvis 
is a blind sac, its internal chamber is small and short, 
practically with the same thickness of f the surrounding 
mantle (Figure 4, ag); on the other hand, that of 
C. orcutti Gerard 1894: “cheminée dorsale”) has a 
deeper empty chamber directed posteriorly ( (Bernard, 
1894: fig. 19, X), more recently, this gland was desig- 
nated “pheromone organ” (Morton, 1981, fig. 10), and 
described with similar characters of C. avalvis. The 
anterior pair of pedal retractor muscles has a branch 
originated from the inner surface of the valves in 
C. avalvis (Figure 14, fr); this is not described for 
C. orcutti (Bernard, iy fig. 20, mp), although men- 
tioned by Morton (1981 The midgut organization of 
C. avalvis (Figures i. is quite different from that 
of C. orcutti (Bernard, 1894: figs 9, 19: Morton, 1981, 
fig. 24) in several details, the main characters are: the 
narrower and longer gastric style sac of C. avalvis, while 
that of C. orcutti is wider and shorter (about 1/3 of vis- 
ceral sac length); the stomach is also narrower and smal- 
ler in C. avalvis than that of C. orcutti; the intestinal 
loops are differently performed in both species, and in 
C. avalvis it is apparently narrower 


Page 2558 


THE NAUTILUS, Vol. 122, No. 4 


Although the living animal of C. avalvis (Figures 1-2) 
was not observed crawling, it is possible that it also 
has an anterior projection of the mantle like that of 
C. orcutti (Bernard, 1894: figs.10, 11; Williams, 1949; 
Morton, 1981), as the mantle arrangement of that region 
is taller and wavy. The presence ofa single siphon close 
to the anus shows that the siphon is excurrent; as 
no incurrent siphon is present, the conclusion that 
water intake takes place through the pedal aperture 
(Morton, 1981). This feature is also found in other 
galeommatids, such as Kellia porculus Pilsbry, 1904: 
Scintilla nitidella Habe, 1962 (Morton and Scott, 1989, 
figs. 3, 18). 


DISCUSSION 


Discovery of the second species in the genus Chlamydo- 
doncha fits the description of the genus by Dall (1554). 
The anatomical characters of the C hlamydoconcha spe- 
cies are quite modified, even if considered under the 
light of the extraordinary suite of modifications exhibited 
by the Galeommatoidea (Woodward, 1893; Morton, 
1981: Bieler and Mikkelsen, 1992). The reduction of 
the shell of Chlamydoconcha is apparently the most 
extreme in all Bivalvia; its interiorization inside the man- 
tle is also found in other genera, e.g., Galeomma Turton, 
1825, Ephippodonta Tate, 1889. (Woodward, 1893: 
Liitzen and Nielsen, 2005), and Divariscintilla yoyo 
Mikkelsen and Bieler, 1989. All these genera and spe- 
cies, however, have proportionally larger valves. The foot 
is an important comparative character in Galeommatoi- 
dea. The “hanging” foot and the flower-like organ are 
some of the main characters (Bieler and Mikkelsen, 
1992; Jespersen and Liitzen, 2006); Chlamydoconcha 
possesses at least the first of these two characters. A 
molecular study (O Foighil et al., 2001) places Chlamy- 
doconcha as terminal taxa inside the Galeommatidae, a 
similar result of the morphological approach (Bieler and 
Mikkelsen, 1992). A dwarf male has been described for 
Chlamydoconcha orcutti (Morton, 1981), however, one 
has not been found so far in C. avalvis. 


ACKNOWLEDGMENTS 


A special thank to Vinicius Padula, Museu Nacional, 
Universidade Federal do Rio de Janeiro, by collect 
and donation of the lot examined herein. This study 
is sponsored in part by FAPESP (Fundagao de Amparo 
a Pesquisa do Estado de Sao Paulo), project no. O4/ 
02333-S. 


LITERATURE CITED 


Bemard, F. 1897. Anatomie de Chlamydoconcha orcutti Dall, 
lamellibranche a coquille interne. Annales des Sciences 
Naturelles, Zoologie et Paléontologie 4: 221-252 + pls. 1-2. 

Bernard, F. R. 1983. Catalogue of living Bivalvia of the Eastern 

Pacific Ocean. Dept. of Fishexes and Oceans. Ottawa, 
102 pp. 

Bieler, R. and P. M. Mikkelsen. 1992. Preliminary phylogenetic 
analysis of the bivalve family Galeommatidae. American 
Malacological Bulletin 9: 157-164. 

Carlton, J. T. 1979. Chlamydoconcha orcutti Dall: review and 
distribution of a little-known bivalve. The Veliger 21: 


Dall, W. H. 1884. A remarkable type of mollusk. Science 4(76): 
50-51. 

Jaspersen, A. and J. Liitzen. 2006. Reproduction and sperm 
structure in Galeommatidae (Bivalvia, Galeommatoidea). 
Zoomorphology 125: 157-173. 

Keen, A. M. 1969. Superfamily Chlamydoconchacea, Dall, 
1884. IN Moore, R.C. [Ed.] Treatise on invertebrate pale- 
ontology. Part N2, Mollusca 6, Bivalvia. The Geological 
Society of America and University of Kansas Press, 
Kansas, 573 pp. 

Liitzen, J. and C. Nielsen. 2005. Galeommatid bivalves from 
Phuket, Thailand. Zoological Journal of the Linnean 
Society 144: 261-308. 

Mikkelsen, P. M. and R. Bieler. 1989. Biology and comparative 
anatomy of Divariscintilla yoyo and D. troglodytes, 
two new species of Galeommatidae (Bivalvia) from sto- 
matopod burrows in eastern Florida. Malacologia 31: 
175-195. 

Morton, B. 1981. The biology and functional morphology 
of Chlamydoconcha orcutti with a discussion on the 
taxonomic status of the Chlamydoconchacea (Mollusca: 
Bivalvia). Journal of Zoology 195: 81-121. 

Morton, B. and P. H. Scott. 1989. The Hong Kong Galeomma- 
tacea (Mollusca: Bivalvia) and their hosts, with descrip- 
tions of new species. Asian Marine Biology 6: 129-160. 

O Foighil, D., R. Jennings, J.-K. Park, and D. A. Merriwether. 
2001. Phylogenetic relationships of mid-oceanic ridge and 
continental lineages of Lasaea spp. (Mollusca: Biv alvia) in 
the northeastern Atlantic. Marine Ecology Progress Series 
213: 165-175. 

Purchon, R.D. 1958. The stomach in the Eulamellibranchia;: 
Stomach Type IV. Proceedings of the Zoological Society of 
London 131 : 487-5235. 

Stasek, C.R. 1963. Synopsis and discussion of the association of 
ctenidia and labial palps in the bivalved Mollusca. The 
Veliger 6: 91-97. 

Williams, W. 1949. The enigma of Mission Bay. Pacific Discov- 
ery 2(2): 22-23. 

Woodward. M. F. 1893. On the anatomy of Ephippodonta 
macdougalli, Tatte. Proceedings of the Malacological 
Society of London 1: 20-26 + pl. 2. 


THE NAUTILUS 122(4):259-260, 2008 


Page 259 


Research Note 


Sinistral Campeloma decisum (Say, 1817) 
(Gastropoda: Viviparidae) from the Fox 
River, Illinois 


Campeloma (Gastropoda: Viviparidae) are a group of 


ovoviviparous, prosobranch snails endemic to North 
America east of the Rocky Mountains (Burch, 1989). 
These snails are known to burrow in mud or sand in 
freshwater streams and lakes and feed on carrion (van 
der Schalie, 1965; Burch, 1989). Campeloma spp. from 
southern North America typically nee sexually but 
those from the northern United States and Canada are 
parthenogenetic due to the scarcity of males in this re- 
gion (van der Schalie, 1965). Shells of Campeloma are 
moderately thick, conical, and imperforate ak a smooth 
surface and rounded whorls, and although normally dex- 
tral, sinistral specimens are occasionally f found (Baker, 
1928: Burch, 1989). In fact, Call (1886) pointed out Rafi- 
nesque’s "type" of Campeloma crassulum Rafinesque, 
1819. from the Ohio River was sinistral, noting that the 
shell had "four whorls of the spire reversed." Sinistral 
individuals show not only a reversal in shell orientation 
but also in organ placement (Savage, 1935). 

Sinistral Campeloma have been found throughout 
eastern North America. Call (1880) recorded C. decisum 
(Say, 1S17) (as C. integrum and C. rufim) from the Erie 

Canal at Mohawk, New York, Pilsbry (1897) ) reported 
sinistral C. decisum from the eee River at Fort 
Edward. New York. and Ancey (1897) commented on 
sinistral C. decisum from New York but gave no specific 
location. Bickel (1966) examined sinistral C. crassuluwm 
from the Ohio River at Louisville, Kentucky, Goodrich 
(1939) discussed sinistral C. geniculum (Conrad, 1834) 
from the Ogeechee River in Georgia, and Lee (2008) 
figured a sinistral C. limum (Anthony, 1860) from 
the Altamaha River at Doctortown, Georgia. Baker 
nee commented on a “reversed” C. decisum (as 

C. rufum) in a private collection, gave dimensions of a 
reversed C. decisum, and figured a sinistral C. decisiwm 

(as C. subsolidum), but did not give locality data for any 
of the specimens examined. Sampson ( (1916) stated that 
he had sinistral C. decisuwm (as C. subsolidum) in his 
collection from Flat Creek in Pettis County, Missouri, 
and Call (1886) referred to sinistral C. descisum (as 

subsolidum) from a slough near Fort Dodge, Lowa, 
and figured a reversed C. decisum (as C. obesum) from 
Lewis. Iowa. Baker (1928) figured sinistral C. decisim (as 
C. integrum, C. rufum, and C. brevispirum) from three 
different locations in Wisconsin (Wisconsin River near 
Merrimack, Sturgeon Bay at Sturgeon Bay, and Mirror 
Lake presumably near Baraboo), and Haas (1939) 
reported on a sinistral C. decisum (as C. integrum) from 
the Kankakee River near Shelby, Indiana. In Illinois, 


sinistral C. decisum (as C. integrum and C. rufum) have 
been reported from the Salt Fork Vermilion River near 
Homer (van Cleave, 1936), the Des Plaines River in the 
Chicago area (Lee, 2008), and Jackson Park Lagoon in 
Chicago ( Hand, 1928; Meyer, 1928; Haas, 1939). 

Sinictral uterine young in Campeloma have been 
reported by several authors, including Call (1880), 
Pilsbry ( (1897), Hand (1928), van Cleave (1936), and 
Haas (1939). Both desteal and_ sinistral Seat 
produce sinistral embryos (Hand, 1928; Haas, 1939 
Sinistrality might re ssult from either embryological is 
turbances (e.g., crowded uteruses or damaged eggs) that 
have no genetic basis, or individual mutations destined 
to disappear in the population because copulation 
between dextral and sinistral snails is impeded by me- 
chanical incompatibility (Call, 1SSO; van Cleave, 1936; 
Cazzaniga and Estebenet, 1990). However, because 
some degree of assortative mating occurs in gastropods, 
sinistral “shail could become reproductiv ely isolated 
(Cazzaniga and Estebenet, 1990); also, because many 
Campeloma lineages are parthenogentic, sinistral snails 
could become unconstrained by mating compatibility 
(Mattox, 1938: van der Schalie, 1965). There appears to 
be a progressive reduction in the percentage of sinistral 
individuals from uterine young to adults sath only a few 
individuals reaching sexual maturity (van Cleave, 1936). 

This high degree of mortality might occur as the result 
of morphologic: al or phy siclogical abnormalities (van 
Cleave, 1936; Bickel, 1966). 

We here report on sinistral C. deciswm from the Fox 
River basin near Algonquin, Illinois. Nineteen specimens 
of C. decisum from “Mill Pond (near creek), Algonquin, 
Ilinois,” were found in the University of Illinois Museum 
of Natural History Mollusk Collection, C ane 
Urbana (UIMNH 1828S): no date was given for this lot 
but the collector, the Rev. W. A. Nason, We din 1921. In 
addition, a relict sinistral C. decisum shell was collected 
by JST while conducting a freshwater mussel survey in 
the Fox River at Buffalo Park Forest Preserve near Al- 
gonquin (42.1486° N, 88.2900° W), Kane County, Ili- 
nois, on 25 July 2007. This specimen was e xtracted from 
silt-compacted gravel in an impounded area of the river 
and has been deposited in the Hlinois Natural History 
Survey Mollusk Collection, Champaign (INHS 31862). 

The Fox River has experienced sub-substandard water 
quality conditions prior to the passage of the Clean Water 
Act and has encountered habitat che inges (e.¢., increased 
siltation and substrate compaction ) \ due to he presence 
of lowhead dams (Santucci et al., 2005; Tiemann et al 
2007). Because these physicochemical changes have 
been shown to alter freshwater snail assemblages 
(Burch, 1989), the population of sinistral C. decisuwm in 
the Algonquin area might be extirpated. Fieldwork will 
continue in an attempt to document live individuals. 


Page 260 


THE NAUTILUS, Vol. 122, No. 4 


ACKNOWLEDGMENTS 


Funds were provided by the Illinois Department of 
Transportation. A. Kuhns (INHS) and J. Griesbaum 
(INHS) assisted in the 2007 sampling. T. Stewart 
(Iowa State University) assisted in identification of the 
2007 specimen. G. Levin (INHS), D. Thomas (INHS), 
B. Tiemann, and two anonymous reviewers offered 
constructive criticism. 


LITERATURE CITED 


Ancey, C. F. 1897. On some sinistral land shells. The Nautilus 
10; 104-105 
Baker, F. C. 1902. The Mollusca of the Chicago area. Part IL. 


The Ce tapode Bulletin of the Chicago Academy of 


Science 3: 131-418 + 9 plates. 

Baker, F.C. 1928. The fresh water Mollusca of Wisconsin. Part I. 
Gastropoda. Bulletin of the Wisconsin Geological and 
Natural History Survey 70(2), i-xx + 1-507 + 5 28 plates. 

Bickel, D. 1966. Campe loma crassula with reversed whorls. 
The Nautilus 79: LO7—108. 

Burch, J. B. 1989. North American freshwater snails. Malaco- 
logic: al Publications, Hamburg (Michigan), vii + 365 pp. 

Call, R. 1880. Reversed Mel vinthones: American Naturalist 
14: ae 
Call, R. E. 1886. On the genus Campeloma, Rafinesque, with 


a revision of the species, recent and fossil. Bulletin of 


the Washburn College 
149-165, 4 plates. 

Cazzaniga, N. J. and A. L. Estebenet. 1990. A sinistral Poma- 
cea canaliculata (Gastropoda: Ampullariidae). Malacolog- 
ical ae 23: 99-102. 

Goodrich, C. 1939. Certain mollusks of the Ogeechee River, 
ee The Nautilus 52: 129-131. 

Haas, F. 1939. Reversed specimens of Campeloma from the 
Chicago area. Zoological Series of Field Museum of 
Natural History 24: 93-94. 


Laboratory of Natural History 1: 


Hand, E. E. 1928. Sinistral Campeloma. The Nautilus 41: 
106-107. 

Lee, H. G. 2008. Jacksonville Shell Club website. http://www. 
jaxshells.org/reverse 1.htm accessed 22 May 2008. 

ae N. T. 1938. Morphology of Campe loma rufum, 
a parthenogenetic snail. Joumal of Morphology 62; 
243-261. 

Meyer, E, 1928. Finding a left-handed Campeloma. The Nau- 
tilus 41: 107. 

Pilsbry, H. A. 1897. Campeloma decisum Say, reversed. The 
Nautilus LO; 11S. 

Sampson, F. A. 1916. Reversed or sinistral shells. The Nautilus 
29: 128-129. 

Santucci, V. J., Jr, S. R. Gephard, and S. M. Pescitelli. 2005. 
Effects of multiple low-head dams on fish, macroinverte- 
brates, habitat, and water quality in the Fox River, Illinois. 
North American Journal of Fisheries Management 25: 
975-992. 

Savage, A. E. 1938. A comparison of the nervous system in 
normal and sinistral snails of the species Campeloma 
rufum. American Naturalist 72: 160-169. 

Tiemann, J. S., H. R. Dodd, N. Owens, and D. H. Wahl. 2007. 
Effects of lowhead dams on unionids in the Fox River, 
Illinois. Northeastern Naturalist 14; 125-138. 

van Cleave, H. J. 1936. Reversal of symmetry in Campeloma 
rufum, a fresh-water snail. American Naturalist 70: 
567-573. 

van der Schalie, H. 1965. Observations on the sex of 
Campeloma (Gastropoda: Viviparidae). Occasional Papers 
of the Museum of Zoology, University of Michigan, 641; 
1-15. 


Jeremy S. Tiemann 

Kevin S. Cummings 

Division of Biodiversity and Ecological Entomology 
Ilinois Natural History Survey 

1816 South Oak Street 

Champaign, [IL 61520 USA 
jtiemann@inhs.uinc.edu 


THE NAUTILUS 122(4):261-263, 2008 


Page 261 


Book Reviews 


Freshwater Mussels of Alabama & the 
Mobile Basin in Georgia, Mississippi © 
Tennessee 


Williams, James D., Arthur E. Bogan, and_ Jeffrey 
T. Garner. 2008. Freshwater Mussels of Alabama & 
the Mobile Be isin in Georgia, Mississippi - Tennessee. 
University of Alabama Press, Tuscaloosa, xv + 1—908, 
including numerous text figures and maps, many in 
color. ISBN-13: 978-0-8173-1613-6 (cloth: alk. paper): 
ISBN-10: 0-S173-1613-6 (alk. paper) 9 x 11.5 inches. 

Hardback; 10 Ibs. $70.00 from publisher and several 
booksellers: possibly less on eBay. 


Over the last decade or two the awareness of the Ameri- 

can populace and its policy-makers with the country’s in- 
digenous flora and fauna has been stirred to an 
unprecedented degree. Government has responded to a 
new culture of concern over environmental change and 
the conservation of natural communities, and one of the 
most important consequences of riding this zeitgeist has 
been the commissioning of scientists to elucidate’ the cur- 
rent state of our biota. ‘Conspicuous among the products 
of this “green revolution” is a watershed of works treating 
the naiad fauna of either a political unit (e.g. state) or a 
major river system. Except possibly for Constantine Rafin- 
esque’s epiph: iny on the banks of la Riviere Ohio has there 
been such a celebration of this natural resource! 


Preceded by recent works treating the biology of 


pearly freshwater mussels of several eastern American 
regions, most conspicuously the state of Tennessee and 
the Appalachicola River system (Georgia, Alabama, 
Florida), Williams, Bogan, and Gamer ice e tackled the 
most extensive fauna yet considered, that of Alabama 
and the entire Mobile Basin, but, based on other works 
of this contemporary genre, as we shall see, the treat- 
ment of those 178 species-level taxa, were it by tradition- 
al measure, only partially accounts for the ‘prodigious 
metrics (e.g. weight) c captioned above. 

The work is organized into a foreword, acknow- 
ledgements, institutional abbreviations, 16 chapters, an 
appendix (North American naiad type catalogues), a 
glossary, bibliography, and index. Certain observations 
can be made as one moves through the work. 

Introductory comments place Alabama and its mussel 


fauna in a broader context and present the grim reality of 


habitat degradation. resource depletion, extirpation, and 
extinction. No less than 23 reviewers are acknowledged 
for vetting this opus: workers in over 30 museums on 
other institutions were cited as collaborators, and dozens 
of field workers contributed their labors. Thanks are also 
offered to molecular geneticists, whose work underpinned 
many of the taxonomic innovations mentioned later. 


Wit A Foreworp By E. O. WILSON 


FRESHWATER 


MUSSELS 


of ALABAMA & the MOBILE BASIN 
IN GEORGIA, MISSISSIPPI & TENNESSEE 


“\_ JEFFREY T. GARNER 


There follows an historical review of naiad work in the 
state. The contributions of the feuding Quakers Isaac 
Lea and Timothy Conrad, of C. T. Simpson, H. H. Smith, 
and H. D. Athearn, the latter two being the dedicatees of 
the book, and many others are presented briefly. 

Chapter 3 spans 25 pages and presents an analysis of 
the inland waters of Alabama and a Mobile Basin, which 
support more aquatic biodiversity than any other area of 
comparable size on the continent. The geography, geology, 
hydrology, and, regrettable degradation of these water- 
ways (damming. canalization, etc.) is discussed in detail. 
The use of archival maps and photographs along with 
present-day images provides a starkly heuristic backdrop. 

Short chapters basically tabulate mussel taxa by con- 
stituent watercourse in the post-European and archaeo- 
logical record. A section on the commercial use ol 
mussels and their pearls is nicely illustrated and again 
reinforces the theme of resource depletion Chapter 7 is 
an historical review of naiad conservation efforts in the 
state, which have been rather extensive, particularly in 
the last decade: it concludes with a tabulation of the 48 
Alabama species listed as endangered or threatened as 


Page 262 


THE NAUTILUS, Vol. 122, No. 4 


* January, 2006, under provisions of the federal 
a ingered Species Act. 

Twenty pages are devoted to the ecology and life his- 
tory of the naiads. The topics are treated with thorough- 
ness and involve aspects of habitat (and its degr aston). 
feeding, predation, competition, parasitism, and the 
unique reproductive and larval strategies if these 
mollusks—including anatomic and behavioral contri- 
vances to optimize host fish infestation. Much recent 
work is brought to bear on these topics. 

Shell morphology and higher (ordinal, suprafamilial, 
familial) classification are dealt with succinctly; the latter 
with the most current systematic insights. 

Chapter 11 explains the format of the accounts in the 
taxonomic section. These headings are uniform and clear- 
ly indicated: Scientific and Common Name (each epithet 
initiated in upper case!); Illustrations; Description of 

Shell, Soft Anatomy, Glochidium, Similar Species; Gene- 
ral Distribution; Alabama and Mobile Basin Distribution 
(a map appears at the end of each entry and is marked 
with black dot for each recorded occurrence); Ecology 
and Biol ogy; Current Conservation Status and Protection: 
Remarks; “ond Synonymy. The latter includes a caveat 
indicating that this is far from a chresony my, being limited 
to the first usage of a species-level epithet ( generic reas- 
signment not ecousider ed) considered in svnonymy. On the 
other hand, it is generously, almost exhaustively, ilhimi- 
nated with type figures, in color when available. 


The over 700- oda pages devoted to the treatment of 


two Unionoidean families, 43 genera, and 175 species- 
level taxa plus short vignettes on five species of hypo- 
thetical occurrence, six non-naiad clams (Sphaeriidae is 
not parsed) including the two non-natives Corbicula 
fluminea and Dreissena polymorpha, and finally a 
newly-diagnosed identity for the spuriously recorded 
(mmslonaline d) Unio decumbens I. Lea, 1861 [Ti “apeoi- 
deus exolescens Gould, 1843) of southeast Asia]. 

The bibliography contains over 1000 titles, and the 
index is inclusive with all topics, terms, person- and 
place-names, and genus-species, species-genus entries, 
and the same reciprocation for the common names. 


A stunning feature of this work is the photography of 


Richard Bryant, who captures the shells of each species- 
level taxon in large format, with crispness and color accu- 
racy. The specimens are almost all of the highest quality, 
sometimes apparently requiring the use of extralimital 
material. The shells are scrupulously posed with the ad- 
ductor scar axis horizontal, poste rior to the left (as was the 
custom of the prolific ables -publisher Isaac Lea). 
Such conventions make it easy on the diagnostic eye. 

The marshalling of information in the taxonomic por- 
tion, particularly in Ecology and Biology and in the 
temarks is staggering and probably indicates a strong 
collaboration among the authors of this work. Other 
features such as the thousands of locality indicators, the 
lifting of hundreds of type figures from classic works, 
give a dimension to this work that is unprecedented, 


especially informative, and indicative of a lot of hard 


There are taxonomic initiatives exercised in this work. 
A major one of these is dealing with the “Pleurobema 
problem.” Tabulations of pages 501-504 indicate the 
profusion of available names for Mobile Basin and other 
Alabama species and the synonymies of four prior 
monographers and in the present work, which has a 
relatively conservative perception of the diversity. Wil- 
liams et al. pare t he list of Turgeon, Quinn, et al. (1998) 
by seven species while adding three classic and one post- 
1998 species. Likewise Hee Elliptio species. are resus- 
citated from synonymy as are a half dozen other species 
in five gener There are three un-named taxa included 
in the work, one sp., Epioblasma sp. cf. capsaefor- 
mis, and Toxolasma sp. Baty is, however, provided with 
vernacular names—a convenient machination. 

Although installed in the literature over the last 
decade, reassignments of long-recognized species to the 
resurrected Pleuronaia Frierson, 1927, and the newly- 
ordained Hamiota Roe and Hartfield, 2005, may sur- 
prise the reader. There are another half dozen generic 
reassignments necessitated by molecular genetic study, 
perhaps the most surprising ‘of which is the Pistolgrip, 
Quadrula verrucosa (placing Tritogonia in synonymy). 

Impl wusible as it may appear im context, there are two 
rather minor detractions which warrant brief mention. 
If I had my say in the creation of this magnum opus, 
T would have asked for a discussion of he geological 
history that provided the state with the colatoin of the 
Mobile Basin system, without which its present naiad 
diversity would have never reached the unassailable 
present- day mark. I see no reference to the Tertiary 
calamity that diverted the Tennessee River from its 
ancient course southwest past Lookout Mt. and into the 
heart of Alabama and thence to the Gulf of Mexico. The 
classic paper by Simpson (1900) on the evolution of the 
relevant naiad faunas and the geological evidence in 
support of it (Johnson, 1905a, “1905b: Adams, 1928) 
seem appropriate for the beginning of Chapter 3. The 
other little vexation is the persistence of gender-bending 
binomina in the naiad literature. It is not entirely clear 
how “Pleurobema stabilis” and “Ptychobranchus subten- 
tum” became entrenched, but the Code and the original 
descriptions indicate they should be rendered Pl. shabile 
and Pt. subtentus. 

Williams, Bogan, and Garner have produced a holistic 
and exhaustive work, carefully executed and seductively 
constructed. The taxonomic scope is unprecedented in 
recent years, covering some 60 percent of the American 
fauna. Aside from being a precious asset to the malaco- 
logical community, it will advance the understanding of 
hiodive srsity, ecology, and conservation in a much wider 
audience. To quote from Edward Osborne Wilson's 
Foreword: “People do care about species of wildlife, 
however, if they see a picture of it, know its name, and 
reac what is known of its distribution and natural his- 
tory. In addition to their contribution in mussel biology, 
this is what the authors have given us.” 

We applaud Williams, Bogan, and Garner, and we 
commend the Alabama Department of Conservation, 


Book Reviews, 2008 


Page 263 


Game and Fish Division as well as Auburn University, 
whose commitment has helped assure that this prodi- 
gious work will be affordable to a wider readership 


LITERATURE CITED 


Johnson, D. W. 1905a. Tertiary history of the Tennesseee 
River. Journal of Geology 13(3): 194-231, map. 

Johnson, D. W., 1905b. The Distribution of freshwater faunas as 
evidence of drainage modifications. Science 21: 588-592. 

Simpson, C.T. 1900. On the evidence of the Unionidae regard- 
ing the former courses of the Tennessee and other south- 
em rivers. Science 12(291): 133-136. 


Turgeon, D. D., J. F. rie Jr, A. E. Bogan, E. V. Coan, 
F. G. ecbes ’ G. Lyons, P. M. Mikkelsen, R. J 
Neves, C. F. E. we r, G. Rosenberg, B. Roth, A. Schel- 
tema, F. G. Thompson, M. Vecchione, and J. D. Williams. 
1998. Common and scientific names of aquatic inverte- 
brates from the United States and Canada: mollusks, 2nd 
edition. American Fisheries Society, ee Publication 


26, Bethesda, ix + pp. 1-509 + 16 pls. ( non-paginated ), 


Harry G. Lee 

4132 Ortega Forest Drive 
Jacksonville, FL 32210 USA 
shells@hglee.com 


Guide to the Freshwater Molluscs of the 
Lesser Antilles 


Pointier. Jean-Pierre. 2008. Guide to the Freshwater Mol- 
luses of the Lesser Antilles. Conch Books, Hackenheim, 
Germany, 127 pp. www.conchbooks.de. Retail price: 35. 


This small volume covers the freshwater molluscan fau- 
na of the islands of the Lesser Antilles. These islands 
form a double are of volcanic islands extending from 
Anguilla in the north to Grenada in the south along the 
east edge of the Caribbean Sea. Dr. Philippe Jarne, of 
Montpellier, France wrote the short preface. A brief 
history of the study of freshwater mollusks in the Lesser 
Antilles is also provided. 

Pointier reports nine families of yale and two 
families of bivalves; 2S gastropod and three bivalve spe- 
cies from the islands. The gastropods are divided into 
19 native gastropod species including the two endemic 
species, Neritilia succinea |Neritiliidae] and freshwater 
opisthobranch Tantulum elegans |Tantulidae], and nine 
introduced species. Lesser Antilles freshwater bivalve 
fauna contains the introduced Mytilopsis leucophaeta 
and two endemic species of Sphaeriidae. 

This volume has a flow chart key to the families of 

gastropods and bivalves and is suppleme nted with a page 
of figures illustrating the key landmarks, shell sculpture, 
and types of ope reula. Species are treated by family units 


in the same order as found in the key with discussion of 


the anatomy and ecology for the members found in the 
Lesser Antilles. Each of the gastropod species accounts 
is accompanied by two to five color illustrations of the 


shells, eggs, live animals, and ecology. The pulmonate 
species accounts incorporate dhistsations of the repro- 
ductive anatomy of each species. A section is devoted to 
aquatic habitats where these mollusks have been collect- 
ed and includes 44 color photogr aphs of habitats and 
The figure captions note some of the snail species 
found on the plants and some plants used as egg-laying 
sites or food. This volume is based on extensive field 
work throughout the Lesser Antilles. Species are well 
illustrated, including close-up color figures of live speci- 
mens of all but the oe opisthobr: anch Tantulum 
elegans. “Guide to the Freshwater Molluscs of the Lesser 
Antilles” is a great companion to the earlier work on the 
freshwater ficllaskes of Cuba (Pointier et al., 2005). 

T have found this volume very useful and recom- 
mend it for anyone working on, or interested in, iden- 
tification of the freshwater mollusks of the Lesser 
Antilles. This book will fit nicely in a backpack and will 
be a handy reference in the field. I would recommend 
this well illustrated book to anyone interested in fresh- 
water mollusks. 


LITERATURE CITED 


Pointier, J.-P, M. Yong and A. Gutiérrez. 2005. Guide to the 
freshwater molluscs of Cuba. Conch Books Hackenheim, 
Germany. 119 pages. 


Arthur E. Bogan 
North Carolina State Museum of Natural Sciences 
esearch Laboratory 


MSC 1626, Raleigh, NC 27699-1626 USA 


THE NAUTILUS 122(4):264, 2008 Page 264 


Notice 


Bulletin of Zoological Nomenclature 65(2): 152. June 2008. 
OPINION 2197 (Case 3341) of the International Commission on Zoological Nomenclature. 


Cardium egmontianum Shuttleworth, 1856 (currently Trachycardium egmontianum; Mollusca, Bivalvia, CARDIITDAE): 


current usage conserved. 


Abstract. The Commission has ruled that the current usage of the specific name egmontianum for a common and widespread 
western Atlantic bivalve Trachycardium egmontianum (Shuttleworth, 1856) of the family CARDIIDAE is conserved by setting aside 


all lectotype designations for Cardium mindanense Reeve, 1844, prior to that by Vidal (1998). 


LITERATURE CITED 


Vidal, J., 1998. Taxonomic revision of the Indo-Pacific Vasticardium assimile species group (Mollusca, Cardiidae). Apex 13: 111-125. 


Errata 


A couple of items published in the most recent issue of The Nautilus (volume 122 issue number 3) need correction. 

In the article by Villalobos-Rojas et al. (2008), page 155, right-hand column, line 5, the first name of Ms. Kirstie Kaiser was 
misspelled. In the same article, the complete reference to Pitt and Kohl (1979) was omitted (see below). 

In the article by Gonzalez-Vallejo (2008), page 180, the image representing the female shell on Figure 3, right, was inverted by 
the authors; that shell is not left-handed. 


The editor apologizes to all parties concerned for these editorial blunders. 


LITERATURE CITED 


Gonzalez-Vallejo, N. E. 2008. Parasitism of Monogamus minibulla (Olsson and McGinty, 1958) (Gastropoda: Eulimidae) on the red 
sea-urchin Echinometra lucunter (Linnaeus, 1758) (Echinodermata: Echinometridae) on the Caribbean coast of Mexico. The 
Nautilus 122: 178-181. 

Pitt, W. and R. Kohl. 1979. A new Panamic Mitrella (Mollusca: Gastropoda). The Veliger 21: 467-468. 

Villalobos-Rojas, F., A. G. Guzman-Mora, Y. E. Camacho-Garcia. 2008. Catalogue of the type material deposited at the Zoology 
Museum, University of Costa Rica. The Nautilus 122: 155-165, 


NAUTILUS 


Volume 122 
2008 


AUTHOR INDEX 


ARRUDA, J. O. ......00 1 D4 MANN GAR) ssscdicescasteyes 217 
Bocan, A. E. ... . 263 MEANS, MiG. TD coc eeccveda she coeieucntossndacavstinircaveadeipeawnmeadielstens 228 
BBROUGHET: P. cccscssccessicsscuscsassscnsvasecanasessesvavsaabaisensesdeuvereeativssadvesens ] MEIER-BROOK, Cy cicéccsscsscoissssssiscecsosssccsvcsvesccsavsacesveccscessoecceine 298 
GAETANO GC. THO Ss. ec cceadecsadeceveva voter sas adenesdicvsiottaaanciteuoipes baste 171 TUTTSEN Sc IM cascscsin caus anvanisniagsudsdec cic oncescdaeceawaumaecsunisuveenvusiccuaevesnacs 90] 
CAMACHO-GARCIA, Yo Ee cececceceeceeeceeteeeret ies eeretieeeeneeeieenees 155, ORDA, Be Ms vee fate vdenessszasvest heliiaes oasdiavsvietadee slawentavyneeeetes 143 
CONCEPCION! Gs P.. sc: s.cissse tein Gnseterane nits nine tits 143 PASTORINO, G. ...cssssssssessesvesssesssssscsvssessessscssssscssevsecsssseess 107, 236 
KOORNEDIL, Fe 6S, av sestengeatiny Poteauarsanes tun ceenaedeneauasuand aves dir teatas santas 143 PETIT, Re Eu ccccocccescsssessscacececevecececevevavevevadsessesisecevivsviveveveverseseees | 
Costa, Po M.S. vesseesee ces eeeiee teenies ieee eesieesiesniesneesecnsieaey 244 IPIMENTAS JA TDs. seitetosnssdoacertitecce aie aonacanvetssacgnten Saute seseen 244 
COUTO, De Ry vessssssssessessssesceeeeeeeeceeeereeeceeeeeeeeececceceeceeeceeseennnty 244 PORTEEL, Ry! Wiles vratamiseccistisin encase cate artesh iets 79 
CUMMINGS, Res Dis. sdeah ie sess dunbac ebagkauneesiebussapieguteiupsetataee eeheueas 259 SANTOS, A. Di. ccccccsccccccoscssccsscssccsscosecssssesssscestesucestseseectcesceeseees 143 
CUNHA, (Gi Mie» sccissiccincitsicacissrcvvisauscevevecdetasaasiesatedecuestaoatecsbeurause 57 Ciccdieis Me ccueenn acter enim, tenth Ae ta c0e fate! ot othe od oa catch A 166 
DuEnr, R. Se aaa ae ae ae 151 DAU: Ts We, 2cucseyelinevseten! devia csusgeh a attiea dusesaatieutiarcrensenieatese 115 
EISPING;: SiS: avcetenectictesesarenyereee We aeceareaveadsdiesscpreneananeerens 143 Sc ; 3 
: i SGARABINO,. Fy yiceccsasesccsissecqatssssscessiensenuotessagscascovinnesassccsensnhons 107 
(GARMAT, MV,. <sssctcccaacascsasdidedvateesededetucasessdetassseensvstipesativaces indossiss’ 19 Srccaannaes A 171 
IGRIGER, IDG. sisecsteaia cts ducks iutateindS AW levens dusGesiesiebinekes Heereaesas 185 ee hm cae he yet OR ee PQ h we peel 8 pa 
= : - SEGADES. Mi. Bs scaiiscusts vost davavies soesvades deecenseansaasansvens exe sacavars ccewwe DOO 
BRR Se IN: cossatectetinacatdizageteccssecetuavecteoiocracacant enh geetceecarastiaadnuae 217 . are 
GONZALEZ-VALLEJO, N. E. ccccccccccssscessesssesessvssessscesessesveseessavees 178 SIMONE, L. R. L, .. 87, 252 
JO, ‘ ; 
ITARDING,, Ui. Ms si sscssetaie sei seiesvens souseusevssurenisasonsoasndsouseet sdaseteens vals SQumes, L. sais a 115 
TEN RIGK Ss [CARS ctsevesctiscceleasavde cresaiset teriavin aston docleecaaael 79 THEMANN, J. 5. -- 59 
HERALDE. FE. M.. DGD ose each ct ete tose este tant claan tends 143 THOME, J. Ww. heals etic acre dunSaiwaah sa saan aeasie puiatuaealey ces acianiceven DEwu ee helnoemm se taieos 94 
FOU ARG! Re eaeteatetn ceria ctesitett easiest ae 99 VALDOVINOS, Cu sascssssssrseeccsssseeecesssiieeecnsnsneecennsinecsanesaeeeeanennns 201 
TTUARTE, Co cocecccececececececececececececeseeeeeeeeeeeececeeetececevevececeseseseses, JIS VIETALOBOS-INOJAS; Po: asicessassanvanssios sas sdesccbendievivaamas vohes savisssi¥sases 155 
KOHN, A. J. cesecssseessssseesssssvessssvecsssseecsssecsesssvessssvesesseecsesseeersess 143 VILLANUEVA, Jo As cecessstseessteeesstseesseeessetecesisecesnsecnteecettisses 143 
Lea, J. |S ae ee inate ann. SO ene ene 166 ZETAVAS DAG, -ceenazsveewesvicviets Santi ytedeecs tees lade teeonedionedeos artenventeaetes 5 
Miers Si, (oe edcceactn agence de ttaceo conicepeeie estecsioetta bates amiga tatosneeues 261 
NEW TAXA PROPOSED IN VOLUME 122 
GASTROPODA 
Admetula affluens Bouchet and Petit, 2008, new species (Cancellariidae) 2... 0... ee 4 
Admetula bathynoma Bouchet and Petit, 2008, new species (Cancellariidae) .. 0... ee 7 
Admetula emarginata Bouchet and Petit, 2008, new species (Cancellariidae) 2.0... ee 4 
Admetula lutea Bouchet and Petit, 2008, new species (Cancellariidae) 2.2... ee 6 
Admetula marshalli Bouchet and Petit, 2008, new species (Cancellariidae) 2... ee 6 
Alvania francescoi Garilli, 2008, new species (Rissoidae) 2... ee 28 
Alvania rosariae Garilli, 2008, new species (Rissoidae) .. 0... 2. ee BG 
Anatoma rapaensis Geiger, 2008, new species (ATatO RIGA) a. i oct6 Age aid hehe ara Seto s oh elane £.GNe aes ad ake dom ar ends igh + 193 
Ariadnaria ainikta Saul and Squires, 2008, new species (fossil, Trichotropidae) .. 0.0.0. 119 
Ariadnaria aldersoni Saul and Squires, 2008, new species (fossil, Trichotropidae) .. 2... 2.0 ee . 12] 
Ariadnaria obstricta Saul and Squires, 2008, new species (fossil, Trichotropidae) . 2.2... 0. ee 122 
Ariadnaria stibara Saul and Squires, 2008, new species (fossil, Trichotropidae) .. 2... 0.0. ee 12] 
Conus alleni Hendricks and Portell, 2008, new species (fossil, Conidae) . 2.0.0... 0 ee Suse Wie Avie, ane aU) 
Conus palmerae Hendricks and Portell, 2008, new species (fossil, Conidae)... 0. 87 
Ergalatax junionae Houart, 2008, new name (Muricidae) ... 22... eee oa 102 
Garzasia Saul and Squires, 2008, new genus (fossil, Trichotropidae) .... 2.0.0 -0-.0.. 00-005. 29 
Garzasia diabla Saul and Squires, 2008, new species (fossil, Trichotropidae) ... 20... 00.0.0... 50.00.05. : 3 
Garzasia intermedia Saul and Squires, 2008, new species (fossil, Trichotropidae) . 2.6... er dteresera tir Oo) 
Leptotrophon atlanticus Pimenta, Couto, and Costa, 2008, new species (Muricidae) ..........0...0.005. 6b tok aa AS 
Lysinae Saul and Squires, 2008, new subfamily (fossil, Trichotropidae) ............. Gas a alys an 2 otad 22 
Lysis jalamaca Saul and Squires, 2008, new species (fossil, Trichotropidae) ...............-. se a 26 
Lysis lomaensis Saul and Squires, 2008, new species (fossil, Trichotropidae) ... 20.0... 5... 00005 ee eee 27 
Lysis mickeyi Saul and Squires, 2008, new species (fossil, Trichotropidae) ........... sed Cd a Buchibee argc k 123 
°Mitrella hayesorum Duerr, 2008, new species (fossil, Columbelliidae) 51 


Page 266 THE NAUTILUS, Vol. 122, No. 47 


°Mitrella phyllisae Duerr, 2008, new species (fossil, Columbelliidae) 2... 0.0. ee 153 
Nipponaphera agastor Bouchet and Petit, 2008, new species (Cancellariidae) 0... 0. ee 12 
Nipponaphera argo Bouchet and Petit, 2008, new species (Cancéllartidae) . gs3 is da oa shes da es bev we dag hak oa MERE 12 
Nipponaphera tuba Bouchet and Petit, 2008, new species (Cancellariidae) . 2.2... ee 13 
Sinezona danieldreieri Geiger, 2008, new species (Scissurellidae) 2... eee 186 
Sinezona wileyi Geiger, 2008, new species (Scissurellidae) 6 c5 oy uw du cag a bead wo Yu bd do wh ee aw eeouieg ae dhe Ae eeee 189 
Trigonostoma ti ybliuum Bouchet and Petit, 2008, new species (Cancellariidae) 2... ee 13 
Trogloconcha lozoueti Geiger, 2008, new species (Larocheidae) ... 2.0... kee eee eee 196 
Trophon columbarioides Pastorino and Scarabino, 2008, new species (Muricidae) oe ee 108 
Trophon fasciolarioides Pastorino and Scarabino, 2008, new species anes ici ba he OG ba bd Hae ee be bed aha dee 110 
Zeadmete bathyomon Bouchet and Petit, 2008, new species (Cancellariidae) 2.2... ee 2 
Zeadmete bilix Bouchet and Petit, 2008, new species (Cancellariidae) 2... 00. eee 8 
Zeadmete physomon Bouchet and Petit, 2008, new species (Cancellariidae) .. 0. ee 8 
BIVALVIA 

Chlamydoconcha avalvis Simone, 2008, new species (Chlamydoconchidae) .. 0... 252 
Dilemma japonicum Sasaki and Leal, 2008, new species (Poromyidae) Panes dr utle Goin Gtin-¥h aide ain cava ahaa doa) neck apt aaah anode eee 167 
Sphaerium cambaraense Mansur, Meier-Brook, and Ituarte, 2008, new species (Sphaeriidae) . 2.2... 229 
Spinosipella agnes Simone and Cunha, 2008, new species (Verticordiidae) .. 2.0. eed 58 
Spinosipella tinga Simone and Cunha, 2008, new species (Verticordiidae) 2... eee 62 
SCAPHOPODA 

Heteroschismoides meridionalis Scarabino and Caetano, 2008, new species (Entalinidae) 2.0... 175 
Heteroschismoides antipodes Scarabino and Caetano, 2008, new species (Entalinidae) 2.0... ee ales) 


REVIEWERS FOR VOLUME 122 


Robert Angelo 
Gary Barker 
Riidiger Bieler 
David Campbell 
Henry Chaney 
Eugene V. Coan 
Timothy Collins 
Robert H. Cowie 
Marta J. deMaintenon 
Thomas J. DeVries 
Tom Duda_ 
Diarmaid O Foighil 
Daniel Geiger 
Martin Haase 

M. G. Harasewych 


Gregory Herbert 
Robert Hershler 
Carole Hickman 
Roland Houart 
Steffen Kiel 

Alan Kohn 
John N. Kraeuter 
Harry a see 
Pierre Lozouet 
Bruce A. Marshall 
Paula M. Mikkelsen 
Helene Mone 


Robert G. Moolenbeek 


P. Graham Oliver 
Guido Pastorino 


Roger W. Portell 
Robert S. Prezant 
Takenori Sasaki 
Ronald L. Shimek 
Luiz R. L. Simone 
Jonathan Todd 
Paul Valentich-Scott 
Janice Voltzow 


Anders Waren 


Richard C. Willan 
Diego Zelaya 


Sponsored in part by the State of Florida, Department 
of State, Division of Cultural Affairs, the Florida Arts 
Council and the National Endowment for the Arts 


pris 

9° % 

RS 2 ©, 

fe) S 
z = NV 
Z a NATIONAL 
an as ENDOWMENT 

c « 

Y y FOR THE ARTS 


a 


INSTRUCTIONS TO AUTHORS 


THE NAUTILUS publishes articles on all aspects of the 
biology, paceniloee and systematics of mollusks. 
Manuscripts describing original, unpublished research 
and review articles will be peace Brief articles, not 
exceeding 1000 words, will be published as notes and do 
not require an abstract. Notices of interest to the mala- 
cological community will appear in a notices section. 


Manuscripts: Each original manuscript and accompa- 
nying illustrations should be submitted to the editor pref- 
erably via e-mail or as hardcopy in triplicate. 

Text must conform to the dimensions of 8% x 11-inch 
paper, double-spaced, and single-column throughout (in- 
cluding literature cited, tables, and figure captions). Au- 
thors should follow the general recommendations of Sci- 
entific Style and Format—The CSE Manual for Authors, 
Editors, and Publishers, available from the Council of 
Science Editors at www.councilscienceeditors.org. The 
first mention of a scientific name in the text should be 
accompanied by the taxonomic authority, including year. 
Latinized names and other words to be printed in italics 
must be underlined; leave other formatting indications to 
the editor. Metric, not English, units are to be used. Geo- 
chronologic modifiers should be capitalized only when 
units are formally recognized: for instance, use Late Cre- 
taceous but early Miocene. Likewise, only modifiers of 
formally recognized chronostratigraphic units are capi- 
talized: use Lower Jurassic but upper Oligocene. 

The sequence of sections should be title page, ab- 
stract, introduction, materials and methods, results, dis- 
cussion, acknowledgments, literature cited, tables, figure 
captions, figures. The title page should include the title, 
author’s name(s) and address(es). If corresponding au- 
thor is not the senior author, please indicate. The ab- 
stract should summarize in 250 words or less the scope, 
main results, and conclusions of the article. Abstracts 
should be followed by a list of additional key words. All 
references cited in the text must appear in the Literature 
Cited section and vice-versa. Please follow a recent issue 
of THE NAUTILUS for bibliographic style, noting that 
journal titles must be unabbreviated. Information on 
plates and figures should be cited only if not included 
within the pagination of cited work. Tables must be num- 
bered and each placed on a separate page. If in doubt, 
please follow a recent issue of the journal for sequence of 
sections and other style requirements. 


Illustrations: Illustrations are rendered either at full- 
page width (maximum width 17 cm) or column width 
(maximum width 8.2 cm). Please take these dimensions 
into consideration when preparing illustrations. Page- 
width illustrations ideally should span the entire width of 
printed page (17 cm). “Tall” page-width illustrations 
should be avoided, square or “landscape” formats work 
better. Please design plates accordingly, such that there 
will be enough space left at the bottom of printed page 
for plate caption. (Digital technology has made this task 
much easier.) 


All line drawings must be in black, clearly detailed, 
and completely labeled. Abbreviation definitions must be 
included in the caption. Line drawings must be high 
resolution files at least 600 dpi (dots per inch) resolution 
at actual size. Standard digital formats for line drawings 
include .tif, .bmp, -psd, eps, and .pdf. 

Photographs may be submitted in black-and-white or 
color, preferably in RGB mode if in color. Standard digi- 
tal formats for photographs include .tif, .psd, .jpg, or .pdf. 
Photographs must be high resolution files at least 300 dpi 
resolution at actual size. 

If more than one figure is included in an illustration, 
all figures are to be consecutively numbered (Figures 1, 
2, 3,..., NOT Figures 1A, 1B, 1C,..., NOR Plate 1, 
Figure 1, .. .). In illustrations with more than one figure, 
make sure that blank areas between figures is kept to a 
minimum, thereby allowing for more area for each indi- 
vidual figure. 

Compressed files (e.g., jpg) may be used to facilitate 
transmission of files during original submission, but may 
not be acceptable at final submission (see below). 


Voucher Specimens: Deposition of the holotype in a 
recognized institutional, public collection is a require- 
ment for publication of articles in which new species- 
level taxa are described. Deposition of paratypes in in- 
stitutional collections is strongly encouraged, as is the 
deposition of representative voucher specimens for all 
other types of research work. 


The Editorial Process: Upon receipt, all manuscripts are 
assigned a number and acknowledged. The editor reserves 
the right to return manuscripts that are substandard or 
not appropriate in scope for THE NAUTILUS. Manu- 
scripts deemed appropriate for the journal will be sent 
for critical review to at least two reviewers. The review- 
ers’ recommendations will serve as basis for rejection or 
continuation of the editorial process. Reviewed manu- 
scripts will be sent back to authors for consideration of 
the reviewers’ comments. The revised version of the 
manuscript may at this point be considered accepted for 
publication by the journal. 


Final Submission: Authors of accepted manuscripts are 
required to submit a final version via e-mail to the editor 
at jleal@shellmuseum.org. Please do not send low-resolu- 
tion or compressed illustration files at this stage. Send any 
files larger than 20 Mb on a CD or DVD to the editor. 


Proofs: After typesetting, proofs will be sent to the au- 
thor. Author should read proofs carefully and send cor- 
rections to the editor within 48 hours. Changes other than 
typesetting errors will be charged to the author at cost. 


Offprints: An order form for offprints will accompany 
the proofs. Offprints will be ordered through the ie 
Aachen with institutional, grant, or other research sup- 
port will be asked to pay for page charges at the rate of 
$60 per page. 


@ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper)