HARVARD UNIVERSITY
Library of the
Museum of
Comparative Zoology
TRANSACTIONS
OF THE
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOLUME 16
1970-1972
Printed from the
W. W. Whitney Publication Endowment
COMMITTEE ON PUBLICATION
CARL L. HUBBS
JOSEPH R. JEHL, JR.
ARNOLD ROSS
CONTENTS
1. Studies on the Tetraclitidae (Cirripedia: Thoracica) : a proposed new
genus for the Austral species TetracUta purpurascens breviscutiim. By
Arnold Ross. 24 February 1970 1-12
2. The shallow water anomuran crab fauna of southwestern Baja California,
Mexico. By Janet Haig. Thomas S. Hopkins and Thomas B. Scanland.
4 June 1970 13-32
3. Comparative biology of American black widow spiders. By B. J. Kaston.
24 July 1970 33-82
4. Eastern Pacific Crown-of-Thorns starfish populations in the lower Gulf
of California. By Thomas Dana and Arthur Wolfson. 24 November 1970 83-90
5. Evolution of Peromyscus on northern islands in the Gulf of California,
Mexico. By Timothy E. Lawlor. 24 February 1971 91-124
6. Lampeira (Eiitosphemis) lethophaga, new species, the nonparasitic deriva-
tive of the Pacific lamprey. By Carl L. Hubbs. 30 April 1971 125-164
7. Recent ostracodes from Clipperton Island, eastern tropical Pacific. By
Edwin C. Allison and John C. Holden. 14 May 1971 165-214
8. Studies on the Tetraclitidae (Cirripedia: Thoracica): a new tetraclitellan
from India. By Arnold Ross. 21 May 1971 215-224
9. Stratigraphy of the Poway area, southwestern California. By Gary L. Peter-
son. 9 July 1971 225-236
10. Herpetofauna of the Pacific coast of north central Baja Cahfornia, Mexico,
with a description of a new subspecies of Phyllodactylus xanti. By Dennis
L. Bostic. 25 August 1971 237-264
11. A new genus of Chthamalidae (Cirripedia) from the southeastern Pacific
island of San Ambrosio. By Arnold Ross. 26 October 1971 265-278
1 2. The larval and pupal stages of four species of Cafius (Coleoptera: Staphy-
linidae) with notes on their biology and ecology. By Gary J. James, Ian
Moore and E. F. Legner. 5 November 1971 279-290
13. The color patterns of downy young ratites and tinamous. By Joseph R.
Jehl, Jr. 15 November 1971 291-302
14. Cenozoic calcareous nannofossils from the Pacific Ocean. By David Bukry.
7 December 1971 303-328
15. An upper Pleistocene marine fauna from Mission Bay, San Diego, Cali-
fornia. By J. Philip Kern, Tom E. Stump and Robert J. Dowlen. 29 De-
cember 1971 329-338
16. The systematic position of Urosalpinx carolinensis Verrill, 1884 with com-
ments on the genus Mohnia Friele, 1878. By George E. Radwin. 12 Janu-
ary 1972 339-342
n ^
MUS. COMP. ZOOL.
LIBRARY
MAR 1 0 1970
HARVARD
UNIVERSITY
STUDIES ON THE TETRACLITIDAE
(CIRRIPEDIA: THORACICA): A PROPOSED
NEW GENUS FOR THE AUSTRAL SPECIES
TETRACLITA PURPURASCENS BREVISCUTUM
ARNOLD ROSS
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 1 24 FEBRUARY 1970
STUDIES ON THE TETRACLITIDAE
(CIRRIPEDIA: THORACICA): A PROPOSED
NEW GENUS FOR THE AUSTRAL SPECIES
TETRACLITA PVRPVRASCENS BREVISCUTUM
ARNOLD ROSS
ABSTRACT. - Epopella gen. nov. is proposed for the Auckland Islands species Tetraclita purpurascens
forma breviscutum Broch, 1922, a solid-walled tesseroporan. Elminius plicatus Gray and E. simplex Darwin
are assigned provisionally to this new genus on the basis of morphological similarities. Epopella. containing
the most primitive tetraclitids, is inferred to have evolved during the early Paleogene, and it is from this
group that Tesseropora and later tesseroporans are derived.
RESUMEN. — Epopella gen. nov. esta propuesto para el especies de las Islas Auckland Tetraclita
purpurascens forma breviscutum Broch, 1922, un tesseroporan que tiene una pared soVida. Elminius plicatus
Gray y E. simplex Darwin son asignados provisionalmente a este genero nuevo en el base de semejanzas
morfologicas. Epopella. conteniendo las tetraclitids mas primitivas se infiera que ha evolucionada durante
del Paleogena, y es de este groupo que Tesseropora y tesseroporans mas tarde estan derivado.
Knowledge of the tetraclitid fauna of Australia, Tasmania, New Zealand, and the
islands comprising the Antipodean Province is limited. Aside from the taxon Broch (1922:
337) described as Tetraclita purpurascens forma breviscutum. the following species are
known from this region: Tesseropora rosea (Darwin, 1854: 335; Linzey, 1942: 280; Pope,
1945: 366; Wisely and Blick, 1964: 166), Tetraclita vitiata (Stephenson, 1968: 51), and
Tetraclitella purpurascens (Darwin, 1854: 337; Linzey, 1942: 279; Foster, 1967a: 83;
1967b: 35).
Tetraclita purpurascens forma breviscutum was collected by the Th. Mortensen
Pacific Expedition (1914-1916) on Auckland Island, the largest of several islands in the
Auckland Islands Group (Fig. 1 ), and more recently it has been found on Rose Island. This
species has neither been reported nor found in collections from any other locality and
appears to be endemic to the Auckland Islands. Unfortunately, there is little known about
the ecology of this tetraclitid.
Hiro (1939: 275) noted differences in the opercular plates of T. purpurascens forma
breviscutum that indicated it was not closely related to the nominate subspecies. However,
he failed to indicate the affinities of this form to other tetraclitid groups. In re-examining
the type specimens I noted several salient wall structures that readily characterize this
taxon at the generic and specific level and suggest that its affinities are to the tesseroporan
rather than to the tetraclitellan lineage (Ross, 1969: 238). Consequently, the ''forma"
breviscutum is elevated to specific rank and the genus Epopella proposed for it and two
other related species.
FAMILY TETRACLITIDAE Gruvel
Remarks. — The familial diagnosis presented earlier (Ross, 1969: 238) is emended to
include those species that lack an inner lamina and have an outer lamina permeated by
cuticular chitin.
SAN DIEGO SOC. NAT. HIST., TRANS. 16(1): 1-12, 24 FEBRUARY 1970
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
166°10' E
ROSE ISLAND
j ^---v- ENDERBY ISLAND
O SARAH'S BOSOM
HARBOR
0 12 3 4 5
50°40'S-
Figure 1. Map of Auckland Islands Ciroup, and its position relative to New Zealand (inset). The known
occurrences oi Epopella breviscutum are Sarah's Bosom Harbor and Rose Island.
1970 ROSS: STUDIES ON THE TETRACLITIDAE 3
KEY TO GENERA OF THE TESSEROPORAN GROUP
1 . Parietal tubes uniformly distributed in one row 2
1 . Parietal tubes uniformly distributed in more than two rows, or lacking 3
2. Parietal tubes bearing transverse septa; scutum lacking depressor
muscle crests (1 sp., eastern Pacific, Pliocene) Tesseroplax
2. Parietal tubes lacking transverse septa; scutum bearing depressor muscle
crests (5 spp., Indo-West Pacific, Recent; Italy, Oligocene) Tesseropora
3. Inner lamina present; longitudinal septa continuous; sheath adpressed,
basal margin not depending (19 spp., tropical, warm temperate,
cosmopolitan. Pliocene to Recent) Tetraclita
3. Inner lamina absent; longitudinal septa discontinuous; sheath free
with basal margin depending (3 spp., southeast Australia, New Zealand,
Recent) Epopella
Epopella gen. nov.
Definition. — Shell large, conic; compartments may or may not be discrete; parietes
effectively solid, permeated with cuticular chitin, and commonly discontinuous plates or
longitudinal lamina depend from inner surface; radii non-tubiferous, narrow or obsolete;
basis membranous; scutum triangular, higher than wide, bearing crests for depressor
muscles; tergum narrow, spur not well separated from basi-scutal angle, truncate basally;
mandible with 4 teeth, basal comb, and spine-like inferior angle; maxilla I with 10-16 spines
comprising medial cluster of cutting edge.
Type species. — Tetraclita (Tetraclita) purpurascens forma breviscutum Broch, 1922,
Recent, Auckland Island.
Etymology. — Named in honor of Elizabeth C. Pope, the Australian Museum, in
recognition of her many contributions to the Cirripedia of the Australian region.
Epopella breviscutum (Broch)
Tetraclita (Tetraclita) purpurascens forma breviscutum Broch, 1922: 337, figs. 71, 72.
Tetraclita ( Tetraclitella ) purpurascens forma, breviscutum: Hiro, 1939: 275.
Material. — Rose Island, Auckland Islands; intertidal; J. C. Yaldwyn, coll., January,
1963; 2 dried specimens lacking appendages and body; in collections of Dominion
Museum, Wellington, New Zealand.
Sarah's Bosom Harbor (Port Ross), Auckland Island, Auckland Islands; under stones
at low tide; Th. Mortensen Pacific Expedition, November 26, 1914; 5 complete specimens;
in collections of Universitetets Zoologiske Museum, Copenhagen, Denmark.
Supplementary Description. — Shell low, conic; grayish-white; parietes deeply
eroded; growth ridges discernible along basal margin only; orifice pentagonal with
peritreme eroded; radii extremely narrow or obsolete, with articular surfaces weakly
crenate; compartments weakly articulated when not secondarily fused; no inner lamina;
longitudinal septa discontinuous basally, not fused, forming separate, smooth, depending
plates (Fig. 2d), in general appearance not much unlike that of Chelonobia testudinaria;
basal margin of sheath free, depending (Fig. 2d). Basis membranous. Measurements (in
mm.) of the lectotype (26-XI-14D), paralectotypes (26-XI-A-C,E), and specimens from
Rose Island (spec. F, G) are presented in Table 1 .
External surface of opercular plates deeply eroded (Fig. 2a, b). Scutum triangular,
commonly slightly higher than wide, articular ridge sinuous, about 2/3 length of tergal
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
K^
'^l^'^-
; \
\
V^
Figure 2. Shell and opercular plates of Epopella brevisculum. a, external view of scutum; b, external view of
tergum; c, internal view of tergum; d, basal view of shell; e, internal view of scutum. Lectotype (26-XI-14D), a-c,
e; paraleclotype (26-XI-14A), d.
margin; adductor ridge erect, undercut, fused apically with articular ridge and extending
nearly to basal margin; adductor muscle depression ovate, deep; 3-4 crests for rostral
1970
ROSS: STUDIES ON THE TETRACLITIDAE
depressor muscle, low, thin, partially hidden by infolding of occludent margin; 4-5 crests
for lateral depressor muscle, high, broad, clearly exposed; overall surface of plate pitted
(Fig. 2e).
Tergum higher than wide, apically eroded; external longitudinal furrow open, shallow,
extending to base of spur; spur truncate basally, width more than 1/2 that of basal margin;
articular ridge erect or inclined, undercut along basal portion; articular furrow broad and
deep; 5-7 crests for depressor muscles, short, broad, erect, bearing close-spaced, thin,
lateral extensions (Fig. 2c).
Table 1. Measurements of Individual Specimens
Specimen
Auckland Id.
26-XI-14A
26-XI-14B
26-XI-14C
26-XI-14 D(lectotype)
26-XI-14E
Rose Id.
F
G
Shell
Opercu
lar Plates
C-RDia.
Width
Height
S. H.
S. W.
T. H.
T. W
29.9
29.2
14.4
9.4
8.7
8.3
5.1
31.1
26.9
12.2
8.3
9.1
8.1
5.3
31.0
30.9
14.3
10.1
9.4
8.7
4.9
28.7
25.9
14.8
9.1
9.2
8.4
5.1
32.9
31.4
12.7
8.2
10.2
8.9
5.2
16.8
19.4
9.4
7.8
7.2
6.4
4.8
16.1
18.5
10.8
6.8
6.5
5.1
4.8
Crest of labrum thick, heavily chitinized, with shallow medial notch (Fig. 30;
multidenticulate, 22-39 simple teeth along crest and in notch (Fig. 3g); short soft setae
along crest and commonly between the teeth.
Palps bluntly rounded distally; superior margin concave, basal convex; distal setae
1/2 longer than superior; both bipinnate.
Mandible with 5 teeth including inferior angle (Fig. 3a); teeth 2-4 with subsidiary
cusps; superior slope of tooth 4 smooth; inferior angle coarsely serrate, 28-42 overlapping,
narrowly triangular teeth.
Maxilla I deeply notched subapically (Fig. 3c); spines along cutting edge in 3 clusters;
2 long, stout and 4-6 short, slender spines above notch; 10-16 long or short slender spines
medially; 7-15 very short and slender spines basally.
Maxilla II taller than broad (Fig. 3e); anterior margin bilobate; basal lobe covered
with cluster of pustules along anterior border.
Rami of cirrus I grossly unequal in length (Fig. 4a); posterior ramus about 1/2 length
of anterior ramus. Rami of cirrus II either essentially equal in length or inner ramus
slightly shorter; intermediate articles of both rami squat, slightly protuberant; setae on
both rami coarsely bipinnate, not comb-like. Rami of cirrus III antenniform (Fig. 4d);
outer ramus approximately 3/5 length of inner ramus; basal segments of both rami armed
with comb-setae lacking basal guards (Fig. 4e). Cirri IV-VI essentially equal in length with
equal rami; 3-5 short, slender setae at each articulation along greater curvature of
intermediate articles; a single row of ctenae occurs along lateral face immediately below
articulation; commonly 4 pairs of setae on cirri IV-V, and 3 on cirrus VI (Fig. 40: at base
of and between each major pair of setae is a cluster of 4-9 long bristles. Cirral counts for
the specimens in the type lot are summarized in Figure 5.
Intromittent organ annulated throughout its length, and sparsely covered with short
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
bf 0 05mm
a 0 25mm
9 0 1mm
0 25mm
0 5mm
cd
■—^^UMvtaL
'>^^^
Figure 3. Trophi of Epopella breviscutum. a, mandible; b, enlarged view of inferior angle of mandible in fig. a; c,
maxilla I; d, palp; e, maxilla II; f, labrum and palp; g, enlarged view of labral crest. Lectotype (26-X1-14D), a-e,
g; paralectotype(26-XI-14B), f.
1970
ROSS: STUDIES ON THE TETRACLITIDAE
^0 05mm
abed 0.5mm
0.25 mm
0.25 mm
Figure 4. Thoracic appendages of Epopella breviscutum. a, left cirrus I; b, right cirrus II; c, right cirrus III; d,
left cirrus III; e, comb seta from segment 5 of outer ramus of left cirrus III; f, intermediate segment of right outer
ramus of cirrus VI; g,h. distal extremity of intromittant organ. Lectotype (26-XI-14D), b, c. f; paralectotypes, a,
d. e. g-h (e,g.d = 26-XI-14C; a = 26-XI-14A; h = 2X-XI-14B).
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
36-
32i
28-
1
0)
-
1- ^.
z 24-
^
r ^J
UJ
,--
2
^
(D
^
^
~^"
m
_
_
^"20-
_
_
1
.
_
li-
1
0
1
1
Sl6-
_
1
1
1
\
1
_
D
\
Z
\
/
12-
-V-
-L
ANTERIOR RAMUS
8-
n=
(8)
(7)
(7) (8) (7) (7)
III IV V
CIRRUS
VI
36-
32-
28-
24-
;
;
/
;
/
;
\
\
\
\
\
\
--
20-
/ -
/
;
/
;
/
\
\
-
-
16-
/
/
/
/
/
12-
/
/
-i-
POSTERIOR RAMUS
8-
n=
(8)
(7) (8) (8) (7) (8)
Ill IV V
CIRRUS
VI
Figure 5. Graph of range (vertical line) and mean values (dotted line) for cirral counts of anterior ramus (left)
and posterior ramus (right) of Epopella breviscutum.
bristles; distal extremity bears two large separate clusters of long setules (Fig. 4g).
Embryos in mantle cavity average 0.10 x 0.22 mm. Stage I nauplii in mantle cavity
average 0.1 2 x 0.20 mm.
Remarks. — Broch illustrated the opercular plates and trophi of breviscutum, but
failed to select a holotype. Because the specimen or specimens he figured are no longer
available, the specimens here figured (opercular plates, figs. 2a-c, 3; trophi, figs. 3a-e, g;
cirri, figs. 4b, c, 0 are designated the lectotype (26-XI-14D), and the remaining specimens,
of which I have seen 4, are designated paralectotypes (26-XI-14A-C, E).
DISCUSSION
Monometric shell growth, non-tubiferous and narrow or obsolete radii, and the orifice
enlarged by attrition rather than diametric growth clearly establish E. breviscutum as a
member of the tesseroporan group (Ross, 1969: 238).
The space between the inner and outer lamina in Tetraclita is filled with a network of
continuous longitudinal septa, which in effect create longitudinal tubes. These are more or
less uniform in section, and occur in rows with the smallest and shortest tubes parallel to
the outer lamina. In Tesseropora and Tesseroplax there is basically but one row of these
tubes. Epopella breviscutum lacks an inner lamina, and thus is effectively solid walled. A
1970 ROSS: STUDIES ON THE TETRACLITIDAE 9
non-tubiferous or solid wall characterizes the geologically earliest chthamalids and
balanids (Ross, 1965: 61; Ross and Newman, 1967:4; Newman, ZuUo and Wainwright,
1967: 167). In the tesseroporan lineage, I interpret the evolutionary trend then as having
been from a solid walled form with diametric growth (Ross, 1969: 240) to a solid walled
form with monometric growth, such as E. breviscutum or an earlier related species, to
Tesseropora with a single row of parietal tubes and not uncommonly secondary tubules,
and terminating with Tetraclita. Tesseroplax, also with a single row of tubes, is an early
derivative of Tesseropora.
Much confusion exists over the systematic position o{ Elminius.\^rgQ\y because certain
of the included species are morphologically similar to the Balanidae on one hand and to the
Tetraclitidae on the other (Darwin, 1854: 346). Those similar to the Balanidae have a
deeply notched or incised labrum (Moore, 1944: pi. 46), and an intromittent organ bearing
a basidorsal point (Nilsson-Cantell, 1930: 225). Those similar to the Tetraclitidae have a
shallow or slightly notched labrum (Broch, 1922: 341-342), lack the basidoral point, have
complex setae on cirrus III (lacking in the Balanidae) that exhibits antenniformy (Moore,
1944: 328), and there are gross similarities in the opercular plates. Additional morpholog-
ical characters, especially in the shell, as noted below, strengthen the inference that at
least two species of Eliminius, namely E. plicatus Gray and E. simplex Darwin, are
tetraclitids rather than balanids. The criteria for forming this group are supported by the
distribution of the species involved, all three occurring within the southeastern Australia-
New Zealand region.
The parietal plates in E. breviscutum are complex, not only because they are a laminate
of calcareous and chitinous materials, but because the inner surface of the wall develops an
elaborate irregular series of depending ridges or longitudinal septa. These undoubtedly
impart rigidity and strength to the wall and provide a broad base of attachment and
vertical support, much as in Emersonius and Chelonobia (Ross and Newman, 1967: 16).
The internal structure of the parietes in E. plicatus is much like that of £. breviscutum, but
in E. simplex the chitinous material occurs in a row of equidistantly spaced thin columns
instead of continuous ribbons.
In E. breviscutum irregularly scattered between the depending ridges are narrow
tubules that in section are either oval, circular, or irregular. Similar surficial depressions
occur in E. plicatus. E. simplex, Tesseropora (at the tips of the secondary longitudinal
septa), Tetraclita (see Pilsbry, 1916: 252) and not uncommonly in Chthamalus. Since
Epopella lacks an inner lamina these "tubules" are not homologous with the parietal tubes
or secondary tubules of other tesseroporans. The functional significance of these tubules
and depressions remains unknown.
Secondary calcification of the parietal tubes in tesseroporans aids in maintaining the
shell in environments where it is subjected to abrasion or corrosion. The shell in
Tesseroplax is strengthened by apical filling of the parietal tubes, much as in Tesseropora
and Tetraclita, and by the formation of transverse septa in the basal portion. In Epopella,
deposition of a layer of calcium carbonate between the youngest series of longitudinal
septa serves the same function. Henry (1957: 36) has suggested that in Tesseropora pacifica
the shell is reinforced through development of elaborate, hollow, spinous processes that
extend into the parietal tube cavities, but further work is needed to substantiate this.
In the Balanomorpha there has been selection both for structural reinforcement of the
shell (Darwin, 1854) and for the development in deep water forms of a protective
mechanism against boring organisms (Newman and Ross, in press). However, Epopella
and other tesseroporans in general differ from these deep water forms in having a relatively
much thicker and more complex wall. The development of a thick, laminated shell in the
10 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
tesseroporans is probably an adaptation for the rigors of the intertidal zone (cf. Paine, 1966).
There is a considerable interval between the earliest known occurrence (Recent) of the
evolutionary more primitive Epopella and of the more complex Tesseropora (Oligocene).
Thus the Epopella lineage must be significantly older than the fossil evidence indicates, a
conclusion that is also suggested by the morphologically complex shell of E. breviscutum.
Therefore, it seems probable that the origin of the tetraclitids should be sought in rocks
dating from the Eocene if not the Paleocene or Cretaceous.
ACKNOWLEDGMENTS
I am indebted to Dr. Torben Wolff, Universitetets Zoologiske Museum, for the extended loan of Broch's
type series of this species as well as other tetraclitids collected during the Th. Mortensen expeditions. Elizabeth
Pope of the Australian Museum kindly searched the collections in her charge on numerous occasions for
specimens, which she has made available to me. Thanks are also due Dr. William A. Newman, Scripps Institution
of Oceanography, for many invaluable discussions relating to various aspects of this and earlier studies on the
Tetraclitidae. For comparative material and other courtesies I thank Dr. Huzio Utinomi, Seto Marine Biological
Laboratory, Dr. Elizabeth J. Batham, Portobello Marine Biological Station, New Zealand, Dr. Victor A. Zullo,
California Academy of Sciences, and Brian Foster, University College of North Wales.
LITERATURE CITED
Broch, H.
1922. Papers from Dr. Th. Mortensen's Pacific Expedition 1914-1916. X. Studies on Pacific cirripeds.
Vidensk. Meddel. Dansk Naturhist. foren. Copenhagen 73: 215-358.
Darwin, C. R.
1854. A monograph on the sub-class Cirripedia, with figures of all the species. The Balanidae (or sessile
cirripedes): the Verrucidae etc.. etc., etc. London, Ray Society, 1-684, pis. 1-30.
Foster, B.A.
1967a. A guide to the littoral balanomorph barnacles of New Zealand. Tuatara 1 5 (2); 75-86.
1 967b. The early stages of some New Zealand shore barnacles. Tane 1 3: 33-42.
Henry, D. P.
1957. Some littoral barnacles from the Tuamotu, Marshall, and Caroline Islands. Proc. U.S. Natl. Mus.
107 (3381): 25-38.
Hiro, F.
1939. Studies on the Cirripedian fauna of Japan. IV. Cirripeds of Formosa (Taiwan), with some
geographical and ecological remarks on the littoral forms. Mem. Coll. Sci., Kyoto Imp. Univ., ser. B,
15 (2): 245-284.
Linzey, J. T.
1942. The balanomorph barnacles of the Kermadec Islands. Trans. Roy. Soc. New Zealand 71: 279-281.
Moore, L. B.
1944. Some intertidal barnacles of New Zealand. Trans. Roy. Soc. New Zealand 73 (4): 315-334, pis. 46-47.
Newman, W. A., and A. Ross
MS. Antarctic Cirripedia: A monograph based on the collections of the USNS Eltanin Expeditions. Amer.
Geophys. Union, Antarctic Res. Ser.. in press.
Newman, W. A., V. A. Zullo, and S. A. Wainwright
1967. A critique on recent concepts of growth in Balanomorpha (Cirripedia: Thoracica). Crustaceana 12 (2):
167-178.
Nilsson-Cantell, C. A.
1930. Thoracic cirripedes collected in 1925-1927. Discovery Repts. 2: 223-260.
Paine, R. T.
1966. Function of labial spines, composition of diet, and size of certain marine gastropods. Veliger 9(1):
17-24.
Pilsbry, HA.
1916. The sessile barnacles contained in the collection of the U.S. National Museum: including a
monograph of the American species. Bull. U.S. Natl. Mus. 93: 1-366.
1970 ROSS: STUDIES ON THE TETRACLITIDAE 11
Pope, E. C.
1945. A simplified key to the sessile barnacles found on the rocks, boats, wharf piles and other installations
in Port Jackson and adjacent waters. Rec. Australian Mus. 21 (6): 351-372, pis. 27-30.
Ross, Arnold
1 965. A new cirriped from the Eocene of Georgia. Quart. J. Florida Acad. Sci. 28(1): 59-67.
1969. Studies on the Tetraclitidae (Cirripedia: Thoracica): Revision of Tetraclita. Trans. San Diego Soc.
Nat. Hist. 15(15): 237-251.
Ross, A., and W. A. Newman
1967. Eocene Balanidae of Florida, including a new genus and species with a unique plan of "turtle-
barnacle" organization. Amer. Mus. Novitates 2288: 1-21.
Stephenson, W.
1968. The intertidal acorn barnacle Tetraclita vitiata Darwin at Heron Island. Univ. Queensland Pap. 1 (3):
51-59.
Wisely, B.,and R. A. P. Blick
1964. Seasonal abundance of first stage nauplii in 10 species of barnacles at Sydney. Australian Jour. Mar.
Freshwater Res. 15(2): 162-171.
Department of Invertebrate Paleontology, Natural History Museum. P.O. Box 1390,
San Diego, California 92112.
MUS. COMP. ZOOL.
LIBRARY
JUN16 1970
HARVARD
UNIVERSITY
THE SHALLOW WATER ANOMURAN CRAB FAUNA
OF SOUTHWESTERN BAJA CALIFORNIA, MEXICO
JANET HAIG, THOMAS S. HOPKINS
AND THOMAS B. SCANLAND
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 2 4 JUNE 1970
THE SHALLOW WATER ANOMURAN CRAB FAUNA
OF SOUTHWESTERN BAJA CALIFORNIA, MEXICO
JANET HAIG, THOMAS S. HOPKINS
AND THOMAS B. SCANLAND
ABSTRACT. - Thirty-five species of anomuran crabs are reported from the 1964 "Mag Bay" Expedition,
19 orv\hich are new records for the outer coast of southern Baja California, Mexico. A checklist and keys are
appended for the 52 species of Anomura nov\ known to inhabit this area.
RESUMEN. — Durante la Expedicion "'Mag Bay" en 1964, se observaron treinta y cinco especies de
cangrejos anomuros, diez y nueve de ellas encontradas por primera vez, en las costas occidentales de la zona
meridional de Baja California. Mexico. Se incluye una lista y las claves correspondientes para las 52 especies
de Anomuros observados hasta la fecha en aquellas regiones.
The purpose of the "Mag Bay" Expedition of 1964 was to study the maritime biota
along the coast of Baja California, Mexico, from Punta San Eugenio (Punta Eugenia) to
the lower entrance of Bahfa Magadalena (Figs. 1, 2). Dr. Carl L. Hubbs, Scripps
Institution of Oceanography, was the originator and leader of the expedition, which was
supported by the Office of Naval Research. The scientific party was divided into three
teams: Team 1, aboard the R/V HORIZON, was responsible for sampling in deep water;
Team 2, in small craft, worked in the mangrove-estuarine environment; Team 3, aboard
the Scripps vessel T-441, was responsible for "on site" fish poisonings, invertebrate and
algal collecting, and otter trawling along the 20- and 40 m depth contours in the area of
study.
Two of the authors (TSH and TBS) were members of Team 2, where SCUBA was
used in depths of one to 30 meters. A concerted effort was made to collect decapod
crustaceans and echinoderms, as well as fish. The specimens were returned to the T-441,
where they were kept alive until color notes and tentative identifications could be recorded.
At the conclusion of the cruise, the anomuran crabs were forwarded to the Allan Hancock
Foundation for study by the senior author (JH).
Thirty-five species of Anomura were collected, of which 19 constitute new records for
the outer coast of southern Baja California (Punta San Eugenio and southward) within the
40 m contour. A checklist and keys to all species known to fall within these geographical
and bathymetrical limits are appended.
This report is a Contribution from the Allan Hancock Foundation, no. - 339 ,
supported by NSF Grant GB-2039, and a Contribution from the Scripps Institution of
Oceanography, supported by NSF Grant GB 2312 (to D. L. Fox).
HISTORICAL RESUME
The following anomuran crabs are presently known from the region under consid-
eration: Miinida mexicana (Benedict, 1902), Dardaniis sinistripes (Rathbun, 1910), Pleu-
roncodes planipes, and Emerita analoga (Schmitt, 1921). In addition, Glassell (1936)
reported on several porcellanids which he collected at Bahfa Magdalena. These included
Petrolisthes hirtipes (Lockington), and the new species Orthochela pumila. Pisosoma erosa
{ = Megalobrachium erosum), and Porcellana magdalenensis ( = Pisidia magdalenensis).
He also treated the porcellanid and hermit crabs from the Templeton Crocker Expedition of
SAN DIEGO SOC. NAT. HIST.. TRANS. 16 (2): 13-32, 4 JUNE 1970
14
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 1. Map of Baja California, Mexico. Area delimited by the box on the western side of the peninsula is
shown enlarged in figure 2.
1936 (Glassell, 1937a, 1937b). However, only one species comes within the scope of the
present work, Paguristes bakeri Holmes, which was collected otT Isla Cedros (Glassell,
1937b.) Schmitt (1939) listed "Paguristes species'' from Bahfa Magdalena, a form still
awaiting description. To the known fauna Haig (1960) added 16 species of Porcellanidae.
1970
HAIG, HOPKINS AND SCANLAND: ANOMURAN CRABS
Figure 2. Map of Bahia Magdalena region, Baja California, Mexico.
The presence of 19 additional species in this fauna, and the larger number of new
records, attests to the effectiveness of SCUBA in an area already well surveyed by dredging
and intertidal collecting.
ANNOTATED SPECIES LIST
Family COENOBITIDAE
Coenobita compressus H. Milne Edwards
Cenobita compressa H. Milne Edwards, 1837: 241.
Coenobita compressus -.Boone, 1931: 145, text-fig. 3; Holthuis, 1954: 16, text-figs. 4a-b.
Recorded Range. — Santa Rosalfa, Golfo de California, to Estrecho de Magallanes.
Islas Revillagigedo; Isla del Coco; Archipielago de Galapagos.
Material. — Punta Belcher; above high tide at night; 2 Feb. 1964; Id', 2 9.
Remarks. — Glassell (1937b: 242-243) stated: "For the most part these terrestrial
hermit crabs inhabit the land bordering on the sea. They select heavy shells for their abode.
They are, in the main, vegetarians, though they do not limit their diet and may at times act
as scavengers, or become carnivorous ... In addition they are good tree climbers."
16 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
On Punta Belcher we observed that these animals are nocturnal. A search for their
homesites was fruitless, although the crabs "appeared" within 17m of the camp just after
dark, and could not be found after daybreak. Probably they had remained inactive in their
shells along the upper tide marks during the day, and thus went unnoticed.
Six to 10 individuals were observed feeding on dead or molted Pleuroncodes planipes
which formed windrows on the beach.
The range of this species is extended to the outer Baja California coast, and 300 km
north along the outer coast.
Family DIOGENIDAE
Dardanus sinistripes (Stimpson)
Pagurus sinistripes Stimpson, 1858: 22i (nom. nud): \S59: S2.
Dardanus sinistripes: Glassell, 1937b: 251.
Recorded Range. — Bahfa Magdalena (Rathbun, 1910) and Isla Tiburon, Golfo de
California, to Bahfa de Sechura, Peru.
Material. — Off Boca de Santo Domingo; otter trawl; 40 m; 27 Jan. 1964; 1 ? . Laguna
de Santa Maria: 1.2 m; 29 Jan. 1964; Id".
Color. — "In alcohol, the carapace is buff with red markings. The chelipeds, purple
and red, with the interior margins of the meri white; the teeth of the fingers white, bordered
with yellow. The ambulatory legs are purple, their dactyli with dark brown setae, their
meri and carpi blotched on a light ground with red" (Glassell, 1937b). There are two
narrow, dark bands, one median and the other proximal, on the eyestalk.
Calcinus californiensis Bouvier
Calcinus californiensis Bouvier, 1898: 380; Glassell, 1937b: 252.
Calcinus californiensis: Chace, 1962: 627, text-figs. 5-6.
Recorded Range. — Isla San Jose, Golfo de California, to Acapulco, Me'xico. Isla
Clipperton.
Material. — Punta Cala; 3 m; 31 Jan. 1964; Id, 1 ?.
Roca de la Vela;6 m; 1 Feb. 1964; 2$. Inside Punta Tosca, in lagoon; 5 m; 4 Feb.
1964; 5c^, 2 9.
Color. — The coloration of this and allied species, in both live and preserved material,
was discussed in some detail by Chace (1962: 628). A broad white band at the base of the
cornea and the solid dark color of the dactyls of the walking legs unmistakably identify the
above specimens as Calcinus californiensis.
Remarks. — The range of this species is extended to the outer Baja California coast,
and 300 km north.
Aniculus elegans Stimpson
Aniculus elegans Stimpson, 1858: 234 («om. nud.): 1859: 83; Boone, 1931: 140, text-fig. 1.
Recorded Range. — Golfo de California (exact locality not stated) to Cabo de San
Francisco, Ecuador.
Material. -Outside Bahfa Magdalena; 18 m; 1 Feb. 1964; 1 juv. Inside Punta Tosca,
in lagoon; 5 m; 4 Feb. 1964; 1 ?. Off Punta Redonda; 15 m; 5 Feb. 1964; 6d", 5?.
Color. — Carapace red, with a pink area on the posterior part of the shield. Eyestalks
tan. Chelipeds pink, with red on the fingers and on the distal half of the chelae. Dactyls of
walking legs dark- red; other segments pink. A broad, submedian red band on the
propodus, a submedian red blotch on the anterior margin of the carpus, and a smaller
median blotch on the anterior margin of the merus.
Remarks. — Off Punta Redonda these hermits occurred under rocks in aggregates of
1970 HAIG, HOPKINS AND SCANLAND: ANOMURAN CRABS 17
five or more. Porcellana paguriconviva Glassell were found in the shells of specimens
collected at the same locality, an association that has not been recorded previously.
The range of this species is extended to the outer Baja California coast, and 300 km
north.
Trizopagurus magnificus (Bouvier)
Clibanarius magnificus Bouvier, 1898: 378.
Clibanarius chetyrkini Boone, 1932: 29, text-fig. 8.
Trizopagurus magnificus: Forest, 1952: 4, 12, text-figs. 2, 11, 18.
Recorded Range. — Golfo de California (exact locality not stated) to Isla de la Plata,
Ecuador. Archipielago de Galapagos.
Material. — Roca de la Vela; 6 m; 1 Feb. 1964; 1 ?. Off Punta Redonda; 15 m; 5 Feb.
1964; 15.
Color. — The species may be readily identified by the large, irregular pale blotches on
a dark background which cover the carapace shield, eyestalks, chelipeds, and walking legs.
Remarks. — The range of this species is extended to the outer Baja California coast,
and 300 km north.
Clibanarius panamensis Stimpson
Clibanarius panamensis Stimpson, 1858:235 (nom nud.): 1859: 84; Holthuis, 1954: 23, text-figs. 7-8.
Recorded Range. — Santa Rosalfa, Golfo de California, to Isla de la Correa, Peru.
Material. — Laguna de Santa Marfa; in +1 m and out of water; 29 Jan. 1964; 8c?, 1 ?.
Color. — Hermits of this species may be recognized immediately by the color pattern
of the walking legs, which consists of longitudinal dark and light stripes, subequal in width,
on each segment.
Remarks. — The range of this species is extended to the outer Baja California coast,
and 300 km north.
Isocheles pilosus (Holmes)
Hoiopagurus pilosus Holmes, 1900: 154;Schmitt, 1921: 127, pi. 17 fig. 2; Ricketts and Calvin, 1939: 189, pi. 39
fig. 2.
Isocheles pilosus: Forest, 1964: 294.
Recorded Range. — Off San Francisco Bay, California, to Estero de Punta Banda,
outer Baja California.
Material. — Punta Abreojos, Bahfa de Ballenas; +.7 m while wading; 29 Jan. 1964;
1 . 2.5 miles west of Boca de Santo Domingo; 16' otter trawl; 14-20 m; 26 Jan. 1964; Id".
Outside Punta Hughes; 20 m; 30 Jan. 1964; 1 2 (juv.). Punta Pequena, Bahfa de San
Juanico; 3 m; 8 Feb. 1964; Id", 1 juv.
Remarks. — Identification of the above specimens with Isocheles pilosus is tentative,
pending a revision of genus Isocheles by J. Forest of the Muse'um National d'Histoire
Naturelle, Paris. Some of the material may belong to /. pacificus Bouvier (see Forest,
1964: 291, text-fig. 11). Neither /. pilosus nor /. pacificus has heretofore been reported
from the outer coast of southern Baja California, and the range is extended 850 km south.
Paguristes bakeri Holmes
Paguristes bakeri Holmes. 1900: 152;Schmitt, 1921: 122, 124, pi. 18 figs. 2, 6; Glassell, 1937b: 243, 244.
Paguristes holmesi Glassell, 1937b: 243, 247.
Recorded Range. — Outside San Francisco Bay, California, southward along the
California and outer Baja California coast (Glassell, 1937b; Parker, 1964), and in Golfo de
California as far north as Punta Baja.
Material. — HORIZON Sta. A-11, 8 miles west of Punta Redonda; Isaacs-Kidd
18 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
midwater trawl, scraped on sand bottom; 106-1 16 m; 29 Jan. 1964; 1 juv.
Remarks. — Studies now in progress by JH indicate that Paguristes holmesi is a
synonym of P. bakeri Holmes. Paguristes bakeri has been recovered from shallow water
(generally in the northern half of its range), but it occurs more commonly in over 40 m and
has been reported from depths as great as 232 m.
Paguristes ulreyi Schmitt
Paguristes ulreyi Schmitt, 1921: 123, 125, pi. 18 figs. 3-5, 7.
Paguristes occator Glassell, 1937b: 243, 244.
Recorded Range. — Monterey Bay, California, southward along the California and
outer Baja California coast, and in Golfo de California as far north as Punta Gorda.
Material. — Bajio Knepper, Punta Abreojos; 17-20 m; 9 Feb. 1964; 1 ? ovig. Bahia
de San Hipolito; 10-13 m; 9 Feb. 1964; 5cf, 3$ovig., 1 juv. Without data; Id", 2 9(1
ovig.), 2 juv.
Remarks. — Examination of specimens in the collections of the Allan Hancock
Foundation indicates that this species is not uncommon in the southernmost part of
Golfo de California, where it occurs in deeper water than it does in higher latitudes.
Paguristes occator Glassell seems to be a synonym.
Paguristes parvus Holmes
Paguristes parvus Holmes, 1900: 151, pi. 2 fig. 26; Schmitt, 1921: 123, 124, pi. 17 fig. 1, text-fig. 83.
Recorded Range. — Known only from White's Point near San Pedro, California.
Material. — Arrecife Sacramento; 13 m, from kelp holdfast; 25 Jan. 1964; IcT.
Remarks. — On the basis of the above specimen, the range of Paguristes parvus is
extended southward 550 km.
Paguristes anahuacus Glassell
Paguristes anahuacus Glassell, 1938:421.
Recorded Range. — Reported only from Punta Penasco, Golfo de California.
Material. — Outside Punta Hughes; 20 m; 30 Jan. 1964; 1 d. Roca de la Vela; 6 m; 1
Feb. 1964; Id", 1 ?. Outside Isla Santa Margarita, west of Punta Tosca; 21-25 m; 3 Feb.
1964; 19. Off Punta Redonda; 15 m; 5 Feb. 1964; Icf, 19. Bajfo Knepper, Punta
Abreojos; 16-20 m; 9 Feb. 1964; 119. Without data; 3 9 ovig.
Color. — Carapace shield with a broad median longitudinal orange stripe on ante-
rior half; orange blotches on median portion of lateral margins. Proximal half of eyestalk
orange; distal half bright purple, with a narrow white band at base of cornea. Antennal
and antennular flagella purple. Pereiopods orange.
Remarks. — Punta Penasco is 1000 km north of Cabo San Lucas on the mainland
side of the Gulf. The range of this species is further extended to the outer coast of Baja
California, and 550 km north.
Paguristes praedator Glassell
Paguristes praedator Glassell, 1937b: 243, 245.
Recorded Range. — Golfo de California, from Bahfa de Santa Ine's and Isla Tiburon
south to Isla Isabel.
Material. — Off Boca de Santo Domingo; 16' otter trawl; 40 m; 27 Jan. 1964; 1 juv.
Remarks. — Examination of a long series of specimens in the collections of the Allan
Hancock Foundation shows that this species occurs infrequently within the 40 m line; it has
been most often dredged in 60 m or deeper. The range is extended to the outer Baja
California coast, and 350 km north.
1970 HAIG, HOPKINS AND SCANLAND: ANOMURAN CRABS 19
Paguristes, undescribed species
Paguristes species, Schmitt, 1939:9.
Material. — Off Boca de las Animas; 16' otter trawl; 40 m; 27 Jan. 1964; Id" (juv.).
Off Boca de Santo Domingo; 1 6' otter trawl; 40 m; 27 Jan. 1964; 2 cT ( 1 juv.).
Color. — Carpus of chelipeds red; chelae white with red blotches, which merge to
form a transverse band about midway along each finger. Walking legs white; propodus and
dactyl each with a proximal and subdistal red ring.
Remarks. — This species, which will be described and illustrated in a future report,
appears to be abundant on the outer Baja California coast. The ovigerous female noted by
Schmitt (1939) was collected by the HOUSTON (Presidential Cruise of 1938) in Bahfa
Magdalena between Punta Belcher and the anchorage, in 20-30 m.
Family PAGURIDAE
Pagurus smithi (Benedict)
Eupagurus smithi Benedict, 1892: 4.
Pagurus smithi: Glassell, 1937b: 256, 259.
Recorded Range. — Golfo de California, from Estero de Tasiota to Punta Piaxtla on
the east side (Parker, 1964) and from Bahia de Santa Ines to Bahia de la Paz on the Baja
California peninsula.
Material. — Off Boca de Santo Domingo; 16' otter trawl; 40 m; 27 Jan. 1964; 2 juv.
Remarks. — This species is well represented in the collections of the Allan Hancock
Foundation from depths greater than 40 m; it seems to occur rarely within the 40 m
contour. The range is extended 150 km south to the outer Baja California coast, 350 km
north along the outer coast.
Pagurus, undescribed species (1 )
Material. — Off Boca de las Animas; 16' otter trawl; 20 and 40 m; 27 Jan. 1964; Id,
2 9ovig., 3 juv.
Remarks. — This small species, represented by considerable material in the collec-
tions of the Allan Hancock Foundation, will be described elsewhere. It belongs to the
group of Pagurus species having multispinulate eyescales.
Pagurus lepidus (Bouvier)
Eupagurus lepidus Bouvier, 1898:381.
Pagurus lepidus: Glassell, 1937b: 256.
? Pagurus lepidus: Chace, 1962: 623, text-fig. 2.
Recorded Range. — Golfo de California, from Puerto Penasco to El Mogote. ?Isla
Clipperton.
Material. — Off Boca de Santo Domingo; 16' otter trawl; 40 m; 27 Jan. 1964; Icf, 1?
ovig. Outside Punta Hughes; 20 m; 30 Jan. 1964; 4cf, 1 ? ovig. Punta Cala; 3 m; 31 Jan.
1964; 1 juv. Outside Bahfa Magdalena; 20 m; 1 Feb. 1964; 2cf. Isla Santa Margarita; 16'
otter trawl; 20 m; 4 Feb. 1964; Id", 1$. Punta Redonda; 15 m; 5 Feb. 1964; id'. Punta
Pequena, Bahfa de San Juanico; 3 m; 8 Feb. 1964; 2d', 1 ?. Bahfa de San Hipolito; 10-13 m;
9 Feb. 1964; 3d", 2 9 (1 ovig.). Bajfo Knepper, Punta Abreojos; 16-20 m; 9 Feb. 1964; 7d",
69(3 ovig.), 1 juv.
Color. — Most of the specimens showed a color pattern on the walking legs like that
in Chace's illustration (1962, text-fig. 2). In a few individuals the longitudinal stripes were
less well developed, and pigment was concentrated at the proximal end of the dactyl to
form a narrow ring. Current studies by JH show that there is a series of eastern Pacific
Pagurus species closely allied to P. lepidus and probably confused under that name; this
20 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
problem, as it concerns the "Mag Bay" material, will be discussed elsewhere.
Remarks. — The range of this species is extended to the outer Baja California coast
and 550 km north.
Pagurus galapagensis (Boone), new combination
N ympagurus galapagensis Boone, 1932: 17, text-fig. 5.
Recorded Range. — Known only from Bahia de Gardner, Isla Hood (or Espanola),
Archipielago de Galapagos.
Material. — Roca de la Vela; 6 m; 1 Feb. 1 964; 1 d', 1 9 .
Color. — Hand under dense tomentum pale orange, with tubercles of darker orange;
fingers white with an intense orange spot at tip of each. Walking legs with longitudinal
orange stripes, overlying a broad median orange ring on carpus and propodus and two
rings on dactyl.
Remarks. — Studies currently in progress by JH indicate that this species occurs
throughout the Panamic faunal province. Examination of the holotype (Cat. No. 12238 in
the American Museum of Natural History, New York) showed that it falls within Pagurus
as that genus is currently (although too broadly) defined. Its affinities are with a natural
group o{ Pagurus species recently discussed and designated as "groupe miamensis" (Forest
and Saint Laurent, 1968: 116).
Pagurus, undescribed species (2)
Material. — HORIZON Sta. A-11; 8 miles west of Punta Redonda; Issacs-Kidd
midwater trawl scraped on sand bottom; 106-1 16 m; 29 Jan. 1964; 2d", 29(1 ovig.).
Remarks. — This species has been collected on several occasions around the southern
part of the Baja California peninsula, never in less than 104 m. It will be described in a
future report.
Pylopagurus californiensis (Benedict), new combination
Eupagurus californiensis Benedict, 1892: 21; Faxon, 1895: 55, pi. 1 1 figs. 2, 2a-f.
Pagurus californiensis: Glassell, 1937b: 256, 257.
Recorded Range. — Santa Catalina Island, California, and Cabo Tepoca, Golfo de
California, to Darfen, Panama. Isla del Coco; Archipielago de Galapagos.
Material. —Outside Bahfa Magdalena; 20 m; 1 Feb. 1964; Id". Without data; Id".
Color. — Carapace shield with longitudinal dark and light streaks anteriorly; solid
color toward anterolateral margins. Eyestalks orange, with broad white ring submedially.
Basal antennal article and acicle orange; acicle darker along outer edge. Manus and fingers
of major cheliped with diffused orange; raised granular areas white; outer margin with
alternating while and orange spots. Carpus solid dark orange except for narrow longitudi-
nal white area along inner margin, and a few very small white spots on dorsal surface.
Merus mostly orange, with small white spots. Manus of minor cheliped with a large
irregular orange area on dorsal surface; a longitudinal orange stripe, not continued far
onto pollex, along outer edge. Carpus with thin orange stripe along each dorsolateral
margin; a broader stripe midway along both outer and inner lateral surfaces; another along
ventral surface. Merus of walking legs with two longitudinal red stripes on outer surface,
one on upper margin, and two on inner surface; carpus with three on outer surface, one on
dorsal margin, and two on inner surface; propodus with two on outer surface, one on dorsal
margin, one on ventral margin, and two on inner surface; dactyl with one each on outer
surface, dorsal margin, and inner surface. With the exception of those on inner surface of
merus, which are incomplete, none of these stripes are interrupted. All these stripes are
imposed on broad transverse bands of white and pale orange.
1970 HAIG, HOPKINS AND SCANLAND: ANOMURAN CRABS 21
Remarks. — This species, and the two that follow, have not been recognized as
members of the genus Pylopagurus and were consequently not included in Walton's (1954)
review of the eastern Pacific forms of that genus. The shape of the major chela and the
presence of paired first pleopods in the female place all three species with Pylopagurus.
Although there are no published records of Pylopagurus californiensis along the outer
coast of Baja California, the distribution between Santa Catalina Island and Bahfa
Magdalena is not interrupted; specimens from many intermediate localities are in the
collections of the Allan Hancock Foundation.
Pylopagurus venustus (Bouvier), new combination
Eupagurus venustus Bouvier, 1898: 383.
Recorded Range. — Known only from Bahia de la Paz, Golfo de California.
Material. — Outside Punta Hughes; 20 m; 30 Jan. 1964; 1 ? ovig.
Color. — Ground color white and pale orange, with darker shades of orange-brown.
Carapace pale brown, with two irregular small brown blotches anteriorly. A narrow band
of pale orange on eyestalks at about level of tip of eyescales. Major chela with irregular
diffused brown except on fingers and distal half of lateral expansion. A few small dots on
fingers. Other segments of major cheliped with irregular blotches; darker on lateral
surfaces. Minor cheliped with two broad bands on manus, one on carpus, one on hand and
basal part of fingers; these bands are brown, edged in darker brown, and with an irregular,
sinuous outline. Walking legs with similar, sinuous-margined bands: two on merus, a distal
one on carpus, a median one on propodus, and a proximal one on dactyl. Non-banded
areas white and very pale orange.
Remarks. — The characteristic broad bands with sinuous margins which decorate the
walking legs unfortunately fade rapidly in alcohol; specimens can then best be distin-
guished from faded specimens of Pylopagurus californiensis by the carpus of the right
cheliped, which in P. venustus is covered dorsally by numerous forward-directed spinules.
The range is extended 150 km south to the outer Baja California coast, and 300 km
north along the outer coast.
Pylopagurus roseus (Benedict), new combination
Eupagurus roseus Benedict, 1892: 22.
Recorded Range. — The type locality was given only as "Gulf of California" by
Benedict. According to the accompanying label, however, the holotype and only known
specimen was collected off Bahfa Adair, in the northernmost part of the Gulf, 1000 km
north of Cabo San Lucas.
Material. ~ North of Punta Belcher; 8 m; 2 Feb. 1964; Id", 1?. Punta Cala; 5 m; 6
Feb. 1964; Ic/, 2 9 ovig.
Color. — Eyestalks white, with broad orange band at about level of tips of eyescales.
Chelipeds orange-brown; carpus with many small white spots. Walking legs orange-brown;
merus with large white blotches; propodus and dactyl each with a broad, subdistal white
ring and a distal narrower one.
Remarks. — The range is extended to the outer Baja California coast, and 300 km
north.
Pylopagurus diegensis Scanland and Hopkins
Pylopagurus diegensis Scanland and Hopkins, 1969: 257, fig. 1.
Material. — Uncertain locality data; 1 9 .
Remarks. — This species has been collected at several localities in southern California
and northern Baja California.
22 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
Family GALATHEIDAE
Pleuroncodes planipes Stimpson
Pleuroncodes planipes Stimpson, 1860: 245; Schmitt, 1921: 163, pi. 31 fig. 2.
Recorded Range. — Monterey Bay, California, and northern Golfo de California, to
about 250 km south of Cabo San Lucas.
Material. — These crabs were very abundant at the surface just inside Punta Entrada
and many were collected during the expedition but were not critically examined. Large
windrows of either dead or molted individuals littered the beach just south of Punta
Belcher and were fed on by Coenobita compressus.
Remarks. — Pleuroncodes planipes normally ranges from Baja California southward;
Radovich (1961: 49-50) discussed the history of its occurrence off the California coast
during p)eriods of high ocean temperatures. Crabs of this species are pelagic, frequently
occurring near the surface in swarms several miles wide, and may be washed ashore in
great numbers; at other times they are not visible near the surface, but may be taken by
mid-depth or bottom trawling (Radovich 1961: 50). Swarming and mass strandings on the
outer coast of southern Baja California were noted by several authors, including Matthews
(1932: 472), Steinbeck and Ricketts (1941: 455), and Radovich (1961: 50).
Family PORCELLANIDAE
Orthochela pumila Glassell
Orihochela pumila Glassell, 1936: 296, pi. 21 fig. 1; Haig. 1960: 14, pi. 18 fig. 1, text-fig. 1.
Recorded Range. — Bahia Magdalena, outer Baja California, and Mazatlan, Mexico,
to Bahfa de Caraquez, Ecuador.
Material. — Inside Punta Hughes; 3-8 m; 29 Jan. 1964; 19d", 21 ? (18 ovig.), 1 juv.
Near Mexican naval establishment at Puerto Cortez, northwest end of Bahia de Almejas;
1.5 m; 5 Feb. 1964; 1 . Punta Pequelia, Bahfa de San Juanico; 3 m; 9 Feb. 1964; Id, 6 9 (4
ovig.). Bahfade San Hipolito; 10-13 m; 9 Feb. 1964; 46^, 3? ovig.
Color. — The specimens collected and described by Glassell were yellow, with red
striations on the carapace and some red areas on the chelae; they were found clinging to
yellow gorgonian corals. During the "Mag Bay" expedition collectors took some yellow
individuals with red markings; other specimens were solid purple; blotched red and yellow;
brown with white spots; and white with rust-colored spots. Each specimen perfectly
matched the color of the gorgonian coral upon which it was found.
Remarks. — Several other porcellanids were found associated with Orthochela pu-
mila on gorgonians. Unlike Orthochela. however, they are not obligatory commensals but
take shelter in a variety of situations.
Orthochela pumila was previously collected by Glassell at Bahfa Magdalena, the type
locality. On the basis of specimens collected during this expedition, the range of the species
is extended northward along the outer Baja California coast 300 km to Bahfa de San
Hipolito.
Petrolisthes sanfelipensis Glassell
Petrolisthes sanfelipensis Glassell, 1936: 281; Haig, I960: 24, 30, pi. 20 fig. 3.
Recorded Range. — Bahfa de San Juanico to BahTa Magdalena, outer Baja Califor-
nia (Haig, I960); Punta Penasco to Guaymas, Golfo de California.
Material. — Olitside Punta Hughes; 20 m; 30 Jan. 1964; 1 cT Ouv.). Near Mexican
naval establishment at Puerto Cortez, northwest end of Bahfa de Almejas; 1.5 m; 5 Feb.
1964; 1?. Punta Pequena, Bahfade SanJuanico; 3 m; 9 Feb. 1964; lc^(juv.).
Remarks. — The specimen from Punta Pequena was taken from a gorgonian. The
1970 HAIG, HOPKINS AND SCANLAND: ANOMURAN CRABS 23
range of this species is now extended slightly northward in Bahia de San Juanico to Punta
Pequena.
Petrolisthes hians Nobili
Petrolisthes hians Nobili. 1901: 17; Haig, 1960: 26, 121, pi. 22 fig. 3.
Pisosoma flagraciliata Glassell, 1937a: 82, pi. 1 fig. 2.
Recorded Range. — Bahfa de Santa Maria, outer Baja California, and Guaymas,
Golfo de California, to Bahfa de Santa Elena, Ecuador. Islas Revillagigedo.
Material. — Inside Bahia Magdalena about 300 m north of Punta Belcher; 6 m; 1
Feb. 1964; 1 ovig. Inside Punta Tosca, in lagoon; 5 m; 4 Feb. 1964; 1 9 .
Remarks. — The specimen from Punta Tosca was taken from a sponge. The only
previous record for this species from the outer Baja California coast is from Bahia de
Santa Marfa (Haig, 1960).
Pachycheles marcortezensis Glassell
Pachycheles marcorlezensis Glassell, 1936: 290; Haig, 1960: 134. 149, pi. 33 fig. 3.
Recorded Range. — Bahfa de Santa Maria, outer Baja California (Haig, 1960); Isla
Xngel de la Guarda to Banco Arena, Golfo de California.
Material. — Off Isla Santa Margarita; otter trawl; 20 m; 4 Feb. 1964; 1 cf.
Pachycheles panamensis Faxon
Pachycheles panamensis Faxon, 1893: 175; 1895: 71, pi. 15 figs. 2, 2a; Haig, I960: 134, 155, pi. 33 fig. 1.
Pachycheles sonorensis Glassell, 1936: 291.
Recorded Range. — Isla Tiburon, Golfo de California, to Bahfa de Santa Elena,
Ecuador.
Material. — Inside Punta Hughes; 6 m; 29 Jan. 1964; 1 cT (]nv.).
Remarks. — The specimen was collected from a yellow gorgonian. The range is
extended to the outer Baja California coast, and 300 km north.
Pachycheles pubescens Holmes
Pachycheles pubescens Holmes, 1900: 1 10; Schmitt, 1921: 175, 177, pi. 33 fig. 4. text-fig. 1 12; Haig, 1960: 133,
162, pl.'34fig. 3.
Recorded Range. — Goose Island, British Columbia, to Cabeza de Thurloe, outer
Baja CaHfornia (Haig, 1960).
Material. — Without data; 1 juv.
Pachycheles holosericus Schmitt
Pachycheles holosericus Schmitt, in Nininger, 1918: 39, text-fig. 18 (nom. nud.) Schmitt, 1921: 175, 177, pi. 33
fig. 3; Haig, 1960: 133, 173, pi. 34 fig. 2.
Recorded Range. — Santa Barbara, California, to Bahfa Magdalena, outer Baja
CaHfornia (Haig, 1960).
Material. —Bahfa de San Hipolito; 10-13 m; 9 Feb. 1964; 3 juv.
Remarks. — The specimens were collected from a gorgonian coral.
Porcellana cancrisocialis Glassell
Porcellana cancrisocialis Glassell, 1936: 292; Haig, 1960: 198, 200, pi. 38 fig. 2, text-fig. 9 (2).
Recorded Range. — Bahfa de Santa Marfa and Punta Tosca, outer Baja California
(Haig, 1960); Punta Penasco, Golfo de California, to Bahfa de Santa Elena, Ecuador.
Material. — Off Bahi'a de San Juanico; 1 6' otter trawl; 40 m; 27 Jan. 1 964; 1 .
Remarks. — Porcellana cancrisocialis is often found associated with large hermit
crabs, but is sometimes free-living as was the above specimen. The range of this species is
24 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
extended northward from Bahfa de Santa Maria 150 km to Bahfa de San Juanico.
Porcellana paguriconviva Glassell
Porcellana paguriconviva Glassell, 1936: 293; Haig, 1960: 198, 203, pi. 38 fig. 1, text-fig. 9 (3).
Recorded Range. — Bahfa Magdalena, outer Baja California (Haig, 1960), and Punta
Penasco, Golfo de California, to Islas Toboga and Taboguilla, Panama (Haig, 1962).
Material. —Off Punta Redonda; 15 m; 5 Feb. 1964; 9d^, 4 9.
Color. — Ground color in life bright lavender, with uniform longitudinal stripes of
bright orange. Chelipeds bright lavender; legs with a white spot on propodus. Ventral side
iridescent, pinkish white; longitudinal stripes on carapace continued on first three segments
of abdomen (Glassell, 1936).
Remarks. — The specimens were found living in shells in association with Aniculus
elegans Stimpson. Porcellana paguriconviva was previously reported in association with
two other large hermits, Petrochinis californiensis Bouvier and Paguristes digueti Bouvier.
Pisidia magdalenensis (Glassell)
Porcellana magdalenensis Glassell, 1936: 295; 1938: 431, pi. 32.
Pisidia magdalenensis: Haig, 1960: 209, pi. 38 fig. 4, text-fig. 10.
Recorded Range. — Bahia de Santa Maria, outer Baja California (Glassell, 1936), to
Bahfa de Santa Elena, Ecuador. Apparently absent from Golfo de California.
Material. — Inside Punta Hughes; 6 m; 29 Jan. 1964; 3^", 2 ?. Outside Punta Hughes;
20 m; 30 Jan. 1964; IJ, 1 ? ovig. Off Isla Santa Margarita; otter trawl; 20 m; 4 Feb. 1964;
Id".
Remarks. — The specimens collected inside Punta Hughes were associated with
yellow gorgonian corals.
Megalobrachium garthi Haig
Megalobrachium garthi Haig, 1957:39, pi. 10; 1960: 213, 220, pi. 16 fig. 7, pi. 39 fig. 4.
Recorded Range. — Isla Turner, Golfo de California, to Puerto Utrfa, Colombia.
Material. — Inside Punta Hughes; 6 m; 29 Jan. 1964; 1 cT. Outside Bahfa Magdalena;
20 m; 1 Feb. 1964; 1 d- Inside Punta Tosca, in lagoon; 5 m; 4 Feb. 1964; 1 c5".
Remarks. — The specimen from Punta Hughes was taken from a yellow gorgonian,
and the one from Punta Tosca from a sponge. The range of this species is extended to the
outer coast of Baja California, and 300 km north.
Megalobrachium tuberculipes (Lockington)
Pachycheles tuberculipes Lockington, 1878: 396, 404.
Pisonella tuberculipes: Glassell, 1938: 437, 440, pi. 34 fig. 1.
Megalobrachium tuberculipes: Haig. 1960: 213, 227, pi. 16 fig. 11, pi. 40 fig. 4.
Recorded Range. — Punta Penasco and San Felipe, Golfo de California, to Bahfa de
Santa Elena, Ecuador.
Material. — Inside Punta Hughes; 6 m; 29 Jan. 1964: 1 cf , 1 ?. Inside Punta Tosca, in
lagoon; 5 m; 4 Feb. 1964; 2 cf. Bahfa de San Hipolito; 10-13 m; 9 Feb. 1964; 19.
Remarks. — Specimens were taken from a yellow gorgonian at Punta Hughes and
from sponge at Punta Tosca.
The range of this species is extended to the outer coast of Baja California, and 550 km
north.
1970 HAIG, HOPKINS AND SCANLAND: ANOMURAN CRABS 25
APPENDIX
The checklist and keys which follow include all species of anomuran crabs known to occur on the west
coast of Baja California from Punta San Eugenio (Punta Eugenia) southward, in depths of 40 m or less.
Nineteen of these species are included as a result of the expedition reported upon in the first part of this paper;
26 species on the basis of published records; and 7 species on the strength of records, as yet unpublished, in the
Allan Hancock Foundation of the University of Southern California.
CHECKLIST OF ANOMURAN CRABS FROM
SOUTHWESTERN BAJA CALIFORNIA, MEXICO
Unpublished records are marked with an asterisk (*)
Family HIPPIDAE
Emerita analoga (Stimpson)
Hippa analoga Stimpson, 1857: 85. Emerita analoga: Schmitt, 1921: 173, pi. 31 fig. 5, text-fig. IjO: 1935:
214, 216, text-figs. 75a, b. Range: Alaska to southwest Baja California; also Peru and Chile. Bahfa de San
Bartolome" (Schmitt 1921); Bahfa Magdalena (Schmitt 1935).
Family ALBUNEIDAE
Lepidopa niyops Stimpson
Lepidops myops Stimpson, 1860: 241. Lepidopa myops: Schmitt, 1921: 172, pi. 31 fig. 4. Range: Southern
California to Cabo de San Lucas, Golfo de California. *Bahia de Santa Marfa.
Family COENOBITIDAE
Coenobita compressus H. Milne Edwards. See p.l5 .
Family DIOGENIDAE
Dardanus sinistripes (Stimpson). See p. 16 .
Calcinus calijorniensis Bouvier. See p.l6 .
Petrochirus calijorniensis Bouvier.
Petrochirus californiensis Bouvier, 1895: 6. Glassell, 1937b: 251. Range: northern Golfo de California to
Ecuador. * Bahfa de Santa Marfa.
Aniculus elegans Stimpson. See p. 16 .
Trizopagurus magnificus (Bouvier). See p. 17 .
Clibanarius panamensis Stimpson. See p. 17 .
Isocheles sp. See p. 17 .
Paguristes bakeri Holmes. See p. 1 7 .
Paguristes ulreyi Schmitt. See p. 18 .
Paguristes digueti Bouvier ^
Paguristes digueti Bouvier, 1893: 18, text-fig. 1-4. Glassell, 1937b: 243. Range: Golfo de California. *Bahia
de Santa Marfa; * Bahfa Magdalena.
Paguristes anahuacus Glassell. See p. 18 .
Paguristes praedator Glassell. See p. 18 .
Paguristes. undescribed species. See p. 19 .
Family PAGURIDAE
Pagurus gladius (Benedict)
Eupagurus gladius Benedict, 1892: 7. Pagurus gladius: Glassell, 1937b: 256, 257. Range: Golfo de
California to Ecuador. * Bahfa de Santa Marfa; *Bahia Magdalena; * Punta Tosca.
Pagurus sniithi (Benedict). See p. 19 .
Pagurus. undescribed species (1). See p. 19.
Pagurus lepidus (Bouvier). See p. 19 ,
Pagurus galapagensis (Boone). See p. 20 .
Pagurus samuelis (Stimpson)
Eupagurus samuelis Stimpson, 1857: 86. Pagurus samuelis: Schmitt, 1921: 129, 139, pi. 16 figs. 2-3. text-
fig. 90. Range: Northern California to northwest Baja California. *Punta San Eugenio; *Punta San
Bartolome; *Bahia de Tortuga; * Punta Asuncion; * Punta Abreojos.
Pylopagurus californiensis (Benedict). See p. 20 .
Pylopagurus venustus (Bouvier). See p.21 .
Pylopagurus roseus (Benedict). See p.21 .
Family GALATHEIDAE
Munida mexicana Benedict
26 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
Munida mexicana Benedict, 1902: 251, 264, text-fig. 13. Range: Northern Golfo de California to
Archipielago de Galapagos. Bahia de Santa Maria (Benedict).
Munida reful^ens Fa.xon
Munida refulgens Faxon, 1893: 177; 1895: 75, pi. 17. Range: Southern Golfo de California to Panama.
*Bahia Magdalena.
Munida tenella Benedict
Munida tenella Benedict, 1902: 252, 274, text-fig. 20. Range: Golfo de California. *Bahia Magdalena.
Pleuroncodes planipes Stimpson. See p. 22.
Family PORCELLANIDAE
Orthochela pumila Glassell. See p. 22 .
Petrolisthes sanfelipensis Glassell. See p. 22.
Petrolisthes edwardsii (Saussure)
Porcellana edwardsii Saussure, 1853: 366, pi. 12 fig. 3. Petrolisthes edwardsii: Haig, 1960: 24, 33, pi. 21.
Range: Southwest Baja California and southern Golfo de California to Ecuador. Bahia de Santa Maria;
Bahfa Magdalena (Haig).
Petrolisthes hirlipes Lockington
Petrolisthes hirtipes Lockington. 1878: 395, 397. Glassell, 1936: 284. Haig, 1960: 26, 60, pi. 24 fig. 3.
Range: Golfo de California and southwest Baja California. Bahfa Magdalena (Glassell).
Petrolisthes manimaculis Glassell
Petrolisthes manimaculis Glassell. 1945: 223, text-fig. 1. Haig, 1960: 28, 77, pi. 27 fig. 1. Range: Northern
California to southwest Baja California. Punta San Eugenio (Haig).
Petrolisthes gracilis Stimpson
Petrolisthes gracilis Stimpson, 1858: 227 (nom. nud.); \%59: 74. Haig, 1960: 28, 79, pi. 27 fig. 2. Range:
Northern Golfo de California to southern Mexico. Bahia de Santa Maria (Haig).
Petrolisthes cabrilloi Glassell
Petrolisthes cabrilloa Glassell, 1945: 225, text-fig. 4. Petrolisthes cabrilloi: Haig, 1960: 28, 88, pi. 26 fig. 3.
Range: Central California to southwest Baja California. Punta San Eugenio; Punta San Bartolome; Punta
Asuncion; Punta Abreojos; Bahia de San Juanico; Bahia Magdalena (Haig).
Petrolisthes crenulatus Lockington
Petrolisthes crenulatus Lockington, 1878: 395, 398. Haig, 1960: 27, 110, pi. 23 fig. 4. Range: Golfo de
California and southwest Baja California. Bahia Magdalena (Haig).
Petrolisthes hians Nobili. See p. 23 .
Pachycheles marcortezensis Glassell. See p. 23.
Pachvcheles spinidactvlus Haig
Pachycheles spinidactylus Haig, 1957: 31, pi. 7; 1960: 134, 153, pi. 33 fig. 2. Range: Southwest Baja
California and southern Golfo de California to Colombia. Bahia de Santa Maria (Haig, 1960).
Pachycheles panamensis Faxon. See p. 23.
Pachycheles pubescens Holmes. See p. 23.
Pachvcheles rudis Stimpson
Pachycheles rudis Stimpson, 1858: 228 (nom. nud): 1859: 76, pi. 1 fi_g. 5. Haig, 1960: 133, 170, pi. 34 fig.
1. Range: Alaska to southwest Baja California. Punta San Bartolome; Bahi'a Magdalena (Haig).
Pachycheles holosericus Schmitt. See p. 23 .
Euceramus transversilineatus (Lockington)
Porcellana transversilineata Lockington, 1878: 396, 405. Euceramus transversilineatus: Glassell; 1938: 426,
pi. 30. Haig, 1960: 188, 190, pi. 36 fig. 2, text-fig. 7(2). Range: Northern Golfo de California to Panama.
BahTa de Santa Maria; BahTa Magdalena (Haig).
Porcellana cancrisocialis Glassell. See p. 23 .
Porcellana paguriconviva Glassell. See p.24 .
Pisidia magdalenensis (Glassell). See p.24 .
Megalobrachium garthi Haig. See p.24 .
Megalobrachium erosum (Glassell)
Pisosoma erosa Glassell, 1936: 289. Megalobrachium erosum: Haig, 1960: 213, 222, pi. 16 fig. 8, pi. 40 fig.
2. Range: Golfo de California and southwest Baja California. Punta Malarrimo; Bahfa de San Juanico
(Haig). Bahfa Magdalena (Glassell).
Megalobrachium tuberculipes (Lockington). See p. 24.
Polvonvx quadriungulatus Glassell
' Polyonyx quadriungulatus Glassell, 1935: 93, pi. 9. Haig, 1960: 233, 236, pi. 41 fig. 2, text-fig. 12(1).
Range: Southern California to southwest Baja California. Punta San Eugenio (Haig).
1970 HAIG. HOPKINS AND SCANLAND: ANOMURAN CRABS 27
KEYS TO ANOMURAN CRABS KNOWN FROM
SOUTHWESTERN BAJA CALIFORNIA, MEXICO
Identifications made with these keys should be considered tentative until sp)ecimens can be checked against
descriptions and illustrations. The two sand crabs falling within the scope of our report were dealt with by Schmitt
( 1 92 1 ). The hermit crabs of the eastern Pacific are currently being revised, but published information available at
this writing is widely scattered and in some cases inadequate; a number of references are given in the preceding
sections of this paper. For the Galatheidae, Pleuroncodes planipes was treated by Schmitt (1921), and
descriptions and illustrations of the remaining species are found in either Faxon (1895) or Benedict (1902). The
eastern Pacific Porcellanidae were monographed recently by Haig (1960).
Sand Crabs
la. Carapace suboval; first pair of legs simple: HIPPIDAE Emeritaanaloga
lb. Carapace subquadrangular: first pair of legs subchelate: ALBUNEIDAE Lepidopa myops
Hermit Crabs
la. Antennular peduncles several times length of eyestalks; antennular flagellum compressed and truncated
at tip: COENOBITIDAE Coenobita compressus
lb. Antennular peduncle less than twice length of eyestalks; antennular flagellum ending in a filament 2
2a. Outer maxillipeds approximated at their bases; chelipeds equal or subequal in size, or left cheliped larger
than right (in Petrochirus. right larger than left): DIOGENIDAE 3
2b. Outer maxillipeds widely separated at their bases; right cheliped always larger than left: PAGURIDAE 15
3a. Chelipeds markedly unequal in size and form 4
3b. Chelipeds equal or subequal in size and form 6
4a. Right cheliped larger than left Petrochirus californiensis
4b. Left cheliped larger than right 5
5a. Major chela smooth; fingertips calcareous Calcinus californiensis
5b. Major chela tuberculate; fingertips corneous and dark Dardanus sinistripes
6a. No paired abdominal appendages in either sex 7
6b. Paired pleopods present on first and second abdominal segments of male, and usually on first ab-
dominal segment of female 10
7a. Fingertips acuminate; antennal flagella heavily setose Isocheles sp.
7b. Fingertips spooned or hoof-shaped; antennal flagella nude or sparsely setose 8
8a. Chelipedsand walking legs with strong, grooved rings Aniculus elegans
8b. No such rings on legs 9
9a. Fingers open horizontally; no white spots on chelae; walking legs with longitudinal dark and light
stripes Clibanarius panamensis
9b. Fingers open obliquely; chelae and walking legs with large white spots Trizopagurus magnificus
10a. Fingertips acuminate; rostrum scarcely developed, broadly rounded;
no paired pleopods in females Paguristes. undescribed sp.
10b. Fingertips spooned; rostrum a well-developed, acute projection; females with a pair of pleopods 11
1 la. Rostrum long, acuminate, extending well'betweeneyescales 12
lib. Rostrum broad, well-produced but falling short of or barely reaching base of eyescales 13
12a. Chelaenarrow, covered with dense tomentum; eyestalks and antennae blue Paguristes anahuacus
12b. Chelae very broad, not tomentose; eyestalks and antennae not blue Paguristes digueti
1 3a. Eyescales with margins entire; in adults, spines on chelipeds
not densely pigmented Paguristes praedator
13b. Eyescales toothed; in adults, spines on chelipeds densely pigmented 14
14a. Antennal flagellum with very long hairs on lower surface; rostrum extending beyond lateral frontal
lobes Paguristes ulreyi
14b. Antennal flagellum with short hairs on lower surface; rostrum about equal in length to lateral frontal
lobes Paguristes bakeri
1 5a. Major chela narrow, not forming an operculum; no paired pleopods in female 16
15b. Major chela broad, forming an operculum; female with a pair of pleopods 21
1 6a. Eyescales with 2 or more spines 17
1 6b. Eyescales with a single spine 18
17a. Tip ofeyescale rounded and bearing 2 or 3 spines Pagurus, undescribed sp.
1 7b. Tip ofeyescale truncate and bearing 4 spines Pagurus lepidus
18a. Carapaceshield wider than long; eyestalks greatly expanded distally 19
1 8b. Carapace shield longer than w ide; eyestalks not greatly expanded distally 20
28 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
19a. Major chela granulate on outer margin, about 2'/2-3 times as long as wide, narrower than carpus
except at base of fingers Pagurus gladius
19b. Major chela with sharp teeth on outer margin, about twice as long as wide and about as wide
as carpus Pagurus smithi
20a. Major chela with a thick fringe of hairs on margins, its dorsal surface
tomentose and spiny Pagurus galapagensis
20b. Major chela not hairy, its dorsal surface granulose Pagurus samuelis
21a. Inner margin of major chela not expanded; no longitudinal ridge on
movable finger Pylopagurus roseus
21b. Inner margin ofmajor chela usually expanded laterally; movable finger with a longitudinal ridge 22
22a. Carpus ofmajor cheliped nearly smooth; walking legs with longitudinal stripes overlying diffuse broad
bands Pylopagurus californiensis
22b. Carpus of major cheliped with small, forward-directed spines covering its dorsal surface; walking legs
with broad bands whose margins are sinuous and sharply defined Pylopagurus venustus
Galatheid Crabs
la. Sides of carapace greatly swollen and visible in dorsal view; chelipeds and walking legs with a thick
fringeoflong, fine hairs on margins. Often pelagic Pleuroncodes planipes
lb. Sides of carapace not visible in dorsal view; no thick fringe of hairs on chelipeds and walking legs.
Bottom living 2
2a. Second, third, and fourth segments ofabdomen armed with spines Munida tenella
2b. Abdomen unarmed 3
3a. Fingers about as long as, or shorter than palm; rostrum with several lateral spines near
its apex Munida refulgens
3b. Fingers much longer than palm; rostrum without spines at apex Munida mexicana
Porcelain Crabs
la. Carapace nearly or quite half again as long as broad 2
1 b. Carapace scarcely or not at all longer than broad 3
2a. Carapace and chelipeds nearly smooth, without hairs; lateral margins of carapace with a series of about
12-15 minute, close-set spinules Orthochelapumila
2b. Carapace and chelipeds rugose, with long, scattered hairs; no marginal spinules on carapace posterior
toepibranchial spine Euceramus transversilineatus
3a. Movable segments of antennal peduncle with free access to orbit 4
3b. Movable segments ofantennal peduncle separated from orbit by a broad projection of basal segment. ... 17
4a. Side walls ofcarapace entire; chelipeds flattened, subequal 5
4b. Posterior portion of side walls of carapace separated by membranous interspace from anterior portion;
chelipedsthick,robust, one distinctly larger than the other 12
5a. Carapace with transverse striations; a row ofspines on anterior margin of merus of walking legs 6
5b. Carapace not transversely striate; anterior margin of merus of walking legs unarmed 7
6a. Carapace with groups of spines on dorsal surface, and a row of spines on lateral margins posterior to
epibranchial spine Petrolisthes sanfelipensis
6b. No spines on dorsal surface ofcarapace, nor on lateral margins posterior to
epibranchial spine Petrolisthes edwardsii
7a. Carpus ofchelipeds armed on anterior margin with strong teeth or tubercles 8
7b. Carpus not armed with strong teeth or tubercles 10
8a. Carpus ofchelipeds with wide-set conical tubercles on anterior margin; chela with a thick fringe of hair
on outer margin Petrolisthes hirtipes
8b. Carpus with strong teeth on anterior margin; chela without a thick fringe of hairs 9
9a. Telson5-plated; outer orbital angle produced into a distinct tooth Petrolisthes hians
9b. Telson 7-plated; outer orbital angle not strongly produced Petrolisthes crenulatus
10a. Carpus ofcheliped about twice as long as wide, a lobe occupying proximal '/4 of
its anterior margin Petrolisthes cabrilloi
10b. Carpus more than twice as long as wide, its margins subparallel 11
1 la. Carapace nearly smooth posteriorly, often granular anteriorly; merus of walking legs with a fringe of
hairs on anterior margin Petrolisthes manimaculis
lib. Carapace nearly smooth anteriorly as well as posteriorly; merus of walking legs nude or with only
traces of hair Petrolisthes gracilis
1 2a. Front with a distinct tuft of hairs 13
1 2b. Hairs sometimes present on frontal area, but not forming a distinct tuft 16
1970 HAIG, HOPKINS AND SCANLAND: ANOMURAN CRABS 29
13a. Manus with a large granulate protuberance at base of pollex; telson 5-plated in both sexes; males with
apairofpleopods;chelipedswitheither long, scattered hairs or short, close-set hairs 14
1 3b. No distinct protuberance on manus at base of pollex; telson 7-plated in males, 7- or 5-plated in females;
male pleopods present or absent; chelipeds with both long, scattered hairs and short, close-set hairs 15
14a. Carpus of chelipeds with a broad triangular lobe; chelipeds with long,
scattered hairs Pachycheles rudis
1 4b. Carpus with a broad, serrate -edged lobe; chelipeds with short, close-set hairs . . Pachycheles holosericus
1 5a. Carpus of chelipeds with a broad lobe cut into 3 or 4 uneven, serrate teeth; males with a
pair of pleopods Pachycheles pubescens
15b. Carpusarmed with 3 (rarely 4) spine-tipped teeth; no pleopods in males Pachycheles spinidactylus
1 6a. Carpus of chelipeds with 2 broad teeth; telson 7-plated; males with a
pair of pleopods Pachycheles panamensis
16b. Carpus with 3-5 narrow teeth; telson 5-plated; no pleopods in males Pachycheles marcortezensis
17a. Carapace broader than long; dactyl ofwalking legs with 4 fixed spines Polyonyx quadriungulatus
17b. Carapace not broader than long; dactyl ofwalking legs with a single terminal claw and several movable
spinules 18
18a. Front prominent, strongly tridentate or trilobate in dorsal view 19
' 18b. Front deflexed, appearing rounded or faintly trilobate in dorsal view 21
19a. Lateral margins ofcarapace unarmed posterior to epibranchial angle 20
19b. Lateral margins ofcarapace with minute spinules Pisidia magdalenensis
20a. Epibranchial angle with 2 or 3 spinules; frontal teeth pointed at tips Porcellana cancrisocialis
20b. Epibranchial angle unarmed; frontal teeth rounded at tips Porcellana paguriconviva
21a. Telson of abdomen 7-plated Megalobrachium erosum
21b. Telson 5-plated 22
22a. Carapace,chelipeds, and walking legs covered with small, shallow pits Megalobrachium garthi
22b. Carapace,chelipeds, and walking legs tuberculate Megalobrachium tuberculipes
REFERENCES
Benedict, J. E.
1892. Preliminary descriptions of thirty-seven new species of hermit crabs of the genus Eupagurus in the
U.S. National Museum. Proc. U.S. Natl. Mus. 15: 1-26.
1902. Descriptions of a new genus and forty-six new sf)ecies of crustaceans of the family Galatheidae, with a
list of the known marine species. Proc. U.S. Natl. Mus. 26: 243-334.
Boone, L.
1931. A collection of anomuran and macruran Crustacea from the Bay of Panama and the fresh waters of
the Canal Zone. Bull. Amer. Mus. Nat. Hist. 63: 137-189.
1932. The littoral crustacean fauna of the Galapagos Islands. Par II. Anomura. Zoologica 14: 1-62.
Bouvier, E. L.
1893. Paguriens recueillis par M. Diguet, sur le littoral de la Basse-Californie. Bull. Soc. Philom. Paris, Ser.
8,5: 18-25.
1895. Sur une collection de crustaces decapodes recueillis en Basse-Californie par M. Diguet. Bull. Mus.
Hist. Nat., Paris 1:6-8.
1898. Sur quelques crustaces anomoures et brachyures recueillis par M. Diguet en Basse-Californie. Bull.
Mus. Hist. Nat., Paris 4: 371-384.
Chace, F. A.. Jr.
1962. The non-brachyuran decapod crustaceans of Clipperton Island. Proc. U.S. Natl. Mus. 1 13: 605-635.
Faxon, W.
1893. Reports on the dredging of)erations off the west coast of Central America to the Galapagos, to the
west coast of Mexico, and in the Gulf of California ... by the U.S. Fish Commission Steamer
"Albatross," during 1891 . . . VI. Preliminary descriptions of new species of Crustacea. Bull. Mus.
Comp. Zool. 24: 149-220.
1895. Reports on an exploration off the west coasts of Mexico, Central and South America, and off the
Galapagos Islands ... by the U.S. Fish Commission Steamer "Albatross," during 1891 . . . XV.
The stalk-eyed Crustacea. Mem. Mus. Comp. Zool. 18: 1-292.
Forest, J.
1952. Contributions a la revision des crustaces Paguridae. I. Le Genre Trizopagurus. Mem. Mus. Hist.
Nat., Paris, Ser. A., 5(1): 1-40.
30 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
1964. Sur un nouveau genre de Diogenidae (Crustacea Paguridea) de I'Atlantique sud-ame'ricain, Loxopa-
gurus gen. nov., etabli pour Pagurus loxochelis Moreira. Zool. Meded. 39: 279-296.
Forest, J., and Michele de Saint Laurent
1968. Campagne de la Calypso au large des cotes atlantiques de rAmerique du Sud (1961-1962). 6.
Crustace's de'capodes: Pagurides. Ann. Inst. Oce'an.45: 47-170.
Glassell, S. A.
1935. New or little known crabs from the Pacific Coast of northern Mexico. Trans. San Diego Soc. Nat.
Hist. 8:91-106.
1936. New porcellanids and pinnotherids from tropical North American waters. Trans. San Diego Soc. Nat.
Hist. 8:277-304.
1937a. The Tempieton Crocker Expedition. IV. Porcellanid crabs from the Gulf of California. Zoologica 22:
79-88.
1937b. The Tempieton Crocker Expedition. XI. Hermit crabs from the Gulf of California and the west coast
of Lower California. Zoologica 22: 241-263.
1938. New and obscure decapod Crustacea from the west American coasts. Trans. San Diego Soc. Nat.
Hist. 8:411-454.
1945. Four new species of North American crabs of the genus Petrolisthes. J. Wash. Acad. Sci. 35: 223-229.
Haig,J.
1957. Four new porcellain crabs from the eastern Pacific. Bull. S. Calif. Acad. Sci. 56: 31-41.
1960. The Porcellanidae (Crustacea Anomura) of the Eastern Pacific. Allan Hancock Pac. Exped. 24: 1-
440.
1962. Papers from Dr. Th. Mortensen's Pacific expedition 1914-1916. LXXIX. Porcellanid crabs from
eastern and western America. Vidensk. Meddel. Dansk Naturhist. Foren. Kjobenhavn 124: 171-192.
Holmes, S. J.
1900. Synopsis of California stalk-eyed Crustacea. Occ. Pap. Calif. Acad. Sci. 7: 1-262.
Holthuis, L. B.
1954. On a collection of decapod Crustacea from the Republic of El Salvador (Central America). Zool.
Verhandel.,Leiden 23: 1-43.
Lockington, W. N.
1 878. Remarks upon the Porcellanidea of the West Coast of North America. Ann. Mag. Nat. Hist., Ser. 5,
2: 394-406.
Matthews, L. H.
1932. Lobster-krill, anomuran Crustacea that are the food of whales. Discovery Rept. 5: 467-484.
Milne Edwards^H.
1837. Histoire Naturelle des Crustaces, Comprenant 1" Anatomic, la Physiologic et la Classification de Ces
Animaux. Vol. 2. Paris.
Nininger, H. H.
1918. Crabs taken at Laguna Beach in the summer of 1916. J. Ent. Zool. 10: 36-42.
Nobili,G.
1901. Viaggio del Dr. Enrico Festa nella Repubblica dell'Ecuador e regioni vicine. XXIII. Decapodi e
stomatopodi. Boll. Mus. Zool. Anat. Comp. Torino 16 (415): 1-58.
Parker, R. H.
1964. Zoogeography and ecology of some macro-invertebrates, particularly mollusks, in the Gulf of
California and the continental slope off Mexico. Vidensk. Meddel. Dansk Naturhist. Foren.
Kjobenhavn 126: 1-178.
Radovich, J.
1961. Relationships of some marine organisms of the northeast Pacific to water temperatures particularly
during 1957 through 1959. Fish Bull. 112. Calif. Dept. Fish and Game.
Rathbun, M. J.
1910. The stalk-eyed Crustacea of Peru and the adjacent coast. Proc. U.S. Natl. Mus. 38: 531-620.
Ricketts, E. F., and J. Calvin
1939. Between Pacific Tides. Stanford University Press, Palo Alto, Calif.
Saussure. H. de
1853. Description de quelques crustace's nouveaux de la cote occidentale du Mexique. Rev. Mag. Zool., Ser.
2,5:354-368.
Scanland, T., and T. Hopkins
1970 HAIG. HOPKINS AND SCANLAND: ANOMURAN CRABS 31
1969. A new species of hermit crab, Pylopagurus diegensis (Decapoda: Anomura), with a key for the genus
in the Eastern Pacific. Pac. Sci. 23 (2): 257-260.
Schmitt, W. L.
1921. The marine decapod Crustacea of California. Univ. Calif. Pubi. Zooi. 23: 1-470.
1935. Crustacea Macrura and Anomura of Porto Rico and the Virgin Islands. Sci. Surv. Puerto Rico 15:
125-227.
1939. Decapod and other Crustacea collected on the Presidental Cruise of 1938 (with introduction and
station data). Smithsonian Misc. Coll. 98 (6): 1-29.
Steinbeck, J., and E. F. Ricketts
1941. Sea Cortez, a Leisurely Journal of Travel and Research, with a Scientific Appendix Comprising
Materials for a Source Book on the Marine Animals of the Panamic Faunal Province. Viking Press,
New York.
Stimpson, W.
1857. Notices of new species of Crustacea of western North America; being an abstract from a paper to be
published in the Journal of the Society. Proc. Boston Soc. Nat. Hist. 6: 84-89.
1858. Prodromus descriptionis animalium evertebratorum . . . Pars VII. Crustacea Anomura. Proc. Acad.
, Nat. Sci. Philadelphia 10: 225-252.
1859. Notes on North American Crustacea, No. 1. Ann. Lye. Nat. Hist., N.Y. 7: 49-93.
1860. Notes on North American Crustacea, in the Museum of the Smithsonian Institution. No. II. Ann. Lye.
Nat. Hist., N.Y. 7: 176-246.
Walton, B. C.
1954. The genus Pylopagurus (Crustacea: Anomura) in the Pacific with descriptions of two new species.
Allan Hancock Pac. Exped. 18: 139-173.
Allan Hancock Foundation, University of Southern California, Los Angeles, Califor-
nia 90007 (JH), and University of California San Diego, Scripps Institution of Oceanogra-
phy, LaJolla, California 92037 (TSH and TBS).
Present address of TSH: Faculty of Biology, University of West Florida, Pensacola,
Florida 32504.
COMPARATIVE BIOLOGY OF AMERICAN
BLACK WIDOW SPIDERS
B. J. KASTON
MUS. COMP. ZOOL.
LIBRARY
DEC 3 "1972
HARVARD
UNIVERSITY
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
Vol. 16, No. 3 24 July 1970
COMPARATIVE BIOLOGY OF AMERICAN
BLACK WIDOW SPIDERS
B. J. KASTON
ABSTRACT. — There are three American species of black widow spiders: Latrodcctiis mactans
and L. variohis in the eastern U.S., and only L. Iiespenis in the western. Although there is much
variation within each species, and to an extent some overlap, they differ in several ways. The egg
sac in L. hesperiis is tan and pyriform. in L. variohis gray and pyriform, and in L. mactans gray
and spherical. Latrodectiis mactans averages fewer egg sacs per female than L. hesperiis. but
more eggs per sac; L. variohis averages the smallest number per sac. The eggs and newly emerged
spiderlings are smallest in L. mactans. and largest in L. variohis: those of L. variolas hatch and
emerge in the shortest, of L. hesperiis in the longest time.
Newly emerged spiderlings of each species are characteristically marked, but as they ap-
proach maturity they become more alike, especially the females. Mature males are more easily
distinguished by their pattern. Red and white marks on the dorsum of mature females tend to be
most prominent in L. variohis (and the texaniis variety of L. hesperiis) less in L. mactans. and
least or even absent in L. hesperiis. The hour-glass mark is always divided in L. variolas, and
generally complete in the others. Latrodectiis variohis takes longest to mature; L. mactans is
quickest, and also is shortest lived.
In all three species the spiderlings usually rtoh only once, occasionally twice, and rarely
more times before emergence from the sac. The sexual behavior is similar, and in each a portion
of the embolus is left behind in the female genitalia after copulation. The palpal organ of L. hes-
periis is more like that of L. variolas generally; but because of the extreme variation, it is not
always possible to separate L. hesperiis and L. mactans by embolus coil morphology. The inter-
vals between successive ovipositions and the proportion of eggs that develop from successive egg
sacs show no trend toward increasing, or decreasing, and are highly variable in any series.
RESUMEN. — En America existen tres especies de aranas viudas negras: Latrodectus mactans
y L. variohis en la region oriental de los Estados Unidos, y L. Iiespenis en el Oeste. Aun cuando
cada especie presenta gran variabilidad en sus caracteres morfologicos y algunos de ellos son
comunes, en cierto grado, a varias especies, es posible diferenciarlas facilmente. En L. hesperiis
el saco ovigero es piriforme y de color marron, en L. variohis es tambien piriforme pero de color
gris, y en L. mactans es gris y esferico. Las hembras de Latrodectus mactans presentan en general
menos sacos ovigeros y mas huevos en cada saco que en L. Iiespenis; mientras que en L. variohis
el niimero de huevos por saco es menor. Los huevos y las aranitas racien nacidas son de talla
menor en L. mactans, y mayores en L. variolas. Las crias de L. variolas tardan menos tiempo
en nacer, mientras que en L. hesperiis la incubacion dura mas tiempo.
Las araiias racien nacidas correspondientes a estas tres especies presentan caracteres diferen-
tes y bien marcados; pero tales diferencias van reduciendose progresivamente a medida que
alcanzan la madurez, llegando entonces a ser tan similares que resulta dificil diferenciarlas, par-
ticularmente tratandose de las hembras. Los machos maduros son sin embargo mas faciles de
identificar, considerando el diseiio particular que presentan. Las marcas rojas y blancas que
aparecen en el dorso de las hembras maduras tienden a ser mas prominentes en L. variolas (y en
la variedad te.xaniis de L. Iiespenis) y menos marcadas en L. mactans. y en L. hesperiis esas mar-
cas se desvanecen hasta llegar a desaparecer. La marca en forma de reloj de arena aparece di-
vidida en L. variolas, mientras que en las otras especies esta completa. Latrodectus variolas es
la especie que tarda mas en alcanzar la madurez, y L. mactan madura pronto ye es de vida corta.
Las aranitas de estas tres especies pasan, antes de salir del saco, por una sola muda, a veces
dos y raramente sufren una tercera muda. El comportamiento sexual es similar en estas tres
SAN DIEGO SOC. NAT HIST, TRANS. 16(3): 33-82, 24 JULY 1970
34 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
especies, y en todas ellas el emholo queda en las genitales de la hembra despues de la copula.
El organo palpal de L. hespcnis es similar al de L. varioliis; pero debido a la extremada varia-
bilidad que presentan los individuos. resulta a veces dificil separar a L. hespenis de L. mactiis
basandose en la morfologia del rizo del embolo. Los intervalos entre Duestas sucesivas de huevos
y la proporcion de huevos que se desarrollan en las series de sacos ovigeros no presentan ninguna
tendencia particular, ya sea en sentido progresivo o regresivo, variando extremadamente su
niimero en cualquiera de estas especies.
Because of the potency of their venom against man, spiders of the genus Latrodectus
are of great importance medically. As a result of the interest generated during protracted
periods in which cases of envenomation were being regularly reported, and specimens were
being collected by many people, several life history studies were made during the 1930's
and 1940's. The taxonomic study by Levi (1958, 1959) stimulated a renewed interest, which
has continued to the present.
Pickard-Cambridge (1902) published a revision of the genus, in which he considerably
reduced the number of valid species. He maintained as distinct however, L. curacaviensis
(MuUer) of the New World, which he separated from L. mactans in part on the basis of the
hourglass mark being different in form. Nearly all later workers generally assumed that the
spider which is most commonly called the ''black widow" represented a single, but highly
variable, species, L. mactans. and the list of its synonyms is long (see Petrunkevitch, 191 1;
Roewer, 1942; Bonnet, 1957). Thus much of the available information on life history,
variation, and so forth, reported for this latter species was "contaminated," being ap-
plicable, in part, to the now recognized related species.
Levi (1958, 1959) not only maintained L. curacaviensis but considered that it was the
species which is widely distributed over the northern and western United States and
Canada, with L. mactans in the southern States. Levi also discussed the hourglass mark
and general pattern of spots, and especially noted instances where both species were
completely devoid of such markings. In 1964, with McCrone, he showed that the "brown
widow," L. geometricus C. L. Koch, and the "red widow," L. bishopi Kaston, appear to be
limited to southern Florida. Also, of the two species of "black widow," L. mactans is more
common in southern, and another species more common in the northern States and
Canadian Provinces. The species which he had previously identified as the West Indian L.
curacaviensis, was in reality L. variolus. The distribution of L. mactans and L. variolus
overlaps, and although in my previous work (1937a, 1937b, 1938, 1948, 1953, 1954) I had
used only the name L. mactans, it is known that both species occur in the southern New
England area, with L. variolus the more common of the two.
McCrone and Levi published some notes on the life history of the Florida populations
of L. mactans and L. variolus. which are sympatric there. They indicated that the same two
species also occur in California and other western States. Shortly afterwards I began my
observations on the post-embryonic development of our California black widow. At the
same time I was also observing the development of a family of L. mactans from Florida. In
noting the differences, particularly in the appearance of the spiderlings, I assumed that my
California specimens, which did not look like the Florida L. mactans, must be L. variolus.
But I found certain discrepancies between my results and those reported for L. variolus by
McCrone and Levi. In addition, there appeared to be some morphological differences from
those specimens of L. variolus which Dr. McCrone was kind enough to have sent me from
Florida, and from those specimens collected by me years ago in Connecticut. As additional
material became available it became apparent that neither L. mactans nor L. variolus
occurs in the west, and that yet another species, L. Hesperus, is represented (Kaston, 1968).
It is my primary aim to compare the biology of this western species with that of the true L.
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 35
mactans, but some comparisons are made with L. variolus as well.
Each of these species is variable so that if one sees enough specimens one encounters
the variations noted by Parrott (1946) with the ctenid, Uliodon piscator. and by MuUer
(1952) with the agelenid, Coelotes atropos. These variations can be morphological, both
exophenotypic and endophenotypic, as well as ethological. And at times they overlap with
the characters shown by one or both of the other two species, so that even with a specimen
in hand one may not be certain to which species it belongs.
While most of the information presented is based upon studies since 1965, some data
were collected in Georgia, in North Carolina, and in Connecticut over many years previous
to 1964.
MATERIALS AND METHODS
Specimens were collected from various areas, the majority from California and hence
mostly L. hesperus. But many live specimens of L. mactans were sent from eastern and
southern States by cooperative workers. Likewise, a relatively small number of £. variolus
were obtained. For the most part mature females were received, but a number of males,
and some half grown individuals, were likewise obtained.
From the egg sacs produced by these female spiders I set out for L. hesperus 29
families (ca. 2600 spiderlings); 18 from California, four from Texas, four from British
Columbia, and one each from Arizona, Oregon, and Baja California; for L. mactans 37
families (ca. 2800 spiderlings): nine from New York, four each from Virginia and
Arkansas, three each from Florida and Illinois, two each from Missouri, New Jersey,
North Carolina and Ohio, and one each from Alabama, Louisiana, Mississippi, Oklahoma
and Tennessee; and for L. variolus 12 families (ca. 900 spiderlings) seven from Missouri,
three from Michigan, and one each from Arkansas and Florida. I received the L. variolus
material late in my study, and unfortunately, because of factors as yet not understood, I
had little success in rearing these latter spiderlings to maturity. Live specimens were also
received from the following areas: Alberta, Georgia, Kansas, Washington and West
Virginia.
Many of the mature specimens were used in studies on the chemical nature of the
hemolymph (as in McCrone, 1968); many of the males were sacrificed before maturity for
studies on the chromosomes. The results of both these studies are being published
elsewhere.
In the laboratory each spider was placed in its own glass container. When egg sacs
were m.ade they were removed and each placed in a vial and given a code number. Thus the
first sac made by female #1052 was #1052-A, the second was #1052-B, the third was
#1052-C, etc. Those used for the study of development and/or for the rearing of
spiderlings were kept in a constant temperature room at 25° C.
When the spiderlings emerged, each was placed in its own labelled vial, and given its
own number, e.g., #1052-A-1, #1052-A-2, etc. When the spiderling became larger it was
transferred to a larger container. The containers were stoppered with polyurethane foam.
All specimens were checked and data taken daily. After the spiders matured they were
removed from the constant temperature room and kept in one of the laboratories at room
temperature.
In the early years of this study the spiderlings were fed entirely on fruit flies,
Drosophila melanogaster. Although some males matured (in the fourth and fifth instars),
the females did not. Apparently nutritional deficiency was involved, possibly an insufficient
supply of one of the essential amino acids. Comparable findings were later reported by
Miyashita (1968) in his attempts to raise specimens of Lycosa. However, when older
36
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 1. a. Female Latrodectus in the normal position hanging in her web; b, Latrodectus mating position; the
female is represented in outline and the male is blackened; c, the venom glands as seen from above in relation to
the entire cephalothorax; d, the left venom gland with its duct, and left chelicera.
spiderlings, as well as the adults, were fed on mealworms, Tenebrio molitor, and other
insects that could be obtained outdoors, chiefly blue bottle and other muscoid flies, many
reached maturity.
DESCRIPTION OF ADULTS
The genus Latrodectus Walckenaer, 1805, is cosmopolitan. It comprises medium-
sized spiders, which are the largest members of the Family Theridiidae, those spiders
bearing a comb of serrated bristles along the ventral surface of tarsus IV. This comb is well
developed, and quite conspicuous in Latrodectus. The lateral eyes of each side are widely
separated, generally a diameter or more apart. The colulus is large and distinct, and the
legs are moderately long, with the first usually longer than the fourth, and the third
shortest. In females the abdomen is usually relatively large, high and subglobose. The
venom glands are large and extend far back into the cephalothorax (Figs. Ic and Id).
Males are much smaller than females, have relatively longer legs, and have the abdomen
lower and narrower, so that it appears somewhat ellipsoidal. Males are also commonly
more brightly colored. The webs made by these spiders are irregular meshes in which the
spider stands in an inverted position (Fig. la). Three species occur in the United States, all
quite variable in markings and color pattern. The descriptions and illustrations here
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 37
supplied represent the colors and patterns commonly met with.
Latrodectus mactam (Fabricius)
Figs. 2, 3, 4a, 4c, 5c, 12, 13a-d, g, i, k
Aranea mactans Fabricius, 1775, Systema Entomologiae, etc. p. 492 (no. 4); Fabricius, 1793, Entomologiae
Systematica, etc., p. 410 (No. 1 1) (exact copy of 1775).
Latrodectus mactans mactans: Chamberlin and Ivie, 1935, Bull. Univ. Utah, 25(8): Biol. Ser., 3(1): 13.
Latrodectus mactans: Kaston, 1948. Connecticut Geol. and Nat. Hist. Surv. Bull. 70: 92 (in part); Levi, 1959,
Trans. American Microscop. Soc, 78:24 (in part).
Levi indicated (1959: 16) that Chamberlin and Ivie gave Massachusetts as the type
locality for L. mactans, and added "where L. mactans has now been found not to occur."
Levi implied that Massachusetts could not have been the type locality as he had not seen
specimens from there. Yet this species may well be found in Massachusetts, most probably
along the southern shores and nearby islands in the Cape Cod region. After all, L. mactans
is known from Connecticut, and I have seen numerous specimens from the part of New
York (Westchester County) bordering on Long Island Sound, a region hardly 25 miles
farther south than the southernmost part of Massachusetts. One can no more assume that
Massachusetts is not the type locality for L. mactans merely because Levi has not found
any there, any more than we can assume that the island of Cura9ao is not the type locality
for L. curacaviensis because he was unable to find even a single individual when he visited
that island especially to search for specimens (McCrone and Levi, 1964)! It is true,
however, that one would hardly have expected L. mactans to be common enough in
Massachusetts for this to be where Fabricius' specimen came from. Many years ago I
wrote Professor Chamberlin about this. In his answering letter he stated, ''Massachusetts
is given as type locality on the basis of the statement by Fabricius himself." Yet in both of
the references cited above I found Fabricius to state only "In America Dom. Lewin," so
that I must agree with Levi that the type locality was "incorrectly stated" in the paper by
Chamberlin and Ivie (1935). Levi considers that the type may have come from the
southeastern United States, or the West Indies.
This is the common species of our southeastern States, but has been found as far north
as southern New York, and southern New England, west through southern Ohio, Indiana,
Illinois, Missouri, to about central Kansas, and south through central Oklahoma and
Texas.
In the female the cephalothorax and legs are shiny black, usually unmarked. The
abdomen is black with the following markings in red. On the venter is the characteristic
hourglass mark, usually consisting of an anterior triangle, and a generally wider posterior
rectangle with rounded corners.
There is much variation, and even in the same individual the markings, including the
hourglass, may at times be more distinct, and at other times be less distinct. Some of the
shapes taken by the hourglass are shown in figure 3. McCrone and Levi (1964) suggested
that the more brightly colored specimens had undergone fewer molts, as I had formerly
supposed. But as indicated above, the evidence seems not to bear this out. Gerschman and
Schiapelli (1943) illustrated a variety of patterns for Argentine specimens, and I find it
significant that they found no correlation between body size and type of pattern. Possibly
they were dealing with more than one species; at any rate Abalos and Baez (1967) and
Pinter (1968) published figures showing the variations present in what they considered to
be four species additional to L. mactans. Along the mid-dorsal line is a row of spots, the
most anterior of which may appear as a short chevron.
Generally, the male is similar to a fifth instar female. The cephalothorax may be all
38
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 2. Latrodeclus mactans. a, dorsal aspect of female from Florida; b, ventral aspect of same female; c,
epigynum of same female from below; d, dorsal aspect of male from Florida; e, ventral aspect of same specimen;
f, lateral aspect of male from Louisiana; g, dorsal aspect of cleared epigynum of female from New York.
black, or the carapace may show a somewhat lighter band on either side of a median black
band. The legs may be all black (Fig. 5c), or have lighter annuli (Fig. 4c) retained from
earlier instars. There is a median row of red spots on the dorsum of the abdomen. On the
venter the hourglass mark is distinct. Encircling the anterior end of the abdomen is a white
band, and on either side farther back are two additional white bands which extend
diagonally down and to the rear. Seen from the side this gives the appearance of three
white bands (Fig. 20- In L. Hesperus there are also three white bands as seen from the side,
but the anterior one is always hooked and procurved near its ventral end, while this is
uncommon in L. mactans. Also, in L. Hesperus the background color is much lighter. In L.
variolus the ground color is as dark as in L. mactans. and there are four bands visible along
the side.
Latrodectus variolus Walckenaer
Figs. 4b, 5a, 6a-f, 7, 14a, b, f
Latrodectus variolus Walckenaer, 1837, Hist. Nat. Insectes Apteres, 1:648.
Latrodectus mactans: Emerton, 1902, Common Spiders of the United States, (in part) fig. 291(j'; Kaston, 1948
Connecticut Geol. Nat. Hist. Surv. Bull. 70;92 (for the most part).
Latrodectus curacaviensis: Levi, 1959, Trans American Microscop. Soc, 78:38 (in part).
Latrodectus variolus: McCrone and Levi, 1964, Psyche 71:13
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
39
Figure 3. Ventral aspect of abdomens of L. maclans females to show variation in the shape of the hourglass
mark, and in the epigynal orifice, a, from Arkansas; b, from New York; c, From New York; d, from Tennessee; e,
from North Carolina; f, from Virginia; g, from New York; h, from New York.
While most workers seem to consider this species distinct from L. mactans.
Gerschman and SchiapeUi (1965) continue to place L. curacaviensis (and presumably also
L. variolas) as a synonym of L. mactans.
This species occurs in the United States in about the same areas as L. mactans, but
apparently is much less common. However, its range extends into more northern states and
adjacent Canadian provinces where L. mactans presumably does not occur.
Illustrations of this species have been published recently by Judd (1965) and Wilson
(1967). The female has the cephalothorax black, and the legs are similar, usually without
faintly brown annuli. There is a row of middorsal red spots on the abdomen. In addition,
there are three pairs of diagonal white stripes on each side, and a narrow white stripe
encircling the front of the dorsum. In some specimens these stripes are yellowish to
pinkish. The hourglass mark is divided, the two halves separated. There is much variation
in the shape and the size of the two halves (Fig. 7), occasional specimens showing only one
of the halves, and occasional specimens lacking the mark entirely (Kaston, 1954). The
male has the cephalothorax and legs black as in the female, but the legs are more likely to
show the brown annuli. The dorsum shows a pattern similar to that of the female, but with
the white stripes generally broader. From the side four such stripes are visible (Fig. 6f)
making this sex relatively easy to distinguish from the males of L. Hesperus and L.
mactans. which have only three light stripes. The hourglass mark is divided.
Latrodectus Hesperus Chamberlin and Ivie
Figs. 5b, 8, 9, 10, 11, 13e, f, h
Latrodectus mactans Hesperus Chamberlin and Ivie, 1935, Bull. Univ. Utah 25(8): Biol. Ser. 3 (1):15 [types from
Salt Lake City. Utah].
Latrodectus mactans texanus Chamberlin and Ivie, 1935, Bull. Univ. Utah 25(8): Biol. Ser. 3(1): 14 [types from
Texas ).
Latrodectus mactans: Gerschman and SchiapeUi, 1943, in, Sampayo "Latrodectus mactans y Latrodectismo" (in
part) fig, 7.
Latrodectus curacaviensis: Levi, 1959, Trans. American Microscop. Soc. 78:38 (in part).
Latrodectus variolus: Levi, 1969, Psyche 76:72.
40
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
y. -^
Figure 4. a, L. maclans female from Missouri showing ventral surface; b, L. variolas female from Connecticut,
showing dorsal spots; c, L. mactans male showing annulate legs.
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 41
This is the only species of black widow found west of about the middle of Texas,
Oklahoma, and Kansas to the Canadian provinces. Levi has recorded both L. mactans and
L. vaholus (sub curacaviensis) from the west but, based especially upon the appearance of
males and spiderlings, I am unable to ascribe to either of these two species any black
widows I have seen from the areas indicated.
The figure 7 supplied by Gerschman and Schiapelli is an exact copy of an illustration
from D' Amour et al. (1936) which is of a male from Colorado and quite definitely L.
Hesperus. O'Rourke (1956) indicated that what he had seen from western Canadian
provinces belonged to L. Hesperus. But Levi (1969) synonymized this species with L.
variolus, presumably on the basis of the fact that the male palpal organ shows two loops of
the embolus.
The female has the cephalothorax and legs black. In most specimens the dorsum of
the abdomen is likewise entirely black. In only a few is there left a remnant or the
middorsal stripe as a small red spot just above the anal tubercle. Also, there may be, on the
anterior portion of the abdomen, which overhangs the carapace, a remnant of the light
transverse band, as a kind of "chevron" pointing downward. Occasionally this "chevron"
is doubled, composed of two closely set thin lines (as in Fig. 10).
Ordinarily the hourglass mark is complete, with a narrow connecting piece between
the two triangular halves, and usually the base of the anterior triangle is wider than the
base of the posterior triangle. Sometimes there is a spot of black in the center of the
connecting piece, or the mark is divided into two parts. There is much variation, some of
the varieties being shown in figure 9. Rarely is the middle part broad, and very rarely is the
posterior half wider than the anterior.
In the variety texanus the dorsum retains more of the white and red, with a central
band, and lateral bands much like juveniles in the sixth instar (Fig. 8c). Often the white
areas become suffused with black pigment so that older females will show only the red
spots surrounded by black. I have this variety not only from Texas, but also from several
localities in California. Interestingly enough, both this and the more typical variety
sometimes were taken in the same place at the same time.
The male has on the carapace a dark gray to black band along the lateral margins, and
a dark band along the middle. The sternum is dark along the lateral borders, but lighter in
the middle. The legs show dark annuli. The hourglass mark is not much constricted in the
middle and is usually yellow rather than red, some specimens showing a slight suffusing of
orange pigment.
The abdominal dorsum shows a characteristic pattern of olive greenish gray
alternating with light tan bands (Fig. 8e, 0- One of the light areas appears as a middorsal
band usually with orange pigment as a thin line down its center. The other light areas
appear as three bands on each side, which extend along the sides diagonally down and
toward the rear. The most anterior of these is hooked and procurved nears its ventral end
(Fig. 8g). The male of L. mactans also shows three bands along the side but only rarely is
the first one hooked and procurved. In L. variolus there are four bands. Moreover, in both
L. mactans and L. variolus the areas between these light bands are much darker than in L.
Hesperus. In the variety texanus the gray areas, instead of being suffused with olive-green
pigment are suffused with pink.
COLOR VARIATIONS
One common variation concerns the background black characteristic of most adults,
especially the females. That the abdomen is in some individuals brownish or sepia in color,
rather than black, was reported by Burt (1935) for Kansas specimens that were probably L.
42
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
.(
%
W"^
■%,
*j
Figure 5. a, L. variolus male; b, L. Hesperus male; c, L. mactans male, with legs devoid of annuli.
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
43
Figure 6. Latrodectus variolus. a, female from Florida, dorsal aspect; b, ventral aspect of same female; c. female
from Michigan, dorsal aspect of abdomen; d, female from Connecticut, dorsal aspect; e, ventral aspect of same
female; f, lateral aspect of male; g, dorsal aspect of penultimate male.
mactans: by Herms et al. (1935) for California specimens that were definitely L. hesperus;
by Minton (1950) for Indiana specimens that were probably L. variolus: and by Wilson
(1967) for Michigan specimens of the same species. As I pointed out (1968) for all three
species, the same specimen can at times appear black, and at other times sepia or even
lighter brown. Some specimens revert to black over a period of several months. My notes
indicate that 111 specimens of L. hesperus turned from black to brown, and 27 turned
black again; twenty-nine specimens of L. mactans turned brown, and 9 turned black again;
and 8 specimens of L. variolus turned brown. I have collected both black individuals and
brown ones of L. hesperus at Brawley, California, within a few feet of one another, where
the ecological situations were seemingly similar. In one backyard lot the ratio of brown
specimens to black collected was 9: 1 .
A variation about which much has been written is the shape of the hourglass mark,
and for each species a few examples from mature females are illustrated (Figs. 3, 7 and 9).
Sometimes the mark disappears from a specimen which formerly showed it well; but it
may also return slowly at a later date.
Pattern variations formed by the red and white marks on the dorsum, have also been
noted. It is generally considered that more of a dorsal pattern, and that larger areas of red
and white, reflect the retention in the adult of a more juvenile pattern. Reese (1940)
44
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 7. Ventral aspect of abdomens of L. variolus females to show variations in the shape of epignyal orifice,
and of hourglass mark, a, from Arkansas; b, from Missouri; c, from Michigan; d, from Missouri; e, from
Missouri; f, from Michigan; g, from Michigan; h, from Illinois.
illustrated some variations in shape of hourglass mark and the arrangement of dorsal
spots. However, he may have included some juveniles, and could have confused L. variolus
and L. mactans, both of which occur in West Virginia. Levi (1959) suggested that Reese's
smaller ones were L. variolus. and his larger ones L. mactans. In my experience, however,
L. variolus averages larger in size than L. mactans. Thorn (1967) gave a brief discussion of
the variation in what is undoubtedly L. Hesperus.
With so much variation occurring in the adult females, the three species at times are
difficult to distinguish. But as Keegan (1955) has indicated "juvenile specimens possess
distinctive markings even when individual adults are alike.'" It was on the basis of these
differences (between Kansas and Michigan juveniles) that Lawson (1933) first suggested
that we were dealing with another species besides L. mactans. This other species has since
come to be known as L. variolus. From what is now known of the appearance of the
spiderlings it seems obvious that the descriptions of Herms et al. (1935), D' Amour et al
(1936), Moles (1916), Bristowe (1945, 1946) and Gonzales (1954) apply to L. hesperus;
those of Lawson (1933), Blair (1934), Muma (1944) and Deevey (1949) apply to L.
mactans: those of Kaston (1937b) to L. variolus.
MEASUREMENTS
The vernacular name "black widow" alludes to the commonly held supposition that
the female always eats the male after mating. This idea is strengthened by the fact that
there is a great difference in size between the two sexes. One often finds in the literature
remarks about the female being twice, or even three times as large as the male. Of course,
such statements are misleading, for they refer to the length of the body, as in the case of a
male 4 mm long and a female 12 mm in length. But the female has thicker legs, and a much
higher, globose abdomen, so that her mass may be many times more than three times that
of the male. Most males of L. hesperus '^Q\g\\ between 8 and 18 mg; most females between 120
and 400 mg. One small male, # 1052-C-203 with a body length of 3.5 mm weighed 5.9 mg,
and a large gravid female, # 1 528, whose body length was 12.5 mm, weighed 944.9 mg, or
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
45
Figure 8. Latrodectus Hesperus, a, female from Arizona, dorsal aspect; b, lateral aspect of same female; c, dorsal
aspect of the variety texanus, from Texas; d, ventral aspect of a light male from El Centro, California; e, dorsal
aspect of the same male; f, lateral aspect of same male; g, dorsal aspect of a darkly pigmented male from
Pasadena, California.
160 times as much as the male!
Of the three species, L. mactans averages smallest for both sexes. Thirty-seven males
ranged from 2.9 to 5. 1 mm in length, with most between 3.2 and 4 mm; 52 females ranged
from 5 to 13.5 mm, with most between 8 and 10 mm. Latrodectus variolas has the largest
males, mostly between 5.5 to 6.5 mm, with a range for 34 specimens of 4.5 to 8.3 mm.
Females of L. variolus are mostly 9 to 1 1 mm in length with a range for 32 specimens of 7.4
to 13 mm. Latrodectus hesperus has the largest females, 59 specimens ranging from 8 to
15.5 mm, with most from 10.5 to 13 mm. Sixty-three males ranged from 3 to 6.5 mm, with
most between 3.8 and 4.5 mm.
While it might appear that within a species the larger individuals would be those that
have gone through more molts, and vice versa, this has not been found to be the case. Some
of the smallest females of both L. mactans and L. hesperus matured in the sixth, the
seventh, and the eighth instars, and one cannot ascertain their instar from the size. These
size discrepancies can be seen among siblings in the same family, and as early as the time of
emergence from the egg sac. By the time they have reached the fifth instar some may be
almost twice the length of their sisters. The same applies to males, which when mature
show a wide range in all three species. Again, this is irrespective of the instar in which they
matured, or of the locality in which found. Two mature males of L. hesperus were collected
quite close together at the same time at Yuma, Arizona. One, #1510, measured 3.4 mm
and the other, #1511, was 6.5 mm long. Similar great discrepancies in size were found
46 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
among males of Nephila inaurata (Walckenaer) reared from the same egg sac by Derouet
and Dresco(1956).
Although there is much variation, for females the first leg is proportionately longest in
L. variolus. shortest in L. mactans, and intermediate in L. Hesperus. In males there is very
little ditTerence between L. Hesperus and L. variolus. both of which have the first leg
averaging slightly longer than in L. mactans.
Besides dilTerences in the length of the legs, there are distinct differences in thickness.
Again, it has not been possible to correlate this with species, though in general it seems that
the first leg is slimmer in L. mactans females than in the two other species. However, in two
females of L. Hesperus collected at Ramona, California, not more than 50 feet apart, one
had a tibial index for leg I of 9.3, and the other, of 11.5. Exactly the same sort of finding
has been noted for two females of L. mactans collected together in New York. This kind of
variation in specimens from the same locality was also observed by Smithers (1944) in L.
indistinctus.
With some spiders there is a direct correlation between width of the carapace and
stage of growth (i.e., instar). As indicated above there is a tremendous variation in size in
Latrodectus even at the time of emergence from the egg sac. Newly emerged spiderlings of
L. mactans usually range in length from 1.2 to 1.3 mm, with some up to 1.5 mm; those of
L. Hesperus from 1.5 to 1.8 mm; and those of L. variolus from 1.7 to 2.0 mm, but
sometimes down to 1.4, with the extremes even from the same egg sac. Furthermore, as
reported by Shulov (1940) for L. tredecimguttatus, many spiderlings molted from the first
to the second instar without having been fed, often within a day or two after emergence; no
growth can be measured. After measuring many spiderlings, and in several families, I had
to conclude, as had Miyashita (1968) for Lvcosa, that the width of the carapace could not
be used for ascertaining the instar.
EPIGYNUM
The epigynum appears externally as a highly arched, heavily sclerotized structure
which bulges ventrally, and has a transversely elliptical opening (Fig. 2c). There is much
variation in the exact shape and relative length of the opening. In some specimens the
anterior lip is developed into a carina, with or without a small median pointed process
pointing toward the rear (Figs. 3, 7 and 9). Although not easily seen in the intact specimen,
there is, on the dorsal wall of the atrium an opening on each side, leading into connecting
ducts, the so-called bursae copulatrices. Examination of the cleared epigynum from the
dorsal side shows that each bursa copulatrix is rather lightly sclerotized and twines around
the heavily sclerotized, darkly pigmented spermatheca of its side. The spermathecae are
dumb-bell shaped and lie with their axes making an angle of about 45 degrees to each other
and their posterior rounded portions separated by a distance about half the diameter of one
of them.
The shape of the atrial opening, and the details of structure vary a little, but the general
appearance is the same in all three species. Levi (1966) indicated that there was much
variation in his material from Israel, and from the studies of Lucas and Bucherl (1965) one
would expect some variation even in sisters.
In L. mactans the connecting ducts have four outside coils (Fig. 2g), while in L.
variolus (Fig. 14f) and L. Hesperus (Fig. 1 1 0 there are only three. This makes the epigynum
of L. mactans appear wider than that of the other two species. Commonly the sperma-
thecae and/or the connecting ducts will contain one or more embolic fragments left behind
by the male after mating (Figs. 2g, 1 Ig, 1 Ih).
According to Bhatnagar and Rempel (1962), who studied the structure of the
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
47
Figure 9. Ventral aspect of abdomens of L. Hesperus females to show variations in the shape of the epigynal
orifice, and of the hourglass mark, a, from Texas; b, from British Columbia; c, from Washington; d, from
California; e, from California; f, from California; g, from California; h, from British Columbia.
epigynum of L. hesperus, the openings of the ducts into the spermathecae lie "on the
middle portion of the spermathecae on the latter's outer lateral margin."
In specimens one molt short of maturity the area where the epigynum will later show
appears much arched. Thus it is possible to recognize one in the penultimate instar. In the
antepenultimate instar the area is usually somewhat lighter than those surrounding areas,
but definite recognition of females in this stage is not easy.
PALPAL ORGAN
Although Levi (1959) greatly reduced the number of species of Latrodectus, partly on
the basis of similarity of palpal structure, he later (1966) admitted that the morphology of
the palpal organ may be the same in two or more species. He accepted for North America
only three species: L. mactans, in which the embolus shows three coils, L. geometricus. in
which it shows four, and L. curacaviensis, in which it shows two. In 1964, with McCrone,
he acknowledged that the true L. curacaviensis did not occur in North America, and that
there were two additional species in which the embolus had two coils. These are L. bishopi.
of south Florida, and L. variolas, which is widely distributed over the United States.
Abalos and Baez (1967) and Pinter (1968) appear to have demonstrated that there are at
least three additional species in the ""mactans group" and one additional in the ""curaca-
viensis group." I consider L. hesperus a good species in the latter group, though some
specimens show traits that would lead to its placement in the former group. This could
account for Levi's distribution lists showing both L. mactans and L. variolus in the western
States, where in my opinion neither one occurs.
In all three species there is a long spirally coiled embolus (Figs. 1 la, b, c, 12a, b, 14a,
b). The origin of the embolus from the radix is broad, and shows a thick curved tooth.
There follows the heavily pigmented trunk of the embolus on the outside of the coil,
paralleled by the membranous pars pendula on the inside of the coil. Near the distal end of
the embolus is a blunt tooth marking the proximal articulation of the apical sclerite. After
copulation the embolus breaks at this point, and the apical sclerite can be found lying
48
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 10. Lalrodeclus hesperus. a, female from California showing anterior chevron marks on abdomen, from
the side; b, same female, showing chevron marks from above and front.
within the parts of the female genitaUa (Figs. 2g, 1 Ig, h). This phenomenon, which in
recent years has been shown to occur in other spiders too, was according to Gerhardt
(1928) first described for Latrodectus in 1902 by Dahl. It was later noted by Smithers
(1944), Abalos and Baez (1963, 1967), Gerschman and Schiapelli (1965), Wiehle (1967)
and Bhatnager and Rempel (1962). The latter's work included detailed studies on the
structure of the palpal organ and female genitalia. Their specimens had come from
Kamloops, British Columbia, and from my own observations on abundant material from
that very same locality it is quite certain that they were working not with L. ciiracaviensis
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
49
Figure 1 1 . Latrodectus hesperus. a, apical aspect of the palpal organ of the type male from Utah; b, the same, for
a male from Arizona; c, same palp, ectal aspect; d, distal end of pedipalp of a male in the preantepenultimate
instar; e, the same from a male in the antepenultimate; f, dorsal aspect of a cleared epigynum of a virgin female
from Oregon; g, the same, from a California female that had mated, showing an embolic sclerite left behind in
each of the spermathecae; h, the same, from another California female showing three embolic sclerites. indicating
that it had mated with at least two males.
as stated, nor with L. variolas as implied in McCrone and Levi (1964), but with L.
hesperus.
In the resting position, the distal end of the embolus usually lies against the conductor,
but extends somewhat beyond it. Adjacent to the conductor is the heavier terminal
apophysis, and just proximad of this is the sickle-shaped median apophysis. The distal
border of the latter is provided with a socket into which fits a heavily sclerotized tooth
borne near the distal end of the cymbium (Fig. 1 3a). Smithers described this tooth as two-
lobed in Z.. indistinctus and L. geometricus, and indicated that these two species differed
with respect to the size and shape of the two lobes. Levi (1959) referred to this cymbial
tooth as a "paracymbial hook" and illustrated it for L. mactans, as well as for other
species. It would seem that his usage of the term is ill-chosen, for the term paracymbium is
already in use for a structure arising from the basal portion of the cymbium. In my studies
on our three species I have found this tooth to be single-lobed, as illustrated by Bhatnagar
and Rempel (1962). That Levi illustrates the tooth as two-lobed indicates a mis-
50
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 12. Latrodectus maclans. a, apical aspect of palpal organ of a male from Florida: b. ectal aspect of same
palp: c, distal end of pedipalp of a mature female: d, distal end of pedipalp of an antepenultimate male: e, distal
end of a pedipalp of a penultimate male: f, dorsal aspect of a penultimate male from Florida; g, ventral aspect of
same male.
inlerprelation. The tooth is strongly sclerotized and pigmented, and from the point of its
attachment at the edge of the cymbium the latter shows the same degree of sclerotization
and pigmentation. This may give the impression of a two-lobed process (Fig. 13a). But if
one views the structure from the apicomesal (Fig. 13b), the mesal (Fig. 13c) or the ectal
aspect (Fig. 13d) its single nature can be seen.
Levi found, that to some extent at least, the character of the embolus could be used to
separate species. In L. variolas it is wider than, and about three-fourths as long as, in L.
mactans. and it makes two coils in L. variolus. but three in L. mactans. Moreover, I noted
that the coils are tighter and less open in L. variolus than in L. mactans.
In the type male of L. Hesperus the embolus (Fig. 1 la) shows two coils (as in the
"curacaviensis group") though it appears longer than in L. variolus, as the coils are more
open. However, the coils are tighter and less open than in L. mactans. But there is much
variation among the many specimens of L. Hesperus that have been studied. In fact I have
seen many specimens that could not be identified on the basis of coil morphology. For
example, figure 13h represents the palpal organ of a L. Hesperus specimen from Los
Angeles, California, and figure 13k the same of a L. mactans specimen from Rutledge,
Georgia. Note the similarity between figure 13g of a Gainesville, Florida specimen of/,.
mactans. and figure 13e of an El Centre, California specimen of L. Hesperus. Adding to the
confusion is the fact that the basal portion of the embolus sometimes extends from its
origin towards the distal end of the palp, sometimes towards the basal, or the mesal, or the
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
51
Figure 13. a, cymbium of pedipalp of a New York specimen of L. maclans; b. cymbial hook as seen from the
apicomesal aspect; c, the same, mesal aspect; d, the same, ectal aspect; e, embolus of L. hesperus male it 1002-A-
51 from El Centro, California; f, the same, of a litter-mate, * 1002-A-48; g, the same, ofZ,. mactans msAt #1005-
B-39 from Florida; h, the same, of L. hesperus male from Los Angeles. California: i, the same, of L. mactans
male #1005-C-29 from Florida;], the same, of L. maclans male #1005-8-49 from Florida; k, the same, of Z..
mactans from Georgia.
ectal sides; and Levi's illustrations likewise show these variations. The problem becomes
one of deciding where to make the coil count, since, of course, it is not a matter of
concentric circles, but of a spiral. Even two brothers may look quite different, and have
been mistakenly considered as belonging to different species, e.g., L. hesperus # 1002-A-48
(Fig. 13f) and #1002-A-51 (Fig. 13e). Obviously, therefore, one must expect some
specimens of L. hesperus to show an apparent three coiled condition. The same is
sometimes noted with brothers of /.. mactans, e.g., #1005-6-39 (Fig. 13g)and #1005-6-49
(Fig. 13j). Note the similarity between the latter and figure 13f of the L. hesperus specimen.
Also note the similarity between the former and the L. hesperus figure 13e. Many other
52 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
variations occur, and the figure 13i shows another L. mactans, brother of the previous two.
Levi's (1959) specimen from Colorado (undoubtedly L. Hesperus) illustrated in his
figure 58 and labelled L. mactans by him appears almost identical with my L. hesperus
from Los Angeles, figure i3h. Similarly, in his figures 40 through 47, representing L.
geuinetricus. some appear to show the four coils characteristic of that species and others
show only three. Likewise, his figure 61 of a Peruvian L. mactans is almost identical with
his figure 47 of a L. geometricus from South Africa; both show three coils over part of the
circumference, and four over part.
It would seem that the number of coils of the embolus cannot be used alone as a
character for separating the species. Yet in effect this is presumably what Levi has done,
and because he has found in Utah (and other western States and Provinces) both three-
coiled and two-coiled specimens he naturally supposed that he had both " L. mactans and
L. curacaviensis [ actually L. varioliis ] both of which are found in Utah," and that
Chamberlin and Ivie failed to distinguish them. As has been previously indicated, I believe
that our western black widow is L. hesperus, and further that the variety texanus is a
variety of L. hesperus, and not of L. mactans in the strict sense.
In nearly all araneomorph spiders the male can be recognized in the penultimate
instar, because the palpal tarsus appears bulbous then (fig. 12e). Because the palpal organ
in Latrodectus is relatively large, its development begins even before the penultimate
instar. Hence the palpal tarsus is slightly swollen in the antepenultimate instar (Fig. 12d) so
that it is possible also at this stage to recognize a male. The swelling may be noticed even in
the preantepenultimate instar (Fig. 12c) as suggested by Bhatnagar and Rempel (1962), but
I have not found this a constant character. On the other hand I have noted a number of
instances in which there are three (not just two) instars preceding the penultimate one, in
which a slight enlargement of the palpal tarsus gives an indication that the specimen is a
male. I have also noticed that the degree of swelling is not the same for all individuals in the
preantepenultimate or antepenultimate instars. There are indications that some expansion
occurs during an instar itself
INTERNAL MALE GENITALIA
Chromosomes were studied in the cell divisions taking place in the testes. The detailed
description of meiosis and of the individual chromosomes is being reserved for a separate
publication elsewhere by Barbara Kaston. Suffice it to say that in all three species she has
found that the sex chromosome situation is of the X] XtO type, so that in the female the
diploid condition shows two chromosomes more than in the male.
The dissection of the testes was carried out by submerging the freshly severed
abdomen in frog Ringer's solution and pinning it venter down in a wax-bottomed dish. A
cut was made along the middorsal line and the dorsal exoskeleton removed. Usually the
heart remained adherent to the exoskeleton, and the testes became visible as two tubes only
slightly kinked, and arched (Fig. 19c) to more or less conform to the curve of the dorsum.
The testes are loosely attached to each other by short bands of connective tissue and at the
posterior end are attached in the region of the spinnerets by a longer ligament. At the
anterior end of each testis is a ductus deferens, which extends forward, then downward and
backward, joining its mate just before the gonopore at the middle of the epigastric furrow.
It should be noted that although Millot (1949) described the testes of spiders as lying in
the ventral half of the abdomen, and below the chylenteron, and his figure 454 shows this
for Scytodes thoracica, in Latrodectus the testes for most of their length lie fairly close to
the heart in amongst the chylenteric ceca. While not as short and straight as given by
Bertkau (1875) for Tegenaria domestica, they are not as long nor as convoluted as in the
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
53
Figure 14. a, apical aspect of palpal organ of male L. variolus from Connecticut; b, same palp, ectal aspect; c,
outline of abdomen of male L. Hesperus from the left side, showing position of testes in situ; d, testes straightened
out as they appear from above; e, sperm cells of L. Hesperus; f, dorsal aspect of cleared epigynum of a mated
female of L. variolus from Michigan, that had mated, showing an embolic sclerite left behind in each of the
spermathecae; g, egg sac of L. Hesperus, natural size; h, egg sac of L. mactans, natural size.
theraphosid illustrated by Melchers (1964). If removed from the abdomen and straightened
out somewhat they appear with gently undulating walls (Fig. 14d), each testis
approximately 2.4 to 2.7 mm long and about 0.25 mm wide. The ductus deferens extends
for about another 0.4 to 0.6 mm from its anterior end. In the fresh condition the testes
appear grayish opalescent in contrast to the chalky white of other adjacent structures.
Upon crushing portions of the testis in frog Ringer's solution and examining on a slide
one can make out the sperm cells. These appear as in figure 14e, with the head piece rather
elongate, and at least slightly curved. Some spermatozoa show the head pieces curved even
more, with much variation all the way to those showing the head in a tight spiral. For L.
hesperus the head piece measures about 19 to 21 microns in length by 3 or 4 in width. The
flagellum appears attached asymmetrically and is about 30 microns in length. In L.
mactans the head piece is 15 to 18 microns long and the flagellum 30 to 35. Very little is
known about spider spermatozoa, but recently Bacetti et al. (1970) working with Pholcus
phalangioides. Agelena labyrinthica, and Pardosa vittata likewise reported a flagellum
length of about 30 microns. The head portion was not described or illustrated, but these
workers commented on the fact that while in the testes the head portion is spirally curved.
54 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
COURTSHIP AND MATING
In nature, mating most often occurs in the late spring and early summer, but in mild
climates, such as along the Pacific coast and in our southern states mating pairs may also
be seen in late summer and early fall as well.
The courtship and mating behavior for L. hesperus was reported by Herms et al.
(1935) and D' Amour et al. (1936). and for L. bishopi by McCrone and Levi (1964). It is
essentially similar to that described by Gerhardt (1928) for L. tredecimguttatus. by Shulov
(1940) for that species and L. pallidus, and for L. indistinctus by Smithers (1944).
Presumably, before the male begins his search for the female, and courtship is to
begin, the male will charge his palpal organs with semen. I have not observed this process
of sperm induction. Herms et al. (1935) reported that they had seen it, but gave no details.
The length of the courtship preliminaries varies, but generally is shorter with young
females than with older ones. For these observations the female was allowed to establish
herself for at least a week in a large glass cage. The male was introduced at the upper
corner farthest from the female.
Almost immediately upon being put into the female's cage the male shows signs that
he is aware of the female's proximity. The abdomen is vibrated rapidly, and with jerky
movements of the legs the male wanders about, every once in a while twanging the threads
as he progresses. Eventually he heads in the direction of the female. Within 10 or 15
minutes he begins a new maneuver, which consists of cutting portions of the female's web.
He continues this cutting as he approaches her, so that the silk is gathered up in
concentrated bands and sheets, instead of appearing spread out as before. Sometimes the
female charges at him. whereupon he hastily retreats. After a short rest with his abdomen
not twitching, he once more approaches her. This charging and retreating may be repeated
several times, but if it continues and the female becomes violent in her rushing, the male
may remain at a distance and eventually discontinue his courtship.
If the female is not too aggressive he may find himself within touching distance of her
within 30 minutes (Fig. 15). With his front legs he strokes and taps her legs, and then her
body. This contact heightens his excitement so that his abdomen twitches more rapidly. If
the female does not kick him away she too may begin to engage in leg stroking activity.
The male then walks over and around the female jerkily, at the same time surrounding her
with silken threads. This "bridal veil," which I have seen used by crab spiders and others,
was observed for Latrodectus by Gerhardt, and by Smithers, as well as by Herms et al., but
was apparently not observed by McCrone and Levi. As Smithers indicated, the threads are
too fine to hinder her when she later decides to break loose, but serve to remind her that
"her partner is in attendance."
Sooner or later in his wanderings over her body, with his pedipalps constantly tapping
her, he locates the epigynal area. Some time may be spent "boxing" this area, apparently
in an attempt to hook into the opening. Finally, after having been in the female's web about
100 to 140 minutes, he assumes the copulatory position, position III of Gerhardt, in which
both sexes face the same way, venter to venter (Fig. 1 b).
Sometimes the right palp is used first, and sometimes the left. When the palp is
hooked into position, the embolus is inserted and the hematodocha distended, then
deflated, indicating that semen is being transferred. Insertions may last from one to 32
minutes, but most often from 4 to 8. Sometimes the female struggles out of her bonds after
only a few minutes, too short a time for the male to have inserted more than one palp. She
may now turn upon him aggressively, so that he is forced to retreat.
Often the male will try again by going through the same ritual of cutting web lines.
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
55
Figure 15. Courtship pairs, a, L. hesperus. b, L. mactans.
twanging threads, and vibrating the abdomen.
If the female remains still after the first palp is withdrawn, the male will insert the
other palp. This occurred in about one third of the matings observed. It is well known that
when the female frees herself from the "bridal veil" the male may be in danger. If she is
hungry she may kill and eat him. With well fed females this is not likely to happen, and I
have on several occasions left the male in the cage with her. In the course of time, two
weeks or more, the male dies untouched by the female.
Upon withdrawal of the embolus the distal sclerite is left behind in the genitalia of the
female. Depending upon whether one or both palps have been used, there will be one or two
56 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
sclerites (Figs. 2g, 1 Ig). But sometimes one epigynum will show three sclerites (Fig. 1 Ih)
indicating that the female had mated with at least two males.
Of 51 attempts with L. hesperus, where courtship was carried out, there were 12
successful cases of insemination. Of 13 attempts with L. mactans, 5 were successful, and of
8 attempts with L. variolus 5 ended successfully. The behavior appeared the same in all
three species. When insemination did not occur, it was usually because the female repulsed
her partner before the act could be completed.
I tried to induce interspecific crosses between L. variolus and L. mactans (4 attempts)
and between L. variolus and L. hesperus (18 attempts) none of which terminated in an
insemination. Three of 27 attempts between L. mactans and L. hesperus were successful,
but there was no development of eggs laid by these females.
CONSTRUCTION OF THE EGG SACS
Whereas the egg sacs of the three species vary somewhat in size, shape, and color, they
all appear to be of tough papery texture, usually only slightly translucent. In this respect
they are similar to the sac of L. bishopi, but quite dilTerent from that of L. geometricus,
which is non-papery and quite translucent. Sacs made by virgin females may be abnormal
(Kaston, 1968). The sacs of L. hesperus and L. variolus are pear-shaped (Fig. 14g), and
often somewhat spread at the top, about 13 or 14 mm in height and about 10 or 12 mm in
diameter. Those of L. hesperus are most often creamy yellow to light tan, those of L.
variolus light tan with most often a tinge of gray. The sacs of L. mactans almost always
show the gray tinge and often are quite decidedly gray, even when freshly made. Also, they
are more nearly spherical, about 1 1 or 12 mm in diameter, and with a conspicuous nipple
at the top (Fig. 14h).
At times the sacs do not show the typical color. McCrone and Levi (1964) indicated
that sacs of L. bishopi dilTer from those of L. mactans in having a soft texture and being
white in color. But the sacs of L. bishopi I have seen are just as papery as those of L.
mactans and are light tan. The sacs of Z,. geometricus are studded with conspicuous pom-
pons, whereas the other species make sacs without these surface features. But Abalos
(1962) and Abalos and Baez (1967) indicate that apparently there is a species in Argentina
(their Lairodectus No. 2) which does have pom-pons on the surface, albeit they are smaller
and less conspicuous than those of L. geometricus. Likewise, occasional sacs made by L.
variolus appear to have very small, irregularly distributed, whitish pom-pons showing up
against the gray surface. This has been observed with specimens from Arkansas, Michigan
and Missouri.
Nearly all of the hundreds of egg sacs made in the laboratory were made during the
night. However, of those made in daylight four of L. hesperus were begun after noon, and
27 before noon; five of Z,. mactans were made in the morning and one in the afternoon; and
for L. variolus four were made before noon. The behavior is about the same in all three
species, and the construction may be conveniently divided into four steps.
The egg sac is begun with the laying down in step I of the canopy, a small circular
sheet, which is gradually enlarged so that its diameter is about that of the completed sac.
As the spider finishes the periphery of the disc she also slowly pulls it into the shape of a
shallow cone with a slight peak. The duration of step I averages about 23 minutes.
Proceeding to step II the spider stands under the canopy and extrudes the mass of eggs.
This is done with rapid upward jerks of the abdomen at the rate of about 1 00 to 1 20 times
per minute. The jerking expels the eggs and coUeterial fluid which cements them together,
and also pushes the eggs higher toward the canopy. This step averages about 1 1 minutes.
The female next begins spinning a transparent layer of gauzy silk around and under
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
57
Figure 16. Egg sacs of L. Hesperus from California, a, sac opened to show eggs shortly after ovipositon; b,
embryos ready to hatch; c, adult fly parasites, Pseudogaurax signala, on egg sac; d, maggots and puparia of
Psuedogaurax in amongst spiderlings.
the egg mass, and about 5 or 6 mm from it. She works from the canopy downward and
finishes at the bottom where the egg mass usually soon comes to lie (Fig. 21b). The egg
mass itself comes to occupy one-half to two-thirds the volume of the sac. This step III
averages about 25 minutes.
The final action, in step IV, consists of covering the sac with tough, more or less
opaque papery silk, and averages about 100 minutes. During the first 10 to 15 minutes of
this period the spider walks around over the sac, drawing out silk with her hind legs and
tapping with her spinnerets at the rate of 60 times per minute. Later, the hind legs are no
58 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
longer used, but rather, the silk is applied directly each time the spinnerets touch the sac,
now at the rate of about 120 times per minute. Since the abdomen is brought up to the sac a
distance of several millimeters it is this length of thread which is applied each time a tap is
made. The rate of tapping may rise to 150 and even 200 times per minute (in one case to
240) as the work continues. At intervals she turns the sac about. During the final 30
minutes or so she stops for several brief rests.
The sac is ordinarily suspended in the snare (Fig. 21a), in or near the retreat if one is
built. Sometimes eggs are laid without any sac whatever. They are merely dropped to the
bottom of the cage. Rarely this behavior is displayed by impregnated females; more
usually by virgins.
NUMBER OF EGG SACS AND FECUNDITY
The interval between copulation and the production of the first egg sac varies
considerably. While Abalos and Baez (1967) found this interval to be 7 days for L.
geometricus. and Miller (1947) found it to be 61 days for L. Hesperus, the period noted by
D'Amour et al. for this latter species was over a year. My own records show for L.
Hesperus that the shortest time was 7 and the longest was 305 days; for L. mactans 16 and
22 days; and for L. variohts 14 and 27 days respectively.
Black widow females are capable of making many egg sacs. The highest number
reported is 29 for L. geometricus by Bouillon (1957a). According to Burt (1935),
Illingworth reported 15 for L. mactans; nine is the maximum reported for L. indistinctus
by Smithers (1944), for L. Hesperus by Chamberlin and Ivie (1935), and for L. curaca-
viensis by Bucherl (1969); and eight for L. tredecimguttatus by Juberthie (1954). My own
observations indicate up to 6 for L. variolus, 10 for L. mactans, and 21 for L. Hesperus.
Since 1 had relatively few specimens of L. variohis, and had them for a relatively short
time, I cannot be sure of the significance of the figure for that species. But there were a
surticient number of specimens of L. mactans, and they were maintained for a sufficiently
long time so that it may be safely said that there are fewer egg sacs made by this species
than by L. Hesperus. However, they lay more eggs per sac on the average, the mean for 185
sacs being 255 eggs, while for L. Hesperus the mean of 464 sacs was 196 eggs.
A maximum of 5761 eggs laid by a single female of L. geometricus was reported by
Bouillon (1957a). For L. mactans I found that the greatest productivity was 2132 eggs in
the nine sacs made by #1221. The largest number of eggs per sac was 919, followed by 530,
435, and only a few sacs with over 300. The most common range was 2 1 5 to 237. There was
one sac with a single egg, and five others with fewer than 100. For L. Hesperus the largest
number of eggs per sac was 598, and the same female produced a sac with 527. The next
highest was 427, and very few sacs had over 300. Commonly the range was 160 to 225.
There was one sac with a single egg, and five others with fewer than 10 eggs each. The
greatest productivity was 3024 eggs for the 12 sacs made by # 1002. The largest number of
eggs laid in one sac by L. variohis was 3 1 5, and the mean for 34 sacs was 1 64 eggs.
Baerg (1945, 1959) reported L. mactans females producing four to nine sacs with the
largest number of eggs in the early ones, and fewer in the later ones, the last one or two
often being empty. I did not find this to be the case. The number varied from one sac to the
next, sometimes smaller, and other times larger.
Bouillon and Lekie (1961) noted that in L. geometricus eggs were laid at intervals of
four days for the first few sacs, but the intervals became progressively longer. On the other
hand 1 found that the intervals between successive sacs varied widely. The shortest period
lor /,. Hesperus was six days between the eighth and ninth sacs made by #2130; the longest
was 332 days between the second and third sacs made by #1069. For L. mactans the
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
59
shortest interval was six days between the second and third sacs of # 1 147; the longest was
190 days between the first and second sacs of # 1 192. For L. variolas the shortest was 13
days between the second and third sacs of # 1376, and the longest was 1 13 days between the
third and fourth sacs of the same spider. No trend for either increasing or decreasing the
interval can be seen from the data, and as can be seen from table 1 the coefficient of
variability indicates wide variation in the spacing of sacs.
Table 1 . Interval between ovipositions.
Number
of
Females
Number
of
Sacs
Inter
val; in
days
Coeff. of
Variation
Range
Mean
S.D.
I L.
variolus
Virgin
4
7
17-86
28.9
± 25.6
88.6
Non-Virgin
11
20
13-114
30.0
± 29.5
98.3
L.
mactans
Virgin
5
10
13-190
40.0
±171.8
429.5
Non-Virgin
43
128
6-190
33.5
± 29.0
86.6
L.
hesperus
Virgin
48
156
1-332
49.5
±175.8
355.2
Non-Virgin
109
672
7-305
34.8
± 29.7
85.3
Naturally one could hardly expect that all the eggs laid would actually develop. Very
few sacs showed a development of 100% of the eggs. However, it would be reasonable to
expect that those sacs produced early in a series would show a higher percentage of fertility
than those made later. Sometimes the first five to seven sacs from a single female showed a
fertility percentage of 98 or 99. Though one would expect that later sacs might show an
increasingly lower fertility this did not follow in any regular manner. For example, for L.
hesperus #2130 the first five sacs showed over 90% development from each. From the
sixth sac only 30% of the eggs developed, yet the next six sacs averaged a development of
over 90 % again ! The 1 3th (and final) sac had 78 % of the eggs developed. There were many
other instances where a sac would show no development whatsoever, then later sacs show a
fairly high percent of fertility. I could find no pattern of increasing or decreasing fertility.
Usually, when a large number of eggs is laid and only a few develop, the spiderlings do
not emerge from the sac. There are exceptions, an outstanding one being the case of egg sac
#1050-B, of L. hesperus. where of the 208 eggs laid only one developed, but the spiderling
emerged.
A Connecticut specimen, presumably of L. variolus, mature when collected on 10
April 1949, produced an egg sac on 17 June 1950 from which spiderlings emerged. Since
the female had not been mated in the laboratory at least 434 days had elapsed from the
time of mating to the time of fertilization. This longevity of sperm cells, which is
considerably greater than that of the male spider that made them, is exceeded in two
instances by L. hesperus. One #1053, collected already mature on 6 December 1965
produced her fourteenth egg sac on 4 February 1967, from which spiderlings emerged.
Thus the sperm cells remained viable for at least 455 days. Also, #1202, collected already
mature on 12 December 1966 produced her fourteenth sac on 24 April 1968. Of the 170
eggs laid, 36 developed into spiderlings, indicating that sperms had lived at least 499 days
after insemination. Actually the correct figure is nearer 600 days, for this female at the
time of collection already had five egg sacs in her web.
60 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
DEVELOPMENT WITHIN THE EGG SAC
In all three species the eggs may be occasionally lavender-pink or mauve. However,
most often they are creamy white to yellow, and sometimes orange. The same sac may
contain eggs of more than one color but the further development bears no relation to the
color. The eggs are spherical, or almost so; those of L. mactans average about 0.95 mm in
diameter; of L. hesperus about 1 . 1 mm, and of L. variolus about 1 .2 mm in diameter.
To obtain data on pre-emergence development many sacs were opened (Fig. 16a)
within a day of oviposition. These sacs were placed in vials of which the plugs were kept
moistened to protect the eggs from total drying. Within a day of oviposition the eggs dried
sufficiently to roll around freely. If dropped on a hard surface they would bounce and roll,
seemingly without injury.
Although Baerg (1945) reported hatching in L. mactans after only 8 days, in my
experience the hatching time for all three species was nearer two weeks. The average time
in days was 13.4 ± 2.0 for L. variolus, 14.2 ± 1.4 for L. mactans. and 14.6 ± 2.0 for L.
hesperus. About a day or so before hatching the membranous surface of the egg becomes
wrinkled, and one can see the outlines of the cephalothorax and appendages as bulges (Fig.
16b). The newly hatched spiderlings are entirely unpigmented, without eyes or hairs, and
they move feebly. Within a day after hatching the first sign of pigmentation appears as a
ring at the periphery of each anterior median eye. The six indirect eyes begin to develop
their pigment a day or so later. Also, about this time, fine black hairs begin to show up on
the dorsum and legs. In addition, the legs show a slight yellowing. That the spiderlings
undergo their first molt inside the sac was first reported by Rau (1924) and this occurs
about three or four days after hatching.
During the next five to seven days, pigment is gradually deposited in the characteristic
pattern. Also, the spinnerets develop to the point where they can function, so that by the
end of this period the spiderling begins to spin silk as it crawls about among its fellows in
the sac. Twenty to 23 days after oviposition the youngsters appear ready to emerge from
the sac. However, the actual emergence does not usually take place for another few days,
during which time there is somewhat more pigment deposited in the pattern. The average
time, in days, from oviposition to emergence was 26.2 ±2.2 for L. variolus, 29.1 ±3.0 for
L. mactans, and 30.3 ± 2.8 for L. hesperus.
EMERGENCE
An emergence hole is visible about a day or so before the spiderlings actually emerge.
The hole is made by cutting with the chelicerae, and possibly also by digesting away the silk
by regurgitated proteolytic enzymes. At least I have seen what appears to be a moistened
area on the silk as the chelicerae are worked around enlarging the hole. Ordinarily a single
hole is cut allowing escape of the spiderlings. This hole, about 1 mm in diameter, may be
made by one spiderling, or by two working together. A few sacs have been found with two
exit holes, and rarely with three.
If, after the escape of one or more spiderlings, the exit hole is now covered over or
plugged with glue, a new hole will be made by the remaining spiderlings Even a third hole
will be made if the second is plugged. On the other hand some sacs remained unopened,
and the spiderlings were later found dead in the sac, e.g., #1374-B of L. variolus from
Michigan, #1380-8 of L. mactans from Missouri, and #14 15- A of L. hesperus from
Texas. In a few sacs spiderlings did not emerge but nevertheless grew and molted, in one
case to within one instar of maturity (Kaston, 1968). In another sac, of L. mactans from
Illinois #1283-8, there was no emergence six months after the eggs were laid. In the sac
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 61
were 8 spiderlings among a large mass of undeveloped eggs. Of the 8 spiderlings one was a
large male, probably in the fourth instar to judge by its size and markings, which by its
palpal development appeared to be in the antepenultimate instar. The other seven looked
like third and fourth instar females. Usually when growth and molting occur without
emergence, there are few spiderlings and a relatively large number of undeveloped eggs;
and the spiderlings obviously have been feeding on the eggs. In fact, when a sac has only a
small number of spiderlings emerging, and a large number of undeveloped eggs, the
spiderlings are often larger and in the second post-emergence instar, making it appear that
they have been feeding on eggs. This phenomenon has been recorded for L. tredecimgut-
tatus by Juberthie (1957) and for L. geonietricus by Bouillon (1957a). It seems, however to
be of even more general occurrence, having been observed by Galiano (1967) in Loxosceles
laeta, and in Gnaphosidae and Clubionidae by Holm (1940). The latter also indicated that
it had been observed by Wagner in Lycosidae, by Becker in Drassodes lapidosus, and by
Lecaillon in Chiracanthium carnifex. Peck {in litt.) informed me that he had observed the
same in C. inclusum.
Often eggs may not develop to the hatching stage, and of course not all spiderlings
emerge even when a hole is cut and the majority escape. Baerg (1954) considered that
drought was a factor in failure to hatch, but in view of the fact that many black widow
spiders live in arid regions, and judging from the studies of Shulov (1940) and Bouillon
(1957b) it is hardly likely that drought is an important factor. Also, Baerg believed that
emergence coincides with the end of the first instar, when in fact the molt to the next instar
generally occurs some time after emergence.
DESCRIPTION OF POST-EMBRYONIC STAGES OF DEVELOPMENT
Among the exotic species the young of L. tredecimguttatus were first described by
Duges (1836), and more recently by Marikovskii (1947) and by Shulov and Weissman
(1959); of A. indistinctus and L. geonietricus by Smithers (1944); of A. revivensis by Shulov
and Weissman ( 1 959); and of L. pallidus by the latter and by Beregovoi ( 1 962).
Some authors have indicated periods in the life cycle by specifying the number of
molts (to maturity, etc.). Others have referred to stages, with the first being the one after
hatching from the egg. This takes place inside the egg sac, as does also one molt before
emergence. Often an additional molt, or two, occur inside the sac as mentioned above.
Ordinarily the spiderling will leave the sac during its second true instar. However, for
convenience I have designated this post-emergence nymphal instar as the first instar, as
was done by Miyashita (1968). For the sake of uniformity in making comparisons
therefore, the results stated by other authors have been converted to this system.
There is a wide variation in appearance among both adults and juveniles. As long as it
was supposed that all North American forms belonged to the one species, L. mactans, this
variation seemed even wider, but even with the acceptance of L. variolus, and now of L.
hesperus. there is still much variation in each.
The drawings for each instar give some idea of the appearance of a common form, but
it must not be expected that all specimens will look exactly like these. Even siblings from
the same egg sac may vary widely, and even by the time they emerge. In addition, I have
raised many females, which upon maturity assumed a pattern quite different from that of
their mother.
Another type of variation is that occasioned by the spiderlings changing their
appearance during a given instar, as originally observed by Moles (1916). In general this
means that some third instar spiderlings may resemble some second, and others may
resemble some fourth instar individuals. Presumably this change is somehow related to
62 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
nourishment during this time. At any rate, those spiderlings which were not fed during the
first instar, and yet molted, changed very Httle.
Still another type of variation is geographic. This had been recently demonstrated for
Peruvian specimens of L. mactans, by McCrone and Levi (1966), and it had been noted by
others as well. Smithers (1944) encountered it in his studies on L. indistinctus in South
Africa, and I have seen many examples of it in our three species from all over the United
States.
To some extent the appearance of the adult is correlated with the instar in which the
individual matured. Those maturing in later instars may be darker than those maturing in
earlier. However, this is not absolute, and I have found that many specimens maturing in
the sixth, seventh, and eighth instars cannot be told apart. McCrone and Levi (1964)
reported ''noting a striking correlation between total length of the spider and the
coloration of different forms, the smaller being brightly colored, the largest ones dark."
This correlation seems high, but it is not perfect. Also, smaller specimens can be lacking in
red markings as much as larger specimens.
Latrodectus mactans
FIRST INSTAR (Figs. 17a, b). — The carapace is mostly yellowish suffused with gray, with
black around the eyes, and with black marginal and median stripes. The sternum is yellow
with a very thin marginal black stripe. The legs are yellowish orange, at most only very
faintly annulate with dusky, but usually somewhat grayer toward the distal ends of the
tarsi.
The abdomen is orange-red, with a pair of anterior white bands, often joined across
the front as a single "chevron" mark, a median white band, and two pairs of diagonal white
bands extending laterally. The median band may be divided into two, three, or four spots,
by encroachments, at intervals, of the orange-red ground color. In many specimens the
white lateral bands are bordered along the posterior edges by a line of black pigment.
There may also be thin black lines along the sides of the median light band. But many in
the same brood of spiderlings show no black at all in this stage.
There is a conspicuous black spot covering the spinnerets and anal tubercle. Many
specimens already show an indication of an hourglass mark, but with others the reddish
orange pigment seems evenly distributed over the venter and sides.
SECOND INSTAR (Figs. 17c, d). — The general impression is that there has been a marked
change from the first instar, with much black pigment deposited. On the carapace the dark
marginal bands are broader, and even the light areas are suffused with gray. The sternum
likewise is darker. On the legs the annuli are now distinct. Leg I has the femur with the
basal half dark, and the distal half light, except for a small dark ring at the apex; the
patella is suffused with gray; the tibia has a dark ring at each end, and one just about at its
middle; the metatarsus and tarsus are without dark annuli. Leg II has the femur yellow,
except for a small dark ring at the apex; the patella has a faint indication of a dark ring
distally; the remaining segments are like those on leg I. Leg III has the femur with a dark
ring at the apex, and a dark ring at each end of the tibia; otherwise it is like leg I. Leg IV
has the femur with a dark area occupying the middle half of the prolateral surface, then a
ring at the apex; the patella has a distal dark ring; and the remaining segments are like
those on leg I.
On the dorsum black bands alternate on each side with creamy white areas (Fig. 17c).
There is almost always a light median spot, followed by a light median band, which
however, may be broken into spots. In some specimens these white areas have a little
orange pigment in their centers. The venter plainly shows the hourglass mark, which is
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
63
Figure 17. Postembryonic development stages of L. maclans. a, first instar, dorsal aspect; b, the same, ventral
aspect; c, second instar, dorsal aspect; d, the same, ventral aspect; e, third instar, dorsal aspect; f, the same,
ventral aspect; g, fourth instar, dorsal aspect; h, the same, ventral aspect; i, fifth instar, dorsal aspect;], the same,
ventral aspect; k, sixth instar, dorsal aspect.
much more constricted in the middle than with L. hesperus.
THIRD INSTAR (Figs. 17e, 0- — The gray areas are now larger on the carapace as well as
on the sternum, where the light area is reduced to a median narrow stripe. The legs are
much darker, though annuli are still visible. The most noticeable change is on the
abdomen, where the black areas are much larger, with the consequent decrease in size of
the light areas. This is particularly noticeable in the mid-dorsal stripe, which is now broken
up into more spots. The hourglass mark is more distinct, and is bordered with black. In
64 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
some individuals it is quite red, in others it is almost white.
There is a great deal of variation. Some individuals look much like those in the second
instar, and others like those in the fourth. The degree of blackness varies widely. Some
have the white areas quite restricted; some show only black and white, but no red on the
dorsum; others show only black and red, with no white on the dorsum.
FOURTH INSTAR (Figs. 17g, h). — The carapace is darker. The midline stripe on the
sternum is narrower. The legs have changed little, but the dark areas are a bit more
extensive. The abdominal dorsum shows the light areas still further reduced in size. Of
those specimens with white, rather than red, spots, most show orange pigment in the
centers of these white spots. As in the preceding instar there is considerable variation,
many specimens looking like third and many like fifth instar individuals. Many males show
by the enlargement of the palpal tarsus that they are in the penultimate instar.
FIFTH INSTAR (Figs. 17i, j). — There is relatively little change from the previous stage,
with a continuation of the overall darkening, as the black pigment spreads. The legs are
still banded. On the dorsum the white diagonal bands are thinner and shorter, and often the
third (or most posterior) pair is absent.
SIXTH INSTAR (Fig. 17k). — The carapace is quite black. The legs are almost all black,
with the former light areas being dark brown. However, some specimens may show the
annuli slightly. On the dorsum the white diagonal bands are very much reduced; the first as
a small chevron mark, and the second as a faint remnant. Sometimes the chevron is
represented as a pair of red spots. The spots of the median row are now bright red. In many
specimens the spots tend to disappear from anterior to posterior so that in some there
remains only the most posterior spot just above the anal tubercle. Some of the females
mature in this instar.
SEVENTH, EIGHTH AND NINTH INSTARS. — Most females mature in these instars. There
is usually only a remnant of the chevron mark at the front of the dorsum. The diagonal
light bands have disappeared entirely, or are at most only very faint, so that the only spots
remaining are the red ones of the median row, and even these may be reduced to just one
above the anal tubercle. The hourglass mark usually consists of an anterior triangle and a
narrower posterior rectangle with rounded corners.
Latrodectus hesperus
FIRST INSTAR (Figs. 1 8a, b). — The ground color of the carapace is dusky grayish yellow.
However, the eye region is black, and there are three black longitudinal stripes. One of
these extends from the median eyes to the rear; the other two are along the lateral margins.
The sternum has similar marginal bands, but is otherwise dusky yellow. The legs have the
same ground color as the carapace. Leg III is much less pigmented than the others, and is
dark only at the distal end of the tarsus. The other legs show a dark ring at the distal ends
of femur, patella, tibia, metatarsus and tarsus, as well as a ring at the proximal end and
middle of the tibia.
The ground color of the abdomen is creamy white, some specimens showing a light tan
to olive-green hue toward the sides. There are two rows of black spots extending along the
dorsum, four to six spots in each row. Between these two rows many specimens show two
rows of very narrow lineate black spots, sometimes only on the anterior half. On the
ventral side one can see two black spots on either side of the spinnerets. Between the
epigastric furrow and the base of the spinnerets is a more or less rectangular yellow area
where the hourglass mark will appear in later instars. At this stage it is barely pinched in at
the middle.
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
65
Figure 18. Postembryonic developmental stages of i. hesperus. a, first instar, dorsal aspect; b, the same, ventral
aspect; c, second instar, dorsal aspect; d, the same, ventral aspect; e, third instar, dorsal aspect; f, the same,
ventral aspect; g, fourth instar, dorsal aspect; h, the same, ventral aspect; i, fifth instar, ventral aspect.
SECOND INSTAR (Figs. 18c, d). — The dark lines on the carapace have become a trifle
wider and the legs a bit darker, with leg III beginning to show annuli like the others. The
greatest amount of change is on the abdominal dorsum where there is a suffusing of gray
pigment on the sides of the dorsum that extends down laterally. In some specimens the
black spots are larger than they were in the previous instar, but in others they have become
incorporated into two faintly discernible olive bands. A similarly colored band begins to
appear encircling the dorsum up front.
THIRD INSTAR (Figs. 18e, 0- — On the carapace the marginal dark bands have widened.
The marginal dark bands of the sternum have become wider than the median yellow area.
66
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 19. Postembryonic developmental stages of L. hesperus. a, sixth instar female, dorsal aspect of a dark
specimen from El Centro, California; b, the same specimen, ventral aspect; c, sixth instar female, dorsal aspect of
a light specimen from Riverside, California; d, fifth instar female, dorsal aspect; e, fourth instar, penultimate
male, dorsal aspect; f, the same, ventral aspect.
The dorsum now distinctly shows the pair of longitudinal bands, olive gray, one on either
side of the midline, with a branch on each side from the anterior end and another such
branch from just behind the middle. Each of these branches extends diagonally back to the
side. The anterior encircling band is now more pronounced, so that from the side one now
sees these bands extending down obliquely to the rear. Besides these there is the band
extending straight back, closer to, and parallel to, the midline. Thus there are three light
areas on each side, as well as the median one. In many specimens a little orange pigment
begins to form along the center of this median one. There are still remnants of the original
black spots, though these are now blended in with the olive gray bands. In the texanus
variety these latter bands are often pink. On the venter the two black spots on either side of
the spinnerets are quite conspicuous. The hourglass mark now shows tinges of orange, and
to each side of it a black line has developed.
FOURTH INSTAR (Figs. 18g, h). — There is a light variety, members of which are hardly
distinguishable from the third instar. Often males are recognizable now as in the
penultimate instar (see below) and they are usually of the light variety (Figs. 19e, 0- For
the dark variety one notes that on the sternum and carapace the areas covered by the black
bands have enlarged, and the median band of the carapace has widened behind.
The abdominal dorsum now shows the pattern of bands much more distinct, with
more gray mixed in with the olive and usually with somewhat more orange pigment along
the midline. In some specimens the light areas between the bands show some orange
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 67
pigment, and the bands are brownish gray. The hourglass mark shows a Httle more orange
pigment along its middle.
Some males mature in this instar.
FIFTH INSTAR (Figs. 18i, 19d). — The carapace is not much different from that in the
fourth instar, though the sternum shows the light central area still further reduced. The legs
still show some of the annuli faintly, but they are getting darker, and on the tibiae the
central and distal rings have come together to form one large ring. On the abdominal
dorsum the dark bands are wider, so that now the median light stripe is narrower than they
are. For the dorsum as a whole there is a much larger surface covered by the gray bands
than by the light areas. The gray areas are outlined in black. A row of orange spots now
appears in the median light stripe. The orange pigment in the hourglass mark is deeper in
the front and back halves, with hardly any in the central portion.
Some males mature in this instar; and those maturing later seem nevertheless to retain
the markings of this fifth instar. They hardly change although a few seem to get a little
darker. I have examined many specimens that matured in this, in the sixth, and in the
seventh instars, and contrary to ideas I formerly held I was unable to note any significant
differences in their appearance.
SIXTH INSTAR (Figs. 19a, b, c). — The darker specimens show more pigment on the
carapace and have the dark areas more extensive than previously. There remains on the
sternum only a narrow central light band. On the legs the dark areas have increased in size.
The abdominal dorsum is mostly covered with dark pigment now, with the only light
areas reduced to a basal transverse band, a row of spots along the midline, and two pairs of
diagonal stripes extending down the sides to the rear. These latter are the areas that had
previously been wide, between the dark areas that had previously been narrow. Each of the
light spots along the midline encloses a reddish spot. The hourglass mark is becoming more
constricted at the middle and has more red pigment.
Some individuals, showing a more or less similar arrangement of spots, have the
pigmented areas lighter. Also, the light diagonal bands extend farther down on the sides,
and the dorsal spots are more orange than red.
SEVENTH, EIGHTH AND NINTH INSTARS. — Most females mature in these instars. While
I had previously supposed that those maturing in the later instars would show more of the
black pigment, and smaller areas of light pigment a comparison of many specimens
revealed that (as was the case with the mature males) there is no significant difference in
their appearance. A female maturing in the seventh instar may look quite similar to one
maturing in the ninth. This, of course, does not preclude the possibility of changes
occurring during the instar. Some males mature in the sixth and seventh instars, and in
general they resemble a female in the fifth or sixth, of the light variety. Some males are
darker. The legs retain the annuli, which are often more conspicuous than those in young
females, and often without the fusion of middle and proximal rings on the tibiae.
Latrodectus variolus
FIRST INSTAR (Figs. 20a, b). — The carapace is reddish orange to brownish orange, with
the eye region black. The median and marginal dark bands so noticeable in the other two
species are lacking. The sternum is about the same color with the margins somewhat
dusky. The legs show the same ground color as the carapace, but there are faintly indicated
annuli. These appear on legs I, II, and IV at the distal ends of patella, tibia and metatarsus,
with those on leg I slightly darker. Leg III is hke leg I but the annulus is absent from the
metatarsus.
68 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
The ground color of the abdomen is reddish orange. There is a large black area around
the spinnerets and anal tubercle, and there are three pairs of large black spots on the
dorsum. The dorsum has white blotches along the middle, and some white extending to the
sides in front of each black spot. Where the abdomen overhangs the carapace is a white
transverse band, which extends diagonally to the rear on each side to a pair of white spots
farther back. There are three pairs of diagonal white bands. The hourglass mark is not
distinct, but some specimens have irregular white blotches in this area.
SECOND INSTAR (Figs. 20c, d, e, f)- — This is very similar to the first instar, with the
carapace somewhat browner, the black spots on the abdominal dorsum a bit larger, and the
hourglass mark showing more white as "cottage cheese" blotches.
THIRD INSTAR (Figs. 20g, h). — The carapace is now more chestnut brown. The legs
appear dull orange to chestnut brown, and the annuli are slightly darker. On leg I the basal
half of the femur is dark, as are also the distal half of the patella, the distal third of the
tibia, and the distal fourth of the metatarsus and tarsus.
The greatest change is seen on the abdomen, which is now black over most of the
dorsum. There remain a white band across the front, a median row of white spots, and
three pairs of diagonal white bands extending down the sides, so that their lower ends are
visible from the ventral aspect. In each median spot is a small spot of orange-red pigment.
The hourglass mark is complete, but relatively faint in the middle area.
Specimens that have recently molted to the third instar often show the black area
brownish instead, except for those places where the original six black spots were. But the
black pigment suffuses into these brown areas so that later in this stage a larger area looks
black.
FOURTH INSTAR (Figs. 20i, j). — The general appearance is much like the third instar.
The carapace and legs, however, are more dusky. On the abdominal dorsum the spots of
the median row are now red. The basal band and the lateral oblique white bands are
narrower and shorter, so that the laterals no longer extend to the ventral side. The
hourglass mark is now red, and in most specimens shows a distinct separation into two
parts.
FIFTH INSTAR (Figs. 20k, 1). — The spiderlings look much like those of the preceding
instar, but with the carapace and legs darker. The annuli still show on the latter. The
abdominal dorsum has the light areas still further reduced. Nearly all specimens show the
hourglass mark divided.
SIXTH INSTAR. — The carapace and legs are dark brown to black, and the leg annuli show
plainly. Some males mature in this stage.
SEVENTH, EIGHTH AND NINTH INSTARS. — Females mature in these stages. Illustra-
tions of this species have been published by Judd (1965), Wilson (1967) and, under the
name L. mactans, by Emerton (1902) and Kaston (1937a, 1948, and 1953).
APOSEMATIC COLORATION
One can see from the above descriptions how different the spiderlings are from the
adults. These changes, as the spiderlings grow and molt, were first noted by Duges (1836)
for L. tredecimguttatus. Duges, and also Marikovskii (1947) pointed out how the many
different appearances could be responsible for authors describing each stage as a different
species, and this, in part, accounts for the long list of synonyms.
Apropos of the changes in color as the spiderlings develop, it should be noted that all
three of our North American species acquire more black pigment, and that the hourglass
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
69
Figure 20. Postembryonic developmental stages of L. variolas, a, first instar, dorsal aspect; b, the same, ventral
aspect; c, second instar, dorsal aspect; d, the same, ventral aspect; e, second instar, variation, dorsal aspect; f, the
same, ventral aspect; g, third instar, dorsal aspect; h, the same, ventral aspect; i, fourth instar, dorsal aspect; j, the
same, ventral aspect; k. fifth instar, dorsal aspect; 1, the same, ventral aspect.
mark on the venter becomes more prominent and a deeper red. Bristowe (1945, 1946)
considered this red mark an indication of aposematic coloration for the spider, presumably
present in the older females only and not in females younger than the sixth instar, nor in
males (which he considered as maturing in the sixth). From his illustrations of the first and
second instar spiderlings, it is clear that he was referring to L. hesperus, not to L. mactans.
Yet, in L. hesperus the hourglass mark is quite clearly developed already in the third instar,
becomes impregnated with orange pigment (and even with red in some specimens) in the
fourth, and is quite definitely present in males, even if these mature in the fourth or fifth
70
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 21. a, Z,. hesperus female from California, with egg sac; b, L. Hesperus female from Arizona making egg
sac; note ball of eggs through semitransparent unfinished sac; c, on left, the egg sac and newly emerged spiderlings
of L. maclans: on right, the same of L. hesperus.
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 71
instar.
In L. variolus the hourglass mark may appear more "complete" in the young, and
usually loses its middle portion in later instars. This has been reported also by Marikovskii
(1947) for L. tredecimguttatus. Likewise, for L. indistinctus Smithers (1944) has shown
that an hourglass mark is present in the spiderlings but disappears completely (or almost
so) by adulthood. According to Beregovoi (1962), L. pallidus has no hourglass mark in any
of its stages. Despite Bristowe's remark that L. geometricus differs from other members of
the genus in not having the hourglass mark, all specimens from Florida and the West
Indies that I have seen do possess this mark, and Smithers has observed it in South African
members of this species. Moreover, this species has been known to cause envenomation in
humans (Finlayson, 1956). Contrary to Bristowe's supposition, it has been known for
many years that young individuals and males do have poison, albeit less than mature
females. Moreover, even those species without a distinct hourglass may have virulent
venom, and may be much feared.
RATE OF DEVELOPMENT AND LONGEVITY
Several observers have supplied information as to the number of molts and the length
of time to maturity. Studying the ctenid Cupiennius salei, Melchers (1963) found that
poorly fed spiderlings matured after fewer molts. Contrariwise, Miyashita (1968) for
Lycosa T-insignata, and Deevey (1945, 1949) for Latrodectus mactans found that those
poorly fed required more molts. Well fed spiderlings not only underwent fewer molts but
matured in a shorter time. But of course feeding is not the sole factor, and as indicated,
wide variations occur even among siblings in the same family when raised under identical
conditions.
Even under uniform environmental conditions, there was considerable variation with
respect to the number of molts, the intervals between molts, and the length of time it took
for the spiders to mature. This variability extended even to "litter-mates'" from the same
egg sac, and is similar to that found by Deevey (1949), and by Witt and Reed (1965). For
example, there is the case of three sisters maturing on the same day, 100 days after
emergence, L. mactans #1132-A-51 in the sixth instar, #1132-A-57 in the seventh, and
#1132-A-59 in the eighth. Sometimes a particular family shows a faster or slower
development, or in some other way is different from the average, e.g., some of the L.
variolus families had all the males ready to mature in the sixth instar, but in other families
it was the seventh instar. By far the greater majority of spiderlings failed to mature. This
was especially the case with specimens of L. variolus, which appear to require a longer
time, and most often a greater number of molts than do the other two species. Most of
them died in the fourth or fifth instar.
More success was obtained with males than with females, probably in part because
males go through fewer instars on the average, and mature earlier. Thus I am unable to
understand the remark by Baerg (1923) that males (of Arkansas L. mactans) were more
difficult to rear than females, nor the statement by Knowlton (1935) that males (of L.
hesperus in Utah) required a longer time to mature than did females.
For what is undoubtedly L. hesperus, data were given by Herms et al. (1935),
ChamberHn and Ivie (1935), Knowlton (1935), and Bhatnagar and Rempel (1962). For L.
mactans proper, data were given by Lawson (1933), Blair (1934), Muma (1944), Deevey
(1945, 1949) and McCrone and Levi (1964). The latter also compared the development of
L. mactans with that of L. variolus. For exotic species Smithers (1944) reported for L.
indistinctus; Bonnet (1938) and Baerg (1954) for L. geometricus. For L. tredecimguttatus
data were given by Juberthie (1954), Shulov (1940) and Marikovskii (1947); and for L.
72 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
pallidus by Shulov (1940) and by Beregovoi (1962).
Table 2 shows that the number of instars passed through to maturity is quite variable,
and about equally variable for L. mactans and L. hespenis. However, for both sexes, with
individuals undergoing the same number of molts, development is somewhat more rapid in
L. mactans. Also, although females of both species mature in the sixth through ninth
instars, more of the L. mactans reach maturity in the sixth and seventh (mode is seventh),
while more of the L. hesperus mature in the eighth or ninth (mode is eighth). The difference
between them, however, is not nearly as great as that found by McCrone and Levi (1964)
between L. mactans and L. variolus. Because of my poor success with rearing of L. variolus,
only relatively few data can be supplied, and these for males primarily. Therefore, for
comparison, I am including in Table 3 data about this species from McCrone and Levi, but
have rearranged the data to conform to the way these are presented in Table 1. Although I
had no males of L. hesperus maturing in the eighth instar, two penultimate males from
California died in their seventh, and would therefore, had they lived, have matured in the
eighth. Also, besides those shown in the table, one L. mactans and three L. hesperus
females died as penultimates in the eighth instar, and would therefore have matured in the
ninth instar had they lived.'
Table 2 shows that for males there is likewise a wide spread of instars, with the fifth
the mode in both species. It would appear that the shortest time in which a male can
mature is 37 days. However, among L. hesperus families for which the records are
incomplete, were one Arizona family, and one from British Columbia, in which males
matured in 33 days, and a family from Texas in which several males matured in 27 and 28
days. Likewise, although from the table, 177 appears as the longest interval in days to
maturity for a L. mactans male and 263 days for a L. mactans female, I have records of
Table 2. Rate of development of L. mactans and L. hesperus.
Instar
in which
Sex Matured
Number of
Individuals
Number of days spent in each
instar
No. of days to
Range Mean
Maturity
S.D.
Species
1
2
3
4
5
6
7 8
4
16
13.6
13.5
28.8
37-139
61.6
±42.4
Male
5
140
13.9
11.6
19.0
48.0
40-196
99.0
±29.0
S
6
88
14.9
10.7
12.6
24.5
42.3
46-210
119.3
±49.0
7
6
17.0
10.5
11.0
21.0
53.5
27.5
61-214
150.6
±73.5
6
5
9.6
8.6
12.0
26.8
70.2
73-185
137.1
±42.4
'-I
7
22
11.8
10.0
11.5
29.5
58.3
26.8
75-239
146.9
±38.7 1
8
49
12.4
10.3
11.3
22.9
48.9
36.8
26.6
74-325
202.3
±63.9 !
9
12
11.7
10.6
14.7
38.0
68.2
50.9
28.6 29.4
102-325
242.0
±81.8
4
13
16.3
13.7
24.2
49-84
54.4
± 9.2
5
82
12.9
10.7
13.4
26.1
38-138
63.7
±11.8
Male
6
48
13.4
9.5
11.9
22.3
32.3
49-161
84.3
±16.9
s
7
14
10.6
9.1
12.7
13.0
28.8
28.6
74-177
87.7
±42.4
a
8
1
11
12
9
6
14
97
17
—
166
—
5
6
11
16.6
10.6
13.0
24.4
35.9
62-134
111.7
±18.1
'J
7
23
12.6
8.6
14.9
15.0
33.3
31.8
64-193
137.9
±41.2
8
13
11.0
8.9
10.7
19.5
53.4
34.0
23.8
97-263
140.5
±85.4
9
1
7
14
9
12
24
16
7 18
—
107
1
While this manuscript was in press a female L. mactans. #1647-A-1 19, matured in the tenth instar, 190 days
after emergence.
1970
KASTON: AMERICAN BLACK WIDOW SPIDERS
73
194 for a male and 378 for a female from North Carolina.
Table 3 shows that spiderlings of L. variolus spend a much shorter time in the first
instar than do spiderlings of the other two species. Table 4 shows that the minimum
number of days spent in a given instar tends to increase as the spiderlings get older. For the
maximum number of days in a given instar I am unable to find any correlation.
Table 3. Rate of development of L. variolus.
Sex
Instar in which
would have
matured
No. of
Ind.
Mean num
ber of days spent
in each instar
I
II
III
IV
V
VI VII
VIII
5
10
9.8
12.3
24.0
Male
6
75
4.8
11.3
13.3
21.9
7
5
5.1
14.6
12.2
18.2
19.2
8
2
4.5
18.5
12.0
15.0
15.0
12.0
6
3
4.0
8.3
9.3
19.7
Female
7
4
3.5
6.5
11.5
17.0
38.8
8
3
4.5
17.5
11.5
18.5
11.5
26.5
Instar in
which
matured
Data below have been taken from McCrone and Levi (1964).
No. of
days to
maturity
6
1
6
12
11
15
87
131
Male
7
9
5.1
15.1
11.8
13.4
32.6
49.4
129.4
8
12
5.2
16.5
12.3
10.7
14.1
31.6
38.3
129.7
Female
8
18
6.4
13.8
10.1
13.4
22.6
41.9
43.2
152.4
9
4
6.3
12.0
9.2
14.0
14.8
25.2
38.5
33.5
153.5
Table 4. Minimum and maximum number of days spent in each instar.
Instar
I
II
III
IV
V
VI
VII
VIII
L. Hesperus 1
49
3; 52
3; 190
5; 147
6: 200
7; 126
11; 99
15; 53
L. mac tans 1
57
4; 55
4; 112
5; 143
7: 138
10; 97
7; 75
L. variolus 1
33
5; 30
6; 48
6; 76
7; 76124
Table 5 shows that in both sexes L. hesperus on the average lives longer than L.
mactans after becoming mature; for females about one and one-third times and for males
about twice as long. Likewise the total life span from the time of emergence is longer, again
by more than one and one-third times in females, and about one and one-half times in
males. Table 6 shows the maximum number of days that any specimen lived after
maturing, and the maximum number for the entire life span after emergence. The female
of/., mactans which lived 849 days after maturing must have been older than the 858 days,
which is the maximum I have noted for another individual. Adding to 849 the 62 days
74
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Table 5. Life span.
Species
Sex
Number of
individuals
Mean
number of days
post-maturity
Number of
individuals
Mean
number of days;
emergence to death
L. Hesperus
Male
255
46.5 ± 34.6
240
146.1 ± 51.0
Female
144
277.5 ± 152.6
87
481.0 ± 199.5
L. mactans
Male
158
21.1 ± 12.6
149
90.8 ± 24.0
Female
62
203.1 ± 141.4
44
369.6 ± 149.0
found minimal for L. mactans females to mature, one obtains for this particular individual
a life span of at least 9 1 1 days.
The longevity of Z,. hesperus and of/,, variolus thus is greater than that of I. mactans.
To ascertain whether this hardiness was also a feature of the young, I set out a family of
each species right after emergence and kept them without food. Shulov (1940) had found
that newly emerged L. pallidus spiderUngs can live up to 19 days without food. Also, that
the spiderlings of L. tredecimguttatus can molt to the second instar without their having
fed.
My three families included 213 newly emerged spiderlings of/,, mactans (from egg sac
1377-A collected in Arkansas), 200 spiderlings of L. hesperus (from egg sac #1352-B
collected in California), and 128 spiderlings of/,, variolus (from egg sac # 1381 -A collected
in Missouri). From the L. mactans family, the first spiderling died two days after
emergence, the last 16 days later. The largest number ( = 54) died on the seventh day after
emergence, when a little over one-half their number had died. None had molted to the
second instar. From the /,. hesperus family, the first to die survived 8 days, the last 32 days.
The largest number ( = 28) died 19 days after emergence, when a little over one-half their
number had died. Fifteen of them had molted to the second instar, one within a day after
emergence and two 10 days after emergence. Those which attained the second instar died
from the 19th to the 24th day after emergence. From the /,. variolus family 36 died the day
after emergence, but one survived 37 days. The largest number ( = 8) to die after the second
day succumbed 19 days after emergence, when a little over half had died. Eighteen molted;
two of them two days after emergence, and two on the eleventh day after emergence. Those
which attained the second instar died from the 1 2th to the 3 1 st day after emergence. Thus it
would appear that L. variolus shows the greatest longevity and hardiness, and L. mactans
the least.
MOLTING
I often saw spiders in the act of molting, but only once was I able to observe the entire
Table 6. Maximum longevity: in days.
L.
hesperus
L.
mactans
L.
variolus
Male
Female
Male
Female
Male
Female
Number of days
post-maturity
196
952
127
849
155
822
Number of days
post emergence
369
1049
235
858
(911)*
1063
*See text for explanation
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 75
process from the beginning. The specimen was a female L. mactans #1 186, molting from
the penultimate instar to maturity. In all essentials the process closely resembled that
described for L. Hesperus by Hagstrum (1968).
The spider assumes a position with spinnerets attached to a thread overhead and all
legs fully extended, hanging from above. There is slow rhythmic up and down movement of
the body, and the carapace splits around its edges. In about five minutes the old carapace
comes off from the cephalothorax. During the next half hour the legs are extricated from
the old skin, and the abdomen likewise slowly emerges, with the old skin being pushed
toward the spinnerets. The shortest legs, II and III, emerge first, and the anterior legs last.
When the legs are all out of the old skin, the spider extends them horizontally, and holds
them in that position for a while. The entire process took about 30 minutes. The old
exuviae are cut out of the web some hours or even days later.
While ordinarily the molt to maturity is the final one, several exceptions have been
encountered, and an account has already been published by Kaston (1968), of five instances
of post-maturity molting. One additional case can be added here. A mature female L.
hesperus #1336 which was collected in San Diego, California, on February 16, 1968
molted, on March 8, 1968.
SEX RATIO
Montgomery (1908) supposed that he could sex spiderlings upon their emergence
from the egg sac. He took the newly emerged young with high, wide abdomens to be
females, those with low and narrow abdomens to be males. On this basis he obtained a
ratio of 8. 1 males to 1 .0 females. He did not rear the spiderlings to verify his prediction.
I too have observed that among the emerging spiderlings some have stout high
abdomens, and others had abdomens perhaps half as high and half as wide; but there were
also some of intermediate size and form. For one large family of/,, mactans from Florida,
# 1005-A, the shape and size was noted for each spiderling that emerged. The development
was followed, and the sex ascertained when old enough. Both males and females developed
from the spiderlings that had the large abdomens, small abdomens, and intermediate
abdomens, so no correlation could be made.
Bonnet (1938) raised a small family of L. geometricus and obtained about twice as
many males as females. However, Bouillon (1958) working with the same species, but a
much larger sample size, found a slightly greater number of females than males. His
statistical analysis indicated that the results were consistent with a ratio of 1:1. Likewise
Deevey (1949) obtained a 1:1 ratio in L. mactans; Herms et al. (1935) had found this to be
the case in L. hesperus, and McCrone and Levi (1964) Hkewise obtained this ratio for both
L. variolus and L. mactans.
In my studies the majority of specimens maturing were male. Deevey indicated that
when the spiderlings were underfed a higher percentage of males matured. This might
possibly be the explanation for some of my results with many spiderlings dying before their
sex could be ascertained. However, in about half of the families raised in all three species,
the ratios obtained were consistent with the hypothesis of a 1:1 ratio when a chi square
analysis at the 5% level of significance was made.
HABITAT AND WEB STRUCTURE
Most members of the genus Latrodectus build their webs close to the ground.
However, L. bishopi builds above ground in palmetto shrubs, and Abalos and Baez (1967)
reported their Latrodectus # 1 as never having been found less than a meter above ground.
They also indicated that L. geometricus seems to prefer human habitations, and this had
76 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
been noted as well by Smithers (1944) and by Lamoral (1968). McCrone and Levi (1964)
found L. variolus in northwestern Florida high off the ground in trees, but this species in the
more northern parts of its range (at least in Arkansas, Missouri, Illinois, Kansas,
Michigan, and Connecticut) will be found in leaf litter on the ground in mesic to xeric
deciduous forests. Fitch (1963) in Kansas, and I in Connecticut have also seen this species
under stones, and in Michigan it has been found under logs, under fence posts, and in the
holes made by small mammals in the ground (Wilson, 1967). It may well be, as suggested
by Bhatnagar and Rempel (1962) that sympatric populations "tend to differ in their
habitat, but the allopatric populations [ of L. mactans and L. variolus] may occur in
identical habitats."
The webs of L. hesperus have been reported similarly from holes of small mammals
(Jellison and Philip, 1935), in other holes of uneven ground, and along roadsides, etc., but
also at times in trash, in sheds, sometimes six or more feet above ground level, and along
the outside of houses close to the ground level. At times the density of individuals may be
quite high where suitable hiding places exist. For example, in one weedy, litter-filled lot in
Brawley, California, 100 specimens, adult or nearly so, were obtained in a couple of hours
collecting, in an area of 120 by 150 feet.
The female is negatively phototropic and generally hangs in an inverted position under
a piece of overhanging board, or clod of earth, or back in her retreat. Thus she is usually
not visible during daylight hours. But after dark the spider may move out over the snare,
taking a position perhaps several inches in front of the overhang.
Latrodectus mactans has been reported in Louisiana (Gowanlock and Leeper, 1935)
and in Maryland (Muma, 1944) from relatively dry situations in piles of stones, in culverts,
fence corners, under steps, in burrows of animals, in housings of service meters, etc. I
myself have found this species in and around human habitations, in tobacco barns in North
Carolina, and corners of rooms and basements in Georgia.
Specimens may be found throughout most of the year, though in those regions with
cold winters the spider remains inactive in a retreat under a stone, etc. Since females quite
often live more than one year, it would be expected that mature females can be found at
any season. On the other hand, males, which have a shorter life, are found mature mainly
during the warmer months. I have records of L. variolus males being taken in April and
May; of L. mactans males from May through October; and of L. hesperus males from
March through October, with the majority in August and September.
As indicated previously, the webs of black widows are of the irregular mesh type.
Nevertheless, the webs are not lacking in organization. They have been extensively studied
by Szlep (1965, 1967) and also by Lamoral (1968). In all three of our species the spider
builds a retreat, or refuge, which in nature would be under a clod of dirt or other
overhanging protective structure.
In L. hesperus the retreat has a horizontal upper border, and a curved lower border,
and leads through a tunnel an inch or so in diameter to a catching sheet. This latter is
usually horizontal, or only slightly inclined. Above the sheet is a loosely woven upper and
outer portion. Below the sheet are a number of oblique and vertical threads connected to
the substratum. These threads are of the "gum-footed" type, with viscid globules arranged
for the most part along the lowest three to five mm, but occasionally extending up as much
as 30 mm. These viscid globules are usually absent from the middle layer, the catching
sheet, and always absent from the retreat itself. While the webs are usually a foot or so
across, and equally as high, they can be larger. One web in an unused wooden shed had a
catching portion about 30 inches above the ground level, with threads extending to a retreat
in the rafters about 12 feet above the ground.
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 77
In L. mactans the retreat is globular and quite dense, and Gaul (1949) noted that it
was never in actual contact with the ground. The middle layer is not as well defined as in L.
hesperus, and there are many polygons formed by the crossing threads below the layer. The
"gum-footed" threads are lacking, or almost so, but there may be some viscid droplets on
the middle layer itself.
Lamoral (1968) indicated that he had observed "gum-footed" threads in webs of L.
mactans. But he was basing the determination of the species (in South Africa) on Levi's
having synonomized L. indistinctus. This difference in the nature of the webs could be one
of the valid reasons for maintaining L. indistinctus separate from L. mactans.
In L. variolus the retreat is dome-shaped, but larger than in L. mactans. The middle
layer is much more extensive than in the other two species, and is provided with some viscid
globules. There are fewer "gum-footed" threads extending to the substratum, than is the
case in L. hesperus.
FOOD AND FEEDING HABITS
Black widow spiders will attack and eat almost any insect that wanders into the snare.
Depending upon whether the web is relatively close to the ground, or higher up, there will
be a larger percentage of crawling or flying forms.
Shulov (1967) reported L. pallidus subsisting more on ants, while he noted that L.
tredecimguttatus consumes tenebrionid beetles, crickets, grasshoppers, and bugs.
McCrone and Levi (1964), finding that webs of L. bishopi are generally free of insect
remains, suggested that this species feeds only on soft-bodied insects. Even if such were the
case the corpses would still be adorning the web for at least a short time, since the spiders
would merely digest out the non-cuticular portions of the prey. The other alternative they
proposed would appear to be more likely; that is, the spider ejects the remains from the
web sooner than do the other species.
Occasionally, there will be found, suspended by the very strong threads of the web,
animals other than insects, and L. hasselti has been reported catching centipedes and
snakes. Roberts (1941) suppHes a photo of a female L. hasselti feeding on two lizards which
were ensnared. McKeown (1943) repeats this account, and also supplies an account, with
illustrations, of an instance where a mouse was the victim. For an itemized list of the prey
species taken by L. hesperus see Exline and Hatch (1934).
According to many reports in the literature once the male matures, he eats little, or
not at all. I did not find this always to be the case. Many males ordinarily caught and ate
prey until very near the end of their life span. It is well known that spiders generally can
endure long periods of fasting. To get some idea as to the capabiUty of black widows in
this regard, 37 femals of L. hesperus were kept without food from the day of their molting
to maturity. The one to succumb soonest died in 36 days; the hardiest lived for 193 days.
Eleven individuals lived over 100 days, and the mean longevity was 89. 3± 12.8 days.
PARASITES AND OTHER ENEMIES
A number of egg sacs collected in the field proved to contain dipterous parasites
within; except for one from Texas, they were from various California localities. This
parasite, the chloropid, Pseudogaurax signata (Loew), (Figs. 16c, d), is really an egg
predator, and has a sparse distribution (Pierce 1942). Even when a sac is parasitized the
ratio of emerging spiderlings to parasitic flies is 5:1. Details on its life history and
development were given by Kaston and Jenks (1937).
Of the hymenopters parasitizing egg sacs, Shulov (1940) reported for L. tredecimgut-
tatus a Eurytoma sp. and Abalos and Baez (1967) reported for L. mactans a eulophid. But
78 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
the hymenopterous parasite about which most is known is the sceHonid, Baeus latrodecti
Dozier, reported by Pierce (1942) from the egg sacs of L. hesperus in CaHfornia. Also,
from sacs of the latter species Herms et al. (1935) reared the egg predator Gelis sp., an
ichneumonid.
As for the predators, one notices in the laboratory occasional instances of a
mealworm, Tenebrio molitor, eating a black widow. Probably this is managed when the
spider is molting (Deevey, 1949), or else moribund, and thus helpless to defend itself.
Pierce (1942) and Branch (1943) for L. hesperus, and Archer (1947) for L. mactans, have
observed that black widow spiders are eaten by the related theridiid spider, Steatoda grossa
(C. L. Koch). Archer also noted that the pirate spider, Mimetus sp. attacked L. mactans in
Alabama, and I have seen Mimetus hesperus Chamberlin feeding on L. hesperus in
California.
Cowles (1937) considered that the San Diego alligator lizard was an effective
predator, but, as pointed out by Irving and Hinman (1935) perhaps the most effective, and
certainly the most wide-spread predator, is the blue mud-daubing wasp, Chalvbion
californicum (Saussure). Rau (1935) had observed in Missouri that this wasp preferentially
provisions its mud cells with L. mactans rather than with other spiders. D' Amour et al.
(1936) had noted the same for L. hesperus in Colorado, and I have observed the same for
L. mactans in Georgia, as well as for L. variolus in Connecticut.
ACKNOWLEDGMENTS
These studies were carried out with the aid of a National Institutes of Health grant No. GM 14623, and a
faculty grant-in-aid from the San Diego State College Foundation, both here gratefully acknowledged. I extend
thanks to those who helped me obtain live specimens from the following States and Provinces: from Alberta, P. E.
Blakely; Alabama, J. D. Unzicker; Arizona, V. D. Roth, F. E. Russell, and W. J. Gertsch; Arkansas, Maxine
Hite; Baja California, J. Y. Sandoval; British Columbia, E. Thorn, and L. C. Curtis; California, M. H. Stewart,
M. E. Thompson. F. E. Russell, and M. H. Stetson; Florida, J. F. Anderson, J. A. Beatty, and J. D. McCrone;
Georgia, J. R. Gorham and B. M. Furlow; Illinois, J. M. Nelson; Kansas, D. E. Gates; Louisiana, K. W.
McCain; Michigan, L. A. Wilson; Mississippi, L. R. Roddy; Missouri, H. E. Frizzell, and W. C. Peck; North
Carolina, J. J. Moore; New Jersey, R. C. Kern; New York, Arthur Bordes; Ohio, C. Oehler; Oklahoma, John
Taylor; Oregon, J. Anderson; Tennessee, B. C. Moulder; Texas. B. R. Vogel, and R. W. Mitchell; Virginia, R. E.
Ailstock, and J. E. Carico; Washington, Wyatt Cone; and West Verginia, W. Shear. The drawings were done
chiefly by M. H. Stewart and B. R. Burnett. I was able to see the types of L. hesperus through the kindness of Dr.
W. J. Gertsch. I owe thanks to Professor L. A. Fetzer for the translation of Russian hterature, and to Steve Sitko,
Joseph Y. Sandoval, and Bruce A. Richardson for their technical assistance.
LITERATURE CITED'
Abalos.J. W.
1962. The egg sac in the identification of species of Lalrodectus. Psyche 69:268-270.
Abalos, J. W. and E. C. Baez
1963. On spermatic transmission in spiders, /bid. 70: 197-207.
1967. The spider genus Latrodectus in Santiago dell Estero. Argentina. In, Animal Toxins, New York,
Pergamon Press, p. 59-74.
Archer. A. F.
1947. The Theridiidae or comb-footed spiders of Alabama. Alabama Mus. Nat. Hist., Paper no. 22:5-67.
Bacetti, B.. R, Dallai and F. Rosati
1970. The spermatozoon of Arthropoda. VIII. The 9 + 3 flagellum of spider sperm cells. J. Cell Biol.
44:681-682.
'A book about black widow spiders often cited is that by R. W. Thorp and W. D. Woodson. 1945,
Black Widow, America's most poisonous spider. Chapel Hill. University of North Carolina Press. 220
p. While it is true that it does give much information, all of it gleaned from the observations of others,
numerous errors and contradictions are included. The authors are entirely without scientific training,
and for various phenomena concerning black widows have supplied peculiar interpretations, some
quite teleotogical.
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 79
Baerg, W. J.
1923. The black widow: its life history and the effects of its poison. Sci. Monthly 17:535-547.
1945. The black widow and the tarantula. Trans. Connecticut Acad. Sci. 36:99-1 1 1.
1954. The brown widow and black widow spiders in Jamaica. Ann. Entomol. Soc. America 47:52-60.
1959. The black widow and five other venomous spiders in the United States. Univ. Arkansas Agr. Exp. Sta.
Bull. 608, 43 p.
Beregovoi, V. E.
1962. Contribution to the study of Latrodectus pallidus O. Cambr. subsp. pavlovskyi Charit. Zool. Zhur.
41:528-538 (in Russian).
Bertkau, P.
1875. Ueber den Generations-apparat der Araneiden. Arch. f. Naturgesch. 41:235-263.
Bhatnagar, H. D. S. and J. G. Rem pel
1962. The structure, function, and postembryonic development of the male and female copulatory organs of
the black widow spider, Latrodectus. Canadian J. Zool. 40:465-510.
Blair, A. W.
1934. Life history o{ Latrodectus mactans. Arch. Intern. Med. 54:844-850.
Bonnet, P.
1938. Elevagede Latrodectus geometricus. Bull. Soc. Hist. Nat. Toulouse 62:171-178.
1957. (Latrodectus) Bibliographia Araneorum. Toulouse, L'Imprimerie Douladoure 2:2364-2383.
Bouillon, A.
1957a. La fecondite'chez Taraignee Latrodectus geometricus C. Koch. Stud. Univ. "Lovanium" Fac. Sci.
Leopoldville. No. 1, 22 p.
1957b. Les fonctions du cocon chez I'araignee Latrodectus geometricus C. Koch. Ibid. No. 2, 30 p.
1958. La sex-ratio chez I'araigne'e Latrodectus geometricus C. Koch. Ibid. No. 3, 8 p.
Bouillon, A. and R. Lekie
1 96 1 . Cycle and rhythm in the ovulation of the spider Latrodectus geometricus. Nature 1 9 1 :620-62 1 .
Branch, J. H.
1943. On the life history and progressive factor in growth of Teulana grossa Koch. Bull. Southern California
Acad. Sci., 42:34-38.
Bristowe, W. S.
1945. (Review of "Black Widow" by Thorp and Woodson). Entomol. Monthly Mag. 81:xxxix-xl.
1946. Some notes about the American black widow spider, Latrodectus mactans F. Ibid. 82:54.
Bucherl, W.
1969. Biology and venoms of the most important South American spiders of the genera Phoneutria.
Loxosceles. Lvcosa and Latrodectus. American Zool. 9:157-159.
Burt, C. E.
1935. A review of the biology and distribution of the hourglass spider. J. Kansas Entomol. Soc, 8:1 17-130.
Chamberlin, R. V. and W. Ivie
1935. The black widow spider and its variations in the United States. Bull. Univ. Utah 25(8): Biol. Ser.
3(1):1-18.
Cowles, R. B.
1937. The San Diegan alligator lizard and the black widow spider. Science 85:99-100.
D'Amour, F. E., F. E. Becker and W. Van Riper
1936. The black widow spider. Quart. Rev. Biol. 11:123-160.
Deevey, G. B.
1949. The developmental history of Latrodectus mactans at different rates of feeding. American Midland
Nat. 42:189-219.
Deevey, G. B. and E. S. Deevey
1 945. A life table for the black widow. Trans. Connecticut Acad. Sci. 36: 1 1 5- 1 3 1 .
Derouet, L. and E. Dresco
1956. Contributions a Tetude du genre Nephila. Sur la variabilite des males de Nephila inaurata (Walck.).
Bull. Soc. Entomol. France 61:9-16.
Duges, A.
1 836. Observations sur les Araneides. Ann. Sci. Nat. Zool., ser. 2, 6: 1 59-2 1 8.
Emerton, J. H.
1902. Common Spiders of the United States. Boston, Ginn and Co., Fig. 291.
Exline, H.and M. H. Hatch
1934. Note on the food of the black widow spider. J. New York Entomol. Soc. 42:449-450.
Finlayson, M. H.
1956. Arachnidism in South Africa. In, Venoms, A.A.A.S.Publ. no. 44, (ed. by Buckley, E. E. and N.
80 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
Porges) pp. 85-87.
Fitch, H. S.
1963. Spiders of the University of Kansas Natural History Reservation and Rockefeller Experimental
Tract. Misc. Publ. Univ. Kansas Mus. Nat. Hist., No. 33:87-88.
Galiano, M. E.
1967. Ciclobiologico y desarroUo de Loxosceles laeta (Nicolet, 1849) Acta Zool. Lilloana 23:431-464.
Gaul, A. T.
1949. Habits and control of the black widow spider. J. Econ. Entomol. 42:700-701.
Gerhardt, U.
1928. Biologische Studien an Griechischen, Corsischen und Deutschen Spinnen. Z. Morph. Oekol. Tiere
10:621-626.
Gerschman, B. S. and R. D. Schiapelli
1943. Revision del genero Latrodeclus Walckenaer 1805, In: R. Sampayo "Latrodectus mactans y
Latrodectismo." Capi'tulo II (sistematica). Mus. Argentine Ciencias Nat. 23 p.
1965. El genero Latrodectus Walckenaer, 1805 (Araneae; Theridiidae) en la Argentina. Rev. Soc. Entomol.
Argentina 27:51-59.
Gonzales, L.
1954. Latrodectus mactans mexicanus subsp. nov. Ann. Inst. Biol. Mexico 24:455-457.
Gowanloch, J. N. and B. F. Leeper
1935. Report on the black widow spider. Louisiana Cons. Rev. 4(7):13-21.
Hagstrum, D. W.
1968. Molting behavior of the black widow spider Latrodectus mactans. Ann. Entomol. Soc. America
61:591-593.
Herms, W. B., S. F. Bailey and B. Maclvor
1935. The black widow spider. Univ. California Agr. Exp. Sta. Bull. no. 591, 30 pp.
Holm, A.
1940. Studien Ciber die Entwicklung und Entwicklungsbiologie der Spinnen. Zool. Bidrag, Uppsala 19:1-
214.
Hutson, R.
1936. The black widow is rare in Michigan. Michigan Agr. Exp. Sta. Quart. Bull. 18:247.
Irving, W. G. and E. H. Hinman
1935. The blue mud-dauber as a predator of the black widow spider. Science 82:395-396.
Jellison, W. L.andC. B. Philip
1935. The biology of the black widow spider, Latrodectus mactans. Science 81:71-72.
Juberthie, C.
1 954. Sur les cycles biologiques des Araignees. Bull. Soc. Hist. Nat. Toulouse 89:299-3 1 8.
Judd, W. W.
1965. The black widow spider (Latrodectus variolus Walck.) in southwestern Ontario. Ontario Field Nat.
19:24-25.
Kaston, B. J.
1937a. The distribution of black widow spiders. Science 85:74.
1937b. The black widow spider in New England. Bull. New England Mus. Nat. Hist. No. 85, 1 1 p.
1938. Check list of the spiders of Connecticut. Connecticut Geol. Nat. Hist. Surv. Bull. 60:175-201.
1948. The spiders of Connecticut. Ibid. 70:92-93.
1953. How to Know the Spiders. Dubuque, W. C. Brown Co., p. 156.
1954. Is the black widow spider invading New England? Science 1 19:192-193.
1968. Remarks on black widow spiders, with an account of some anomalies. Entomol. News 79:1 13-124.
Kaston, B. J. and G. E. Jenks
1937. Dipterous parasites of spider egg sacs. Bull. Brooklyn Entomol. Soc. 32:160-165.
Keegan, H. L.
1955. Spiders of the genus Latrodectus. American Midland Nat. 54:142-152.
Knowlton, G. F.
1935. The black widow spider. Utah Agr. Exp. Sta. Leaflet No. 57, 4 p.
Lamoral, B. H.
1968. On the nest and web structure of Latrodectus in South Africa, and some observations on body
coloration of Z.. geometricus. Ann. Natal Mus. 20:1-14.
Lawson, P. B.
1933. Notes on the life history of the hourglass spider. Ann. Entomol. Soc. America 26:568-574.
Levi, H. W.
1958. Number of species of black widow spiders. Science 127:1055.
1970 KASTON: AMERICAN BLACK WIDOW SPIDERS 81
1959. The spider genus Latrodectus. Trans. American Micr. Soc, 78:7-43.
1966. The three species oi Latrodectus found in Israel. J. Zool. (London) 150:427-432.
1969. Notes on American theridiid spiders. Psyche 76:68-73.
Lucas, S. and W. Biicherl
1965. Importancia dos orgaos sexuais na sistematica de Aranhas. Mem. Inst. Butantan 32:89-94.
Marikovskii, P. I.
1949. Contribution to the ecology of the juvenile stages of the venomous spider. Latrodectus tredecimgut-
tatus R., 1790. Zool. Zhur. 26:531-538 (in Russian).
McCrone, J. D.
1968. Biochemical differentiation of the sibling black widow spiders, Latrodectus maclans and L. variolus.
Psyche 74:212-217.
McCrone, J. D. and H. W. Levi
1964. North American widow spiders of the Latrodectus curacaviensis group. Ibid. 71:12-27.
1966. Postembryological development of spiderlings from two Peruvian Latrodectus populations. Ibid.
73:180-186.
McKeown, K. C.
1943. Vertebrates captured by Australian spiders. Proc. Roy. Zool. Soc. New South Wales p. 17-29.
Melchers, M.
1963. Zur Biologic und sum Verhalten von Cupiennius salei (Keyserling), einer amerikanischen Ctenide.
Zool. Jahrb., Abt. Syst. 91:1 -90.
1964. Zur Biologic der Vogelspinnen. Z. Morph. Oekol. Tiere 53:517-536.
Miller, I. M.
1947. Amateur research on the black widow spider. Pest Control and Sanitation 2( 1 1 ):22-33.
Millot.J.
1949. Araneae. In, Traite de Zoologie, ed. by P. Grasse. Paris, Masson et. Cie. 6:589-743.
Minton, S. A.
1950. Injuries by venomous animals in Indiana. Proc. Indiana Acad. Sci. 60:315-323.
Miyashita, K.
1968. Growth and development oi Lycosa T-insignata Boes. et Str. under different feeding conditions. App.
Entomol. Zool. (Japan) 3:81-88.
Moles, M. L.
1916. The growth and color patterns of spiders. J. Entomol. Zool. Pomona College 8(4): 129- 157.
Montgomery, T. H.
1908. Sex ratio and cocooning habits of an aranead, and the genesis of the sex ratio. J. Exp. Zool. 5:429-
452.
Muller, L.
1952. La variabilite morphologique de Coeloles atropos Walck. Bull. Soc. Nat. Luxembourg, N. S., 45:26-
35.
Muma, M. H.
1944. The black widow spider in Maryland. Univ. Maryland Ext. Bull. No. 103, 6 p.
O'Rourke, F.J.
1956. The toxicity of black widow spider venom. In, Venoms, A.A.A.S. Publ. no. 44, (ed. by Buckley, E. E.
andN.Porges)p. 89-90.
Parrott, A. W.
1946. The eyes as taxonomic characters in spiders with special reference to Uliodon piscator (Hogg). Rec.
Canterbury Mus. 5(2):95-103.
Petrunkevitch, A.
1911. A synonymic index-catalog of spiders of North, Central, and South America . . . etc. Bull American
Museum Nat. Hist. 29:181.
Pickard-Cambridge, F. O.
1902. On the spiders of the genus La/ro^ecr«j Walckenaer. Proc. Zool. Soc. London 1:247-261.
Pierce, W. D.
1942. Utilization of the black widow parasite, and further data on spiders and parasites. Bull. Southern
California Acad. Sci. 41:14-28.
Pinter, L. J.
1968. Species of widow spiders in Northern Argentina. Psyche 74:290-298.
Rau, P.
1924. Some life history notes on the black widow spider (Latrodectus mactans). Ibid. 31: 162-166.
1935. The wasp, Chalvbion cvaneum Fabr., preys upon the black widow spider, Latrodectus mactans.
Entomol. News 46:259-260.
82 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Reese, A. M.
1940. Variations in the markings of the black widow spider. J. Hered. 33: 1 18.
Roberts, N. L.
1941. Some notes on Australian Spiders. Proc. Roy. Zool. Soc. New South Wales p. 36-41.
Robinson, M.
1947. A new food supply for Latrodectus maclans. Entomol. News 58:258.
Roewer, C. F.
1942. (Latrodectus). Katalogder Araneae. Bremen 1:424-429.
Semans, F. M.
1941. Black widow spider distribution in Ohio. Ohio J. Sci. 41:380.
Shuiov, .\.
1 940. On the biology of two Latrodectus spiders in Palestine. Proc. Linn. Soc. London 1 52:309-328.
1967. Biology and ecology of venomous animals in Israel. Mem. Inst. Butantan. 33:93-99.
Shuiov, A. and \. Weissman
1959. Notes on the life habits and potency of the venom of the three Latrodectus spider species in Israel.
Ecology 40:5 15-5 1 8.
Smithers, R. H. N.
1944. Contribution to our knowledge of the genus Latrodectus in South Africa. Ann. So. African Mus.
36:263-312.
Szlep, R.
1965. Web-spinning process and web-structure of Latrodectus tredecimgultatus. L. pallidus. and L.
revivensis. Proc. Zool. Soc. London 145:75-89.
1967. The web structure o^ Latrodectus variolus Walck. and L. bishopi Kaston. Israel J. Zool. (for 1966)
15:89-94.
Thorn, E.
1967. Preliminary distributional list of the spiders of British Columbia. British Columbia Prov. Mus. Ann.
Rept. (for 1966) p. 23-39.
Wiehle, H.
1967. Steckengebliebene Emboli in den Vulven von Spinnen. Senckenb. Biol. 48: 197-202.
Wilson, L. F.
1967. The northern widow spider, Latrodectus variolus. in Michigan. Michigan Entomol. 1:147-153.
Witt, P. N.andC. F. Reed
1965. Spider-web building. Science 149:1190-1197.
Zoology Department, San Diego State College. San Diego. California 921 15
MUS. COMP. 2:OQL.
LIBRARY
WAR 121971
HARVARD
UNIVERSITY
EASTERN PACIFIC CROWN-OF-THORNS
STARFISH POPULATIONS
IN THE LOWER GULF OF CALIFORNIA
THOMAS DANA AND ARTHUR WOLFSON
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 4 24 NOVEMBER 1970
EASTERN PACIFIC CROWN -OF-THORNS
STARFISH POPULATIONS
IN THE LOWER GULF OF CALIFORNIA
THOMAS DANA AND ARTHUR WOLFSON
ABSTRACT. — Populations of Acanthaster ellisii (Gray) were investigated on three islands in
the southern Gulf of California. Average density (0.0045/m' or 1/225 m") exceeded that given
in several definitions of normal densities for A. planci populations in the Indo-Pacific. Small
patches of Porites were the most frequent food item; other hermatypic scleractinians, gorgon-
ians, and algae were also fed upon. Estimates of coral coverage and growth rates, and Acan-
tluister feeding rates, indicate that /I rw/u/iai^rcr predation is a significant source of coral mortality
but that corals are not being eliminated from the areas studied. Gonad analysis suggests an ex-
tended spawning season rather than a short synchronous one. Size-frequency data do not neces-
sarily lead to the conclusion that populations of Acanthaster are expanding.
RESUMEN. — Se estudiaron las poblaciones de Acanthaster ellisii (Gray) en tres islas de la zona
meridional del Golfo de California. La densidad de dichas poblaciones presentaba un promedio
de 0.0045 por m^, o sea de 1 por 225 m\ que vienen a ser concentraciones mas elevadas que las
consideradas normales para. A. planci en el Incio-Pacifico. Pequenas agrupaciones de Porites con-
stituyen el alimento mas frecuente de estos equinodermos, aunque tambien se observe que se ali-
mentan de otras escleroactinias hermatipicas. gorgonias y algas. Las determinaciones sobre la
cobertura de corales y los valores de crecimiento, asi como los datos relacionados con la alimen-
tacion de Acanthaster indican que la predacion de este equinodermo es uno de las causes princi-
pales en la mortalidad del coral, aun cuando los corales no aparecian exterminados en las zonas
estudiadas. El analisis de las gonadas sugiere que la epoca de puesta no es corta y sincronica,
sino prolongada. Datos sobre la frecuencia de tallas no indican. al parecer, que las poblaciones de
Acanthaster amplien su area de dominancia.
INTRODUCTION
The presence of conspicuous populations of the eastern Pacific Crown-of-Thorns
starfish Acanthaster ellisii (Gray) on three islands just north of La Paz, Baja California,
Mexico, was recently brought to our attention. In the central and western Pacific in areas
of luxuriant reef development the closely related starfish Acanthaster planci (Linnaeus) is
reportedly undergoing population explosions (Barnes 1966; Weber 1969; Chesher 1969,
1970). Reputed consequences of these "infestations" range from economic disaster for
small isles and atolls of Oceania, destruction of fisheries upon which the inhabitants of
Oceania depend for almost all their protein, severe land erosion by storm waves, to the
extinction of madreporarian corals in the Pacific (Chesher 1969). More recently the
assertions that Acanthaster aggregations represent a massive environmental upheaval,
which seems to have no recorded precedent, have been challenged (Newman 1970; Weber
and Woodhead 1970; Dana 1970). However, since no complex coral reef structures
comparable to those of the Indo-Pacific are to be found in the Gulf of California, the
presence of populations of A. ellisii apparently exceeding densities given as normal by
Chesher (1969) for A. planci posed intriguing questions as to the ecological relationships
between eastern Pacific corals, A. ellisii. and reef formation. Goreau (1964) has even
suggested that under certain conditions Acanthaster might be an important factor limiting
the growth and development of coral reefs. This prompted a short but intensive survey of
SAN DIEGO SOC. NAT. HIST.. TRANS. 16(4): 83-90, 24 NOVEMBER 1970
84
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
111°
10°
ISLA
SAN JOSE
S(o(.on -Cjfl SAN FRANCISCO
1-5 '♦
O
25°
24°
Figure 1 . Map of study area in lower Gulf of California, Mexico showing station locations.
Isla San Francisco (24°55'N, 110°35'W) on 23 and 25 April, Isla San Jose (24°55'N,
1 10°35'W) on 24 April, and Isla del Espiritu Santo (24°35'N, 1 10°25'W) on 26, 27, and 28
April, 1 970, to investigate various aspects of the ecology of those populations (Figure 1 ).
METHODS
Area, depth, and per cent coral coverage of all surveyed areas were estimated. Usually
1970 DANA AND WOLFSON: Acanthaster 85
in making surveys two divers were towed at slow speeds ( 1 to 2 knots), one on each side of a
12' skiff. In early stages of the survey all A. ellisii located were investigated for active
feeding; later only occasional individuals were checked. Frequently, when visibility and the
width of suitable substrate prohibited a thorough survey by towing, free or SCUBA diving
was employed to more completely cover the area. At several stations both day and night
observations were made. Only diving was utilized for night surveys, during which special
emphasis was placed on locating juvenile A. ellisii (none were found). Specimens were
collected from selected areas and individuals were kept in a large opaque aquarium on
board ship. A variety of living corals were presented to these specimens. All the
Acanthaster collected were measured (disk diameter) and gonad samples taken.
RESULTS
Isla San Francisco. — The area adjoining nearly the entire western half of the island
was surveyed. Detailed observations were, however, limited to the southwestern sector.
Station 1 was the submerged portion of a spit composed of small boulders (<0.5m in
diameter) at the southern end of a small sandy embayment. The area investigated
measured some 120x10m, with depths ranging from 0.5 to 2m. All observations were
made while snorkeling. Coral coverage was 2 to 3%, consisting of small patches of Porites
(3-6 cm in diameter) and scattered individual heads of Pocillopora. Seven A. ellisii, all in
the open, were located; most were feeding on small Porites patches during the day. There
was evidence of occasional feeding on Pocillopora, but none of these coral heads were
completely eaten. (Density of A. ellisii: 0.006/m- or l/171m-.)
Station 2 was located along a rocky shoreline across the sandy embayment from
Station 1 and included the point at the northwestern end of the bay. The substrate
consisted of large boulders (>lm in diameter) that had tumbled down onto a flat sandy
bottom. These boulders were almost completely covered with algae. The area surveyed
stretched for about 315m along the shore and varied in width from 8m on the inner end to
15m at the northwest point. Depth of the water to sand bottom gradually increased from 5
to 15m at the point. Day observations were made towing, free diving, and with SCUBA.
Coral coverage was estimated to be less than 1 %, except at the point where it was between
2 and 3%. Small encrusting patches of Porites and small heads of Pocillopora were
present. Several larger heads of Pocillopora and patches of Porites (>30 cm in diameter)
were found in shallow water at the point. A total of 27 A. ellisii (including 7 taken by
Faulkner on 18 April) were scattered throughout the area. Nearly all were in water
between 1 and 5m deep and were feeding on patches of Porites. A single individual which
was not feeding was found in 12m of water at the northwest point. (Density of A. ellisii:
0.009/m-or 1/1 17m'.)
Station 3 began on the north side of the point where Station 2 terminated and
continued for some 200m into an adjacent cove. The substrate was similar to that of
Station 2 except the boulders were smaller and less algal covered. The rocky area was 5 to
8m wide, ending on a smooth sand bottom in 3 to 4m of water. Coral coverage was
estimated to be less than 1%. Four Acanthaster were seen but were not checked for
feeding. (Density of^. e//m/; 0.003/ m' or l/400m-.)
Station 4 was located on the north side of a small cove opposite Station 3. The
substrate along the inner 75m of the cove was a grey, vesicular basalt dipping seaward
gently for about 18m to a depth of 3 to 4m, and then more sharply to a smooth sand
bottom at a depth of 9m. This area was surveyed by towing and free diving during the day.
The bottom over the remaining 3 10m length of the area consisted of large boulders and was
surveyed by day towing and a SCUBA dive at night. The entire station was densely covered
86 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
with algae. Coral cover was between 1 and 2% and consisted of small patches of Porites
and occasional small heads of Pocillopora. A total of 24 Acanthaster (including 12
collected by Faulkner on 19 April) were found at this station. Most individuals observed
during the day were feeding in the open on small patches of Porites. Identical behavior was
observed during the night dive with the additional observations of one completely cleaned
colony o'( Pocillopora and a single Acanthaster feeding on a gorgonian {Pacifigorgia sp.)
in a crevice. (Density of A. ellisii: 0.006/ m- or l/160m-.)
Station 5, located along the southern side of the cove adjacent to Station 4, comprised
an area of 300 x 10m. The bottom was an algal-covered sloping rock outcrop with a few
boulders scattered at its seaward extremity. Sand replaced the rocky substrate in about 3m
of water. Observations were made by towing and free diving in daylight. Coral coverage
was 8 to 10%, consisting principally of encrusting to submassive patches of Porites, a few
heads of Pocillopora, and an occasional patch of Pavona. (Density of ^. ellisii: 0.002/m-
or l/600m-.)
Isla San Jose. — A single station (Station 6) was made at this island — around a
linear rock outcrop well out into the mouth of the large bay on the southwestern extremity
of the island. The substrate consisted of large algal covered boulders, and water depth to
sand bottom ranged from 6m at the northern end of the outcrop to 14m at the southern
end. The area surveyed was about 375m long and 10 to 12m wide. Corals present were
Porites, Pocillopora. and Tubastrea. and cover for most of the area was about 1%,
increasing slightly at the southern tip where a strong current prevailed. Observations were
made towing and with SCUBA. During the day 5 Acanthaster were found scattered along
the western side of the outcrop, all in less than 4m of water. Three were in the open, fully
exposed but not feeding, another was feeding on a gorgonian {Pacifigorgia sp.), and one
had its stomach everted over a clump of Padina sp. (a lightly calcareous brown alga). All of
these specimens were collected and no additional individuals were located that night.
(Density of/4, ellisii: 0.00\ /m~ or l/750m-.)
Isla del Espiritu Santo. — Two stations were occupied in the northwestern sector of
the mouth of an embayment on the western side of the isthmus. The first. Station 7, was
around a small rock outcrop a short distance out into the bay. The surveyed area was
approximately 500m-. All observations were made snorkeling during the day. Algal cover
was much sparser than at previous stations and coral coverage was between 4 and 5%.
Small patches of Porites were present, a single clump of Psammocora (Stephanaria) was
noted, and a number of colonies of Pocillopora (up to 0.75m in diameter) were scattered
about. Eight Acanthaster were observed in 1 to 3m of water. (Faulkner also collected one
individual from this locality on 15 April.) Of the 8, 7 were feeding on tiny patches of
Porites, and one was under a large head of Pocillopora — a small portion of which had
been eaten. (Density of A. ellisii: 0.0\6/m~ or l/63m-.)
A small point opposite Station 7 was selected for Station 8. An estimated 1 100m of
rock outcrops and boulders were investigated by free diving. Algal and coral coverage, as
well as the kinds of corals, were similar to Station 7. Of the 8 Acanthaster seen during the
day, 6 were feeding on Porites and one on Psammocora (Stephanaria). (Density of A.
ellisii: 0.001 /m-OT l/138m'.)
Two stations were made at the northwestern extremity of Bahia de San Gabriel,
located in the southwestern sector of Isla del Espiritu Santo. The first. Station 9, was
located just outside and to the north of the bay at Punta Prieta and covered approximately
600m- of rock ledges and boulders in water less than 5m deep. Coral coverage was between
3 and 4%, principally Pocillopora, and under ledges, Tubastrea. No Acanthaster were
found during the day or night.
1970 DANA AND WOLFSON: Acanlhaster 87
Station 10 was located just inside the bay where a fringing reef is forming in shallow
water. Coral growth terminated on a sand bottom in less than 2.5m of water. Squires
(1959) described a series of coral patches aligned as a barrier across the central portion of
this same bay; however, that area was not investigated. Four species of Pocillopora were
the principal reef builders with occasional scatterings of Pavona, Psammocora (Step-
hanaria), and Pontes. At one end of the reef structure there was an extensive patch of
Pontes in very shallow water. Approximately 1500m'^ were thoroughly searched by
snorkeling during the day. A single specimen o{ A. ellisii, the largest located during the
survey, was found under a large head of Pocillopora that had a freshly killed portion
comparable in size to the disk area of the starfish. No other Acanthaster were found at this
station. However, occasional small white patches were noted on branch tips o{ Pocillopora
clumps. Closer examination revealed that the regular five-armed sea star Pharia pyrami-
data (Gray) was everting its stomach in a manner similar to Acanthaster and removing
coral tissue. Steinbeck and Ricketts (1941) reported Pharia to be common in coral areas in
the Gulf of California, but our observation is the first to indicate that they feed on coral.
Thirty specimens o{ A. ellisii were collected. Disk diameters ranged from 62mm to
142mm with a mean of 97.9mm (Figure 2). No juveniles were found. All of these individuals
fit within the size range of specimens available to Case (1962), although our mean is slightly
greater.
Gonad samples taken from 14 males and 12 females were analyzed by Dr. John S.
Pearse of the Kerckhoflf Marine Laboratory. Eighteen individuals were ripe, including the
largest and smallest collected — both females. Numerous mature spermatozoa and a thick
layer of spermatogenic cells in the 1 1 ripe males, and the presence of various-sized,
growing oocytes alongside abundant, fully developed ones in the 7 ripe females, suggests
that gametes are produced over a considerable period of time, or that the samples were
taken during the peak of reproductivity. Four females contained several sizes of maturing
oocytes but few full-grown ones. One female had recently spawned and appeared to be
beginning a new cycle of gametogenesis. Three males were not ripe but were either
maturing or perhaps had recently spawned and were beginning a new cycle of gametogen-
esis.
DISCUSSION
The behavior o{ A. ellisii differed from that described for A. planci by Goreau (1964)
and Chesher (1969). Rather than hiding by day and feeding at night, A. ellisii was almost
always conspicuously out in the open, and usually feeding, during the day. All A. ellisii but
one were seen in water shallower than 4m. Their limited distribution was undoubtedly
related to the narrow distributional limits of suitable food organisms. There was no
apparent clumping oi Acanthaster on a scale of a few tens of square meters.
Hermatypic scleractinian corals appeared to be the preferred food item for A. ellisii —
particularly small encrusting patches of Porites estimated to be no more than 2 years old.
Feeding experiments tended to support this observation. Goreau (1964) noted that in the
southern Red Sea A. planci selected smaller coral heads more frequently than larger ones.
There was no field evidence that Acanthaster feeds on the ahermatypic coral Tubas trea,
and this coral was avoided in feeding experiments. However gorgonians of the genus
Pacifigorgia were fed on occasionally, and one Acanthaster was seen in normal feeding
attitude on a clump of the alga Padina.
All areas surveyed except the northwestern portion of Isla San Francisco were in the
lee of the prevailing northwesterlies (November to May) and southeasterlies (June to
October) (Roden, 1964). Such normally sheltered locations have been reported to support
88 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
larger average Acanthaster population levels (Chesher, 1970; Weber, 1970).
Ninety Acanthaster (including the 20 taken by Faulkner) were located within the
20,250m- surveyed in detail. The average density obtained, 0.0045/m- or about l/225m-,
exceeds several of Chesher's definitions of normal population densities for A. planci: 2 or
3/ 1000m-, 4 or 5/km of reef, 1/hour of search, and 20/20 minutes of search but usually no
more than 8 (Chesher, 1969, 1970). For several stations densities approached, and at one
station exceeded, the density reported for the infestation of Double Reef, Guam (886
animals on 90,000m- of reef, Chesher, 1969). The effect of the starfish on coral formations
in the lower Gulf of California is certainly problematical, especially considering the
general lack of reef development and sparse distribution of corals in that area (Squires,
1959). Excluding the 2 stations from Bahia de San Gabriel, coral coverage, in terms of
projected images of individual colonies, averaged about 3% over 18,150m-. There are then
approximately 6.1m- of coral standing crop available for each Acanthaster, although the
actual feeding surface is certainly greater. Using a consumption rate of twice the area of
the disk per day (Chesher, 1969, for A. planci), approximately 5.3m- of coral would be
consumed by an average size A. ellisii in a year's time. Such a feeding rate would require a
replacement rate of coral standing crop — in terms of areal coverage — of 87% annually.
However, this feeding rate, considering the effects of temperature differences on metabolic
rate (Kinne, 1963), is probably an overestimate (surface temperatures in the lower Gulf of
California range from 17 to 31°C with an annual mean of 24.7°C, while the tropical
western Pacific remains nearly uniform at 28°C; see Roden, 1964).
The relationship between increase in weight and increase in area of the projected
image of a coral colony is difficult to estimate and depends in a complex manner on such
factors as growth form, degree and mode of branching, and skeletal density. Nevertheless,
growth data giving annual increments of increase as per cent gain in weight does not seem
an unreasonable means of approximating a coral replacement rate. In Hawaii, Edmondson
( 1 929) found an average annual weight increase for a number of colonies of various sizes of
two species of Porites to be 60.7% and 90.4% and of three species of Pocillopora to be
148.0%, 137.5%, and 103.9%. Since the Hawaiian Islands are on the border of the tropics,
coral growth data from there seem appropriate for comparison, even though Hawaiian
growth rates certainly exceed those in the lower Gulf of California. Despite the com-
plications, it appears that under present conditions coral growth alone should be sufficient
to provide enough tissue to satisfy the energetic requirements of current population levels
of A. ellisii.
The gonad analysis indicates that in the Gulf of California Acanthaster has at least a
protracted, if not continuous, spawning season. This does not agree with the report from
Green Island (about 16°S) on the Great Barrier Reef of a highly synchronous breeding
season in December and January for A. planci (Endean, 1969), nor with the contention of
Chesher ( 1 969) of a breeding season for A . planci at Guam (about 1 6°N) during November
and December. Our data, however, agree with analyses by Pearse on specimens from
Guadalcanal, Guam, Ifaluk, and Wolei (Eldredge, 1970), and with Mortensen's observa-
tion (1931) from off Java (about 6°S) that, for A. planci, the sexual products are not shed
all at once but in portions at different times. Furthermore, continual influx of young, or
recruitment extending over many months, could account for the lack of modes represent-
ing year classes in the size-frequency distribution of the populations of A. ellisii observed.
Since no growth rate data are available for A. ellisii, age structure of the populations
cannot be inferred from their size distribution. However, one important point about the
shape of the size-frequency curve, as it relates to population increases, should be made: the
peak at intermediate sizes (see Figure 2) does not necessarily indicate an unusually large
1970
DANA AND WOLFSON: Acanthaster
89
12
1 1
-
10
-
9
-
8
-
7
-
6
5
4
3
-
2
1
' 61-7o' 71-eo' 81-9o' 91-IOOM01-Ito')n-I2ol|21-130M31-I4o'l4t-I5o'
SIZE (mm)
Figure 2.
mm).
Size frequency histogram based on 30 specimens. Range, 62-142 mm; mean, 97.9 mm (median, 95.5
recent influx of young. Several combinations of survivorship curves coupled with non-
linear growth could give size-frequency distribution curves of the shape observed even
when annual recruitment is relatively constant over a period of several years. Probably the
populations observed contain individuals in several year classes, and any contention for a
recent population increase would be highly speculative.
Lacking adequate knowledge of recruitment, settling requirements, survivorship,
spawning periods and behavior, growth rates, rates of mortality from various sources, and
longevity of both corals and Acanthaster, as well as information concerning past popu-
lation levels and fluctuations, we consider drawing any conclusions as to the consequences
of present levels of predation on corals in the lower Gulf of California by A. ellisii tenuous
at best. However, the feeding pressure exerted by A. ellisii, when coupled with suboptimal
temperatures for corals resulting in relatively slow growth rates, an observed abundance of
boring organisms, and paucity of coralline algae to serve as a binding agent, may
contribute significantly to the almost total absence of reef formation in the Gulf of
California.
ACKNOWLEDGMENTS
We would like to thank Dr. D. John Faulkner for bringing the situation in the Gulf to our attention. Garth
Nicholson for help with the field work. Dr. John S. Pearse for analysis of the gonad samples. Dr. William A.
Newman for encouragement and critical reading of the manuscript, and the Foundation for Ocean Research for
travel and ship time.
LITERATURE CITED
Barnes, J. H.
1966. The crown of thorns starfish as a destroyer of coral. Australian Mus. Mag. 15: 257-261.
Bayer, F. M.
1951. A revision of the nomenclature of the Gorgoniidae (Coelenterata: Octocorallia), with an illustrated
key to the genera. Wash. Acad. Sci. 41: 91-102.
90 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
Caso, M. E.
1962. Estudios sobre Asteridos de Mexico: Observaciones sobre especies Paciticos del genero Acanthaster y
descripcion de una subespecie nueva, Acanthaster ellisii pseudoplanci. Ann. Inst. Biol. Univ. Mexico
32:313-331.
Chesher, R. H.
1969. Destruction of Pacific corals by the sea star /lca«r/ia.y/er/7/a«c/'. Science. 165:280-283.
1970. Acanthaster planci: Impact on Pacific coral reefs. U.S. Dept. Interior. Publ. 187631.
Dana, T. F.
1970. Acanthaster: a rarity in the past? Sciences. 169:894.
Edmondson, C. H.
1929. Growth of Hawaiian corals. Bernice P. Bishop Mus., Bull. 58: 1-38.
Eldredge. L. G.
1970. Acanthaster Newsletter, no. 2, from the Marine Laboratory, Univ. of Guam, Agana, Guam.
Endean, R.
1969. Report on investigations made into aspects of the current Acanthaster planci (Crown-of-thorns)
infestations of certain reefs of the Great Barrier Reef. Brisbane, Queensland Dept. of Primary
Industries, Fisheries Branch.
Goreau, T. F.
1964. On the predation of coral by the spiny starfish Acanthaster planci (L.) in the southern Red Sea. Israel
South Red Sea Expedition, 1962. Rept. no. 2, Bull. Sea Fish. Res. Sta., Haifa 35: 23-26.
Kinne, O.
1963. The effects of temperature and salinity on marine and brackish water animals. Oceanogr. Mar. Biol.
Ann. Rev. 1:301-340.
Mortensen. Th.
1931. Contributions to the study of the development and larval forms of echinoderms: I-II. D. Kgl. Danslce
Vidensk. Selsk. Skrifter, naturvidensk. og mathem. Afd., 9 Raekke, IV 1 : 1-39.
Newman, W. A.
1970. /lca«?/ia.?rfr.- a disaster? Science 167: 1274-1275.
Roden, G. I.
1964. Oceanographic aspects of the Gulf of California. In Marine Geology of the Gulf of California, A
Symposium. Memoir 3. Amer. Assoc. Petrol. Geol. Edited by T. H. van Andel and G. G. Shor, Jr.
Squires, D. F.
1959. Results of the Puritan-American Museum of Natural History Expedition to Western Mexico: Corals
and coral reefs in the Gulf of California. Bull. Amer. Mus. Nat. Hist. 118: 371-431.
Steinbeck, J., and E. F. Ricketts
1941. The Sea of Cortez, a Leisurely Journal of Travel and Research, with a Scientific Appendix
Comprising Materials for a Source Book on the Marine Animals of the Panamic Faunal Province.
New York, Viking Press.
Weber, J.N.
1969. Disaster at Green Island — other Pacific islands may share its fate. Earth and Mineral Sci. 38: 37-41.
Weber, J. N., and P. M. J. Woodhead
1970. Ecological studies of the coral predator Acanthaster planci in the South Pacific. Marine Biology 6: 12-
17.
Scripps Institution of Oceanography, University of California-San Diego, La Jolla,
California 92037
^ MUS. COMR ZOOL.
^ LIBRARY ,
MAR 121971
HARV"
UNlVEKoi 1 ii
EVOLUTION OF PEROMYSCUS ON NORTHERN ISLANDS
IN THE GULF OF CALIFORNIA, MEXICO
TIMOTHY E. LAWLOR
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 5 24 FEBRUARY 1971
EVOLUTION OF PEROMYSCUS ON NORTHERN ISLANDS
IN THE GULF OF CALIFORNIA, MEXICO.
TIMOTHY E. LAWLOR
ABSTRACT. — Mice of the genus Peromyscus on northern islands of the Gulf of California and
adjacent mainland areas were examined to trace the divergence of populations there. A total of
27 qualitative characters of the osteology, pelage, phallic morphology, soft anatomy, serology,
and karyology was examined in detail. Morphometric characters and dental patterns also were
studied, and matings of pertinent forms were attempted with limited success.
The island and mainland forms were treated numerically according to the above qualitative
characters, as follows: ( 1) Two Prim Networks were computed, utilizing different combinations of
characters. Each indicated that P. stephani (Isla San Esteban) is closely related to P. boylei, and
that those two species and P. crinitiis are only distantly related to the remainder of the island
and mainland forms. (2) A dendogram (Wagner Diagram) was computed for the latter, using
the quantitative phyletic method. P. eretuicus was considered ancestral on morphologic and
zoogeographic grounds. P. giiardia (Islas Angel de la Guarda, Granito, and Mejia) is the most
divergent of the eremicus-Wkc forms and cladistically is closest to P. merriami. P. interparietalis
(Islas San Lorenzo Sur, San Lorenzo Norte, and Salsipuedes) also is relatively far removed from
the hypothetical eremicus-\'\V.Q ancestor. Populations from the Baja Californian and Sonoran main-
lands and Isla Tiburon (P. eremicus), and Isla Turner {P. collotus), are closely related and
should be considered conspecific. On zoogeographic grounds, the populations on western Gulf
islands (giiardia, interparietalis) probably are derived from a Baja Californian eremiciis-hke
progenitor, whereas eastern island forms (collatiis, eremicus tihiironensis) and stephani prob-
ably are derived from Sonoran eremicns-hke and boylei-Uke forms, respectively. Evidence from
morphology, amount of gene flow between islands and between islands and the mainland, and
time of formation of the islands, suggests that the time interval since initial formation of the
islands has been the principal factor affecting divergence of the island populations.
Trends in the evolution of certain characters among Gulf Peromyscus suggest that complex
features may result from simple conditions in the phallus and dentition, and that acrocentric chro-
mosomes derive from a bi-armed condition. The data suggest that the subgenus Haplomylomys,
which consists of eremicus-Uke species, contains primitive members of the genus.
RESUMEN. — Se estudiaron los ratones del genero Peromyscus en las islas septentrionales del
Golfo de California y zonas adjacentes del continente, con objeto de determinar las divergencias
que presentan sus poblaciones. Se examinaron con todo detalle un total de 27 caracteres morfo-
logicos, relacionados con la osteologia, pelaje, organos sexuales externos y otros caracteres ana-
tomicos, serolologicos y citologicos. Tambien se analizaron los caracteres morfometricos y la
denticion, intentandose ademas cruces entre las formas pertinentes, obteniendo exitos muy limitados.
Las formas encontradas en las islas del Golfo de California y en el continente se analizaron
numericamente en cuanto a los caracteres morfologicos arriba mencionados, en la forma siguiente:
1 ) Se efectuaron dos "Prim Networks," utilizando diferentes combinaciones de caracteres. En
cada caso resulto que P. stepliani (isla de San Esteban) aparecia como pariente proximo de
P. boylei, y estas dos especies con P. crinitiis resultan parientes lejanos de las formas restantes
que habitan estas islas y el continente. 2) El diagrama denditico (diagrama de Wagner) se com-
pute para P. crinitiis, utilizando el metodo filogenetico cuantitativo. P. eremicus aparece asi
como una especie ancestral, basandonos en la morfologia y la zoogeografia. P. guardia (islas
Angel de la Guarda, Granito y Mejia) es la especie que diverge mas de las formas del tipo
eremicus, y la mas proxima en la escala a P. merriami. Peromyscus interparietalis (Islas de San
Lorenzo Sur, San Lorenzo Norte y Salsipuedes) aparece como una segregacion lejana del ascen-
diente hipotetico tipo eremicus. Las poblaciones de P. eremicus de las zonas continentales de
Baja California, Sonora y de la isla Tiburon, y las de P. collatus de la isla Turner aparecen muy
SAN DIEGO SOC. NAT. HIST, TRANS. 16(5): 91-1 24, 24 FEBRUARY 1971
92 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
relacionadas entre si, por lo que podrian considerarse co-especificas. Bajo el punto de vista
zoogeografico, las poblaciones de las islas occidentales del Golfo (guardia, interparietalis) derivan
probablemente de un progenitor tipo eremiciis de Baja California; mientras que es posible que
las formas de las islas orientales (collatus, eremicus, tihuronensis) y stephani procedan respec-
tivamente del tipo eremicus y del tipo boylei. Las caracteristicas morfologicas evidencian un flujo
importante de genes de unas islas a otras y entre estas y tierra firme durante el perido de forma-
cion de dichas islas; lo cual sugiere que el lapso de tiempo transcurrido desde que se inicio la
formacion de esas islas constituye el factor principal responsable de la divergencia encontrada en
las poblaciones insulares.
La tendencia o curso en la evolucion de ciertos caracteres en los Peromyscus del Golfo sugiere
que estructuras complejas pueden resultar simplemente de las condiciones de los organos sexuales
externos y la denticion, y que los cromosomas acrocentricos derivan de una condicion bifurcada.
Los datos obtenidos indican que el subgenero Haplomylomys, que incluye las especies tipo
eremicus, contiene los miembros primitives del genero.
The ecologic and geographic characteristics of islands make them particularly suited
for studies concerning differentiation and adaptation in natural populations. In general,
climatic stability, decreased ecologic diversity, and increased isolation distinguish
islands from mainland areas. In the sense of Preston (1962), the plant and animal
populations of an island form a complete ("canonicaT') system as a result of these
peculiarities, while mainland populations represent only a "sample" of a more widely
distributed and more diverse biota. Thus, effects of isolation are more pronounced on
islands than on continental areas.
In addition, islands presumably are subject to colonization by organisms undergoing
primary radiation on continental areas. This seems like a reasonable assumption, although
the reverse situation undoubtedly occurs to a lesser degree. Consequently, insular popu-
lations may constitute unique control groups in which to examine patterns of evolution and
divergence of particular groups of organisms.
Mice of the genus Peromyscus are widely distributed in North America in insular and
mainland situations. They are nearly ubiquitous on island and mainland areas in and
surrounding the Gulf of California. No less than 18 species of two subgenera are
recognized there, of which ten are island endemics. It appears that at least five of the seven
non-endemic species were important for radiation of the group onto the islands of the Gulf.
The following account is an assessment of the morphologic, serologic, and karyologic
divergence of the island populations o{ Peromyscus relative to one another and to those on
the mainland of Baja California to the west and Sonora, Mexico, to the east.
The geographic area of study consists of the northern group of Gulf islands (Figs. 1
and 2). These islands form an irregular chain from one side of the Gulf to the other, thus
affording several possible access routes to and from the mainland. In addition, the cham is
separable into deep- and shallow-water islands. The latter group consists of islands
(Turner, Tiburon) occupying waters within the 1 10 meter depth contour, the level to which
the sea is thought lo have been lowered by eustatic changes during the Pleistocene, whereas
the former group (San Esteban, Salsipuedes, the Lorenzos, and Angel de la Guarda and
nearby islands) consists of islands that attained their present configuration as long ago as
Pliocene (Anderson, 1950). Thus, certain of the islands are chronologically much younger
than others by virtue of their relatively recent separation from the mainland.
One would expect a greater degree of morphological and genetic differentiation in
peromyscines inhabiting distant and deep-water islands as a result of more effective
isolation than in those mice on islands in close proximity to the mainland and in shallow
water. The latter islands could be subjected to repeated invasions by mice from mainland
populations, resulting in suppression of morphological or genetic differences that might
1971
LAWLOR: Peromyscus
93
Figure 1 . Map of the northern part of the Gulf of California, Mexico, and surrounding areas. The area enclosed
in dotted lines is enlarged and presented in detail in Fig. 2. Numbers identify localities discussed in text and
specified in "Specimens Examined," and are as follows: 1 — Turtle Bay: 2 Barril: 3 — Bahfa de los Angeles; 4
— San Francisquito; 5 — El Marmol; 6 — San Telmo: 7 — Escondido: 8 — Punta Penasco; 9 — Tucson; 10 —
Imuris; i 1 — Puerto Libertad; 12 — Punta Sargento; 13 — Bahia Kfno; 14 — Isla San Pedro Nolasco; 15 —
Presa Obregon.
Otherwise have arisen. However, two other alternatives seem plausible: ( 1) the environment
on the proximal, shallow-water islands may more closely resemble that on the mainland,
and (2) the time interval since the initial formation of the shallow-water islands may not
have been sufficiently long for a large amount of differentiation to have taken place. The
amount and trends of variation in the mice from the different islands and the two mainland
areas provide sufficient data for determining which of the above factors is relevant.
The affinities of the species of Peromyscus on the northern islands have not been
94
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Sonora
. GRANITO
I. ANGEL DE LA GUARDA
PUNTA SARGENTO
BAHIA DE LOS
ANGELES
a
I. PARTIDA
\. RAZA
\. TIBURON
BAHIA KINO
SALSIf^UEDES* j— .
I. SAN LORENZO NORTe'V^ J-^
I. SAN LORENZO SUR
Baja California
0|. TURNER
I. SAN ESTEBAN
Figure 2. Map of the northern islands of the Gulf of California, Mexico, and adjacent mainland areas enlarged
from the insert in Fig. 1 .
thoroughly documented. Currently populations of five species are recognized, all of which
are included in the subgenus Haplomylomys and are considered closely related to P.
eremicus:
P. eremicus tiburonensis (Mearns, 1897); Isla Tiburon
P. collatus Burt, 1932; Isla Turner (= I. Datil)
P. stephani Tovjnsend, 1912; Isla San Esteban
P. guardia guardia Townsend, 1912; Isla Angel de la Guarda
P. guardia mejiae Burt, 1932; Isla Mejia
P. guardia harbisoni Banks, 1967; Isla Granito
P. interparietalis interparietalis Burt, 1932; Isla San Lorenzo Sur
P. interparietalis lorenzo Banks, 1967; Isla San Lorenzo Norte
P. interparietalis ryckmani Banks, 1967; Isla Salsipuedes
Taxonomically these forms have remained virtually unchanged since their description,
except that tiburonensis is now considered a subspecies of eremicus (Osgood 1909; cf.
Mearns, 1897), and interparietalis is considered specifically distinct from guardia (Banks,
1967; cf. Burt, 1932). Hooper and Musser (1964b) have suggested, on the basis of phallic
morphology, ihdi stephani m-dy be closely related to species of the subgenus Peromyscus.
MATERIALS AND METHODS
Specimens examined in this study were collected during trips to the Gulf of California
and surrounding areas in the summers of 1967 and 1968, and in January, 1969, or were
borrowed from the following institutions: California Academy of Sciences (CAS); Dickey
Collection, University of California, Los Angeles (UCLA); San Diego Natural History
Museum (SD); Department of Zoology, University of Arizona (UA); Museum of
1971 LAW LOR: Peromyscus 95
Vertebrate Zoology, University of California, Berkeley (UC); Museum of Natural
History, University of Kansas (KU); Museum of Zoology, University of Michigan
(UMMZ); and the United States National Museum (USNM). Both live and preserved
examples of each of the island populations were obtained.
Except for analysis of variation in dental patterns, only adult wild-caught mice were
treated for purposes of studying morphologic, serologic, and karyologic features. All age
groups were examined in the former, although specimens with excessive tooth wear were
omitted. Adult status was determined according to the methods of Lawlor (1964) and
HotTmeister (1951). Briefly, an animal was considered an adult if there was at least
moderate wear on the lingual cusps of M' and M- (the M^^ generally is well worn at this
age), and if the specimen was in advanced (''adult") pelage. Specimens examined for
serologic and karyologic properties were considered adult after retention in the laboratory
for at least two months.
in osteological considerations I dealt with quantitative and qualitative measures of
cranial and other skeletal features. Measurements, in millimeters, were taken with dial
calipers. Post-cranial features were examined from whole skeletons except that the number
of caudal vertebrae was determined from X-rays. The latter technique provides an accurate
means of counting vertebrae and avoids potential error in vertebral counts of whole
skeletons owing to vagaries of preparation. External characters include tield-taken body
measurements and pelage features.
Dental patterns were examined according to a modified scheme of the procedure
specified by Hooper (1957). Lophs and styles were considered present only if they
comprised a prominent element of an enamel valley. Even so, considerable variation
accrues in the development and appearance of these structures. Variation is particularly
evident in the shape and placement of styles, but no rigorous attempt was made to
determine homologies.
Phalli of freshly killed mice were extracted and fixed in 10% formalin. After everting
the prepuce over the proximal portion of the glans, the following procedure was used for
clearing and staining:
2 % KOH solution ca. 60 minutes
Alizaren red stain (in 2% KOH) 1-2 hours
Distilled water wash 1 minute
Solution of 2 pts. HOH, 1 pt. glycerin ca. 24 hours
Solution of 1 pt. HOH, 2 pts. glycerin ca. 24 hours
Glycerin permanent storage
The procedure for dried specimens dilTered slightly. Good results were obtained by
shortening the clearing and staining times by about one-half. This reduces the chances for
sloughing of the epithelial layer, a common occurrence if the glans was kept in KOH
solutions for long periods. The remaining steps were the same.
Karyotypes were examined by means of a bone marrow technique (Patton, 1967). An
average of 10 metaphase cells was counted to determine the diploid number of each
individual. The fundamental number ("Nombre FundamentaP' of Matthey, 1951) was
determined as the total number of autosomal arms (excluding the sex chromosomes). The
system for describing the chromosomes (Patton, 1967) was as follows:
Chromosome type Arm ratio
Metacentric Less than 1: 1. 1
Submetacentric 1:1.1 to 1:1.9
Subtelocentric 1:2 or greater
Acrocentric (telocentric) One arm only
96 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Eleclrophoretic analyses were made according to the modified method of Smithies
(1955) used by Brown and Welser (1968).
The taxonomic designations that are applied below to island and mainland popu-
lations concerning character diagnoses, and the discussion of phylogenetic and zoogeogra-
phic relationships that follows, correspond to the currently held taxonomy of those forms
(see above). They are used only to facilitate interpretation by the reader; they do not reflect
any taxonomic evaluation made prior to construction of the phylogeny and taxonomic
conclusions.
Statistics and construction of the dendrogram and phenograms were calculated by use
of an I BM 360/67 computer at the University of Michigan Computation Center.
MORPHOLOGIC VARIATION
MORPHOMETRIC CHARACTERS
A total of 20 external and cranial dimensions was examined. Six of these (total length,
length of tail, greatest length of skull, zygomatic breadth, length of rostrum, and length of
maxillary toothrow) are presented in Figs. 3-5. The remainder are either relatively
invariable or exhibit similar geographic changes as the ones treated in detail here.
Mice of the guardia islands (Angel de la Guarda, MejTa, and Granito) show no
consistent trends of inter-island variation in size, although significant differences (P<.05)
are evident in certain dimensions. However, interparietalis (Isla Salsipuedes, San Lorenzo
Norte, and San Lorenzo Sur) exhibits a trend of increasing size in nearly all dimensions
from Salsipuedes in the north to San Lorenzo Sur in the south. An exception is zygomatic
breadth, and the relative constancy of this dimension together with the variation for
greatest length of skull gives interparietalis from Salsipuedes a shorter, broader-headed
appearance than its southern island counterparts. P. stephani (Isla San Esteban) differs
from other island forms in having cranial dimensions that usually average larger.
In general, forms from Islas Turner (collatus) and Tiburon {eremicus) were similar to
mainland eremicus on the basis of morphometric characters. Other mainland populations
o{ eremicus in the Gulf area do not differ importantly from the two samples given in Figs.
3-5 (for example, see Lawlor, in press).
The large differences in certain dimensions evident between insular populations of the
same species (e. g., in interparietalis and guardia) suggest that isolation of these mice on
islands has resulted in the retention of morphometric differences that generally are
eliminated by higher rates of gene flow in continental populations. For example,
differences of significant proportions usually are not evident in morphometric data for
mainland populations o^ eremicus {Ibid.).
DENTAL PATTERNS
Dental topography in all of the island forms is relatively simple. The enamel valleys
between major cusps generally are unobstructed except for styles. The most common
accessory tooth structures are ectostylids on the lower molars (Mi and Mi) and
mesostyles on the uppers (Figs. 6, 7; Table 1 ); the latter are more variable in frequency and
are found uncommonly on the second molars. Mesolophs occur at high frequency in the
M ' only in mice from Islas Granito and San Esteban, whereas entolophs, mesolophids,
and ectolophids were not observed in any specimens. A ''pseudomesolophid'' (Hershkov-
itz, 1962) was observed in several specimens oi interparietalis on the lower first molar (two
specimens |5.6%| from Isla San Lorenzo Norte and four [1 1.7%] from Isla San Lorenzo
Sur). The mesoloph and mesostyle are rarely fused. This condition was observed on the M '
in only two (7.1 %) specimens o{ stephani and one (4.8%) o{ eremicus from Bahfa de los
1971 LAWLOR: Peromyscus 97
BAHIA DE LOS ANGELES
28
ZZl ISLA GRANITO
3
12
II I ANGEL DE LA GUARDA
17
ID I SALSIPUEDES
20
SAN LORENZO NORTE C
21
I, SAN LORENZO SUR
28 28
ZZl I TIBURON I ~WW 1
17
m PUNTA SARGENTO
160 180 200 220 80 100 120
TOTAL LENGTH LENGTH OF TAIL
Figure 3. Geographic variation of two external dimensions of Peromyscus on northern island and mainland
areas in and adjacent to the Gulf of California, Mexico. The solid rectangles represent two standard errors on
either side of the mean: hollow rectangles refer to the range of variation. Sample sizes are indicated for each plot.
Angeles.
Mice from the guard ia and interparietalis groups of islands exhibit the simplest dental
topography in the upper molars, owing to the absence or low frequency of mesostyles,
particularly in the M-. Populations of guardia differ from interparietalis by the nearly
complete absence of mesostyles and entostyles on the M ' . In the lower molars there is little
variation in frequency of ectostylids, but the mice from Isla Granito differ from the
remainder of the island and mainland Peromyscus by virtue of the high frequency of
mesostylids there.
On the basis of dental structures Peromyscus from Islas Granito and San Esteban are
the most distinct of the island forms. In addition to possession of mesostylids, the mice
from Isla Granito have a high frequency of mesolophs on M' . Also, a mesoloph on the M-
was noted in 20.8% of the specimens; except for its occurrence at very low frequency in the
population from Bahia de los Angeles, this structure was not observed on the M- in
specimens from other localities. The mice from San Esteban resemble those from Granito
in having a high frequency of mesolophs on the M' . However, the population differs from
that on Isla Granito by the absence of mesolophs on the M" and the presence of mesostyles
on the M' in 25.0% of the specimens. In general, populations from Islas Turner and
Tiburon resemble mainland populations o{ eremicus.
Little phylogenetic information can be derived from the variation in dental patterns.
For comparison, two mainland populations of eremicus from Bahfa Kino and Punta
Sargento, Sonora, show as much variation in frequencies of mesostyles, entostyles, and
ectostylids as do all other island and mainland populations studied; yet, these mainland
localities are only 30 miles apart. Note also the variation in dental structure among the
98 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
1
■■
BAHIA
DE LOS ANGELES L^ 1
31
32
1
m
1 ISI A CRANirn 1 ■
i:
13
13
1 MEJIA 1 pH
1
^■1
14
14
1
■■
1 ANHFi nr 1 A nilARRA 1 ^H 1
19
21
L SALSIPUEDES
V
19
m
1
21
^^
1
■■
1
L SAN LORENZO NORTE d
W 1
17
24
25
1 1
P
1 SAN LORENZO SUR 1
■
1
op
1. TURNER
1. TIBURON
PUNTA SAR
1 , 1
V
9 10
□ 1 SAN ESTEBAN '-^^'^
16
m
1
1 ^
J
30
29
^
1
17
14
i
■■
GENTO !_■_
1
1 , 1
1
1,1.
23.0 25.0 27.0 11.0 13.0
LENGTH OF SKULL ZYGOMATIC BREADTH
Figure 4. Geographic variation of two cranial dimensions of Peromyscus on northern island and mainland areas
in and adjacent to the Gulf of California, Mexico. For explanation of plots see Fig. 3.
three populations ofguardia (Figs. 6 and 7, Table 1). Similar variation was noted in several
species of Peromyscus by Hooper (1957). His results for eremicus correspond closely to
those presented for populations of that species here.
QUALITATIVE CHARACTERS
Osteology. — Variation in osteological characters among Peromyscus commonly is subtle,
and distinct character differences often are difficult to detect in closely related species. The
skulls of six island and mainland examples studied here are illustrated in Fig. 8. Eight
cranial features of taxonomic importance were discernible in the island and mainland
forms. Many of these features were observed by Banks (1967). The characters and their
character states are as follows:
1971
LAWLOR: Peromyscus
99
15
BAHIA DE LOS ANGELES
15
32
ISLA GRANITO
32
14
19
^=f=
21
13
I. MEJIA
I ANGEL D£ LA GUARDA
L SALSIPUEDES
L SAN LORENZO NORTE
13
14
19
21
17
30
18
SAN ESTEBAN
25
10
TURNER
I TIBURON
PUNTA SARGENTO
17
^¥^
30
18
8.0 10.0
LENGTH OF ROSTRUM
4.0
LENGTH OF MAXILLARY
TOOTHROW
Figure 5. Geographic variation of two cranial dimensions o{ Peromyscus on northern island and mainland areas
in and adjacent to the Gulf of California, Mexico. For explanation of plots see Fig. 3.
(I) Shape of frontal bone (Fig. 8). — The posterior margin of this bone is curved
(coded 0) in most of the island and mainland mice, but in certain populations (stephani,
boylei) it usually is sharply angular (coded 1 ).
(II) Position of nasal bones. — In stephani and boylei the nasals extend posteriorly to
or beyond the premaxillaries (0) (Fig. 9A'), while in all other populations the nasal bones
do not reach the level of the posterior extension of the premaxillaries (1) (Fig. 9A).
(III) Shape of posterior margin of nasals. — The posterior margin of the two nasal
bones is rounded or bluntly pointed (0) (Fig. 9A'), or squared (1) (Fig. 9 A). This character
is variable among mice on the eastern Gulf islands and among mainland populations of
eremicus. Squared nasals are particularly prominent among the three populations of
interparietalis.
(IV) Shape of interparietal bone. — Mice from populations of guardia exhibit a
100
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
UPPER MOLARS
Ml M2
LOWER MOLARS
Ml M2
100
50
n
a
BAHIA DE
LOS ANGELES (21)
nl ms el es ml ms el es
ml ms el es ml ms el es
lOOrn
50
^ 0
O 100
Q.
I GRANITO (24)
50
100
n
J]
n
50
>-
I ANGEL
DE LA GUARDA (19)
_a_
o
100
50
0
100
50
n
I SALSIPUEDES (35)
H
_n
ml ms el es
nl ms el es
I SAN LORENZO
NORTE (36)
ml ms el es ml ms el es
Figure 6. Frequencies of styles (stylids) and iophs (lophids) among Peromyscus on northern island and
mainland areas in and adjacent to the Gulf of California, Mexico. Sample sizes are indicated in parentheses. Data
for additional mainland populations are presented in Table 1.
relatively small, triangular interparietal bone (0) (Fig. 9B), while the bone in remaining
insular forms generally is strap-shaped ( 1 ) (Fig. 9B' and B").
(V) Lateral bony extensions of interparietal (Fig. 9). — Populations of guardia,
boylei, and stephani commonly have small bony extensions of the interparietal bone that
extend outward on each side toward the squamosals (1). The remaining island and
mainland populations almost always lack these elements (0).
(VI) Position of squamosals. — P. interparietalis, stephani, and boylei from Isla San
Pedro Nolasco have relatively flattened skulls. The squamosal bones in these forms are
slanted dorso-medially (1). This trait is not evident (0) in other populations examined and
the skulls are inflated. This trait is indicated in Fig. 8 by the enlarged appearance of the
squamosals.
(VII) Shape of mesopterygoid fossa. — In most island and mainland forms the
pterygoid bones adjacent to the fossa are straight (0) (Fig. 9C), but in guardia (Fig. 9C') the
1971
LAWLOR: Peromyscus
101
lateral pterygoid margins of the fossa are usually concave and as a result the fossa appears
larger and is expanded laterally (1). The mainland population of eremicus from Presa
Obregon, Sonora, also exhibits the latter feature. The occurrence of an expanded
mesopterygoid fossa varies geographically within species.
(VIII) Position of incisive foramina. — The incisive foramina in populations of
interparietalis, coUatus, and certain eremicus commonly extend posteriorly beyond the
level of the first molars (1) (Fig. 9C), while in other populations the posterior termination
of the foramina is usually anterior to the molar toothrow (0) (Fig. 9C'). The values for this
character are quite variable geographically.
Table 1. Frequencies of occurrence of styles (stylids) and lophs (lophids) among some mainland
populations of Peromyscus eremicus not included in Figs. 6 and 7. Entolophs (upper molars) and
mesolophids and ectolophids (lower molars) were not observed in any specimens. Localities are
arranged in order according to their number designation in Fig. 1. Numbers in parentheses identify
sample sizes.
Upper Molars
Lower Molars
X!
u
4>
•o
"O
o
B
B
">.
CO
■S!
t/i
-•-•
1
O
CO
1
o
CO
s
4>
V
■*-^
<o
■*-»
Locality
S
S
c
S
Baja California
Turtle Bay (6)
Ml
0.0
83.3
0.0
0.0
50.0
M2
0.0
16.7
0.0
0.0
50.0
Barril (10)
Ml
0.0
80.0
0.0
0.0
90.0
M2
0.0
70.0
10.0
0.0
60.0
San Francisquito (14)
Ml
7.1
100.0
21.4
0.0
85.7
M2
0.0
64.3
7.1
0.0
85.7
El Marmol (10)
Ml
0.0
70.0
0.0
0.0
100.0
M2
0.0
30.0
0.0
0.0
100.0
San Telmo (7)
Ml
14.3
85.7
0.0
14.3
71.4
M2
0.0
85.7
0.0
0.0
71.4
California
Escondido (13)
Ml
0.0
92.3
1.1
0.0-
84.6
M2
0.0
69.2
0.0
0.0
46.2
Sonora
Puerto Peiiasco (19)
Ml
0.0
89.5
36.8
0.0
94.7
M2
0.0
73.7
26.3
0.0
89.5
Imuris (9)
Ml
0.0
66.7
0.0
0.0
88.9
M2
0.0
11.1
0.0
0.0
88.9
Puerto Libertad (18)
Ml
0.0
56.7
50.0
0.0
61.1
M2
0.0
38.9
38.9
0.0
100.0
Presa Obregon (11)
Ml
0.0
63.6
9.1
0.0
100.0
M2
0.0
9.1
0.0
0.0
91.0
Mean values for coded character states are presented in Table 2. Although mean
values of certain characters exhibit considerable geographic variation (see above) they
have been included here to demonstrate the osteological variation that exists among
different populations of certain species relative to that between species. Criteria for
weighting such features are discussed below.
It is evident that mice from Isla San Esteban are readily distinguishable from the
remaining island populations on the basis of osteologic features. Further, stephani seems
closest in these characters to boylei. Mice from the Guardia island group (guardia) are
distinguishable chiefly by the triangular shape of the interparietal bone and the prominent
102
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
UPPER MOLARS
Ml M2
LOWER MOLARS
Ml M2
100
50
100
50
ID
X 100
CL
O
Q
<:
50
CO 0
LU
^ 100
I—
CO
^ 50
■>-
o
100
50
0
100
50
I SAN LORENZO
SUR (34)
ml ms el es ml ms el es
ml ms el es ml ms el es
II
JZI
I SANESTEBAN (28)
I TURNER (30)
JZL
n
_□ I TIBURON (36)
PUNTA SARGENTO (16)
\\
ml ms el es
BAHIA KINO (8)
ml ms el es
ml ms el es ml ms el es
Figure 7. Frequencies of styles (stylids) and lophs (lophids) among Peromyscus on northern island and
mainland areas in and adjacent to the Gulf of California, Mexico. Sample sizes are indicated in parentheses. Data
for additional mainland populations in Table 1.
lateral bony extensions of the interparietal, while those from the Lorenzo island group
{interparietalis) differ from other island and mainland forms chiefly by the squarish
posterior margin of the nasals and the flattened braincase. The forms from Islas Turner
{collatus) and Tiburon {eremicus tiburonensis) closely resemble mainland eremicus in all
features.
Post-cranial skeletons of all island and several mainland populations were examined,
but no important variation in shape or position of bones was evident. However, differences
in number of caudal vertebrae were observed (Fig. 10). Mice from mainland populations
are more variable in this feature than those from the islands. In certain populations (e. g.,
guardia) the number of caudal vertebrae seems to be fixed. However, sample sizes
generally are small, and conclusive statements must await additional data.
Pelage. — Three pelage features were discernible. These characters and their character
1971
LAWLOR: Peromyscus
103
EREMICUS
COLLATUS
INTERPARIETALIS
10 mm
GUARDIA
STEPHANI
BOYLEI
Figure 8. Dorsal skull views of six examples of Peromyscus from areas in and adjacent to the northern part of
the Gulf of California, Mexico.
States are as follows: (IX) extent of tegumentary attachment on the tail (skin tightly
attached to underlying tissue, 0; skin loosely attached and easily removed, 1 ); (X) hairiness
of the tail (scantily haired, 0; well haired, 1); (XI) occurrence of gray facial coloration
(absent, 0; present, 1).
104
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
2 mm
4mm
Figure 9. Views of: the dorsal aspect of the frontal region of the skull in Peromyscus interparietalis (A) and P.
stephani (A'); interparietal bones of P. guardia (B), P. interparietalis (B'), and P. stephani (B"); and midventral
region of the skull in P. interparietalis (C) and P. guardia (C). For characters pertinent to these illustrations see
text.
1971
LAWLOR: Peromyscus
105
too
^ 2 o
(D -^ 3
:£'2 o
S c
c g —
'o ?5 O
S 2i c
<-> ^ fll
5 o JJ
^" a
P^ c
m.
o 2 3 o
on — o ^
C C t- rt
E 3 Si«
c =« f^ 5
^ ° s •-
00—. u ^^
C ' rt -n
« SB '^
-2 o S
u
•2 2i S S
u, o ^^
6t^ o
o ni 2 S
_ rt c o
<= • S^
C3 c« c'j •i;
0.£^.
<u
E
■si
I"
3 O
c o
- 3
>
c
<u
> o
1^
C/5
V3
c
o
'^ <u
00 u-
■7, <^
O O
A3oioAjH;>i Z
X3010J3S
CM-
(lAX) uininoEq jo adBqs
£ (IIIX) sjaddBi
IBJJU3A JO 3DU3JjnD0O
00
00
"o
(IX) 33BJ
UO XbjS jo 3DU3JJnD30
(IIIA) EUIIUBJOJ
3AISIDUI JO UOIJISOJ
(IIA) ^ssoj
pio§/tJ3idBS3ui JO adBqs
(lA)
siEsouiBnbs JO uoijisoj
(a) IBJ9UHdj3JUI
JO SUOISU3JX3 Auoq
[BjajBJ JO 3DU9JjnD0O
(AI) IBJ3UBdj3JUI
joadEqs
(III) siBSBU JO utSjbui
jouajsod JO adBqs
(II) SJESBU JO UOIJISOJ
(I) iBjuojj JO adBqs
T3
g =^
TO *-»
"5 IS
•— O
o O
O Jj
C/2
(S)'-*^
(S<S— 1
_(
— 1 -H(S(S
•)—
<s
— i(s
rj --' -h'
rifn rf
-<'
rfn"'--ri
<N
-n'rf
>r> r^ m
r~- -^ li^
^
v-i (S -^ oo
■1—
o\
CTi oo
ooo
ooo
ooo
o
q
oooo
</-! o o o
o
o
o
o
oo
oo
odd
>— 1 I— 1 ^H
--
d -: —■ -;
d
d
dd
ooo
ooo
VO O O
ooo O
o
o
ooor^
ooo —I
o
o
o
q
OO
d>d)d:
ddd
d
dddd
d
^^
»— 1 —
Os -^ >n
r--- vo 00
VC
■*m -^ VO
t~
Cl
cncs
OOO
q q q
OOO
o
q
O (^ O -H
O oo CT\ VO
d
q
o
o
OO
oo
^ ^ -^
—.—<—(
"-■
—'ddd
*-H
d
dd
\C <N '^
(VI — —
0\ rt v^
-^ (N (N
OS
t^ VO O ON
VO
—
OOtT
oo —
O VD O
o r- o
o
q
(^ r^ o m
0\ vO (N •/^
O
o
o
O rn
O ro
didid)
^d^
-^
dddd
d
d
dd
iri O O
r-oo
ooo
ooo
0\
oooo
oooo
(S
o
o^
O — '
d-^-i
ddd
d
dd-^d
d
d
dd
ooo
ooo
ooo
o
o
oooo
oooo
o
o
o
q
o o\
O CTv
d>dd
— ' -^ ^^
d
dddd
d
' — '
— d
wi o r~-
Tt O v-i
>/iO o
ooo
o
o
oooo
oooo
VO
o
o
O VO
d'-^d
ddd
d
dddd
d
'"'
dd
oor-
ooo
goo
o
q
ooo r-
OO O 00
o
q
o
q
O >/-i
OOv
ddd
— ( ^H -H
-~
^^ _:d
^^
^^
^d
ooo
ooo
OO oo
o OOO
■*
>^ O or-
VO
o
o
OO
OO
ddd
^^d
d
dddd
d
d
dd
ooo
ooo
q q q
o
q
OOOO
q q q q
o
q
o
o
OO
oo
-^ ^H — 1
-H — 1 «
•— '
^" — ' '-" -^
^^
d
dd
OOO
ooo
OOO
ooo
o
o
ooor-
o o O <N
o
o
o
q
o ^
O fS
ddd
ddd
d
dddd
d
"
^d
r4 (^ ^
m ^ —
o\ -^ >/-)
— <N rJ
r-
00 VOO >/-i
r-
o
c3
T3
oj
3
o
03
3
a.,
^
o -o
o o
N N
C C
>-. u
o o
c c
&0 C/5
CiO
c
3
H
CO
"o
00
c
§<
00 en
1) o
= ^,2 '^^2
^ ?; ? 'n I-'
C ;— 03 t, ro
c
o
00
4>
Ui
s:
c
2 c
o
"o*
O 00
1-. CJ
T3 3
o
o
d
o
o
VO
o
o
d
VO
as
d
o
d
o
o
d
o
o
o
o
d
d
o
o
VO
*
.2
'S
u
O
, 0, t*j
^ 03 (U
o
?3 s-:^
T3
(U
c
E
03
K
u
on
C
<u
E
•o
CJ
u
c
a.
(/5
E
d
c
X
o
<L>
c
C/5
o
C
1)
CJ
E
o
v:
a
w
«
a.
O
C/5
M
c
c
o
_o
c
'ca
C
00
u
'c/5
C/3
U
1>
T3
u
c/3
>.
,,
-*-*
c
^^
o
03
en
CJ
o
O
3
'■"'
H
<u
V)
k>
03
'o
u
c
a
E
kn
o
o
u.
PhU,
106
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
c
A
c
c
C
C
c
T
T
K
L
K T
L
B
R
B
B
E
E
E
G
G
G
A
G
G
M
A
N
S
N
S
N
1
S
N
1
S
S
1
S
S
1
D
29 30 31 32 33 34 29 30 31 32 33 34
NUMBER OE CAUDAL VERTEBRAE
Figure 10. Number of caudal vertebrae in populations of P. eremicus and P. collalus (A), P. boylei and P.
stephani (B), P. guardia (C), and P interparietalis (D). Letter designations are as follows: A — Isia Angel de la
Guarda; B — P. boylei: CI. Turner; E I. San Esteban; G — I. Granito; I — I. San Lorenzo Sur; K — Bahia
Kino; L — Bahia de los Angeles; M — I. Mejia; N — I. San Lorenzo Norte; S — I. Salsipuedes; T — I. Tiburon.
Pelage characters serve chiefly to distinguish stephani and boylei from other forms.
Only in stephani and boylei is the tail well haired and loosely attached. Pelage coloration is
similar among the island and mainland mice, except that mice from Isla Mejia and Presa
Obregon, where dark substrates occur, are darker than other populations. Few other
pelage diff'erences are found among the remaining populations, although certain popu-
lations (e.g., mainland eremicus) show considerable variation in the expression of gray on
the face (Table 2).
Morphology of phallus. — The island populations exhibit both complex and simple
peromyscine phallic types (Figs. 11 and 12). Phalli of specimens from Isla San Esteban
{stephani) are complex and closely resemble boylei in all features. Those of the remaining
island forms and mainland eremicus are relatively simple. Accouterments, such as ventral
and dorsal lappets, protrusible tip of glans, and cartilaginous tip of baculum, are poorly
developed or absent.
Specimens from the guardia group of islands have phalli that are morphologically
intermediate between complex and simple types. A protrusible tip is present but not well
developed, and dorsal lappets and a small cartilaginous tip also are present. Phalli of forms
from Islas Tiburon and Turner, and the Lorenzo island group {interparietalis), closely
resemble mainland eremicus, except that phalli of interparietalis are larger and six of seven
interparietalis from Isla Salsipuedes and one of six eremicus from Bahfa de los Angeles,
Baja California, have ventral lappets (see Fig. 1 1 and Table 2). It is not clear whether these
structures in the two latter populations are homologous, however, because the lappets in
the specimen from Bahi'a de los Angeles are separated from the adjacent tissue by a simple
1971 L AWLOR: Peromyscus 111
for eremicus, crinitus, and some populations of maniculatus by Brown and Welser
(1968). The mobility for the three populations of interparietalis {ca. 94) also differs from
the other forms studied here. In addition, the mobility obtained for albumin in boylei
and stephani (90) does not correspond to the value (84) given for one individual of boylei
by Brown and Welser, suggesting that an albumin polymorphism may exist in that species.
Jensen (pers. comm.) has noted polymorphisms of albumin in boylei from northern
Arizona. No intra-population variation was noted in this study.
Although direct comparisons are not possible, the positions of the albumin band in
interparietalis and guardia correspond favorably to densitometer tracings of this band
(Brand and Ryckman, 1969) except that those authors report a difference between albumin
of interparietalis from Isla Salsipuedes and from the San Lorenzos (a mixed sample from
San Lorenzo Sur and San Lorenzo Norte). Further investigation of this discrepancy is
required.
Karyology. — All members of the genus Peromyscus so far examined have a diploid
number of 48 chromosomes regardless of the proportion of acrocentrics in the complement
(Hsu and Arrighi, 1966, 1968). The populations studied here are no exception. There also is
considerable variation between species as regards morphology of the chromosomes. The
populations examined here differ in the following characters (sample sizes given in Table
2):
(XXV) Number of autosomal acrocentrics. — There are no acrocentric chromosomes
in merriami, eremicus. interparietalis, and collatus (0); most are either submetacentric or
subtelocentric. P. guardia has one pair of small acrocentrics (1), while stephani and boylei
each has 20 pairs (2).
(XXVI) Morphology of the X chromosome. — In most populations, including
merriami, eremicus, interparietalis, stephani, boylei, and guardia, this chromosome has
unequal arms (0). Most have a large submetacentric X chromosome, but in guardia from
Isla Mejfa it is a large subtelocentric. P. collatus has a large metacentric X chromosome
(1). The morphology of this chromosome is subject to some variation both locally and
geographically. For example, in eremicus it occasionally appears almost as a metacentric
{cf Hsu and Arrighi, 1968), whereas in guardia it varies from a submetacentric to
subtelocentric condition. Although this character is employed beyond for purposes of
assessing overall similarity, additional data may prove it to be unsuitable for taxonomic
use.
(XXVII) Morphology of the Y chromosome. — This chromosome is a medium-sized
acrocentric (0) in guardia, has unequal arms (1) in eremicus and merriami (medium
subtelocentric), interparietalis and collatus (medium submetacentric), and is a medium
metacentric in stephani and boylei (2). Hsu and Arrighi (loc. cit.) reported that the one
individual of eremicus from Isla Tiburon they examined had a small acrocentric Y
chromosome. However, examination of a photograph of that karyotype reveals that this
chromosome is a subtelocentric according to the classification used here.
The fundamental numbers of chromosomes in the island and mainland forms are:
guardia 90, interparietalis 92, collatus 92, eremicus 92, merriami 92, stephani 52, boylei 52.
Karyotypes are illustrated in Figures 15 and 16.
BREEDING
Attempted matings between different island and mainland forms are given in Table 3 .
The breeding colony of interparietalis from Islas Salsipuedes and San Lorenzo Norte was
obtained in 1967. Consequently more matings of those populations were made. Unless
otherwise noted, results of crosses in the following discussion refer also to reciprocal
112 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
l\A no OA no A/i rtA OA
Aft />A AA rtA Art A/\ no
/SA A>^ '^♦^ f^^ ^^ "^
X y
JJH XA ^x ""«" JJ
BOYLE! (TUCSON)
U
H U i)h Aii AA I^A ftA
M AA Aii A^ A<S )»5 ^A
ftK ^A AJi xj( A«x.
X Y
XX ^^ ^^ Art GUARDIA (I MEJIA) ft ^
GUARDIA (I. GRANITO)
u
Figure 1 3. Karyotypes of P. slephani and P. guardia from Isla Meifa; and sex chromosomes of P. boylei. and P.
guardia from Isla Granito.
1971 L AW LOR: Peromyscus 113
H n H H liil y
jiji H hh n M KA ^/:
AA i(X KK «A HA
X Y
||( M ^X XX INTERPARIETALS iJ ^
X Y
n u n Hi'
I II III IV
I Figure 16. Karyotype of P interparietalis and sex chromosomes of P. eremicus from Bahfa de los Angeles (I)
and Isla Tiburon (II), P. collaius (III), and P. merriami (IW).
matings. All island forms o{ interparietalis bred freely among themselves, and one cross of
interparietalis (?) and collatus (c?) was successful. In each case the offspring were viable.
There was no success at breeding stephani, boylei, guardia, Tiburon eremicus, or collatus
(except with interparietalis), even among controls. P. eremicus crosses, including one of
Kino (?) X Bahfa de los Angeles (cf ), produced viable offspring in all cases.
These results correspond well with data on attempted matings reported by Brand and
Ryckman (1969); they were able to breed interparietalis and eremicus, but had very little
success with guardia. The data indicate that certain island and mainland populations of
eremicus, interparietalis, and collatus, are interfertile and are closely related. No con-
clusive statements can be made concerning the negative breeding evidence for populations
o{ guardia, eremicus tiburonensis, stephani, and boylei. Morphological features, such as
those of the phallus, may act as physical barriers to hybridization with certain forms, P. e.
tiburonensis, however, is obviously closely related to mainland eremicus and collatus; yet
no mated pairs produced offspring.
Evidence regarding laboratory breeding must be viewed with caution, since premating
isolating mechanisms may break down under laboratory conditions. Nevertheless, since
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
certain forms have the potential for interbreeding in the wild, close relationship of those
populations is evident.
Table 3. Attempted matings of island and mainland Peromyscus. Numbers in parentheses repre-
sent reciprocal crosses.
S3
<
(/>
<u
cd
w
O
60
c
<
1)
1
3
o
o
z
o
3
CO
o
c
a
c
on
1
o
7,
c
0
o
C
«J
tn en
N
N
o
c
3
;^ (U
c
c
o
to
D
a "O
<u
<1>
Ui
c>
1 =
c
o
3
Si
H
"2
2
M
C
<
o
'c
a
O
v5
4?
nterpariet
Salsipue
o
C
C3
CO
O
c
CO
5
3
1
T3
1>
CU
C
03
on
3
H
03
C
^
o
Oo
•—
■^
■Jo
FEMALES
eremicus:
Bahia Kino
Bahia de los Angeles
Tiburon
coUatus: Turner
guardia:
Angel de la guarda
Granito
Mejia
interparietalis:
Salsipuedes
San Lorenzo Norte
San Lorenzo Sur
stephani:
San Esteban
boy lei:
San Pedro Nolasco
near Tucson
1 2(1) 1(0) 2(1) — 1(1) _ — __ — — —
3 _ 1(1) _ 0(2) — — 2(1) 0(1) — — —
2 1(1) _ — — — _ — — — —
1 _____ 1(1) _ _ _
1 1(0) _____ — —
2 1(1) — — 1(1) _ _ —
1 ______
3 3(1) 2(1) _ — —
6 2(3) _ — —
2 0(1) 1(1)
1 —
2
EVOLUTION OF THE ISLAND FORMS
PHYLETIC RELATIONSHIPS
A total of 27 serologic, karyologic, osteologic, and other morphologic characters was
treated by numerical taxonomic methods, first by a phenetic clustering technique, and then
by the quantitative phyletic method (Kluge and Farris, 1969). The first step is the
construction of a Prim Network (Prim, 1957; Kluge, in press; see also Edwards and Cavalli-
Sforza, 1964) in which only phenetic differences {sensu Farris, 1967) between the island
and mainland populations, or OTU's (Operational Taxonomic Units; see Sokal and
Sneath, 1963), are determined. Two Prim Networks are presented (Figs. 17 and 18). The
character states used to describe the OTU's are sample means (data for every character
were not available for all individual specimens). Distances between OTU's (interval lengths)
represent the sums of character dilTerences between OTU's. The Prim Network connects
the OTU's with minimum total interval lengths. There is no directionality implicit in the
network, and angles of branching events are arbitrary. The first network (Fig. 17) includes
data derived from all but serologic and karyologic characters. The second network (Fig.
18) includes only those populations for which complete data were obtained (sufficient data
regarding characters of chromosomes and blood proteins were not available for crinitus).
Data regarding phalli for crinitus and mainland boylei and blood proteins and
1971
LAWLOR: Peromyscus
115
9 I San Pedro Nolasco
1 30
I San Esleban (slephani)
2 16
4.95
4.51
o o
c c
ra to
c/) </>
-•;♦
2 10
0 88
Kinol
Pfesa Obregon
f
1 82
1.02
I'* ' •
^ I Tiburon
I. Turner (collalus
6.02
t Mejia
^H Granito
1.03 036 1.161
Bahia de los Angeles
1.58
I Angel de la Guarda
■'boylei'
interparietalis
'"eremicus"
guardia
Figure 17. Prim Networlc computed from data derived from osteology, phallic morphology, pelage, and soft
anatomy in Peromyscus on northern island and mainland areas in and adjacent to the Gulf of California, Mexico.
The interval lengths represent unweighted measures of the sums of character differences between OTU's. The
network length is 38.04.
karyotypes for merriami were obtained from different populations than were data for other
features in those species. Their inclusion is justified for comparative purposes because
the above characters are relatively invariable geographically, and because even consid-
erable variation in the above structures alters only the interval length and not the
branching sequences. Consequently, the data are assumed to be representative.
Four distinct clusters are evident in each diagram: (1) a group consisting of
populations of mainland eremicus and insular forms from Tiburon and Turner (labeled
"eremicus"); (2) a cluster comprised of the three island populations o{ interparietalis; (3) a
group composed of the three island forms of guardia; and (4) a group consisting of
mainland and island boylei (Tucson and Isla San Pedro Nolasco) and the population from
Isla San Esteban (stephani) (labeled "boylei"). The populations of crinitus (Fig. 17 only)
and merriami are located intermediate to boylei- and eremicus-Uke forms on the networks.
The degree of phenetic similarity between the boylei-Vike forms and the remaining
island and mainland forms clearly separates the former populations from the latter. They
evidently are distantly related. Without doubt, stephani exhihhs closest affinities to boylei.
To my knowledge, boylei glasselli (Isla San Pedro Nolasco) and stephani comprise the
only two island derivatives of boylei in the Gulf. Note the differences in the Prim
Networks, especially for populations of interparietalis and eremicus. that result from the
addition of data on serology and karyology.
In the quantitative phyletic method a Wagner Diagram (Farris, 1970) was used to
depict interval lengths (patristic differences; sensu Farris, 1967) and branching events (Fig.
19). The Wagner Diagram differs from the Prim Network in three ways: ( 1 ) each character
is weighted a priori by the mean value of the reciprocal of the intrapopulation standard
deviation over all OTU's (i.e., conservative characters are more heavily weighted; see
Farris, 1966; Kluge and Farris, 1969); (2) hypothetical intermediates are generated to
minimize total interval length (i.e., to maximize parsimony); and (3) a hypothetical
ancestor is chosen, thus giving directionality to the diagraml The intervals on the diagram
represent the sums of weighted character differences between OTU's.
Populations representing eremicus, merriami, interparietalis, collatus, and guardia
were examined for purposes of ascertaining phylogenetic relationships. These forms
exhibit close morphologic and zoogeographic similarities and probably form a mon-
'The character standard deviations, weighted character state values, and character state values for the hypothetical
ancestor and generated intermediates, are filed with the National Auxiliary Publication Service of the American Society
for Information Science, and may be obtained by ordering NAPS Document 01267 from ASIS National Auxiliary
Publication Service, CCM Information Corp., 909 Third Ave., New York, N.Y. 10022, remitting $5 per photocopy or
$2 per microfiche copy.
116
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
interpanetalis
^1, San Pedro Nolasco
1.30
^1 San Esteban {stephanO
2 16
o O
c c
o o
— c c
crt to to
17.10
103 0 36
Presa Obregon
\
• •! Turner (collatus)
1 84
3 10
I 88
2 16
•
mernami
► Bahi'a de los Angeles
902
0 80
I Mena
1 68
1 Angel de la Guarda
I Granito
"boyler"
guardia
Figure 18. Prim Network computed from data derived from all coded characters (features in Fig. 17 plus data
from serology and karyology) in Peromyscus on northern island and mainland areas in and adjacent to the Gulf
of California, Mexico. The interval lengths represent unweighted measures of the sums of character differences
between OTU's. The network length is 48.63.
ophyletic group. They are evidently only distantly related to boylei, crinitus, and other
continental species not studied in detail here. Nineteen of the original 27 characters were
used in this analysis. The other eight characters (II, IX, X, XVIII, XIX, XX, XXII, and
XXIV) serve only to distinguish eremicus-Yikt forms from boylei-Wko, forms. Characters of
Peromyscus eremicus were chosen as ancestral for eremicus-WkQ forms for the following
reasons: ( 1 ) zoogeographically, eremicus represents the only species of Peromyscus that is
present on the mainland of both sides of the Gulf, and it seems reasonable to assume that
the island populations {QxcQ\)\.\r\g stephani) resulted from isolation of a mainland eremicus-
like progenitor; and (2) the species shares the most characters in common with all the
insular forms with the result that populations of this species are located centrally to other
similar forms on the Prim Network. Assuming that evolution from a primitive ancestor
takes place in more than one direction (i.e., it is radiative) and at approximately similar
rates in major phyletic lines, then a population (or populations) located near the center of
the Prim Network would seem to be the best approximation to the ancestral condition in
the absence of unequivocal evidence. Thus, eremicus, or more likely a progenitor of similar
characteristics, is here considered the ancestral type. Mean values of the character states
for mainland populations o{ eremicus were given to the hypothetical ancestor.
From the available data, it is not possible to ascertain which mainland eremicus are
most like the ancestral form; all populations, and particularly those of P. e. eremicus,
which occurs on the coastal areas surrounding the northern portion of the Gulf, are very
similar morphologically. The close phenetic similarities of mainland populations result in
the compact cluster on the Prim Networks (Figs. 17 and 18). On zoogeographic grounds,
however, it seems likely that western island populations are derived from Baja Californian
eremicus, whereas populations on the eastern Gulf islands are probably derived from
Sonoran eremicus. The affinities of other mammals on eastern and western Gulf islands
correspond closely to mainland species of the eastern and western sides of the Gulf,
respectively (Table 4). Similar relationships are shown by peromyscines on other Gulf
islands (Lawlor, in press), and by the amphibians and reptiles in the Gulf (Soule and Sloan,
1966).
It is clear from the phylogeny presented in Fig. 19 that, with the exception of
merriami. interparietalis and guardia are the most divergent of the eremicus-Wkc forms.
Populations from Tiburon {eremicus tiburonensis) and Turner (collatus) are not far
1971
LAWLOR: Peromyscus
117
I. Meiia
0,38
I. Angel de la Guarda
"'—0.18
Granito
I Salsipuedes(23)
0.92
I San Lorenzo Sur
'22-0.19
Bahia Kino (21)
19-0.33
. 0.30
^0 55
005
Hypothetical Ancestor
Figure 19. Wagner Diagram depicting the phylogeny of closely related Peromyscus on northern island and
mainland areas in and adjacent to the Gulf of California, Mexico. The interval lengths represent weighted
measures of the sums of character differences between OTU's. Character state values for hypothetical
intermediates generated during computation of the phylogeny are on file with NAPS (see text). The total length of
the dendrogram is 24.08.
removed from the hypothetical eremicus ancestor. P. guardia differs from other eremicus-
Hi«ce forms in characters that are relatively invariable within species. Examples are the
presence of band C hemoglobin, a triangular interparietal bone, and characters of the
phallus. Conversely, interparietalis differs chiefly in features that often exhibit a high
variance within species, such as the position of squamosals and shape of the posterior
margin of the nasals. For instance, an inflated braincase owing to the position of the
squamosals is observed in one of the two boylei populations examined (Table 2). Because
of these differences in character state variation, guardia is more divergent from eremicus
than is interparietalis.
The phylogeny presented here as a working model can be used to examine evolution-
ary changes in certain characters. For example, according to Hooper and Musser (1964a)
and Hershkovitz (1962) simple conditions of the phallus (in Peromyscus these include the
absence of a protrusible tip, lappets, and a cartilaginous tip on the baculum) and of the
dentition (e.g., absence of accessory styles and lophs on the molars) are generally thought
to result from loss of structures present in a more complex progenitor. Patterns of overall
historical changes in these structures among island Peromyscus may not parallel trends in
rodents in general. Nevertheless, the trends do differ from current views on the subject: (1)
In guardia there evidently has been selection favoring both decreased and increased
complexity of the teeth. There has been a virtual loss of mesostyles in all populations while
in two populations mesolophs are present, and on one island (Granito) the frequency is
100%. Selection has evidently acted to change the two structures independently and in
opposite directions. (2) P. guardia also exhibits a relatively complex phallus (see above). A
protrusible tip and dorsal lappets, although poorly developed when compared with those
structures in boylei or crinitus, are present, which suggests that complex phalli can evolve
from simple phaUi. Also, mice from Isla Salsipuedes (interparietalis) have developed
118 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
ventral lappets. The evidence further indicates that acrocentric chromosomes are derived
from bi-armed chromosomes in GuU Peromyscus. \n guardia, the Y chromosome and one
pair of autosomes are acrocentric. No other island or mainland forms studied, other than
crinitus (Hsu and Arrighi, 1968), boylei. or stephani, have acrocentric chromosomes, and
in the latter species nearly the entire complement of chromosomes are acrocentric (only the
sex chromosomes and three pairs of autosomes are bi-armed).
Whether the trends noted above are characteristic of evolutionary changes of these
features in other peromyscines or in other rodents is a moot question, but this may be the
case in the development of accouterments in the phallus of P. eva (Lawlor, in press). The
trends may represent reversals from the normal pattern of change in Peromyscus. In any
event, the overwhelming evidence based on overall similarity indicates that the phalli,
dentition, and chromosomes of guardia and interparietalis did evolve in the above ways. If
the alternative hypothesis is invoked, namely that guardia and interparietalis are consid-
ered derivatives of an ancestor having a complex phallus and dentition and a chiefly
acrocentric chromosome complement (e.g., a crinitus-Wkt form), then convergences of
many other characters must have occurred (e.g., osteologic and pelage characters, etc.).
The latter seems highly unlikely. Moreover, the probability is quite low that such
convergences occurred in all three populations of guardia while in stephani, which occurs in
seemingly similar habitat on an island that is as well isolated and is of approximately the
same age, none are observed. I regard the similarities of guardia and eremicus as indicators
of genetic relationship and view the derivation of guardia in the most parsimonius manner,
namely that it is derived from an eremicus-Wkt progenitor.
Evidence from morphology of chromosomes and male accessory reproductive struc-
tures suggests that eremicus and closely related species may share characters that are
primitive for Peromyscus. Members of the subgenus Haplomylomys (excepting crinitus)
are the only species having a complete complement of male accessory reproductive features
(Linzey and Layne 1969). In all other species of the genus one or more elements are absent
or vestigial. Except in guardia, acrocentric chromosomes are absent in mice of the
subgenus. Although practically nothing is known about chromosome evolution in Per-
omyscus, particularly in view of the fact that Robertsonian fusion cannot be invoked (Hsu
and Arrighi, 1966; 1968), the data at least are not inconsistent with the view that the
presence of acrocentrics is a derived condition. Hsu and Arrighi (1968) presented a
hypothetical phylogeny of Peromyscus that describes the evolution of chromosomes as
resulting from a primitive acrocentric condition, but they noted (p. 437) that the phylogeny
was presented in that manner principally for convenience, stating that chromosome
evolution in Peromyscus may have occurred in either direction. Information on muscula-
ture (Rinker, 1963) also supports the view that Haplomylomys may be a primitive
peromyscine group. Most of the conditions of the musculature that Rinker considered
primitive are present in that subgenus. The evidence presented in this study suggests that
complexities of the teeth and phallus derive from simple conditions and that acrocentric
chromosomes derive from a bi-armed condition, at least in the species examined. Most of
the Haplomylomys studied herein exhibit simple conditions of those structures. These data
and those presented above support the contention of Linzey and Layne (1969) that
Haplomylomys contains primitive members of the genus.
HISTORICAL PERSPECTIVE
The deserts of western North America, with which the origin and divergence of P.
eremicus and related forms are closely associated, resulted chiefly from rain shadows
produced by extensive mountain building in that area beginning in the Triassic and
I
1971 LAWLOR: Peromyscus 119
continuing to the Pleistocene (King, 1958). However, adequate conditions to support
lowland desert forms like eremicus probably did not exist prior to the formation of the
North American deserts in mid-Pliocene (Axelrod, 1948). Undoubtedly these deserts were
further modified by glacial advances and retreats during the Pleistocene, so that relatively
stable desert conditions probably did not arise until early or middle Pleistocene, when
successive glacial maxima became milder and interglacial periods were characterized by
increasingly drier conditions. Displacement of desert elements by the Madro-Tertiary flora
(e.g., thorn-scrub) during glacial advances in the early Pleistocene probably resulted in the
separation of prototypes of merriami and eremicus and accounts for the differences in their
habitat preferences today (Lawlor, in press).
The history of the Gulf of California is not well documented. Although certain authors
(e.g., Durham and Allison, 1960) consider the Gulf to be as old as the Cretaceous orogeny
in North America and that it reached its present configuration by the beginning of the
Pliocene, recent investigations of the southern Gulf floor (Larson et al., 1968; Moore and
Buffington, 1968) suggest that the majority of crustal movement occurred since middle or
late Pliocene. A proto-gulf is indicated, however, by earlier Pliocene fossil beds located in
northern parts of the Gulf. In any case, the northern deep-water islands in the Gulf may not
have originated until late Pliocene or early Pleistocene. For example, sedimentary beds of
relatively recent deposition are known from the Lorenzos (early Pliocene) and Angel de la
Guarda (late Phocene) (Anderson, 1950), indicating that the islands were submerged in a
shallow water embayment or saline lake at the time. The geologic relationships of these
islands to adjacent submarine troughs suggests that the islands may have resulted partly
from elevation along faults (Shepard, 1950). Part of this uplift was probably Pleistocene
(Ibid.). The present separation of Islas Tiburon and Turner from the Sonoran mainland
was likely attained with the last glacial retreat {ca. 1 5,000 years ago).
P. eremicus and related desert forms probably did not originate until formation of the
deserts in mid-Pliocene. Consequently, evolution and radiation of this group on mainland
and island areas has been relatively recent and no doubt has been substantially affected by
displacement and expansion of the lowland deserts during the Pleistocene. In this
connection, the suggested origin and radiation of these mice corresponds closely to that
described for the lizard genus Uta (Ballinger and Tinkle, in press).
ZOOGEOGRAPHIC RELATIONSHIPS
Relationships between guardia, interparietalis, and eremicus are consistent with
Banks' (1967) contention that guardia and interparietalis probably had separate origins
from a mainland eremicus-Wkt stock. Furthermore, Peromyscus has not been taken on two
islands (Isla Partida and Raza) that are located between the two groups of islands
supporting guardia and interparietalis, although several people have collected on each (I
have collected only on Isla Partida). This suggests that guardia and interparietalis do not
represent isolates of a form once continuously distributed among these islands, but rather
that they are of separate mainland origin (Ibid. ).
The time interval between initial isolation of the island populations from the mainland
is probably the principal factor affecting the degree of divergence of northern island forms
in the Gulf of California. This seems to be the case for the following reasons: (1)
Morphological divergence is at least broadly related to temporal differences in island
formation. Angel de la Guarda and its satellite islands, the Lorenzo group of islands, and
San Esteban have been separated from the mainland for a considerable length of time.
Islas Tiburon and Turner most certainly are no older than late Pleistocene. P. eremicus-
like forms on the older groups of islands (guardia and interparietalis) are more divergent
120
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
than populations on younger islands (eremicus tiburonensis and collatus). (2) Apparently
gene flow is minimal or absent among islands and between islands and the mainland.
Differences persist among populations separated by very short distances, and because of
habitat similarities and high population densities these differences probably are not
attributable to presently existing differential selection coefficients or genetic drift. For
example, it was pointed out above that populations from Islas San Lorenzo Sur and San
Lorenzo Norte differ significantly (P<.05) in many morphometric characters; yet these
islands are separated by 100 yards at most, and at low tide half-submerged rocks project
above the water for almost the entire distance. Furthermore, ventral lappets on the phallus
persist in most individuals of interparietalis from Isla Salsipuedes, while this feature
evidently is absent in other nearby populations of that species. Isla Tiburon is separated
from the Sonoran mainland by as little as two miles and by shallow water (only six meters
at certain places); yet differences in phalli (and perhaps in karyotypes) are evident between
the mice there and on the adjacent mainland (e.g., BahTa Kino); similar differences also
persist between mice on Tiburon and those on Turner. In guardia. dramatic differences in
dental patterns and morphometric characters are evident. It appears that distance effects
owing to differential gene flow, although perhaps important in early colonization and
establishment of the island populations, have been relatively unimportant in shaping
present characteristics of the island forms. (3) Differences among island populations
seemingly are not explicable only in terms of habitat differences. The northern part of the
Gulf is marked by overall floral uniformity (Shreve and Wiggins, 1951; Felger, 1966). In
addition, xeric rocky habitats are characteristic of all the islands inhabited by Peromyscus.
Table 4. Zoogeographic relations of species of mammals on northern islands in the Gulf of California.
Only island species having mainland relatives on one side of the Gulf are compared. No mammals other
than Peromyscus stephani and Rattiis norwegicus are known from Isla San Esteban.
65
S
.5
1
•S
1
1
.1
1
E
Cn
Cg
Co
ex
^
^
•I-*
S
s
S
4i
^•l*
Q
6n
c
•s
•s
•5
a
.Si,
■^
Locality
1
r
<3
1
a
1
1
"3
to
a
a!
a!
a!
^
:?:
Q
■-J
Western islands:
Granito
?
?
Mejia
X
?
Angel de la Guarda
X
X
Salsipuedes
?
San Lorenzo Norte
X
San Lorenzo Sur
X
Eastern islands:
Tiburon
X
X
X
X
X
Turner
x>
?
X'
Mainland:
Baja California
X
X
Sonora
X
X
X
X
X
ifhis population originally was described as P. penicillatus (Burt, 1932) but evidently is intermedius (Patton, pers.
comtn.)
^Called N. varia, but closely related to albigula (Burt, 1932)
1971 LAWLOR: Peromyscus 121
On the basis of its degree of divergence and its phylogenetic relationships to merriami,
guardia evidently has long been separated from an eremicus-WkQ ancestor. A prototype of
merriami is thought to have arisen in early Pleistocene (see above) and the cladistic
relationship between guardia and merriami indicates that they probably share a common
ancestry. Thus, guardia probably represents a derived form of a stock that gave rise to
merriami and that also colonized Angel de la Guarda and satellite islands in the early
Pleistocene. P. interparietalis evidently is more recently derived from a mainland ere-
micus-Yxko, form, possibly in middle to late Pleistocene. Mice from Islas Turner and
Tiburon undoubtedly arose as a result of isolation of the two islands when the last major
increase in sea level took place. P. stephani presumably has been isolated for some time
(probably as long as interparietalis). It probably reached San Esteban from the eastern
mainland during a glacial maximum in the Pleistocene when Isla Tiburon was part of the
continent. Considering the present distribution of boylei, the initial colonization of Isla
San Esteban by a boylei-\\V.Q form was probably also associated with more mesic habitats
at that time.
Early colonization and evolution of the island forms was likely erratic and unstable,
and effects of distance between islands and the mainland, island area, and population size
on genetic change were doubtless substantial. Once established, however, it appears that
the island populations maintained their morphologic (and presumably genetic) integrity,
and that the low rates of gene flow that obtain between the different island and mainland
populations are unable to effect major changes in morphologic features.
The presence of a particular species on an island appears to me to result from
historical accident. I cannot explain the absence of an eremicus-Wko. form from an island
like San Esteban, with its xeric, rocky habitat and floral composition similar to other
Sonoran Gulf islands (Felger, 1966). Perhaps eremicus and boylei, or forms closely related
to them, are competitors. Circumstantial evidence concerning the status of mice on Isla
San Pedro Nolasco, where both boylei glasselli and an eremicus-Yxko, form (pembertoni)
are known, suggests that boylei may be competitively superior io pembertoni. I collected
there twice in the summer of 1967 and was unable to obtain pembertoni, although Burt
(1932) look pembertoni and boylei in about equal numbers. This fragmentary information
suggests that boylei may be supplanting pembertoni there, although the habitat, consisting
of open slopes with cacti and low brush, and ravines of dense grass, is one of the most
diverse of the northern islands.
If the above evidence is indicative of a competitive superiority of boylei-Vike forms,
then P. stephani, owing to its occurrence on Isla San Esteban, may have acted as a barrier
to dispersal of eremicus-VikQ forms across the Gulf. Distinct morphologic differences do
exist between eastern and western island eremicus-Wke forms (see above), suggesting few
such crossings have been made.
TAXONOMIC CONCLUSIONS
I concur with Banks (1967) in considering interparietalis and guardia distinct from
one another and from eremicus. Although interparietalis is evidently much less removed
from the presumed eremicusAike ancestor than guardia, on morphologic and zoogeogra-
phic grounds it seems worthy of specific status. On the other hand, collatus (Isla Turner) is
very similar to mainland and Tiburon eremicus. Excepting the difference in the X
chromosome, differences that separate the two species are subtle and are reminiscent of
geographic variation exhibited by mainland populations of eremicus (Lawlor, in press).
The mice from Isla Turner should bear the name P. e. collatus. The relationships of
merriami and eremicus, based on osteology and morphology of the phallus, are discussed
122 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
elsewhere (Lawlor, in press; see also Hoffmeister and Lee |1963], and Commissaris [1960]).
The additional information regarding blood proteins and karyology and the phenetic and
phyiogenetic relations of the two species presented here support earlier conclusions that
merriami, although morphologically distinct, exhibits close affinities to eremicus and is
probably derived from a progenitor similar to that species. All of the above species are
members of the subgenus Haplomylomys.
P. stephani clearly is a close relative of boylei and should be placed with boylei in the
subgenus Peromyscus. In my view stephani should be retained as a species.
ACKNOWLEDGMENTS
For the loan of specimens 1 am indebted to the following individuals and their respective institutions: Dr. J.
Knox Jones, Jr., Museum of Natural History, University of Kansas: Dr. E. Lendeli Cockrum, Department of
Zoology, University of Arizona; Dr. Seth Benson, Museum of Vertebrate Zoology, University of California,
Berkeley; Drs. Richard H. Manville and John L. Paradiso, United States National Museum; Dr. Joseph R. Jehl,
Jr., San Diego Natural History Museum; Dr. Thomas R. Howell, University of California, Los Angeles; and Dr.
Robert T. Orr, California Academy of Science. For manuscript review and counsel I am especially grateful to
Drs. William H. Burt, Claude W. Hibbard, Donald W. Tinkle, Morris Foster, Emmet T. Hooper, and Arnold G.
Kluge of the Museum of Zoology, University of Michigan; to Carl Welser, Mammalian Genetics Center,
University of Michigan; and to Albert Bennett, intrepid explorer. I also extend thanks to my wife, Bertie, for help
in preparation of the manuscript.
Support for research and field investigations was provided in part by a U.S. Public Health Service
predoctoral fellowship (No. l-Fl-GM-37, 761-01), by a grant to the University of Michigan from the National
Science Foundation for research in Systematic and Evolutionary Biology (NSF GB-6230), and by a research
grant from Sigma Xi.
SPECIMENS EXAMINED
Specimens employed for analysis of blood proteins and karyology were collected alive and maintained at the
University of Michigan. Most of these mice are preserved in the Museum of Zoology, even though not cited
below. Sample sizes and localities from which these specimens were obtained are given in the appropriate tables
above. Only specimens obtained for purposes of examining morphologic features of the pelage, phalli, osteology,
and soft anatomy are listed below. Where appropriate, numbers in parentheses identify the localities on the map
(Fig. 1).
P. boylei.— ARIZONA: Marble Park, Catalina Mts., Pima Co. (9), 24 (UMMZ). SONORA: Isla San
Pedro Nolasco (14), 7 (UMMZ).
P. crinitus.— CALIFORNIA: Paiute Creek, Inyo Mts.. Inyo Co. 28 (UMMZ).
Z'. er?m/CM.?.— BAJA CALIFORNIA: Turtle Bay (1), 10(3 SD, 3 USNM, 4 UMMZ); Barril (2), 10 (SD);
Bahi'a de los Angeles (3), 23 (2 SD, 1 UCLA, 20 UMMZ); San Francisquito (4), 16 (USNM); El Marmol (5), 14
(CAS); San Telmo (6), 7 (UMMZ). CALIFORNIA: Escondido, San Diego Co. (7), 2 1 (KU). SONORA: Puerto
Penasco (8), 20 (17 SD, 3 UA); Imuris (10), 9 (KU); Puerto Libertad (II), 20 (2 KU, 18 SD); Punta Sargento
(12), 24 (UCLA); Bahfa Kino (13), 14 (2 UA, 6 KU, 6 UMMZ); Presa Obregon (15), 13 (10 KU, 3 UMMZ); Isla
Tiburon,45(4CAS, 3 KU, 1 SD, 19 UC, 6 UCLA, 12 UMMZ); Isla Turner, 37 (1 CAS, 4KU,3SD, 15 UCLA,
14 UMMZ).
P. guardia. BAJA CALIFORNIA: Isla Angel de la Guarda, 28 (II SD, 10 UCLA, 7 UMMZ); Isla
Granito,40(7SD, 33UMMZ); Isla Mejia, 17 (5 CAS, 3 SD, I UCLA, 8 UMMZ).
P. interparietalis. - BAJA CALIFORNIA: Isla Salsipuedes, 48 (I CAS, 13 SD, 34 UMMZ); Isla San
Lorenzo Norte, 41 (19 SD, 22 UMMZ); Isla San Lorenzo Sur, 46(7 UA, II CAS, 16 SD, 7 UCLA, 5 UMMZ).
P. merriami.-^ SONORA: Presa Obregon ( 15), 20 ( 17 KU, 3 UMMZ).
P. stephani. SONORA: Isla San Esteban, 37 (2 CAS, 1 SD, 19 UCLA, 15 UMMZ).
The locality specified as "near Tucson" for boylei (Table 2) refers to Marble Park, Catalina Mts., Pima Co.,
Arizona, and Molino Canyon, 18 mi, NE Tucson, Catalina Mts., Pima Co., Arizona. The same designation
(Table 2) for merriami refers to 3/4 mi. SE San Xavier Mission, Pima Co., Arizona.
LITERATURE CITED
Anderson, C. A.
1950. 1940 E. W. Scripps cruise to the Gulf of California. Part 1. Geology of islands and neighboring land
areas. Mem. Geol. Soc. Amer. 43: viii + 53 pp.
1971 LA'WLOR: Peromyscus 123
Axelrod, D. I.
1948. Climate and evolution in western North America during the middle Pliocene time. Evolution 2: 127-
144.
Ballinger, R. E.. and D. W. Tinkle
The systematics and evolution of the genus Uta (Sauria: Iguanidae). Misc. Publ. Mus. Zool., Univ.
Michigan, in press.
Banks. R.C.
1967. The Peromyscus guardia-interparietalis complex. J. Mammal. 48: 210-218.
Brand, L. R., and R. E. Ryckman
1969. Biosystematics of Peromyscus eremicus. P. guardia. and P inlerparietalis. J. Mammal. 50: 501-513.
Brown, J. H., and C. P. Welser
1968. Serum albumin polymorphisms in natural and laboratory populations o{ Peromvscus. J. Mammal.
49: 420-426.
Burt. W. H.
1932. Descriptions of heretofore unknown mammals from islands in the Gulf of California, Mexico. Trans.
San Diego Soc. Nat. Hist. 7: 161-182.
Commissaris, L. R.
1960. Morphological and ecological differentiation of Peromyscus merriami from southern Arizona. J.
Mammal. 41: 305-310.
Durham, J. W., and E. C. Allison
1960. The biogeography of Baja California and adjacent seas. Part I. Geologic history. The geologic history
of Baja California and its marine faunas. Syst. Zool. 9: 47-9 1 .
Edwards. A. W. P., and L. L. Cavalli-Sforza
1964. Reconstruction of evolutionary trees. Syst. Assoc. Publ. 6: 67-76.
Parris, J. S.
1966. Estimation of conservatism of characters by constancy within biological populations. Evolution, 20:
587-591.
1967. The meaning of relationship and taxonomic procedure. Syst. Zool. 16: 44-51.
1970. Methods for computing Wagner Trees. Syst. Zool. 19: 83-92.
Pelger, R. S.
1 966. Ecology of the gulf coast and islands of Sonora, Mexico. Ph. D. Thesis, Univ. Arizona, 460 p.
Hershkovitz, P.
1962. Evolution of neotropical cricetine rodents (Muridae) with special reference to the phylotine group.
Pieldiana Zool. 46: 1-524.
Hoffmeister, D. P.
1951. A taxonomic and evolutionary study of the pinon mouse, Peromyscus truei. Illinois Biol. Monog. 21:
X + 104 p.
Hoffmeister, D. P., and M. R. Lee
1963. The status of the sibling species Peromyscus merriami and Peromyscus eremicus. J. Mammal. 44:
201-213.
Hooper, E. T., and G. G. Musser
1964a. The glans penis in Neotropical cricetines (Pamily Muridae) with comments on classification of muroid
rodents. Misc. Publ. Mus. Zool., U. Michigan 123: 1-57.
1964b. Notes on classification of the rodent genus Peromyscus. Occas. Papers Mus. Zool., Univ. Michigan
635:1-13.
Hsu, T. C, and P. E. Arrighi
1966. Chromosomal evolution in the genus Peromyscus (Rodentia: Cricetidae). Cytogenetics 5: 355-359.
1968. Chromosomes of Peromvscus (Rodentia: Cricetidae) I. Evolutionary trends in 20 species. Cytogenet-
ics 7: 417-446.
King, P. B.
1958. Evolution of modern surface features of western North America. Publ. Amer. Assoc. Adv. Sci. 51:3-
60.
Kluge, A.G.
The evolution and geographical origin of the New World Hemidactylus mabouia-brookii complex
(Gekkonidae, Sauria). Misc. Publ. Mus. Zool., Univ. Michigan, in press.
Kluge, A. G., and J. S. Parris
1969. Quantitative phyletics and the evolution of anurans. Syst. Zool. 18: 1-32.
Larson, R. L., H. W. Menard, and S. M. Smith
1968. Gulf of California: a result of ocean-floor spreading and transform faulting. Science. 161: 781-784.
124 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Lawior, T. E.
1964. The Yucatan deer mouse, Peromyscus yucatanicus. Master's Thesis, Univ. Kansas, 45 p.
Distribution and relationships of six species of Peromyscus in Sonora and Baja Cahfornia, Mexico.
Occas. Papers Mus. Zool., Univ. Michigan, in press.
Linzey, A. V., and J. N. Layne
1969. Comparative morphology of the male reproductive tract in the rodent genus Peromyscus (Muridae).
Amer. Mus. Novitates 2355:1-47.
Matthey, R.
1951. The chromosomes of the vertebrates. In: M. Demerec (ed.). Advances in genetics 4: 159-180.
Mearns, E. A.
1897. Descriptions of six new mammals from North America. Proc. U.S. Natl. Mus. 19: 719-724.
Moore, D. G., and E. C. Butfington
1968. Transform faulting and growth of the Gulf of California since the late Pliocene. Science 161: 1238-
1241.
Osgood, W. H.
1909. Revision of the mice of the American genus Peromvicu.y. N. Amer. Fauna 28: 1-285.
Patton, J. L.
1967. Chromosome studies of certain pocket mice, genus Perognathus (Rodentia: Heteromyidae). J.
Mammal. 48: 27-37.
Preston, F. W.
1962. The canonical distribution of commonness and rarity: Parts I and II. Ecology 43: 185-215 and 410-
432.
Prim, R.C.
1957. Shortest connection networks and some generalizations. Bell Syst. Tech. J. 36: 1389-1401.
Rasmussen, D. I., J. N. Jensen, and R. K. Koehn
1968. Hemoglobin polymorphism in the deer mouse, Peromyscus maniculalus. Biochem. Genetics 2: 87-92.
Rinker, G. C.
1963. A comparative myological study of three subgenera of Peromyscus. Occas. Papers Mus. Zool., Univ.
Michigan 632: 1-18.
Shepard, F. P.
1950. 1940 E. W. Scripps cruise to the Gulf of California. Part III. Submarine topography of the Gulf of
California. Mem. Geol. Soc. Amer. 43: viii -f 32 p.
Shreve, F., and I. L. Wiggins
1951. Vegetation of the Sonoran desert. Publ. Carnegie Inst. Wash., 591: xii + 192 p.
Smithies, O.
1955. Zone electrophoresis in starch gels: group variations in the serum proteins of normal human adults.
Biochem. J. 61: 629-641.
Sokal, R. R.,and P. H.Sneath
1963. Principles of numerical taxonomy. W. H. Freeman and Co., San Francisco, 359 p.
Soule, M., and A. J. Sloan
1966. Biogeography and distribution of the reptiles and amphibians on islands in the Gulf of California,
Mexico. Trans. San Diego Soc. Nat. Hist. 14: 137-156.
Townsend, C. H.
1912. Mammals collected in Lower California with descriptions of new species. Bull. Amer. Mus. Nat. Hist.
31: 117-130.
Department of Biology, Humboldt State College. Areata, California, 95521 .
LAMPETRA (ENTOSPHENUS) LETHOPHAGA,
NEW SPECIES, THE NONPARASITIC DERIVATIVE
OF THE PACIFIC LAMPREY
CARL L HUBBS
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 6 30 APRIL 1971
LAMPETRA (ENTOSPHENUS) LETHOPHAGA, NEW SPECIES,
THE NONPARASITIC DERIVATIVE OF THE PACIFIC LAMPREY
CARL L. HUBBS
ABSTRACT. — The Pacific lamprey, Lampetra (Enlosphenus) tridentata, is now shown to agree with
most parasitic species of the Petromyzoniformes in having evolved into a nonparasitic derivative, L. (E.)
lethophaga. Although the parasitic form ranges widely, from central Baja California around the North
Pacific periphery to southern Japan, varying greatly in adult size, the dwarfed nonparasitic form seems to be
confined to the contiguous drainage basins of the Pit River (a Sacramento River headwater) in northeastern
California, both above and below the Pit River Falls, and to the upper Klamath River system in south-
central Oregon. These two drainage basins harbor additional endemic fishes, and have certain other faunal
features in common. The distributions of the three nonparasitic lampreys in the drainage basins around the
North Pacific appear to be complementary.
L. lethophaga contrasts rather sharply with the dwarfed, probably resident types of L. tridentata in the
Klamath system, as well as with the large, sea-run populations. However, a specimen from Willow Creek in
the Lost River system of Oregon is possibly intermediate between L. lethophaga and the dwarf parasitic
types in the Klamath River system; and a parasitic form of the same group, of Miller Lake, in a disjunct
section of the Klamath River system, is reported to be even more dwarfed than L. lethophaga. Some
intergradation between the parasitic and nonparasitic stocks is not excluded.
The dentition of the nonparasitic form exhibits features both of reduction and of increased individual
variation, probably along with some geographical differentiation.
Like other lampreys, the new species no doubt exists for several years in the larval (ammocete) stage
before metamorphosing in the autumn. The gonads ripen as the gut atrophies. The dwarf adults after
overwintering appear on circumstantial evidence either ( 1 ) to undergo the typical nuptial metamorphosis to
spawn in the following spring, or (2) to attain maturity neotenically while retaining the prenuptial state of
pigmentation and body form, and to spawn over the summer months, or even after overwintering again.
There are indications that lamprey species are subject to regional diversity, and that some of the
speciation has been of a mosaic type.
RESUMEN. — Se demuestra que la lamprea del Pacifico, LMmpetra (Entosphetius) tridentata concuerda en su
evolucion con la mayor parte de las especies parasiticas de Petromyzoniformes, produciendo un derivado no
parasitico, L. (E.) lethophaga. Las formas parasitas presentan una amplia distribucion geografica, exten-
diendose a lo largo de la zona periferica del Pacifico Norte, desde la parte central de Baja California hasta la
zona meridional del Japon. Los adultos de estas formas ofrecen una gran variacion de tallas. Las formas
enanas libres, no parasiticas, estan al parecer confinadas a las cuencas fluviales contiguas del rio Pit (uno de
los tributarios de la parte alta del rio Sacramento) en la zona nordeste de California, a ambos lados de las
cataratas del rio Pit, y en la parte alta del sistema del rio Klamath en la zona centro-meridional de Oregon .
Estas dos cuencas fluviales albergan tambien otros peces endemicos, presentando asi mismo otras
caracteristicas faunisticas comunes. Las tres lampreas no parasiticas que habitan las cuencas fluviales que
bordean el Pacifico Norte, presentan al parecer una distribucion complementaria.
L. lethophaga contrasta notablemente con las formas enanas, probablemente tipos residentes de L.
tridentata en la red fluvial del Klamath, asi como tambien con las especies de talla grande correspondientes a
poblaciones oceanicas. Sin embargo, un ejemplar procedente de Willow Creek, en la red fluvial del rio Lost,
en Oregon, es posiblemente una forma intermedia entre L. lethophaga y los tipos parasiticos y enanos del
sistema del rio Klamath. Una forma parasitica de este mismo grupo aparece en el lago Miller (seccion
disyuntiva de la red fluvial del rio Klamath). Se ha establecido ya, que dicha forma es aun mas pequeiia que
L. lethophaga. Desde luego no puede excluirse la posibilidad de que exista una intergraduacion entre las
poblaciones parasitas y libres (no parasiticas).
Las formas libres presentan variaciones individuals de reduccion o de incremento en la denticion,
caracteristicas probablemente relacionadas con otras diferencias geograficas.
SAN DIEGO SOC. NAT. HIST., TRANS. 16 (6): 125-164, 30 APRIL 1971
126 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Al igual que sucede en otras lampreas, esta nueva especie permanece induablemente durante varies aiios
en la fase larval {amoceto)antes de llegar a la metamorfosis, que tiene lugar en el otono. Al madurar las
gonadas, el tubo digestive se atrofia. Los adultos enanos despues de pasar el invierno, evidencian cualquiera
de las siguientes circunstancias: 1) que pasan la metamorfosis nupcial tipica para asi desovar en la
primavera siguiente, o 2) alcanzan una madurez neotenica, es dicir, retienen la pigmentacion y forma del
cuerpo de la fase prenupcial, desovando entonces durante los meses de verano, o aun mas tarde, despues de
pasado el invierno.
Las observaciones obtenidas indican que las especies de lampreas presentan dlversidad regional, y en
algunos casos la especiacion corresponde al tipo de mosaico.
Although I discovered a dwarfed, nonparasitic derivative of the Pacific lamprey,
Lampetra (Entosphenus) tridentata (Richardson) in 1934, in the Pit River system of
northern California and in the upper Klamath River system in southern Oregon, and
although it has been distinguished by Bond (1961: 14) in key form from L. tridentata. from
the same river systems, it has not yet been assigned a species-group name. Bond merely
designated it "Klamath brook lamprey, Lampetra sp." With the particular need of making
the name and the status of this form available for a forthcoming treatment of the
distribution, phylogeny, and taxonomy of lampreys (Hubbs and Potter, in press), it is now
belatedly made known as:
PIT-KLAMATH BROOK LAMPREY
Lampetra (Entosphenus) lethophaga, new species
Entosphenus tridentatus (misidentification). — Rutter, 1908: 120 (material listed from
"South Fork Pitt River" only).
Lampetra planeri (misidentification). — Hubbs, 1925: 594 (size of recently transformed
specimen from "North Fork of Pitt River").
Lampetra sp. — Bond, 1961: 14 ("Klamath brook lamprey"; "Klamath and Pit River
systems").
Holotype, U. Mich. Mus. Zool. 130648, and paratypes, UMMZ 130649, from source
of Fall River, a tributary to Pit River, in Shasta County, California (as specified under
Location 2, below).
This species is illustrated in Figures 1, 2 A-B, and 6; its range and habitat in Figures 3
and 4; its size in Figure 8. Figures 2 C-D, 3, 5, 7, and 8 pertain in part or in toto to related
forms.
Diagnosis. — The following diagnosis largely follows the sequence of characters utilized by
Hubbs and Potter (in press) in their analysis of the lampreys of the world.
A petromyzonid lamprey agreeing with Lampetra (sensit lato) in having: the extraoral
teeth not in regular alate rows, the lateral and posterior fields of disc essentially toothless
between circumorals and marginals, the teeth of the anterior field few and scattered, none
of the teeth villiform, the supraoral markedly dilated, the anterior circumorals normally 5,
the total anterior and lateral circumorals usually 13, and the lateral circumorals more or
less dilated. Agreeing with subgenera Lethenteron and Entosphenus in having the laterals
connected by the posterior circumorals, and agreeing with Entosphenus in having 4 lateral
circumorals on each side, one or more outer posterior circumorals often bifid, the
supraoral often with a median cusp, the transverse lingual lamina almost rectilinear and
with median cusp not strongly enlarged, and the marginals and posterior circumorals often
in an irregular file. Differing from the complex now passing as Lampetra tridentata in
being nonparasitic (not feeding or growing after the fall metamorphosis, but developing the
gonads as the gut atrophies prior to spawning in the next spring or summer, or even later,
and then dying), and in being much reduced in size at maturity (less than 170 mm), and in
some places (including the type locality) breeding in prenuptial coloration and body form;
1971 HUBBS: A NEW NONPARASITIC LAMPREY 127
also differing from L. tridentata in having the mouth small (disc length less than 5 percent
of total length) and usually much puckered, the median cusp of supraoral often weak or
absent, the cusps on the lateral circumorals often reduced by 1 on any of the four teeth
from the formula 2-3-3-2, the posterior circumorals reduced in number (9 to 15), and
the anterior intermediate disc teeth, between anterior circumorals and marginals, very few
(only 4 in specimen shown for dentition as Figure 6).
MATERIAL
The considerable amount of material (Table 1) referred to Lampetra lethophaga has
come from various places in the Pit River system of northeastern California and in the
Klamath River system in south-central Oregon (Figure 3). The available information on
the habitats at the 1 1 localities, 5 in the Pit system and 6 in the Klamath, and on the
associated fish species and the circumstances of the collecting, is detailed because of the
bearing that this information has on the interpretation of the distribution, environment,
variation, and life history of the species. The localities are listed separately for the Pit and
Klamath systems, in each basin from upstream downward.
Material used in this study has been deposited in the following institutions: CAS,
California Academy of Sciences; CU, Cornell University; OS, Oregon State University;
SIO, Scripps Institution of Oceanography; SU, Stanford University (material now
transferred to California Academy of Sciences); UMMZ, University of Michigan Mu-
seum of Zoology; USNM, United States National Museum.
LOCATIONS IN PIT RIVER SYSTEM, NORTHERN CALIFORNIA
1. North Fork of Pit (formerly "Pitt") River at mouth of Joseph Creek, near Alturas, Modoc County,
collected by Cloudsley Rutter and Fred M. Chamberlain, September 4, 1898. These data are taken from the label,
but the specimens may have come instead from the South Fork of Pit River, for Rutter (1908: 120) failed to list
the North Fork among the collections entered for "Entosphenus tridentatus." but did include it for "South Fork
Pitt River (South Fork P.O., Jesse Valley"); also collected by Rutter and Chamberlain(the location of "Jess
Valley," as now mapped at altitude of ca. 1585 m is located by a question mark on the distributional map. Figure
3). In any event, it seems almost certain that Rutter's record was based on L. lethophaga.
The 2 specimens (UMMZ 55316) making up this collection, received from Stanford University, comprise a
female 142 mm long, in early stage of transformation, with eggs few enough to indicate a nonparasitic form, and
an ammocete 105 mm long, with minute ova. The female was recorded as 138 mm long, under the
xxwi\AQni\'acdiiox\ o{ Lampetra planeri. by Hubbs(1925: 594).
Associated species reported by Rutter are, for North Fork, Catostomus occidenlalis Ayres, Rhinichthys
osculus (Girard) subsp. (as "Agosia robusta"). and Salmo gairdnerii Richardson (as "S. irideus" ): and, for South
Fork, Salmo gairdnerii. Rhinichthys osculus subsp. (as "Agosia robusta"), Gila bicolor (Girard) subsp. (as
"Rutilus bicolor"), and Cottus pitensis Bailey and Bond (as "C gulosus"). This Coitus record has been referred
by Bailey and Bond (1963: 20) to their new species, C. pitensis. which is endemic in the Pit and Little Sacramento
river systems.
2. Head of Fall River, in the west-central part of T 38 N, R 4 E, near the northeastern corner of Shasta
County, close to the settlement of Dana and about 5 km north of Fort (Soldier) Mountain; altitude ca. 1020 m.
This sizable stream (in the river proper about 50 m wide and uniformly about 0.7 m deep), flowed with a slight to
moderate current. It originated in a partly forest-bordered, naturally ponded pocket of springs (Figure 4). Above
the spring-fed origin of the river, the stream course (known as Bear Creek, though labelled "Fall River" on some
maps) is intermittent; it was dry when examined in the very dry year of 1934. Locally we heard it claimed that the
big springs arise from Tule Lake (presumably not the small "Tule Lake" close by to the east) and Lost River
(both in the Klamath River system far to the north). However, it seems plausible that the source lies at least in
part in the extensive lava beds immediately to the northwest, in southeastern Siskiyou County.
This large cold stream has doubtless been a holdout, during periods of desiccation, of relict species. The
stream, within 0.5 km of the springs, yielded, in addition to the lampreys, the endemic sculpin Coitus macrops
Rutter (1908: 146-147, fig. 4) and C. asperrimus (misspelled "asperrima") Rutter (1908: 144-145, fig. 3), both
closely related to endemic species of the Klamath River system. Rutter's list also included Salmo gairdnerii.
along with dried remains of Catostomus occidentalis. from about a lateral spring, that indicated a prior breeding
run of this sucker. The sculpin that Rutter (1908: 146) reported from Fall River as "C. gulosus" has been referred
by Bailey and Bond to their C. pitensis.
128
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 1. Types of Lampelra lethophaga. from head of Fall River, Shasta County, California (Location 2): A,
holotype, UMMZ 130648, a mature, neotenic male 128 mm in total length, in side view. B, same specimen, in
oblique view, with abdominal wall pinned aside, to show enlarged, lobular testis and atrophic gut bearing signs of
hemorrhages. C, paratype, in series UMMZ 130649, a fully mature, neotenic female 1 16 mm long, in side view.
D, same specimen, in oblique view, with abdominal wall pinned aside, to show celome packed full of ripe ova
aligned in alate rows.
1971
HUBBS: A NEW NONPARASITIC LAMPREY
129
Figure 2. Nuptial males of genus Lampetra. subgenus Entosphenus: A, Lampetra lethophaga. OS 2856
(specimen KOOlO), 154 mm in total length, in side view; from Crooked Creek, Klamath County, Oregon
(Location 11). B, same specimen, in ventral view of head region, enlarged. C, Lampetra sp., seemingly
intermediate between L. lethophaga and precocious forms of L. tridentata: SIO 65-144, 176 mm long, in side view;
from Willow Creek, tributary to Clear Lake Reservoir, Modoc County, California. D, same specimen, in ventral
view of head region, enlarged; with mouth pressed open.
130 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
The water was so crystal clear that the bottom seemed to rise ahead. Vegetation comprised patches of
Myriophyllum and very thick clumps of moss on lava rocks. The bottom in the stream was fine gravel and sand,
with scattered lava rocks. The temperature was 11 .4° C in the current from a spring and 1 3.3° C in the river, when
the air temperature was 29.4° C. The collection (M34-135) was made by Carl L. Hubbs and family on August 17,
1934, using 4-foot and 6-foot Common Sense woven-meshed seines. One full-grown ammocete and 3 transformed
adults came from weeds in the river; one adult was in muddy sand along the bank; the others, all adults, were
taken under flat stones lying on clean, coarse gravel in the current from a lateral spring, mostly from under one
stone where, when the stone was first turned, they looked like a breeding pod.
The specimens taken at this station are the only ones designated as types. The holotype (UMMZ 130648),
128 mm in total length, is a mature male (Figures lA-B). The paratypes (UMMZ 130649) comprise one male
ammocete 130 mm long, 4 males and 5 females, nearly to quite ripe, 1 16-142 mm long, and one male that was
taken partly decayed, within the same size range. A fully mature female, the smallest specimen, is illustrated
(Figures IC D).
On the basis of his field work Rutter (1908: 1 10) described Fall River and the adjacent part of Pit River as
follows:
The upper Pitt River, above the mouth of Fall River, was nearly dry in August, 1898. The water it
contained was of a slightly milky color. The rocks at the bottom were covered with a spongy slime . . .
At Fall River Mills, Pitt River receives Fall River, a stream about 100 feet wide and 4 feet deep, with a
strong current, but only about 15 miles long. Fall River takes its rise in two or three large springs near
Dana, and flows several times as much water as Pitt River above their union. The water is clear and cool
and the bottom gravelly, making an excellent spawning stream for salmon, but difficult to attain on account
of the steep rapid at its mouth, as well as the fall in Pitt River [see map. Figure 3].
Above the mouth of Fall River for a few miles, Pitt River is broad and deep, but without any perceptible
current. Below the mouth of Fall River the character changes entirely. It is broad but shallow, very swift,
with many rapids, and makes a rapid descent to the falls [3 km southwest of the mouth of Fall River]. Pitt
River Falls, which are 65 feet high, are thought by many to rival in beauty any to be seen in Yosemite
Valley. The middle portion is a sheer fall, but each side is broken by ledges, so that it is possible in high
water for fish to pass. A fish ladder has been blasted out of the rock near the left bank, and salmon now go
over the falls in considerable numbers.
The falls do not delimit the distribution of Lampetra lethophaga (nor of the endemic Cottus pitensis Bailey
and Bond, 1963: 20 25, figs. Id, 3b, 4d) in the Pit River system, but other Pit River endemics, Catostomus
microps Rutter (1908: 120 121, fig. D), Cottus asperrimus, and C macrops do appear to occur only above these
falls.
3. Lower Hat Creek, below Highway 299 bridge, over a stretch of about 6 km, above Lake Britton (an
artificially ponded section of Pit River), in northeastern Shasta County; altitude ca. 850 m. On October 4, 1968,
lampreys by good fortune were taken and preserved during a massive poisoning by the California Department of
Fish and Game, for the removal of "rough"" fish, presumably in the hope of controlling predation on and
competition with the favored gamefish. Dr. Roger A. Barnhart, Leader of the California Cooperative Fishery
Unit at Humboldt State College, who participated in the operation, preserved the fine series of specimens that he
has made available for the present study. Dr. Barnhart reported (pers. comm., 1970) that "the lamprey turned out
to be quite numerous in this section of Hat Creek. . . . We turned up 2-3 brook lampreys in our fall electrofishing
census last fall so apparently we did not obtain a complete kill of lamprey" (again by good fortune).
The collection furnished by Dr. Barnhart comprises 2 ammocetes 91 and 144 mm and 107 transformers 134-
199 mm long, of which 12 transformers (SIO 71-8) are retained at Scripps Institution. Nine other specimens (2
ammocetes 56 and 91 mm long and 7 transformers 146- 178 mm long; CAS 13391 ) were collected by Leonard O.
Figure 3. Natural lakes and streams of the entire Pit River drainage basin and the upper part of Klamath River
system, showing all known Locations, numbered 1-11, for the nonparasitic Lampetra lethophaga; also some
waters inhabited by parasitic forms of the same subgenus in the Klamath basin. The collection stations for
samples of the precocious stocks of L. tridentata utilized in this report are shown at A, for Shasta River near
junction with Klamath River; B, for Klamath River at Klamathon; and C, for the Copco Lake impoundment of
Klamath River. Shown also are nearby waters of contiguous drainage basins. Two of the largest of the many
marshes in the area are Klamath Marsh (KM) and Sycan Marsh (SM).
Map based largely on the United States Geological Survey 1:500,000 state maps of Oregon and California
and on the following National Topographic Maps of the 1:250,000 series: Medford, Crescent, Klamath Falls,
Weed, Alturas, and Susanville (1955-1963). The natural limits of South Klamath Lake, Tule Lake ("Rhett Lake"
on some old maps), and Clear Lake (of the Klamath system) and of the seldom attained outlet stage of Goose
Lake are taken chiefly from three old one-degree U.S.G.S. topographic sheets 1:250,000: Klamath, Oregon
(1894), and Alturas and Modoc Lava-Bed, California (1892). The Map of the Lake Region of Southeastern
Oregon by Snyder (1908a) was also used.
1971
HUBBS: A NEW NONPARASITIC LAMPREY
131
Fisk and W. E. Schafer of the California Department of Fish and Game during the same poisoning.
4. Pit River at Pit 4 Powerhouse, in northeastern Shasta County, 20 km northwest of Burney and 1 km south
of Oregon line; altitude ca. 650 m. One transformed female (CAS 25959), 155 mm long, with developing eggs;
collected by W. Rowley with electric shocker on June 2, 1953.
5. This number comprises two collections, only approximately located, in the same general area along Pit
River, in Shasta County; altitude ca. 550 m:
UMPQUA RIVER
123°
COLUMBIA RIVER
122°
INTERIOR DRAINAGE
21° 120°
ca
>
<
o
Q.
3
>
cr
3
(J
o
cc
5
kJ
(-
CO
>-
en
>
cr.
<
<
O
O
X)
>
>
O
m
>
in
SACRAMENTO RIVER SYSTEM
10 0 20
INTERIOR DRAINAGE
40 MILES
10
20
40
60 KILOMETERS
132
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 4. Spring source of Fall River, in naturally ponded pocket of springs immediately above origin of stream
flow (the type locality of Lampetra lethophaga); Location 2 on distribution map (Figure 3). Photograph by Laura
C. Hubbs, August 17, 1934.
5A. Near Big Bend, collected May 3. 1944 (no further data): 7 ammocetes (CAS 13392), 52-124 mm long;
ova developing in largest one.
5B. Between Pit 5 Powerhouse and Pit 5 Dam, collected June September, 1953 by William Rowley, Brian
Curtis, and W. O. Cheney, of California Department of Fish and Game, by electric shocker: 1 ammocete (CAS
25968), 63 mm long (identification presumptive).
LOCATIONS IN KLAMATH RIVER SYSTEM, SOUTHERN OREGON
6. North Fork of Sprague River, in east channel, about 1 km above junction with South Fork (prior to
extensive disruption of stream course for irrigation), just east of east boundary of Klamath Indian Reservation,
near center of west border of T 36 S, R 14 E, eastern Klamath County (Sprague River joins Williamson River
just before that stream enters Upper Klamath Lake); altitude ca. 1340 m. Water moderately clear (bottom
visibility about 1 m), shaded by 2-m banks, in pasture; some vegetation in patches; temperature cool; current
moderate to swift; width ca. 5-8 m; depth ca. 0.7 m. The collection, M34-I20b, by Carl L. Hubbs and family,
on August 9, 1934, with 25-foot bag seine, contained one adult male (UMMZ 130573) with maturing testis, with
tail 43 mm long (front end missing; estimated original total length about 143mm) and one male ammocete, 91
mm long, secured by much stranding of bottom material. Associated fish species were: Salmo gairdnerii.
Catostomus snyderi Gilbert, Rhinichthys osculus klamathensis (Evermann and Meek), Gila caerulea (Girard),
and Gila b. hicolor.
1 . Sprague River opposite Ferguson Butte (in narrows of a broad valley), 6.5 km inside Klamath Indian
Reservation, in T 36 S, R 13 E, Klamath County; altitude ca. 1325 m. Water moderately clear (bottom visibility
ca. 1 m); water buttercup and other plants in dense patches; bottom mostly sandy, becoming dirty in weeds, some
gravel, mostly tine, few stones; temperature cool; current mostly slight to moderate; width uniformly ca. 12 m;
1971
HUBBS: A NEW NONPARASITIC LAMPREY
133
depth to 1.2 m. The collection, M34-121, by Carl L. Hubbs and family, on August 10, 1934, with a 6-foot
woven-mesh seine, contained a recently transformed male 145 mm long (UMMZ 130576), with testis developing
and gut reduced, taken in dense vegetation. Associated species were Salmo gairdnerii. Rhinichthys osculus
klamathensis. Gila caerulea. Gila b. bicolor. and Coitus klamathensis Gilbert.
8. Tributary, near mouth, to upper course of Sycan River (affluent to Sprague River), at Pikes Crossing, 3
km south of Currier Camp, near center of T 33 S, R 1 5 E, eastern Klamath County; just above a major canyon in
river course; altitude ca. 1760 m. Water described as white, very slightly turbid, odorless; some green algae;
bottom of sand, coarse gravel, and stones; 23.5° C (air 24.5° C); shore a sage flat, with meadow and timber; current
swift in part; nearly 7 m wide in places and to 1.5 m deep. The collection, M 39-18, by Robert Rush Miller and
Ralph G. Miller, on June 27, 1939, with 9-foot and 15-foot seines, contained an ammocete (UMMZ 136683) 132
mm long, with small testis, and an adult female 106 mm long, somewhat bobtailed, with nearly ripe ova.
Associated species, taken both in tributary and river, were Salmo gairdnerii and Rhinichthys osculus klama-
thensis.
9. Sycan River where it enters Sycan Marsh, at ZX Ranch, near center of T 32 S, R 14 E, in western Lake
County; altitude ca. 1525 m. Water clear, whitish-brown, odorless; without vegetation; bottom of silt, rocks, and
brush, largely scoured; 19.5° C (air 13°); willow thickets along shore, margining meadow; current none to slight;
width to 5 m in pools; depth to 0.5 m. The collection, M 39 17, by Miller and Miller, on June 26, 1939, with 6-
foot and 9-foot seines, contained (UMMZ 136678) an ammocete 121 mm long, with minute gonad, and a female
110 mm long, with large ova. Associated species were Catostomus snyderi. Rhinichthys osculus klamathensis,
and Gila b. bicolor.
Dr. Robert Rush Miller was told by personnel at ZX Ranch that the expansive Sycan Marsh (SM, Figure 3)
had no open springs and was not known to contain fish. However, it presumably passes fish in high water.
10. This collection, comprising 2 spawning males (Cornell University 10296), 125 and 145 mm long, is
labelled "Oregon, 5 mi. W. of Beatty, spring on S. side of road, Apr. 6, 1942, A. H. Wright." This places the
station approximately 3—4 km south of midlength of Sprague River, near mid-west border of T 36 S, R 13 E,
Klamath County; altitude ca. 1280 m. Dr. Wright stated (in letter of October 1, 1942) that:
In a swampy area near a small streamlet west of Beatty, Oregon, 1 happened to find two clear, sandy
areas about five or six feet deep. The swampy stretch was so treacherous that someone had laid boards
across it and as 1 looked in the clear areas, . . . among the boiling sand were these two lampreys. It was a
very striking spring with a very pronounced boiling sandy bottom.
1 1. Klamath State Fish Hatchery, in the Klamath Indian Reservation, on Crooked Creek, a short spring-fed
stream that joins Wood River close to Agency Lake; in Section 6 of T 34 S, R 7 Vi E, 4 km northerly from
Table 1 . Mateiial of Lampetra lethophaga of different stages, arranged chronologically by day
of collection
Date of
Collection
Locality
no.
No. of specimens (and length in mm) at each stage
Ammocetes Transformers
Maturing and
mature adults'
1(142)
114(134-199)
6(130-160)
6(132-154)
1(137)
2(125-145)
1(155)
1(110)
1(106+)
Feb. 16(1961) IIC —
Mar. 13(1970) IID —
Mar. 20(1970) HE —
Apr. 6(1942) 10 —
May 3 (1944) 5A 7(52-124)
May 16 (1970) IIF 6(88-191)
June 2 (1953) 4 —
June26(1939) 9 1(121)
June27(1939) 8 1(132)
June-Sept. (1953) 5B 1(63)
Aug. 9(1934) 6 1(91)
Aug. 10(1934) 7 —
Aug. 13(1934) llA 34(18-155)
Aug. 17(1934) 2 1(130)
August (1949) 11- 4(70-107)
Sept. 4(1898) 1 1(105)
Oct. 4(1968) 3 4(56-144)
Oct. 20(1952) IIB 91(37-205)
'Maturity indicated by boldface type for nuptial and postnuptial stages; by italic type for very definitely maturing
stages, including, for the August 17 type series, some fully mature but not in nuptial color and form; and by
roman type single specimens in earlier stages.
l(ca. 143)
1(145)
11(116-142)
134 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Klamath Agency, Klamath County: altitude 1280 m. Water very clear, arising in springs on hatchery grounds,
close to upper part of Crooked Creek; with thick clumps of submerged vegetation; bottom of sand and pumice
stones, with a little muck mixed with sand in the vegetation; shore grassy with some willows, in meadowland;
7.8° C (hatchery personnel reported virtually no fluctuation); current moderate to, mostly, swift; width of rather
straight course 4 8 m; depth to 0.6 m.
This ecological description is based on observations on August 13, 1934, when Carl L. Hubbs and family
collected 34 ammocetes (UMMZ 130606), 18 155 mm long, with a 6-foot fine-woven-mesh seine, by vigorously
working through thick weed beds, muddy-sand bars, etc. None of the specimens showed any sign of
metamorphosis, which may well take place late in this very cold water. The hatchery superintendent (W. I.
How land) provided evidence that "runs of eels" do not occur in this or other local streams, and gave information
on the local occurrences of lampreys. Some of the larger ammocetes show some development of ova and of testis.
It was therefore concluded, on this initial contact, that the local lampreys are nonparasitic dwarfs.
Associated with the ammocetes in the 1934 collection (M 34 126), in addition to Salmo gairdnerii and
Salvelinus fontinalis (Mitchill) were sculpins, of the Klamath cold-water endemic species Cottus tenuis
(Evermann and Meek), which was common. The superintendent had 3 or 4 large adults of Catostomus snyderi.
which he said runs up Wood River and Crooked Creek in early spring. He indicated that still larger suckers,
which from his description seemed to be Chasmistes. run chiefly up Williamson River, early in the spring, to
spawn, and some go up Wood River. Eggs of these large suckers, he said, cover the bottom near wiers across
Williamson River to a depth of several inches. When they come, the trout run ceases.
Dr. Carl E. Bond (pers. comm., 1970) has received similar testimony regarding the local lampreys from
personnel of the Klamath Hatchery. He has kindly provided me with additional specimens (listed in Table 2 as
from Locations 1 1 B F), taken by and for them at the hatchery, in the ponds and their discharges ("after passing
through the ponds, water is channeled into two ditches that run a short distance to Crooked Creek."). The 13
adults examined in these lots, from Locations 1 ICE, are all in nuptial or postnuptial color and form (one, a
partly spent male 154 mm long, the largest in 1 1 D, is illustrated as Figure 2A-B). Habitat data for the collection
of March 20, 1970 (CEB 70-2; listed in Table 2 as from Location 1 IE), by Dr. Bond, Mr. Kan, and Richard
Wilmot, by "sculpin net (frame net)" are: water clear, with Ranunculus, mostly at edges and behind stones;
bottom of sand and line gravel, with few large stones; temperature 6.7° C; shore of masonry or stone; current
moderate to slow; width 2.5 5 m; depth to 0.6 m. Lampreys taken in this collection were 2 ammocetes and 4
adults (I alive and 3 dead), but the one adult received for study is a ripe male 137 mm long.
ZOOGEOGRAPHICAL CONSIDERATIONS
Like most but not all of the nonparasitic forms of lampreys, L. lethophaga lives within
the range of its assumed parental type (see discussion of Life History, and Hubbs and
Potter, in press). So far as known, it is limited to the upper parts of the Pit River system of
the Sacramento River drainage in northeastern California and of the Klamath River
system, adjoining, in south-central Oregon (Figure 3). This form, and/or parallel-derived
nonparasitic types, may yet be discovered elsewhere within the wide range (Figure 5) of L.
tridentata. but the only nonparasitic lampreys previously known to occur around the North
Pacific are the derivatives of Lampetra (Lampetra) ayresii (Giinther) in the northeastern
Pacific drainages (Vladykov and Follett, 1958, 1965), occurring as far south as the Santa
Table 2. Material of Lampetra lethophaga from Klamath State Fish Hatchery received from
Carl E. Bond and Ting T. Kan
Locality
Date
collected
Coll.
no.
(OS)
Collector
Number (total length, mm)
Nuptial &
Ammocetes postnuptial
IIB
lie
IID
HE
IIF
Oct. 20, 1952
Feb. 16, 1961
March 13, 1970
March 20, 1970
May 16, 1970
2860
2855
2856
2858
2859
Kenneth Cochrun
Do.
Ore. Game Comm.
Bond, Kan, Wilmot
Bond, Johnson, Kan
25(77-189)'
6(130-160)
6(132-154)
1(137)=
6(88-191)
'Kan measured 66 additional ammocetes, as 37-205 mm in total length, from this collection, which was sup-
posedly taken by electrofishing, and 4, of 70-107 mm, collected at the same hatchery in August, 1949. These
measurements have been included in the size-frequency graph (Figure 8).
•Collection sheet lists for this set 2 ammocetes and 4 adults (1 alive and 3 dead).
1971
HUBBS: A NEW NONPARASITIC LAMPREY
135
Ana system of streams in southern California, and the derivatives of Lampetra (Lethente-
ronj japonica (Von Martens), ranging from northern China and southern Japan through
the coastal regions of Siberia to Alaska (and in northeastern North America). Sufficient
material is known to render it highly probable that any other regional occurrences of any
nonparasitic derivative of Lampetra ( Entosphenus) tridentata are at most few and
limited.
The known distribution of the nonparasitic lampreys around the North Pacific
appears to be complementary. The ranges of the widespread nonparasitic representatives
of the subgenera Lethenteron and Lampetra apparently do not overlap, and although L.
lethophaga of the subgenus Entosphenus occurs about midway in the range of the Pacific-
drainage representatives of subgenus Lampetra, no trace of that subgenus has been found
in the Pit or Klamath systems, either by me, or by Carl E. Bond (pers. comm., 1971).
It is noteworthy that no nonparasitic forms of the Entosphenus complex have been
discovered in other parts of the long range of Lampetra tridentata around the periphery of
the North Pacific (Figure 5), which extends southward from Bering Sea and Unalaska
(Jordan and Gilbert, 1899: 434; McPhail and Lindsey, 1970: 58), and from Bering Island
(Svetovidova, 1948; Berg, 1948, Addenda; 1962: 494). The limits of the known distribution
of the parasitic form (or forms) have been expanded southward on both sides of the Pacific.
On the American side it has been taken in streams as far south as southern California and
in the ocean off Baja California, Mexico (Hubbs, 1967). On the Asiatic side there are
several records from Japan, stated below. There seems to be no valid report of L.
tridentata from the mainland of Asia (Lindberg and Legeza, 1959: 17-18 and 1967: 20-
21), where L. japonica holds forth (the record of "Entosphenus tridentatus" from
Kamchatka by Jordan and Evermann, 1900: 3231, pi. 1, fig. 4, was apparently based on the
ammocete that was listed by Jordan and Gilbert, 1899: 434, from a river near Petropaulski,
Kamchatka, as "Entosphenus camtschaticus," though on circumstantial grounds it seems
more probable that it was an example of L. japonica). Okada and Ikeda (1938: 140 141)
\ i
1
\^
"^^^
^^^^&
S^^
^T
\
^^^
'°v
^/(
^
%
m
\ \
^^^^
'7 /
\v
7
'.
V
1
\
/ /
X\X
^°'fCr F 7/1 — i~~i~-4~ /
-
1
uttU
-f
' N.
-tt
/
— C
\
'°°h~-Lj_M
h
^
1
\
\
-\
0
My
LLl^
■
1
1
i
u
/
0
^\L\
120° 130' 140" 150° 160* 170*
.AMBERT-S AZlMuTHflL EOUAL-AREA PROJECTION
180°
170" 160° 150° 110°
130° 120°
110°
100°
90°
S0°
Figure 5. Distribution of Lampetra tridentata around margin of North Pacific Ocean. Assumed usual range
stippled; record stations beyond these limits ringed; area shown in Figure 3 indicated.
136 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
initiated the Japanese records of L. tridentata by listing a specimen from Yahutugawa
(river) in the Okhotsk Sea drainage of Hokkaido. Nemoto (1955: 69-70) stated the range
of the species as "the broad region from the Arctic as far south as southern California and
down to about 35° N Latitude in the western side of the Northern Pacific," but gave no
supporting documentation for either the Arctic or for the southwestern limit, other than
the questionable basis of finding, in the western North Pacific, whales bearing scars
showing the tooth marks of I. tridentata. Aoyagi (1957), however, reported the capture of
a specimen of this species in central Honshu, near 36° N latitude, in Kinugawa (river) at
Atsutamura in Tochigi-ken. A further extension of range of the species has now come to
light: Dr. O. Okamura has obtained a specimen from Yoshinogawa (river) on Shikoku
Island in southern Japan; Tamotsu Iwai (pers. comm., 1970) has verified the identification.
The occurrence of the nonparasitic representative of the Entosphenus group in the
adjacent basins of the Pit and Klamath rivers is not unique, for these stream systems
harbor a number of other endemic fishes, some of which are sympatric with Lampetra
lethophaga. Klamath endemics were described by Gilbert (1898) and by Evermann and
Meek (1898). One of these species, Catostomus rimiculus Gilbert (1898: 3) was described
from the Klamath River system only but it was later found (Snyder, 1908b: 161) to inhabit
also the Rogue River system, which adjoins the Klamath River drainage basin (Figure 3);
it may well have crossed over the divide by some fluvial connection. The peculiarities and
endemism of the Klamath and Pit river systems were summarized by Hubbs and Miller
(1948: 67 71). Catostomus microps is a Pit endemic (Rutter, 1908: 120 121) and Cottus
pitensis Bailey and Bond (1963: 20 24) is known only from drainages of the Pit River and
the contiguous Little Sacramento River. An additional indication of residual endemism in
the fish fauna of the area under consideration appears to be coming to light: Behnke (1970:
241) has referred to "a group of previously undescribed trout native to several desiccating
basins in southern Oregon extending to the Pit and McCloud rivers of northern Califor-
nia."
DESCRIPTION AND COMPARISONS
The specifications, here adopted, of meristic and morphometric characters, involving
definitions and methods, are essentially those proposed by Hubbs and Trautman (1937: 27
-43, figs. 15). They have been adopted also by Hubbs and Potter (in press) in their
account of the distribution, phylogeny, and taxonomy of lampreys.
Chief concern pertains to the designation and to the method of counting of the lingual,
oral, and disc teeth, which have been illustrated for Lampetra ( Entosphenus j tridentata by
Hubbs (1947, fig. 3; 1963, fig. 2), by Vladykov and Follett ( 1958, fig. 1; 1965, fig. 1; 1967,
fig. 2), and by Hubbs and Potter (in press, fig. 7). Special points regarding the cusps on the
lingual laminae (one transverse and two longitudinal) and on the oral laminae (the
supraoral and the infraorbital) are discussed below, in the description of the dentition.
The concept of the circumoral row or series of teeth proposed by Hubbs and
Trautman, primarily on the basis of the generalized dentition of Ichthyomyzon, seems
quite applicable to the Entosphenus group, particularly because the posterior circumorals
are so definitely aligned with the lateral circumorals, just outside the infraoral lamina.
Furthermore, the lateral and posterior circumorals intergrade, through the frequent and
unique bicuspid structure, and often through the increasing dilation outward, of one or more
of the most lateral and most anterior of the posterior circumorals. Although the alignment
of the posterior and lateral elements in a circumoral row is clear, the alignment and
method of counting of the anterior connective is complicated by the tendency of all the
anterior teeth in this group to alternate (in quincunx), so that a rather arbitrary distinction
is involved, as is described below. The alignment and nomenclature of the inner disc teeth
1971 HUBBS: A NEW NONPARASITIC LAMPREY 137
championed by Vladykov and Follett contrasts with the system of Hubbs and Trautman, in
that the anterior circumorals are treated as the inner "anterials,"' the lateral circumorals as
the "inner laterals" or "endolaterals/' and the posterior circumorals as the inner
"posterials."
Because dentition has traditionally and rightfully been emphasized in the systematics
of lampreys, with added stress by Hubbs and Trautman (1937), by Vladykov and Follett
(1967), and by Hubbs and Potter (in press), the dental laminae and teeth are here treated
first.
DENTITION
Distinctive features of the dentition o{ Lampetra lethophaga outlined in the Diagnosis
seem to make clear the relationships as well as the distinctness of this nonparasitic
representative of L. tridentata. Exhibited are some features of reduction and some of
increased variability. Reduction (often a concomitant of dwarfism) is indicated by the
frequent degeneration, or loss (Table 3), of the median cusp of the supraoral (the tricuspid
supraoral has usually been emphasized — often overemphasized — as the main feature of
the genus or subgenus Entosphenus); by the occasional reduction of infraoral cusps to 4; by
the frequent reduction (Table 4) by I cusp on any of the four lateral circumorals, from the
normal Entosphenus formula of 2 3 3-2 (Figure 7); and by the low number (9-15) of
posterior circumorals. The number of cusps in the transverse lingual lamina also seems to
be reduced. Furthermore, the teeth tend to be reduced in size; the lingual and oral laminae
and, in particular, the lateral circumorals, are all less dilated than in typical L. tridentata.
and the other teeth tend to be smaller and less robust. Increased variability (commonly
associated with degeneration) is shown strikingly by the number of cusps on the supraoral
and, less certainly, by the number of infraoral cusps (Table 3), and, definitely, by the
number of cusps on the lateral circumorals (Table 4).
The small size of L. lethophaga and the weakness of its dentition render cusp counts at
times somewhat difficult. Adequate magnification with strong illumination supplemented
by a fine jet of compressed air may be called for.
Although the full development of the teeth is a relatively transient feature, the cusps at
early stages are sharp. In fact, it is difficult to determine from the teeth, at prime
development, whether a specimen represents a parasitic or a nonparasitic form. The lingual
laminae atrophy first, in concordance with the elimination of feeding. Of the disc teeth, the
outer ones, between the circumorals and the marginals, appear to be the first lost. The
degeneration of the anterior circumorals seems to follow soon; they become unrecog-
nizable while the posterior circumorals remain sufficiently developed to be seen. The lateral
circumorals are among the last to disappear, or to fragment. Completely spent individuals
retain very little of their dentition, and the teeth do not seem to fuse into a cornified mass,
as they do in L. tridentata.
Lingual laminae. — In correlation with the reduced size of the laminae, the cusps are
small — often minute, weak, and crowded. The median tooth of the transverse lamina is
usually only weakly to moderately enlarged, and is somewhat variable: it is occasionally
partly fused with an adjoining cusp on either side, and is rarely doubled. The transverse
lamina is nearly rectilinear, with only the outer ends curved backward. The number of
cusps in this lamina is probably reduced, totally only 12-19 in the 11 countable laminae,
with a mean of 15.6 (12 17, averaging 14.75 in 8 from Pit River, 19 in two from the Sycan
River, and 16 in one from Crooked Creek). The counts for the 15 specimens from the
Klamath River near Klamathon are higher, 21-27 (mean 23.3), and the one from Copco
Lake has 22. However, the counts for 437 macrophthalmiae of L. tridentata from the
mouth of Shasta River, not far distant, are intermediate: 14 23 (mean 1 8. 1 ). McPhail and
Lindsey (1970: 57) described the lamina of L. tridentata as having "about 15-25 fine
138
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
points, the median one scarcely larger than the others."
The cusps on the longitudinal lamina are also minute and relatively very numerous.
They were not counted. McPhail and Lindsey ( 1 970: 53) stated that "£". tridentatus has 50-
63 fine points on each longitudinal lingual tooth plate, in contrast to 0-26 points in all
Lampetra species examined," and they regarded this distinction as one basis for the
recognition of Entosphenus as a distinct genus. The number of cusps no doubt varies, and
the "O" counts presumably represent laminae that have been shed, probably postnuptially.
The other basis given for the recognition of Entosphenus was the number and arrangement
of the velar tentacles — a feature (not checked by me) that hardly seems of generic
significance, though seemingly trenchant on the species level.
<s> "^
jj-'--
Figure 6. Dentition of Lampetra lethophaga. from mature female paratype 1 16 mm long, shown in Figures 1 C,
D; note oral papillae as well as fimbriae; development of bicuspid posterior circumorals is extreme in this
specimen.
Oral plates (Table 3). — The oral plates are somewhat more delicate and less dilated
than in the parasitic forms oi Entosphenus.
As noted above, the supraoral plate (or tooth) often fails to exhibit the strongly
tricuspid form traditionally used to diagnose Lampetra tridentata, for, in each river
system, the median tooth is more or less reduced in size, down to a mere rudiment, or is
altogether missing. Bicuspids and tricuspids are about equal in frequency, and quadricus-
pids are occasional. Thus, the supraoral-cusp pattern is much more variable than is usual
in Entosphenus. Ordinarily, in L. tridentata, the tooth is tricuspid, as it is in every one of
the 437 macrophthalmiae counted from the mouth of Shasta River, in all 1 5 adults from the
Klamath River, in the one from Copco Lake, and (Mr. Ting T. Kan, pers. comm., 1971) in
all 86 adult specimens of the dwarf race from Miller Lake. In L. lethophaga, as a further
indication of variability, the third cusp is occasionally well to one side of the midline, and
1971 HUBBS: A NEW NONPARASITIC LAMPREY I39
the 3 quadricuspid plates exhibit different cusp patterns: 2 + 2, 1 + 1+2, and 2+1 + 1 (left-to-
right).
Bond (1961: 14) distinguished the nonparasitic form (his ""Lampetra sp.") from L.
tridentata too sharply, as having "teeth dull, supraoral lamina with two widely separated
cusps" rather than having "all teeth sharp and functional, supraoral lamina with 3 cusps."
This seems to be the commoner condition only in Crooked Creek (Table 3) from which Dr.
Bond had specimens of the new form.
The cusps on the infraoral plate average nearly as numerous in L. lethophaga as in the
precocious stocks of/., tridentata from the Klamath River system. Six among 56 specimens
have only 4 cusps, whereas reduction below 5 was not encountered among the 453
J-.'^"-*^-^''-*-"" ■■■■■■
.:S'V, Q\^ r^' '- -
K^>
■?
Figure 7. Generalized illustration of dentition of Lampetra tridentata, drawn, with mouth somewhat puckered,
by Elizabeth M. Kampa; used as basis for figures in Encyiopaedia Britannica (Hubbs, 1947, fig. 3: 1963, fig. 2).
specimens of Z,. tridentata listed in Table 3, nor in any of the counts for the Trinity and Eel
rivers in northern California. The cusp count was increased in L. lethophaga to 6 or 7 in
only 7% of the specimens, but in the L. tridentata specimens here tallied, to 12% in those
from Shasta River and to 33% in those from Klamathon; and the one from Copco Lake
has 6 infraoral cusps. Counts higher than 5 may result either from a regularly spaced series
or from the interpolation of a small supernumerary cusp toward one end of the plate. The
outermost cusp of each side is strengthened but is never doubled, as it typically is in
subgenus Lampetra. The specimen of doubtful identification from Willow Creek and (Ting
T. Kan, pers. comm., 1971) the Miller Lake lampreys have the usual cusp pattern for the
oral teeth (3 and 5, respectively).
Circumoral teeth and cusps (Tables 4-6). — Some of the sharpest distinctions of L.
lethophaga involve these teeth and their cusps, on the lateral and posterior fields of the
disc.
140
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Anterior circumorals.— As in L. tridentata. the anterior circumorais are typically
countable as 5, on the criterion that any anterior tooth is interpreted as a circumoral that
approximately reaches or passes behind the imaginary arcuate line passing through the
centers of the teeth of the definitely inner-marginal row. The tendency of the anterior oral
teeth to alternate so as to approach a quincunx arrangement renders the inclusion or
exclusion of a given tooth from the anterior-circumoral series somewhat difficult and
rather arbitrary. Another uncertainty is introduced by the tendency for the anterior disc
teeth to atrophy rapidly in this species. The counts recorded are 5 for 14 specimens,
Table 3. Counts of cusps on oral plates in Lampetra lethophaga and in Klamath River preco-
cious populations of L. tridentata
Oral plate
Species
River system (No.)'
Locality (No.)'
Number of cusps per oral lamina
7 Mean
Supraoral
L. lethophaga
Pit (10)
Spraeue (5)
Crooked (13)
Total (28)
L. sp.
Willow ( 1 )
L. tridentata
Klamath (453)
Shasta R. (437)
Klamathon (15)
Copco L. ( 1 )
3
2
8
13
5
3
4
12
437
15
1
2.90
2.60
2.46
2.64
3.0?
3.000
3.00
3.0?
Infraoral
L. lethophaga
Pit (10)
Sprague (5 )
Crooked (13)
Total (28)
L. sp.
Willow ( 1 )
L. tridentata
Klamath (453)
Shasta R. (437)
Klamathon (15)
Copco L. ( 1 )
9'
2
10
21
382
41
10
5
1
- 4.90
1 5.40
1 5.15
2 5.11
5.0?
14
5.158
5.33
6.0?
'Number of specimens.
^Value for holotype.
doubtfully 5 for 7, 6 for 3, and 7 for 2. The anterior circumorais are counted as 5 also in the
Willow Creek specimen of doubtful identification.
Lateral circumorais (Table 4). — A striking feature of L. lethophaga, already alluded
to, is the strong tendency for the number of cusps to decrease by 1 in each of the
consistently 4 lateral circumorais, on each side, from the standard, usually almost
invariable, formula of 2-3-3 2 in L. tridentata. The ratio of reduced counts to the full
1971
HUBBS: A NEW NONPARASITIC LAMPREY
141
E
T3
C
>
C
E
c
6«
^
^
^
r^
kJ
Q
c
D.
iT
c/)
•a
a.
^
'c«
3
JZ
U
(J
03
O
o
o
^_
C3
L-
O
E
3
u
_u
"o
55
^^
a
c«
*—
u
<u
^
^-t
B.
.^
f
■«^
u
C3
>~i
lU
Ui^
c
C
o
«
c
CO
a.
_o
C/5
^-»
3
C3
o
3
<4-i
a,
o
o
1/1
G.
c
C/5
3
3
C
.^
O
'J
CJ
OJ
^'
Q.
_o
Ui
s
4J
03 .C
Hoi
a.
3
u
c
c
H c
c
03
(N
Q.
o ^
5
v-j O 00 O r--
I/-, VD — ' "^ O
(N CN (N (N rn
-^•^ -^os ri
0\ in 00 rs
O "^ £"•
q T-- o
m r<^ m
_ Tt
m i/~i r-1
oo
O O (N O <^•
— 1^ 00 o\ C
o
O C !-•
o — o
(N — m
00 ITi 00 --^ (N
-* r- (N
00
-^ -^ oo
fN
O
<N a
o
O 'Ton
O ^ g--
CucoU
O ■i'
^ ~ t^ S ^ w.
c S E ^- o
^4 -j
O O -^ r^
O 0\ \0 00
o
rN — ' — — tN
o r- Ti- —
(S OOO
rJ
•-i "^ '-^
<N <N <N
(S
o\r^
VO — fN
oo
lO O OS v-i
q
fN r-i <S fS
en
— -"t (N 00
0\ VC O -t
-H fv4 Ti-
es
o -^ t^-
qrsiq
m ro fi
rJ r~-
Csl — <S
oo
o
<u
c
o
.1
J3
O
Co
OO (N
£ §
o
I
o
x; c3 2 o
- E "
u
c
3
O
(/I
OJ
T3
a
u
x>
E
3
C
«
a
>>
^^
o
03 03
E^- O
03 !/5t*0
>-J
u
<u
3
z
■3 o
x:
3
>
142
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
complement, for each of the teeth, counted from the front on each side, for each of the
three stream systems, is as follows:
First tooth 0:20 (Pit), 4: 10 (Sprague), 4:22 (Crooked).
Second tooth — 9:20 (Pit), 5: 10 (Sprague), 1 8:22 (Crooked).
Third tooth — 18:20 (Pit), 6: 10 (Sprague), 20:22 (Crooked).
Fourth tooth 0:20 (Pit), 2:10 (Sprague), 8:22 (Crooked).
The Willow Creek specimen agrees with L. tridentata in the formula of 2-3-3-2.
The same formula, with little variation, holds for the dwarfed, reportedly parasitic Miller
Lake lamprey (Carl E. Bond and Ting T. Kan, pers. comm., 1971).
Posterior circumorals (Tables 5, 6). — One of the dentitional features that most
clearly points to the derivation of A. lethophaga from L. tridentata is the frequent bicuspid
structure of the more lateral of its posterior circumorals (Figures 6, 7), involving teeth
occasionally as far from the end as the seventh (Table 5). Lampetra tridentata is the only
previously known lamprey of the Lampetra type that has such bicuspid teeth in the
posterior commissure, and this tendency is notably characteristic of the precocious Klamath
River types that may well have been ancestral to L. lethophaga. In fact, in those types the
Table 5. Counts of cusps on individual posterior circumoral teeth in Lampetra lethophaga and
in the Klamath River precocious populations of L. tridentata
Species
River system (No.)'
Locality (No.)'
Tooth No. V
U/B Mean'
Tooth No. 2'
U/B— Mean'
Tooth No. 3'
U/B Mean'
Tooth No. 4^
U/B Mean''
L. lethophaga
Pit (20)*
Sprague (10)
Crooked (18)
Total (48)
L. sp.
Willow (2)
L. tridentata
Klamath (906)
Shasta R. (874)
Klamathon (30)
Copco L. (2)
8/12—1.60
8/2—1.20
17/1—1.06
33/15—1.31
7/13—1.65
8/2—1.20
18/0 1.00
33/15—1.31
14/6 1.30
8/2—1.20
18/0 1.00
40/8 1.17
18/2—1.10
8/2—1.20
18/0 1.00
44/4 1.08
0/2—2.0?
213/661 — 1.756
1/29—1.97
0/2—2.0?
1/1—1.5?
549/325—1.372
1/29—1.97
0/2—2.0?
2/0 1.0?
835/39—1.045
5/25—1.83
0/2-2.0'?
2/0 1.0?
835/39—1.045
14/16 1.53
0/2—2.0?
Species
River system (No.)'
Locality (No.)'
Tooth No. 5=
U/B— Mean-'
Tooth No. 6'
U/B Mean^
Tooth No. r-
U/B— Mean^
Tooth No. 8=
U/B Mean'
L. lethophaga
Pit (20)*
Sprague (10)
Crooked (18)
Total (48)
L. sp.
Willow (2)
L. tridentata
Klamath (906)
Shasta R. (874)
Klamathon (30)
Copco L. (2)
18/2—1.10
10/0 1.00
18/0 1.00
46/2—1.04
19/1 — 1.05
10/0—1.00
18/0 1.00
47/1 1.02
19/1 — 1.05
10/0 1.00
18/0 1.00
47/1 — 1.02
20/0—1.00
10/0 1.00
18/0 1.00
48/0—1.00
2/0 1.0?
872/2—1.002
25/5—1.17
0/2 2.0?
2/0 1.0?
873/1 — 1.001
26/4 1.13
0/2—2.0?
2/0 1.0?
874/0 1.000
28/2—1.07
1/1 — 1.5?
2/0—1.0?
874/0 1.000
29/1 — 1.03
0/2—2.0?
'Number of sides counted (2 per specimen).
'The teeth are numbered from the lateralmost and foremost; the ninth tooth is bicuspid on one side of the one
specimen from Copco Lake.
'U/B indicates the number of unicuspids and bicuspids, respectively, for each given tooth number. The numbers
(and the means) are in a sense hypothetical, for it is assumed for all teeth more centrad than the fourth on
either side that the tooth, unless bicuspid, would have been unicuspid had such a tooth been present: the total
number of posterior circumorals may be as low as 9 in L. letliophaga and as low as 15 in the Klamath River
precocious populations of L. tridentata: hence the tooth count on either side often passes beyond the midline into
the series from the other side; the tooth number used assumes the arrangement of unicuspids and bicuspids that
would have existed, had there been as many as 9 teeth on each side.
*The holotype has the outermost 2 teeth on each side bicuspid.
1971 HUBBS: A NEW NONPARASITIC LAMPREY 143
lateral teeth often grade, in position, size, and structure, almost imperceptibly into the
outer members of the posterior series. The proportion of the outer posterior teeth that are
bicuspid is greater in the Pit River sample than in the collections from Sprague River and
Crooked Creek. Oddly, the degree of bicuspidity averages very distinctly higher in the
dwarf adults of L. tridentata from the Klamath River near Klamathon than in the
macrophthalmiae from the Shasta River near its junction with the Klamath, not far below
Klamathon; and the one specimen from Copco Lake has 8 bicuspids on each side (a record
number). In compensation, the unilateral posterior circumorals average fewer in the
Klamathon lot than in the Shasta River specimens, and the one from Copco Lake has only
3. One of the sharpest distinctions of L. lethophaga from the samples of L. tridentata from
the Klamath River system lies in the lower total number of posterior circumorals, with
very little overlap (Table 6). The alignment of the posterior circumorals tends to be slightly
irregular in some specimens, though at the end on each side the series lines up very well
with the posteriormost (fourth) lateral circumoral.
In the numbers of circumoral teeth and cusps the one specimen of doubtful
identification from Willow Creek shows some correspondence with L. tridentata and some
features of seeming intermediacy between the two species. The total number of posterior
teeth in the series (17) is 2 higher than any count for L. lethophaga and below the mean for
the L. tridentata series. The number of unicuspid posteriors (14) is extreme for L.
lethophaga and near the mean for L. tridentata. The number of bicuspid posteriors (total
3) is not definitive.
Outer disc and marginal teeth. — As in L. tridentata, the disc is consistently toothless
between the circumorals and the marginals, except for a few rather scattered teeth (not
counted) in the anterior field. The pattern of the marginal teeth, as in L. tridentata, in some
specimens, weakly suggests that the marginal series may retain elements from the very tips
of the recurved ends of the original alate rows — particularly in that some of the teeth tend
to be rather larger and less completely at the disc edge than those preceding and following
(see figures by Vladykov and FoUett and by Hubbs and Potter and the discussion by the
latter authors). The marginal series, however, remains essentially complete and intact,
whatever its origin may have been.
In the Copco Lake specimen the marginals between the first and second and between
the second and third lateral circumorals are much dilated, and are considerably inter-
polated between the laterals.
ORAL PAPILLAE
The oral papillae (Figure 6, Table 7), of presumed sensory function, were counted
because it was thought that they might be reduced in number in the relatively very small
disc of L. lethophaga. These structures, which occur around the periphery of the disc
among lampreys in general, are almost always distinguishable from the finbriae. They stem
from the groove just ringing the slight pad from which the radially transverse fimbriae
arise. They are conical and pointed, rather than being truncate with fimbriate edge. They
are often irregularly spaced along each side, with a wide intervening separation posteriorly
and with a narrow gap anteriorly, normally broken by a more or less precisely median-
anterior papilla (rarely missing, doubled, or trebled). Slight uncertainty in counts arises
from the rather rare apparent or real intergradation between papillae and fimbriae, at least
in superficial aspect. In one specimen from Klamathon, irregularities, involving adven-
titious creases and folds around the outer part of the disc, render the count useless. A fine
jet of air aids in the count.
In mean numbers and in the range of variation, the papillae are rather similar in L.
lethophaga and L. tridentata (also in the specimen from Willow Creek of doubtful
144
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
c
■a
c
o
D.
O
D.
_o
'o
O
CJ
4>
>
6
•a
c
a
-c
o
c
o
E
o
c
3
O
U
X)
e
D
C
h
O O 00 r«^
vo "^ r*~i -^
1)
(N rJ -^ <n'
rJ
o
>x>
O
\0 O ^■
o o o
o6 00 o\
r4
e\
o ^
Z d
^Z
E-
^ fe c
c
(N
r4
r4
(N -^ r) m
m rJ r) r-
— I r-i i~n
I I
r4 in -^
0\
o
ON <^
as ■
Co
to-* — .
■ 03 O c«
;- O -5^
o
5il
X! <^ 2 c
r- rt OJ "if*
_« ^/5 * U
o
u
•o
CO
3
o
X)
O
>n
o\
o
OO ^ fS
c^-
— OC--
r^ \o — ON
o
Tf -t O
rr, ^H o —
(N r- NO
NO
r)
Si >^
<N
n
CO ^ oo m
oo (s
ON
NO
to rz
>^—
c/) 03
C/5
^o
_a>
> J
OJ
5
D,
C/5
m
>/~i
Oo
_
c
^
^
^~^
*—
c
x:
c.
o S «^-
3:
«
o
— 1 to-f --
^
ID
E
2
d
C/5
5
i
^
NO
r^ ^ ^^
. G .
03 <^ O
t^ E ^
^- o
■~J
1971
HUBBS: A NEW NONPARASITIC LAMPREY
145
a
c
O O w-i m
5
a\ oorJ t-;
odo — ^O
9>^
■K«
.^i
o
o
1
1 1 11
V)
OS
1 1 1 1
C
_o
^-t
iH
oo
3
D.
1
III
O
tA
r~
3
.2
O
B
\c
O
3
u
v~,
1 1 1
u
I-.
o
D.
u
Uri
_o
1)
>
'C
^
u
c
s
1
fN — fs> w->
^
n.
^^
CS
."2
E
'E.
(N
3
i^
1
— <N Tf r-~
C
'S
.-^
• —
3
T3
U-i
C
O
rt
^— ^
L-
J2
7
r^j — <N VO
2
e
ON
3
C
-5:
^H
*^
C«
<4j
-t-J
00
O
1
— —in
CS
h
t^
1
i~
■«^
^
Ci.
r»
£:
.,
a
'■^
-J
1
n 1 1 fS
_c
^
<u
(U
■<-•
Tt
1
1
(^
1 1 1
O
E
3
a
r-)
1
o
u.
O
I-.
u
tZ)
O
a
vt-i
c
tA
-*— »
c
3
O
U
■^
^
0 ^
■a
2
^6
3
eS
, — .
c
w >.
§= -'23-
o
y5 ."S
S' ""^"^
t«
—
o -
> —
^•£:lu?
c^
c^
h-
y.
■^"^
o
\0 vo O
O
O
Tt vD
>/-i -^
^ ^
^ ~
« o
c
^ c:^
<u
J3
3
o
I-
I-.
_o
'C
<u
O
a,
."2
'S.
on
3
O
"S
3
a
>^ .
o5
o «
. ^"3
; "X:
. (U I-
; Q. 3
if I
'-§ S
' t c
•a
c
a
(L>
lU
o
o
E
o
u
c
2 c
1^
c^ ">
!C ^
°-£
3 C
T3 J2
X) j=
n! ■«-
"c5 E
> C
«i
c|
~ V.
a .
c c
o_E
E ci.
— ^ c/:
rt
f~- E
4>
C
<U C«
E =
2 ^
ii Q.
C
<
(U ■-
= x:
^ 1)
C/5
c
Q.
C/5
iri
— I 00
^ —
-H "^
ON
ON
O
m
oo
u
E
I
<N
I
NO
r^
c
3
O
o
<+^
O
u
M
C
a;
NO
o
o
^ -
_ nC
NO
::; «-
c
3
O
u
O
c
146 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
pertinence), though the counts for the nonparasitic form may be slightly the more variable
and slightly the lower on the average, because of some low-count variants.
MYOMERES
The myomeres were counted, as recommended by Hubbs and Trautman (1937: 28),
from the first segment that wholly (or almost wholly) lies behind the groove around the last
gill-opening, to the one whose lower posterior angle lies in part or wholly above the cloacal
slit.
The number, for both ammocetes and adults, has proved to be highly variable in
Lampetra lethophaga. with a suggestion of some regional diversity, though with a broad
overlap. For the three stream systems the counts follow:
Pit River (29 specimens): 63-69; mean 65.62±.31.
Sprague River (8): 58-66; mean 63.50±.94.
Crooked Creek (46): 63-73; mean 67.20±.29.
Bond (1961: 14) gave the number for ammocetes, presumably from Crooked Creek,
as about 65-70. Ting T. Kan (pers. comm., 1970) counted 63-69 (mean 65.71 ±.16) in
95 ammocetes from Crooked Creek, in part overlapping the specimens 1 have counted.
These numbers roughly approximate, in mean value, those found for Lampetra
tridentata, in which, however, there is much regional variation in this character. On the
basis of myomere counts Creaser and Hubbs (1922: 6) erroneously separated that species
into two subspecies: " Entosphenus t. tridentatus" from Unalaska to the Columbia River,
with 68 74, and "E. t. ciliatus," from southern Oregon to southern California, with 57 -
67 myomeres. For a series from Coyote Creek in central California, however, Hubbs
(1925: 592) gave the range as 67 - 76, and Hubbs (1967: 307) listed 60 - 70 for 5 specimens
from southern California and from off northwestern Baja California. Other, unpublished,
counts have mostly approximated 70.
PROPORTIONAL MEASUREMENTS
The proportional measurements (Table 8) of body parts (expressed as permillage of total
length) are fairly consistent among the adults referred to L. lethophaga from the different
stream systems, but are in part somewhat different from the values for the dwarf Klamath
River stocks of L. tridentata. Outstandingly different is the size of the oral disc, which, as
would be expected, is the smaller in the nonparasitic form (36-49 vs. 74-96); the Willow
Creek form (of uncertain species) is strikingly intermediate (64). A similar relation, as to
be expected, holds for the snout length, though with some overlap. Little difference is
indicated for eye length of adults, except that the eye is largest in the Willow Creek
specimen. Length over gill-pores seems to average only slightly lower in L. lethophaga than
in the Klamath precocious stocks of L. tridentata: the value for the Willow Creek example is
a bit higher than the average for the L. tridentata series. Virtually the same relation holds
for body depth, again with much overlap. The tail, also with much overlap, averages longer
in the L. lethophaga than in the L. tridentata series, but is distinctly longer in the Willow
Creek specimen.
Measurements of ammocetes and transformers as well as adults of L. lethophaga
indicate some average ontogenetic changes in body proportions. Consistent differences
with age and stage, however, are not clearly shown for the tail length. As is usual in
lampreys, the body depth increases and the point of greatest depth shifts from near the last
gill-slit to just before the origin of the dorsal fin. The distance over the gill-pore series is
shorter in the transformers and in the adults than in the ammocetes. The eye, snout, and
length over gill-pores appear to average proportionately larger in adults than in trans-
formers.
1971
HUBBS: A NEW NONPARASITIC LAMPREY
147
2
c
c
c
3
c
a.
3
_o
O
o
<u
>
E
CO
•a
c
«
c
to
u.
c
v.
(L>
V5
in
<U
C
^
<U
lU
u
rt
p
c/^
rt
_c
u
^
c«
<<£-.
<u
3
w
00
>
_u
C
rt
-O
<u
s
b g
3^5
o to
c c
c/0 «J
x:
>. to
> c
'3j
c/:
c/5 (U
Si >
D.
r~-
>/-> t~- On t^
r<-j
^ M- ^ ^
^.^
^^•^ oocc
^^ ir^ i^i IT) i/~i
1 1 1 1
r-~
M-
1
I
1 1 1 1
^ n-i (N \0
rj
m -^ ■'^ m
ON
I
/--v
^/~i
OO r-1 O ^
vo r- oo r^
ON
1
^ 00 00 ^^
^ r^ oo o\ o\
1 1 1 1
ON
<N
1
o
1 1 1 1
-H rj >^. —
^O vC ^ '•O
00
NO
o
rj ci —
-^ n —
pj >/-. r-
r<-i r-l r^i
7 7 7
— — r-
ON rj ON
o o —
\C \D \0
r^ r-\ oo
NO NC NC
I I I
nC oo — c
>/-, i/-i ly-,
ON
o
n
00
NO
E
r-
C
o
o
O O oo ON
ri n — —
n -* — -^
(vl <^J rJ pj
1 1 1 1
(N
O
1 1 1 1
oo NO "^ in
r~-
^^o---
m
NO oo O oo
o
ON On ^- On
' — '
'^ O 00 o
00
(N
r^ia ^ ~ —
o
(N
On — — — --
^^
' — '
1 1 1 1
— ON 00 —
r<-j
oo 00 00 oo
ON
NO
OO c ^ —
NO t~- r- r--
m
oo
NCO NC ^
00 oo oo oo
1 1 1 1
NO
ON
1
1 1 1 1
On O nC ON
1/-, NO NO ^''i
1
O 00 r-~
ON
r- nc — o
<N
ON — ON
o
On ^ C O
00
n r-. ol
m
rl r<-i r«-, r<-,
rj
■ — - . . .
. ^
^^ ■ — ■ - . -
\.^^
iTi rj ON
r^-j
r^j r^. NC 00 00
•— <
o
ON ri —
^
00 (^) (^1 (^4 (N
^
o
rj r^, <-<-,
n-j
(N m r^' <^ <^
1 1 1 1
rr-,
1
r-J -^ ri
1
1 1 1 1
r-4 O iTi in
1
O
00 — >/-,
0\
r-- O >/-i >/-i
r~-
ri ro rj
rl
rl r<-i r^j ri
r)
> c
o ^
B^
rl
,ooU
<1>
3
•4-1
00
3 ^^
— C i-^
-^, ■(i: i£ -^ ■(£:
U V5 tri ^ ,,-
£ §2
O. u C
o
NO
148 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
In his key to the lampreys of Oregon, Bond (1961: 14) entered "body rather stout and
deep" for L. le t hop haga {his "'Lampetra sp."), which would suggest a contrast with L.
tridentata, though he included no alternative for the parasitic species. He had, I assume,
adults of L. lethophaga only from Crooked Creek, where the mature specimens exhibit
nuptial features, with the body often turgid, whereas in general L. lethophaga is much
slenderer than the precocious populations of L. tridentata in the Klamath River system.
Other contrasting characters assigned by Bond presumably reflect the nuptial
modifications of L. lethophaga in Crooked Creek.
COLOR
Life colors were annotated in the field on maturing to mature adults, comprising the
male and female type specimens (UMMZ 130648 and 130649) from Fall River (Location
2), and on one maturing male (UMMZ 130576) from Sprague River opposite Ferguson
Butte (Location 7); collected respectively on August 17 and August 10, 1934. Three types
taken in vegetation displayed only a trace of the silvery color of macrophthalmiae. They
were slaty above to bright brassy-silvery below. The fins were clear waxy-yellow. Of those
taken under a stone, the males seemed to approach the silvery color typical of macroph-
thalmiae more than did the females. The one male from Sprague River, taken in dense
vegetation, was deep purplish-brown above the lighter belly, and showed no trace of the
silvery phase.
Preserved adults not in full nuptial condition, whether early in development or of
stocks that do not attain typical nuptial characters, are dusky purplish over most of the
surface, becoming pale yellowish on lower surfaces of head and trunk and on the ventral fin
fold behind the anus; the mid-dorsal ridge is weakly lighter; and the second dorsal is dusky
on the extreme base only. Specimens in full nuptial development are blackish-purple on the
darker areas and on the basal part of both dorsal fins, and the region about the cloaca is
conspicuously paler.
The caudal fin in adults is variably darkened, but is generally darkest along the edges
of the muscle mass and often lighter near the edge. In high-nuptial adults the caudal area
becomes very deeply pigmented, and is almost black along a basal strip, especially on the
lower side.
Younger ammocetes are almost uniformly darkened, barely lighter below. The caudal
fin is at first largely clear, except in the narrow dark streak margining the muscle mass,
about equally above and below. With increasing size the caudal fins grow darker,
progressively outward, as the lighter margin becomes narrow. At any stage, however, the
paler border varies much in width and intensity, as does the basal dark streak. Occasion-
ally, a submedian dark streak develops on either lobe, and the dark area may be blotched.
Transformers are rather evenly pigmented, and the caudal area is largely dark, with
the paler border and the dusky basal streaks varying much in width and intensity. Recently
transformed specimens contrast sharply with those of L. tridentata, including the pre-
cocious Klamath types, in color, in the same way that the macrophthalmiae of L.fluviatilis
and L. planer i differ (Hardisty, Potter, and Sturge, 1970: 385, pi. 1).
At all stages the pigment in the caudal fin area is often marked along the axial line by
small dusky blotches, from which close-set and very oblique melanophore files extend
across the muscle mass, above and below, margining the myomeres.
All of the color features, and in the stated variation, as described above, were seen in
both ammocete and adult stages of the parasitic stocks of the Klamath River system, and
elsewhere. Therefore, I fail to confirm any pigmentary difl'erentiation that may be implied
in the statements, in the key to Oregon lampreys by Bond (1961: 14), that the ammocetes
of the nonparasitic form that he designated ""Lampetra sp." have "dark pigmentation
1971 HUBBS: A NEW NONPARASITIC LAMPREY 149
outlining tip of tail," and that the larvae of L. tridentata have "a dark line above and below
tip of tail."
Vladykov (1950, 1960) described in detail sharp pigmentary differences at various
ammocete stages between Lampetra lamottenii (Leseuer) and Petromyzon marinus, but J.
R. Nursall and D. G. Buchwald (pers. comm., 1970) have found that Lampetra lamottenii
in this respect agrees essentially with the closely related parasitic L.japonica.
LIFE HISTORY TYPE
An outstanding reason for the interpretation of Lampetra lethophaga as a distinct
species of the Entosphenus complex is its alignment among the nonparasitic lampreys,
which have traditionally been accorded specific rank.
It is now recognized that nonparasitic forms have repeatedly evolved from parasitic
lampreys. The original discovery, now well analyzed, was that of Lampetra fluviatilis
(Linnaeus) and L. planeri (Bloch), and a parallel case involving Pacific forms of the
subgenus Lampetra has been documented (Vladykov and FoUett, 1965). A few parasitic/
nonparasitic pairs were implicit in the revision of northern lampreys by Creaser and Hubbs
(1922) and the repeated origin of nonparasitic forms from parasitic ones was definitely
indicated by Hubbs (1925) for northern lampreys in general and by Hubbs and Trautman
(1937) for three separate lines within the genus Ichthyomyzon, one of which had been
treated earlier (Reighard and Cummins, 1916); the dual origin of nonparasitic forms from
one parasitic species in this genus was indicated by Raney (1952). The speciational aspect
of the repeated origin of nonparasitism in lampreys was mentioned by Hubbs (1940: 203;
1941: 188 - 189). Other authors, in particular Zanandrea (1951 - 1962) treated and
expanded on the problem of "paired species of lampreys." Alvarez del Villar (1966)
discovered the nonparasitic Tetrapleurodon of Mexico, and Potter (1968, 1970; Potter,
Lanzing, and Strahan, 1968; Potter and Strahan, 1968) described, as a full species, the
nonparasitic form of the Southern Hemisphere genus Mordacia. Hardisty (1963, 1969) and
others have also dealt with this problem. The systematic status and frequency of the
"paired species" is being discussed by Hubbs and Potter (in press) and the biological
interrelations are being treated by Hardisty, Potter, and Sturge (in press).
Lampetra lethophaga parallels the other nonparasitic lampreys in the rapid maturing of
the gonads, which attain full maturity soon after metamorphosis. No difficulty is
experienced in sexing either transformers or early-stage adults. As usual in fishes the testis
at comparable early stages can be distinguished from the ovary by the circumstance that it
is a slenderer, thinner, whiter, and more opaque band. The testis of this lamprey was
observed to become markedly lobular as it rapidly enlarges during metamorphosis. The
penis at full maturity remains small (Figure 2A). In the holotype, the penis does not
extrude.
When Lampetra lethophaga was first encountered in August, 1934, and for a long
time thereafter, no doubt was felt regarding its interpretation as a nonparasitic species, the
first to be recognized in Entosphenus. The discovery of the Miller Lake lamprey, a
reportedly even more dwarfed yet parasitic form of the same complex, however, has called
for a more thorough appraisal of the evidence for the ehmination of feeding by the adults
of L. lethophaga.
Not one of the considerable series of adults of L. lethophaga, taken throughout much
of the year (Table 1) and representing a full range of stages in maturity was found to
contain any food in the gut. A few had the intestinal wall darkened by apparent
hemorrhage and a few had lumps of some material in the gut, but these appeared to be
cysts and indeed one when opened discharged a larval nematode. Nor did any of the 114
transformers from Hat Creek (Location 3), collected on October 4, 1968, show signs of
150 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
having eaten.
Ammocetes longer than about 100 mm, taken May 3 16, June 26-27, August 13-
17, September 4, and October 4 20, showed that early stages in the maturing of the
gonads are represented through this long period (Table 1). Transformers, taken on
September 4 (one specimen) and on October 4(114) show that the gonads undergo further
maturing in that stage: the ova in many of these specimens, even before the elaboration and
cornification of the teeth, were estimated, by visual inspection, to be from one-fourth to
one-half full size, and to be far fewer than in the dwarfed parasitic form of the Klamath
system. Bare traces were observed of the postmetamorphic macrophthalmia stage (charac-
terized by much enlarged eyes and silvery color), such as is prominently shown by the
parasitic populations of L. tridentata, both dwarfed and full-sized. The incipient macro-
phthalmiae of L. lethophaga contrast with the typical macrophthalmiae of L. tridentata
just as do the corresponding stages of L. fluviatilis and L. planeri, as well depicted by
Hardisty, Potter, and Sturge (1970: 385, pi. 1).
All adults taken from February 16 to August 17 (Table 1) have gonads in various
stages from early to full maturity, though they are, with only moderate overlap, smaller
than the transformers (Figure 8). The smaller size of the adults presumably resulted from
the actual shrinkage that is known to occur in lampreys during transformation, with, in the
case of nonparasitic forms, no later resumption of growth. However, the transformers and
the adults did not come from the same place.
Pertinent testimony was secured from the personnel of the Klamath State Fish
Hatchery on Crooked Creek (Location 11), where several series of this lamprey were
collected at various times of the year. On August 13, 1934, the superintendent of the
hatchery informed me that considerable numbers of small lampreys about 5-8 inches long
which are removed from the screens of the hatchery ponds each year about June contain
eggs "about the size of whitefish eggs, showing through the belly along almost the entire
length of the body." He added that there are definitely no "runs of eels" in this or in other,
nearby streams. On February 5, 1945, the late Dr. Paul R. Needham reported (pers.
comm.) that the superintendent had not seen any of the lampreys. On recent occasions the
hatchery employees have provided Dr. Carl E. Bond (pers. comm., 1970) with corrobora-
tive testimony.
Fish management studies on lower Hat Creek (Location 3) have led Dr. Roger A.
Barnhart (pers. comm., 1970) to conclude that the lamprey of that stream is a nonparasitic
"brook'" form. Nor have I found any evidence of parasitic lampreys in the drainage basins
of Pit, Sprague, or Sycan rivers, or in Crooked Creek, from which the material of
Lampetra lethophaga was obtained.
SEX RATIO
By gross examination it was readily feasible to ascertain the sex ratio of ammocetes
larger than 90 mm total length and of all transformers, as well as the adults. The sex of
some ammocetes and a few of the transformers of earliest stage was not listed until a small
piece of the gonad had been teased apart or crushed between slides, so as to distinguish the
testicular tubules from early ova. For only two of the 1 15 transformers examined was any
hesitation encountered in the sexing.
For all 187 specimens sexed, of all stages, the females were moderately outnumbered
by the males 87: 100. The ratio of females per 100 males seems to decrease with the stage of
development: from 128 for 23 female and 18 male ammocetes longer than 90 mm, through
79 for 51 female and 64 male transformers, to 72 for 13 female and 18 male adults. The
significance of this indicated change in sex ratio, and indeed its validity, call for further
study. Marked fluctuations have been indicated for the sex ratio of Lampetra planeri
1971 HUBBS: A NEW NONPARASITIC LAMPREY 151
(Hardisty, 1944, 1954; Zanandrea, 1951). A preponderance of males has been found for
Petromyzon (Applegate, 1950, App. E).
ETYMOLOGY
The name lethophaga, figuratively referring to the elimination of feeding as adult, is
formed by combining the latinized expressions leth-, from the root of \-q9-q, a forgetting or
forgetfulness; the normal connective -o- in words of Greek origin; phag-, from the root of
<}>ay£iv, to eat; and -a, from the feminine of the adjectival suffix -os.
RELATION OF NONPARASITIC LAMPETRA LETHOPHAGA
TO DWARFED PRECOCIOUS PARASITIC FORMS
REFERRED TO LAMPETRA TRIDENTATA
Although Hardisty and Potter (in press) hold to the opinion that the genes do not
interflow between the members of the respective parasitic/nonparasitic pairs, some
indications have been emerging that within several of the "paired species" of lampreys the
typical large parasitic form may to some degree intergrade with its dwarfed nonparasitic
representative. Intermediacy in size is indeed shown by the "praecox" forms in several
species, such as have been discussed by Berg (1931, 1948, 1962) and others. The reduction
in size is thought to be due to a shortening of the parasitic cycle, which is also a sign of
intermediacy. The high frequency of the "paired species" (Hubbs and Potter, in press)
strongly suggests speciational plasticity, and begets the idea that nonparasitic populations
may be polyphyletic even within any species complex.
There have even been some suggestions that the nonparasitic types should be accorded
only subspecific status. Thus, it has been proposed (Hubbs and Lagler, 1958: 36-37; 1964:
36-37) that the American brook lamprey be distinguished only subspecifically as
Entosphenus lamottenii lamottenii, since "in Alaska it appears to intergrade with the
typical, often anadromous parasitic {ormjaponicus.'" This action, although drawing some
support from studies by Heard (1966) and by J. R. Nursall and D. G. Buchwald (pers.
comm.), was probably premature, but the problem of systematics within the Lethenteron
group (now probably best treated, along with Entosphenus, as a subgenus of Lampetra), is
definitely open. In some recent studies difficulty has been encountered in the identification
of certain specimens of Ichthyomyzon, where /. bdellium (Jordan) and /. greeleyi Hubbs
and Trautman are sympatric (Ernest A. Lachner, pers. comm., 1971). However, for the
present at least, it seems advisable on both practical and theoretical grounds to maintain
the nonparasitic forms at the full specific level.
Suggestions that the parasitic and nonparasitic representatives may intergrade stem in
considerable force from studies of the Entosphenus complex in the related drainage basins
of the Pit and Klamath rivers, from which the nonparasitic form is herein being made
known. Indeed, intermediacy between the two trophically contrasting types was probably
first suggested (Hubbs, 1925: 589, fig. 16) for populations of "Entosphenus tridentatus" in
these two river systems. A race of this species in Goose Lake of the Pit River system was
shown as straddling the intervening line on the chart, on the basis of an examination of
material in the United States National Museum collected by Barton Warren Evermann,
and it was stated that: "Of a series of small adults, all taken on trout in this lake, the males
showed a decided approach toward the brook type of lamprey in the close approximation
of the dorsal fins, relatively blunt teeth, atrophy of the intestine and precocious sexual
development. The females, oddly, were not so altered, but resembled the normal parasitic
young of the species." It was added that "Some specimens from Klamath Lake, not far
distant from Goose Lake, but in a distinct stream system, also show evidence of
degeneration." I may have been dealing, however, merely with stocks in these lakes that
152 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
were maturing at a small size, perhaps particularly as males, and were assuming the
nuptial characteristics that may be essentially similar in the two trophic types.
I have found other evidence of the existence in the Klamath River system of a
presumably indigenous and landlocked local form (or of forms) intermediate in size
between the large sea-run Lampetra tridentata and the endemic nonparasitic dwarf, L.
lethophaga. Indeed, as is indicated below, there is reported to have been, in Miller Lake, an
isolated parasitic form at least as small as L. lethophaga.
Fifteen maturing adults only 214-274 mm long, no doubt at approximately the
maximum size they would have attained, from Klamath River at Klamathon, have been
studied. These specimens (SU 28783 and 35555) were collected at a fishery research
installation, respectively by E. A. McGregor, in the fall of 1922, and by Charles H. Gilbert
(who died in 1928). They are distinctive in the high number of cusps on the teeth. In this
respect they resemble a series of 374 specimens in the macrophthalmia (early-transforma-
tion) stage taken nearby, on February 24 to April 5, 1931, in Shasta River at its junction
with the Klamath, in wiers set across the stream to tally downstream salmon migrants
(Brown, 1938). Tooth and cusp counts taken on all specimens of both lots are herein
tabulated, along with those o{ Lampetra lethophaga (Tables 3 - 6) as these lots are taken to
represent or to approximate the ancestral form of L. tridentata from which the nonpara-
sitic species originated. The counts for these two lots of L. tridentata are in fair agreement,
with the unexplained exception that the Klamathon series yielded much the higher counts
of bicuspid posterior circumoral teeth. This discrepancy seems particularly strange, when
it is noted that the two series were taken only about 1 1 km apart.
Stranger yet are the characters of a single specimen, even more deviant than those of
the Klamathon series. It was taken from Copco Lake, only about 18 straight-line
kilometers farther upstream, near the Oregon state line. This specimen, CAS 25987,
collected by Millard H. Coots, was adhering to the tongue of a sucker, Catostomus
rimiculus, that had been caught in a gill-net set overnight, June 3^, 1953. It is a subadult
female, with the gut turgid with food and with the enlarging ova too few for the large sea-
run type. It is 241 mm long, about as large as the Klamathon specimens, and its permillage
measurements are included within the range for the Klamathon series in Table 8, but the
dentition is sharply divergent: in particular, the total number of bicuspid posterior
circumorals (16) is higher than in any of the 452 other specimens of L. tridentata tallied
(Table 6), and some other counts are aberrant. The dentition of this specimen is as follows:
longitudinal lingual cusps 29-29 = 58 (high); anterior lingual cusps 12 + 1 + 13 = 26
(high); supraoral cusps 3 (normal); infraoral cusps 6 (aberrantly high); anterior circum-
orals 5 (usual); lateral circumorals 2-3-3-2 — 2-3-3-2 (normal); posterior cir-
cumorals 19 (very high) with 16 bicuspid (absolutely extreme) and only 3 unicuspid (next
to lowest number; the seventh, tenth, and eleventh of the 19 teeth beginning with the
anteriormost and foremost on the right side); 2 marginals on each side greatly dilated; total
marginals 57; total teeth 104 (high); total cusps 220 (exceptionally high). Oral papillae 12
+ 1+9 = 22 (high).
The differences between the lampreys comprising the Shasta River, Klamathon, and
Copco Lake series exemplify the tendency toward high local variability of resident
lampreys, on a mosaic pattern. Small wonder that Lampetra lethophaga displays some
local differences.
Additional material of the "praecox" type of L. tridentata from the Klamath River
and other systems, particularly from Goose and Klamath lakes, are currently under study
by Dr. Carl E. Bond of Oregon State College and his graduate student Ting T. Kan. More
or less comparable precociously spawning forms now referable to L. tridentata have come to
my attention from southern California (Hubbs, 1967) and northward to Vancouver Island,
\
1971 HUBBS: A NEW NONPARASITIC LAMPREY 153
British Columbia. Dwarf, nonmigratory races have been discussed by McPhail and
Lindsey (1970: 58 59). There appear to be numerous forms that seem to be comparable to
the races of salmon of diverse and distinctive size at maturity (in each case just before
reproduction and death).
Limited material at hand from the drainage basins of Clear Lake and of Lost River, in
the Klamath system, may bear on the problem of possible intergradation of parasitic and
nonparasitic representatives of the Entosphenus group. This is particularly true of a single
specimen, a postnuptial male only 176 mm in total length (SIO 65 144) that was collected
by William Johnson and Edward J. O'Neill on May 13, 1965 in Willow Creek, tributary to
Clear Lake, Modoc County, California. Originally, as shown on the one-degree 1:250,000
U.S.G.S. Modoc Lava-Bed Sheet of 1892, and on the accompanying distribution map
(Figure 3), Willow Creek was the upper, southern headwater of Lost River, of the Klamath
River upstream complex, but with an intermittent, presumably flood inflow into Clear
Lake. Currently, the flow is directed into this lake, which thus has been enlarged as a
reservoir. This specimen was first regarded as referable to Lampetra lethophaga, then was
thought to represent, more likely, a greatly dwarfed parasitic race. A third possibility,
suggested by some recent testimony, is that the specimen in question is merely an
exceptionally dwarfed example of a moderately dwarfed resident population. Mr. O'Neill
has informed me (pers. comm., 1971) that a number of lampreys 10-12 inches long have
been taken adhering to crappies {Pomoxis sp.) in Willow Creek, and that many of the
''rough fish" of this stream have shown lamprey scars.
Neither by tooth and cusp counts (entered on Tables 3-6 in the row labelled "^L. sp."),
nor by other characters, have I found it feasible to decide to which of these possibilities this
Willow Creek fish can be assigned. In general, such postnuptial specimens are often
difficult to refer to trophic type. On comparison with L. lethophaga and the precocious
Klamath forms of Z.. tridentata this particular specimen is conspicuously intermediate in
several respects, as follows:
The Willow Creek specimen (Figures 2 C, D) is definitely smaller than any known
mature parasitic adult from the Klamath River system, or elsewhere (other than the
representatives mentioned below of the tiny parasitic form of Miller Lake, a disjunct part
of the Klamath basin); yet is 16 mm longer than the largest transformed adult at hand of L.
lethophaga (Figure 8). It is 23 mm shorter than the largest specimen in transformation, but
lampreys shrink considerably during and just after metamorphosis.
Particularly notable for the trophic indication is the measurement of the buccal disc
(Table 8), which is intermediate, without overlap: 64 thousandths of the total length, vs. 36
- 58 (mean 47) for L. lethophaga and 74 96 (81) for the L. tridentata series. The
correlated snout length is also definitely intermediate, but with slight overlap. Body depth
is probably also intermediate, but tail length and eye length are higher than in either type
under comparison, and length over gill-pores is likely also high (Table 8).
The regular formulae for the cusps on the two oral teeth (supraoral 3 and infraoral 5)
and for the four lateral circumorals of either side (23-3-2) tend to align the Willow
Creek specimen with the parasitic type (Tables 3, 4). Furthermore, the teeth are rather less
degenerate than in breeding examples of L. lethophaga, and the median cusp on the
supraoral is relatively large and sharp, instead of being reduced or absent as it usually is in
the nonparasitic form. The total count of posterior circumoral teeth, however, seems
intermediate: 2 higher than any count for L. lethophaga and on the low side for the
parasitic lampreys from Klamath River (Table 6).
The general appearance of the Willow Creek specimen approximates that of the
mature adults of Z.. lethophaga from Crooked Creek, so much so as to suggest consanguinity:
in each the color is dark, the entire face is strongly turgid, the whole form is robust, the
154 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
dorsal fins are much expanded and in contact, and their edges are minutely frayed.
However, the similarities may largely reflect a similar stage in sexual development.
In conclusion, it does not seem justified to align the Willow Creek specimen with
either L. lethophaga or with the Klamath River precocious populations of L. tridentata. Its
general intermediacy, along with some extreme features, indicate it to be a representative
of a somewhat distinct local form.
Data possibly bearing on the uncertain status of the Willow Creek lamprey are
furnished by two specimens taken on June 24, 1965 by Edward J. O'Neill and James Keith
on the Clear Lake National Wildlife Refuge, during banding of White Pelicans. They were
among 44 lampreys spewed up by one young pelican. These may have been captured by the
parent(s) in Clear Lake (now used as a reservoir and a refuge), but may have come from
elsewhere, as White Pelicans sometimes forage many miles from their rookery. Whatever
their source, these 2 specimens, although very considerably damaged, seem to represent a
stock different from that of the one Willow Creek specimen. They are larger: one not
sexable measures about 220 mm in total length and the other, a female with nearly ripe
ova, about 240 mm, vs. 176 mm. Permillage proportions, though hardly precise, seem to
diff'er: tail length, 317 and 327 vs. 340; eye length 15 and 17 vs. 24; snout length 73 and 86
vs. 93. Teeth and cusps number: transverse lingual cusps, 12+1 + 12 and 12+1 + 14
(higher than in L. lethophaga); cusps on supraoral 3 and on infraoral 5 in each, as usual;
posterior circumorals 16, apparently all unicuspid. Clearly these two specimens seem to
represent a dwarfed population of parasitic lamprey, probably similar to the precocious type
sampled from the Klamath River near Klamathon and perhaps similar to the form or
forms occurring in Shasta River and the Klamath lakes.
The most surprising circumstance bearing on the relationship between Lampetra
lethophaga and the parasitic forms referred to L. tridentata is the discovery by Dr. Carl E.
Bond of a parasitic form indicated as even smaller than L. lethophaga. This form seems to
have been endemic in the small drainage basin (shown on the distribution map. Figure 3) of
Miller Lake (named Fish Lake on some old maps), which basin is a disjunct, endorheic
unit, ending in a marsh, at the north end of the Klamath drainage system. Dr. Bond has
stated (pers. comm., 1971) that:
The evidence for predation in the Miller Lake lamprey is strong. The little beasts prevented the
maintenance of a trout fishery in the lake. They would kill trout and tui chub [Gila bicolor] and then mine
out the soft parts, leaving the perforated skin and the skeleton on the bottom. Spawned-out lampreys were
also devoured — even on the spawning beds. Miller Lake is in T 27 S, R 6 V2 E, Sections 11-14 and in a
distrupted portion of the Klamath River drainage.
The lamprey is now extinct, a[tragic]victim of a toxaphene operation designed by the Oregon State
Commission to eradicate it. I had hoped that it had survived in the outlet. Miller Creek, but no specimens
were taken there through extensive and thorough electrofishing by Harry Lorz of the Oregon State Game
Commission in 1970, many years after the extinction of the lamprey.
Adults from the spawning beds range from 72 to 129 mm, mostly between 80 and 105 mm. Many of the
near-term ammocetes and some of the non-spawning adults are longer than the spawning adults. Some of
the lampreys spawned in the very cold creeks that are tributary to Miller Lake, but the major spawning
areas were along the lake shore.
I have thought much during the past 15 years about the significance of the Miller Lake lamprey as a
transitional form in a progression to nonparasitic habits. The creatures were locked into a
system with only the tui chub, if indeed the chub was not planted later, and had adapted to the paucity of
food by cutting a year or two off their lives, so that they metamorphosed in the fall and spawned in the
spring and summer — feeding fiercely if food were available, but not growing beyond the length of the
ammocoetes. Even when the Game Commission planted trout yearly the lampreys did not grow beyond the
range I mentioned - although the trout plants were wiped out each winter.
To me, the outstanding attribute of this form is that parasitism was not obligatory and that the
population finally consisted of the offspring of ancestors that could feed fiercely if prey were on hand, but
apparently could mature and spawn on a starvation diet. I suspect that if any native fish other than the
lampreys were in Miller Lake, the populations were kept low by the lampreys — much as in the same
1971 HUBBS: A NEW NONPARASITIC LAMPREY 155
manner that hatchery plants of trout were virtually wiped out. One year class could feast to the extent that
the next would have little or nothing to eat, unless the Game Commission planted more trout.
The alternative name of Fish Lake, found on various old maps, suggests that the tui
chub was probably a native associate of the Miller Lake lamprey.
A detailed comparison of L. lethophaga with this parasitic midget awaits the
completion of the study by Dr. Bond and Mr. Kan.
Clearly these data on the small lampreys of the Klamath River system are pertinent
not only to their systematic appraisal, but also to the general problem of the relationships
between parasitic forms of lampreys and their nonparasitic relatives, probable derivative.
It is certainly conceivable that the two types do in some way intergrade, with or without
active exchange of genes, and the strong possibility remains that some or even all of the
nonparasitic types may be polyphyletic.
GROWTH
Data are inadequate to indicate clearly the life span of the ammocetes of Lampetra
lethophaga but it appears probable that the period is at least four years (Figure 8) —
comparable to the evidence for other lampreys (Loman, 1912; Meek, 1916; Okkelberg,
1921, 1922; Hubbs, 1925; Schultz, 1930; Ivanova-Berg, 1931; Leach, 1940; Knowles, 1941;
Hardisty, 1944- 1969; Churchill, 1947; Applegate, 1950; Horn and Bailey, 1952; Dendy
and Scott, 1953;Seversmith, 1953;Zanandrea, 1951, 1954b; Hardisty and Potter, in press). In
addition, I have unpublished, original, confirmatory observations for Petromyzon marinus
Linnaeus, Okkelbergia aepyptera (Kirtland), Lampetra tridentata, and L. lamottenii. The
graph for L. lethophaga (Figure 8) covers all localities and all dates, but most ammocetes
measured were collected in August (40) and October (95); only 17 others were taken over
the time span of May 3 to September 4. The sharp mode at 20 - 29 mm length presumably
represents young-of-the-year, all of which were collected on August 13 (Table 1). The four
modal size classes in the total-length range of 90 129 mm obviously represent at least one
older year-class. The intervening size classes likely represent another. On the basis of life-
history studies of other lampreys, it is highly probable that the 7 ammocetes longer than 1 59
mm, all taken on October 20, would have undergone another year of larval life.
The time of metamorphosis from the ammocete stage is adequately indicated only for
the large collection of 114 individuals in early to late stage of transformation taken on
October 4, 1968, by poison in Hat Creek near its mouth into Pit River, at Location 3. The
only other transformer examined was the specimen taken on September 4, 1898 in the
North (or South) Fork of Pit River, at Location 1 (Table 1).
Entosphenus tridentatus, as well as other lampreys, probably also metamorphoses in
the fall. Along with a large sample of ammocetes of that species taken in Trinity River at
Lewiston, California, on November 8, 1945, are 2 males and 2 females in a late stage of
transformation, but with the teeth remaining in pads or only partly and variably exposed.
These transformers are 102, 106, 108, and 1 15 mm long, within the dominant size classes of
the macrophthalmiae taken either at the same place, or in the lower Shasta River (Figure
8).
The size frequencies of the transformers of Z,. lethophaga form a normal curve (Figure
8), which lies almost entirely higher than the sizes of either the transformed and
transforming examples, just cited, or the precocious Klamath River type of/,, tridentata.
An incompletely transformed specimen from Coyote Creek, at San Jose, California, was
listed as of intermediate size, 141 mm, by Hubbs (1925: 594). It has been shown that
nonparasitic lampreys metamorphose at a larger size than do their larger, parasitic
relatives. This relation has been so stated for the paired species of Mordacia (Potter, 1970:
497) and is being indicated as a generalization by Hubbs and Potter (in press) and by
156
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 8. Total-length measurements of all material studied of the nonparasitic Lampelra lethophaga. of the
specimen (L. sp.) of doubtful pertinence from Willow Creek, and of the parasitic, but praecox, form of L.
iridentala. from Klamath River. The measurements were made to the nearest mm, but are grouped by cm (10-19,
20-29, etc.). Number of specimens shown on ordinate.
Hardisty and Potter (in press).
In concordance with other evidence of size shrinkage during metamorphosis in
lampreys, perhaps particularly in nonparasitic species, the modal size of the transformers
seems lower than the size of the largest ammocetes, which are the only ones that could be
expected to metamorphose. However, the ammocetes and the transformers were not taken
at the same locality.
It is probably also significant, and in line with expectation for a nonparasitic lamprey,
that the postmetamorphic, maturing and mature adults of L. lethophaga are, with little
overlap, smaller than the transformers of the same species, but are, to a comparable extent,
larger than the recently transformed macrophthalmiae of the lower Shasta Creek
population of the parasitic L. tridentata (Figure 8). It has been observed for several
lampreys that growth during transformation is negative.
It appears (Table 1 ) either that the time of breeding is unusually variable in Lampetra
lethophaga. or that full sexual development may be long delayed. Specimens taken in
y
1971 HUBBS: A NEW NONPARASITIC LAMPREY 157
Crooked Creek (Location 1 1) and in a spring near Sprague River (Location 10), the only
ones exhibiting definite nuptial modification, obviously had spawned, or would have
spawned, in late winter or spring, for they were collected from February 16 to April 6
(Table 1). The other adults, at least some seemingly neotenic (see next section), were
collected over the summer, from June 2 to August 17. The 1 1 adult types, taken on August
17 at Location 2, exhibited a wide variation in maturation, thus suggesting prolonged
spawning unusually late in the year, or possibly a partial or even complete suspension of
sexual development over the next winter. The single adults taken at five locations from
June 2 to August 10 also varied widely in degree of maturation, further suggesting
prolonged sqawning over the summer.
The great difference between the growth patterns of two resident types in the Pit-
Klamath area is that L. lethophaga almost surely does not grow as adult, whereas the
presumably resident, dwarfed forms of L. tridentata appear to double their size during
their adult, parasitic life — even though they reach only about half the length attained by
the larger sea-run populations of L. tridentata.
NEOTENY
Lampetra lethophaga exhibits, apparently in some populations only, definite in-
dications of what may be considered as neoteny, other than that of merely reproducing
soon after the postammocete metamorphosis.
The only specific reports of neoteny in a lamprey that I have found are by Zanandrea
(1956, 1957a, 1958a, 1961) for the nonparasitic Lampetra zanandreai Vladykov. (This
species is regarded by Hubbs and Potter (in press) as an isolated member, confined to the
Po River drainage of Italy, of the subgenus Lethenteron, which has hitherto been regarded
as restricted to the Arctic Ocean drainage from Europe to North America, to the North
Pacific tributaries of Asia, and to northeastern North America.) Zanandrea (1961: 530)
found at one locality 12 female ammocetes in an advanced ("third") stage of ovarian
development, one of which "showed well-developed principal secondary sexual characters,
namely, enlargement of the two dorsal fins, development of the anal pseudo-fin, and the
transparent appearance of the body wall, through which the eggs can be seen . . . characters
. . . normally associated only with adults that are about to spawn." He obtained at another
locality other neotenic female ammocetes, constituting about one-fifth of a series of 221 in
the larval stage. He suspected that neoteny in each place may have been induced by tannery
pollution, but a test performed to check this suspicion was inconclusive.
The type of neoteny attributed to L. lethophaga involves the maturing at some
locations of apparently all individuals of both sexes in the prenuptial condition. This is
most strikingly shown by the adults from Fall River (the type station, at Location 2). They
had passed through the ordinary, prejuvenile metamorphosis (transformation), but al-
though some are in full maturity (witness a female turgid with large ova — Figure ID),
none has developed the ordinary nuptial attributes: melanistic pigmentation and the
"principal secondary sexual characters" outlined above in the quotation from Zanandrea.
These attributes are seen, well-developed, in 13 adults from Crooked Creek (Figure 2 A, B),
in two males from the Sprague River system (Locality 10), in the specimen of uncertain
species from Willow Creek (Figure 2 C, D), and in the 15 specimens, in early to late stages
of maturing, of the dwarfed parasitic form (referred to L. tridentata) from Klamath River
at Klamathon. These are the normal attributes of the nuptial stage of lampreys in general,
attained at what may be called the second or nuptial metamorphosis. That transformation
seems to have been elided at the head of Fall River (type locality of L. lethophaga), and is
not evident in other specimens from the Pit River system. Series from the Klamath River
complex other than at Locations 10 and 1 1 seem to be developing like the Fall River lot
158 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
(maturity in prenuptial appearance). Therefore, the retention or ehmination of the normal
nuptial metamorphosis does not appear to provide a sound basis for the systematic
distinction of nonparasitic lampreys of the Klamath complex from the Pit River form
(typical L. lethophaga).
The stocks that are neotenic in the sense of developing without the usual nuptial
attributes retain to a very large degree, through maturity, the features displayed in the late
stages of the ordinary, prejuvenile metamorphosis. The body remains trim and non-turgid,
and pale; the dorsal fins remain well separated, with at most a slight connecting ridge, and
stay thin and non-turgid, low, and unfrilled at the margin; the other fins stay rather similar;
even the anal is generally little enlarged, though moderately enlarged and turgid in the
female shown in Figure ID; the cloacal margins are little swollen; and the preanal fin fold
is scarcely enlarged.
The neoteny was notably evident at the type locality (Location 2), where the water was
cold (summer readings of 11.4-13.3° C), but the low temperature was presumably not a
factor suppressing nuptial development because in Crooked Creek (Location 11), where
the nuptial characters are well developed, the water was even colder (7.8° C in August).
REGIONAL DIVERSITY
There is considerable evidence of local diversity in Lampetra lethophaga other than
the retention or loss of the nuptial metamorphosis, just discussed, but this observed
diversity does not seem to warrant specific or subspecific distinction between the popu-
lations of the two main stream systems, or between populations within either system. In the
analysis of variation the Crooked Creek population is contrasted with the populations
sampled from the Sprague (including the Sycan) River system, both in the Klamath
complex.
There seems to be some regional difference in the frequency of cusp number on the
oral plates (Table 3). Loss of the median supraoral cusp is less frequent in Fall River
specimens (the only fully adult ones from the Pit River system) than in those from Crooked
Creek, but the few examples from the Sprague River system are intermediate. Some
increase in number of infraoral cusps beyond the Entophenus standard of 5 was found in
material from Crooked Creek and the Sprague River system, but not in the Fall River
specimens.
There appear to be differences between the samples from the three stream systems in
the frequency of reduction in cusp number on the four lateral circumorals from the typical
Entosphenus pattern of 2-3-3-2 (Table 4). The frequency of bicuspid posterior
circumorals runs higher in the Pit River sample than in the Crooked Creek specimens,
whether tallied by individual teeth numbered from the side (Table 5) or by total number
(Table 6), and again the specimens from the Sprague River system seem intermediate.
There may be average differences in number of trunk myomeres: lowest in the
Sprague River system, highest in Crooked Creek, intermediate in the Pit River system.
There are some indicated average differences in proportional measurements (Table 8).
In the larger ammocetes tail length and length over the gill-pores average longest for the
Sprague River specimens, but only 2 are available. Very slight differences among the
adults may be related to the expression of nuptial features in 2 of the 5 adults from the
Sprague River system and in all 13 adults from Crooked Creek.
The more or less definite indications of local diversity in Lampetra lethophaga are
consistent with the differentiation, seemingly mosaically arranged, that has been observed
among lampreys in general, and among the nonparasitic forms in particular (Hubbs, 1925:
590). Some citations for the genus Lampetra are as follows: For subgenus Entosphenus —
Creaser and Hubbs, 1922: 6, 10-11; Hubbs, 1925: 589; 1967. For subgenus Lethenteron
1971 HUBBS: A NEW NONPARASITIC LAMPREY 159
— Creaser and Hubbs, 1922: 12; Jordan and Hubbs, 1925: 98-99; Hubbs, 1925: 589;
Berg, 1931: 92-93, 98-105; 1948: 35-42; 1962: 29-37; Hubbs and Lagler, 1958 and
1964: 36; Heard, 1966; Hubbs and Potter, in press. For subgenus Lampetra — Creaser and
Hubbs, 1922: 13; Hubbs, 1925: 590. For all three subgenera Hardisty, 1963: 20.
From a partial survey of the literature and from some original material I strongly
suspect (see Hubbs and Potter, in press) that some of the rather confusing treatment of the
local forms of Eudontomyzon reflect strong local diversity more complex than the simple
alignment of the forms into two paired species, the parasitic E. danfordi Regan and the
nonparasitic E. vladykovi (Zanandrea), plus the reputedly unpaired nonparasitic E. mariae
(Berg).
A more detailed and more critical analysis and interpretation of the seemingly
heterogeneous local populations of lampreys seems to be definitely in order.
ACKNOWLEDGMENTS
Many have contributed ideas, notes, specimens, and other assistance for this report.
Particular acknowledgment is offered to those named alphabetically below, and to the
National Science Foundation, which has generously supported my continuing researches on
fishes, currently by grant GB 13319. Dr. Roger A. Barnhart, Leader of the California
Cooperative Fishery Unit at Humboldt State College (Areata, California) sent on loan and
gift the large collection of transforming specimens that he collected and preserved during
fish-management operations on Hat Creek (Location 3); and he provided information on
this collection. Particular acknowledgment is due Dr. Carl E. Bond, along with his
graduate student Ting T. Kan, for extensive field and laboratory data on collections of L.
lethophaga from Crooked Creek (Location 11). They have patiently foregone describing
and naming the subject of this paper, and they have also provided information on the
remarkably dwarfed parasitic lamprey of Miller Lake, Oregon. Dr. Alexander J. Calhoun,
Chief of the Inland Fisheries Branch of the California Department of Fish and Game,
provided needed information on collecting localities and on available material, notably the
large series of transforming brook lampreys from Hat Creek. Through kindly cooperation
and assistance. Dr. William N. Eschmeyer, W. L Follett, Lillian Dempster, and staff' of the
Division of Ichthyology of the California Academy of Sciences, significantly augmented
the material for this study. In 1934, Mr. W. I. Howland, then superintendent of the
Klamath State Fish Hatchery on Crooked Creek, Oregon provided significant information
on lamprey runs locally and in the surrounding area. Laura C. Hubbs participated in the
collection of the types and other specimens and has extensively assisted in the entire
research, not only during the preparation of this report but also during intermittent studies
of lampreys for half a century. Dr. Tamotsu Iwai, of the Department of Fisheries, Kyoto
University, provided information on the distribution o{ Lampetra tridentata in Japan. Dr.
Elizabeth M. Kampa painstakingly drew the dentition of Lampetra tridentata (Figure 7).
Dr. Robert Rush Miller of the Museum of Zoology, University of Michigan, collected
specimens for this study, annotated the habitats sampled, loaned much material for this
and related studies, and provided pertinent field data. The late Dr. James W. Moffett, then
in charge of the United States Fish and Wildlife Service laboratory at Stanford University,
was largely responsible for the securing of large numbers of fyke-net collections of the
dwarfed parasitic stocks o{ Lampetra tridentata from the Klamath River system. Others
who cooperated in securing specimens of this species from northern California were the
late Dr. Paul R. Needham, Dr. Leo Shapovalov, Dr. Stanford H. Smith, and the late Dr.
A. C. Taft. Mr. Edward J. O'Neill, Biologist of the Tule Lake National Wildlife Refuge,
sent me, at the suggestion of Mr. William Johnson of the U.S. Public Health Service, the
fine postnuptial specimen from Willow Creek, tributary to Clear Lake, that is notable as
160 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
being intermediate in some respects between L. lethophaga and the dwarfed parasitic L.
tridentata of the Klamath River system. Mr. CNeill also provided dwarfed parasitic
specimens from a White Pelican nest on Clear Lake. Dr. Ian C. Potter, of Bath University
of Technology, in England, coauthor with me of the revision in press of the lampreys of the
world, contributed many ideas and references that have been utilized in the present
research. Mr. Howard G. Shirley has been patient and skillful in the final drafting of the
distribution maps (Figures 3 and 5) and the graph of length measurements (Figure 8). The
late Dr. Albert Hazen Wright of Cornell University, with the cooperation of Dr. Edward
C. Raney, made available two fine adult specimens that he collected, along with
information on the peculiar habitat.
This paper is a contribution from Scripps Institution of Oceanography, University of
California, San Diego.
LITERATURE CITED
Alvarez del Villar, Jose
1966. not "1964" . Ictiologfa michoacana, IV; Contribucion al conocimiento biologico y sistematico de las
lampreasde Jocona, Mich. Anal. Escuela Nac. Ciencias Biol., 13; 107 144, figs. 1-10.
Aoyagi, Hyoji
1957. General notes on the freshwater fishes of the Japanese Archipelago. Daishukan Shoten, Tokyo: 1-272
+ 22.
Applegate, Vernon C.
1950. Natural history of the sea lamprey ( Petromyzon marinus) in Michigan. U.S. Fish and Wild. Serv.:
Spec. Sci. Rept.; Fish., 55: i-xii, 1-237, frontisp., figs. 1 65.
Bailey, Reeve M., and Carl E. Bond
1963. Four new species of freshwater sculpins, genus Cottus. from western North America. Occ. Pap. Mus.
Zool. Univ. Michigan, 634: 1-27, figs. 1^.
Behnke, Robert J.
1970. The application of cytogenetic and biochemical systematics to phylogenetic problems in the family
Salmonidae. Trans. Amer. Fish. Soc, 99 (1): 237-248.
Berg, L. S.
1931. A review of the lampreys of the northern hemisphere. Ann. Mus. Zool. Acad. Sci. URSS, 32: 87-1 16,
pis. 1 8.
1948. Ryby presnykh vod SSSR i sopredefnykh stran. Faune SSSR. Akad. Nauk SSSR, 27, 4th ed., vol. 1:
1^66, figs. 1-281. (Translated, 1962.)
1962. Freshwater fishes of the U.S. S.R. and adjacent countries, vol. 1; 1-504 (= 1-465 of 4th ed., 1948 + 2-p.
addenda to 4th ed. +1317-1325 from vol. 3 of 4th ed.), figs. 1-281. (Translation of 1948 ed.)
Bond, Carl E.
1961 . Keys to Oregon freshwater fishes. Agric. Exp. Sta. Oregon Sta. Univ., Corvallis, Tech. Bull. 58: 1^2,
figs. 1 10.
Brown, Merrill W.
1938. The salmon migration in the Shasta River (1930 1934). California Fish and Game, 24(1): 60-65, figs.
17-22.
Churchill, Warren S.
1947. The brook lamprey in the Brule River. Trans. Wisconsin Acad. Sci., Arts and Letters, 37, for 1945:
337-346, charts 1-2.
Creaser, Charles W., and Carl L. Hubbs
1922. A revision of the Holarctic lampreys. Occ. Pap. Mus. Zool. Univ. Michigan, 120: 1 14, pi. 1.
Dendy, Jack S., and Donald C. Scott
1953. Distribution, life history, and morphological variations of the southern brook lamprey, Ichthyomyzon
gagei. Copeia, 1953 (3): 152 162, figs. 1-2, pi. 1.
Evermann, Barton Warren, and Seth Eugene Meek
1898. A report upon salmon investigations in the Columbia River basin and elsewhere on the Pacific Coast
in 1 896. Bull. U.S. Fish Comm., 1 7 (2): 1 5-84, figs. 1 6, pis. 1 2.
1971 HUBBS: A NEW NONPARASITIC LAMPREY 161
Gilbert, Charles H.
1898. The fishes of the Klamath River basin. Bull. U.S. Fish Comm., 17 (1); 1-13, 6 figs.
Hardisty. M. W.
1944. The life history and growth of the brook lamprey (Lampetra planerij. J. Anim. Ecol., 13 (2): 1 10-122,
figs. 1-6.
1951. Duration of the larval stage in the brook lamprey (Lampetra planeri). Nature, 167: 38-39, 1 fig.
1954. Sc\ TdUo in spawning populdlions of Lampetra planeri. Nature, 173:874-875.
1961. The growth of larval lampreys. J. Anim. Ecol., 30: 357-371, figs. 1-5.
1963. Fecundity and speciation in lampreys. Evolution, 17(1): 17-22.
1969. A comparison of gonadal development in the ammocoetes of the landlocked and anadromous forms of
the sea lamprey Petromyzonmarinus. J. Fish Biol., 1 (2): 153 166, figs. 1-5.
Hardisty, M. W.. and I. C. Potter
In press. The behaviour, ecology and growth of larval lampreys. In The biology of lampreys, ed. by M. W.
Hardisty and I. C. Potter. Academic Press, London.
Hardisty, M. W., I. C. Potter, and R. Sturge
1970. A comparison of the metamorphosing and macrophthalmia stages of the lampreys Lampetra
fluviatilis and L. planeri. J. Zool., London, 162: 383^00, figs. 14.
Heard, William R.
1966. Observations on lampreys in the Naknek River system of southwest Alaska. Copeia, 1966 (2):
332-339, figs. 1-2.
Horn, Edward C, and Joseph R. Bailey
1952. Investigations on the life cycle of the Tennessee brook lamprey. J. Elisha Mitchell Sci. Soc, 68: 144.
Hubbs, CarlL.
1925. The life cycle and growth of lampreys. Pap. Michigan Acad. Sci., Arts and Letters, 4, for 1924:
587-603. figs. 16-22.
1940. Speciation of fishes. Amer. Nat., 74 (752): 198-21 1 (reprinted in Biological Symposia, ed. by Jaques
Cattell, Jaques Cattell Press, Lancaster, Pa., 1941: 7-20).
1941. The relation of hydrological conditions to speciation in fishes. In A Symposium on Hydrobiology,
Univ. Wisconsin Press: 182-195.
1947. Cyclostomata or Marsipobranchii. Encylopaedia Britannica, 6: 921-922B, figs. -12.
1963. Cyclostome. Encyclopaedia Britannica, 6: 941-944, figs. 1-6.
1967. Occurrence of the Pacific lamprey, Ento.sphenus tridentatus. off Baja California and in streams of
southern California; with remarks on its nomenclature. Trans. San Dieeo Soc. Nat. Hist., 14 (21):
301 311.
Hubbs, Carl L., and Karl F. Lagler
1958. Fishes of the Great Lakes region. Cranbrook Inst. Sci. Bull. 26: i-xiii, 1-213, frontisp., figs. 1-251, col.
pis. 1^4, end-paper map.
1964. Fishes of the Great Lakes region/ With a new preface. Univ. Michigan Press, Ann Arbor: i-xv, 1-213,
frontisp., figs. 1-251, col. pis. 1-44, end-paper map.
Hubbs, Carl L., and Robert R. Miller
1948. II. The zoological evidence/Correlation between fish distribution and hydrographic history in the
desert basins of western United States. In The Great Basin, with emphasis on Glacial and Postglacial
times. Bull. Univ. Utah, 38 (20) = (Biol. Ser.) 10(7): 17-166, figs. 10-29, map 1.
Hubbs, Carl L., and Ian C. Potter
In press. Distribution, phylogeny, and taxonomy. In The biology of lampreys, ed. by M. W. Hardisty and I. C.
Potter. Academic Press, London.
Hubbs, Carl L., and Milton B. Trautman
1937. A revision of the lamprey genus Ichthvomvzon. Misc. Publ. No. 35, Mus. Zool. Univ. Michigan:
1-109, figs. 1-5, map 1, pis. 1-2.
Ivanova-Berg, M. M.
1931. Uber die Lebensdauer der Larve von Lampetra planeri aus dem Gebiete des Finnischen Buses. Zool.
Anz., 96 (11/12): 330-334, fig. 1.
Jordan, David Starr, and Barton Warren Evermann
1900. The fishes of North and Middle America. Bull. U.S. Nat. Mus., 47, pt. 4: i-ci, 3137-3313, pis. 1-392.
Jordan, David Starr, and Charles Henry Gilbert
1 899. The fishes of Bering Sea. In Fur seals and fur-seal islands of the North Pacific Ocean, pt. 3: 433^92, 5
figs., pis. 42-85.
162 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Jordan, David Starr, and Carl Leavitt Hubbs
1925. Record of fishes obtained by David Starr Jordan in Japan, 1922. Mem. Carnegie Mus., 10 (2): 93-348,
fig. 1, pis. 5-12.
Knowles, F. G. W.
1941. The duration of larval life in ammocoetes and an attempt to accelerate metamorphosis by injections of
an anterior-pituitary extract. Proc. Zool. Soc. London, 1 11 ( = ser. A) 3: 101-109, figs. 1-3, pi 1 .
Leach, W. James
1940. Occurrence and life history of the northern brook lamprey, Ichthyomyzon fossor. in Indiana. Copeia,
1940(1): 21 34, figs. 1-3, pi. 1.
Lindberg, G. U., and M. I. Legeza
1959. Ryby Yaponshogo morya i sopredel'nykh chastei Okhotskogo i Zheltogo morei. Part 1. Amphioxi
Petromyzones Myxini Elasmobranchii Holocephali. Akad. Nauk SSSR, Moskva-Leningrad: 1-198,
figs. 1- 108. (Translated 1967.)
1967. Fishes of the Sea of Japan and the adjacent areas of the Sea of Okhotsk and the Yellow Sea. Part I.
Amphioxi Petromyzones Myxini Elasmobranchii Holocephali. Translated by Israeli Program for
Scientific Translation, Jerusalem. (Translation of 1959 ed.)
Loman, J. C_^C.
1912. Uber die Naturgeschichte des Bachneunauges Lampetra planeri (Bloch). Zool. Jahrb., Suppl. 15 (1):
243-269, fig. A, pi. 12.
McPhail, J. D., and C. C. Lindsey
1970. Freshwater fishes of northwestern Canada and Alaska. Fish. Res. Bd. Canada Bull. 173: 1-381, illus.
Meek, Alexander
1916. The migrations of fishes. Edward Arnold, London: 1^27, frontisp., figs. 1 128.
Nemoto, Takahisa
1955. White scars on whales / (I) Lamprey marks. Sci. Repts. Whales Res. Inst., no. 10: 69-77, figs. 1-10.
Okada, Yaichiro, and Hyozi Ikeda
1938. Contribution to the study of the freshwater fish fauna of Hokkaido, Japan. Sci. Repts. Tokyo Bunrika
Daigaku (Sect. B), no. 55: 133-162, figs. 1-6.
Okkelberg, Peter
1921. The early history of the germ cells in the brook lamprey Entosphenus appendix (Gage) up to and
including the period of sex differentiation. J. Morph., 35 (1): 1-151, figs. A-D, pis. 1-12.
1922. Notes on the life history of the brook lamprey, Ichthyomyzon unicolor. Occ. Pap. Mus. Zool. Univ.
Michigan, 125: 1-14, figs. 1-4.
Potter, I.e.
1968. Mordacia praecox, n. sp., a nonparasitic lamprey (Petromyzonidae), from New South Wales. Proc.
Linn. Soc. New South Wales, 92 (3): 254-261 , figs. 1 -3, pi. 14.
1970. The life cycles and ecology of Australian lampreys of the genus Mordacia. J. Zool., London, 161 (4):
487-51 1, figs. 1-10, pi. 1. '
Potter, I. C, W. J. R. Lanzing, and R. Strahan
1968. Morphometric and meristic studies on populations of Australian lampreys of the genus Mordacia. J.
Linn. Soc. (Zool.), 47 (3 1 3): 533-546, figs. 1 ^, pi. 1 .
Potter, I. C, and R. Strahan
1968. The taxonomy of the lampreys Geotria and Mordacia and their distribution in Australia. Proc. Linn.
Soc. London, 179 (2): 229-240, pis. 1-2.
Raney, Edward C.
1952. A new lamprey, Ichthyomyzon hubbsi, from the upper Tennessee river system. Copeia, 1952 (2):
95-99, pi. 1.
Reighard, Jacob, and Harold Cummins
1916. Description of a new species of lamprey of the genus Ichthyomyzon. Occ. Pap. Mus. Zool. Univ.
Michigan, 31: 1-12, pis. 1-2.
Rutter, Cloudsley
1 908. The fishes of the Sacramento San Joaquin basin, with a study of their distribution and variation. Bull.
U.S. Bur. Fish., 27: 103-152, figs. 1-4, pi. 6.
Schultz, Leonard P.
1930. The life history o^ Lampetra planeri Bloch, with a statistical analysis of the rate of growth of the larva
from western Washington. Occ. Pap. Mus. Zool. Univ. Michigan, 221: 1 35, figs. 1 10, pis. 12.
1971 HUBBS: A NEW NONPARASITIC LAMPREY 163
Seversmith, Herbert F.
1953. Distribution, morphology, and life history of Lampetra aepyptera. a brook lamprey, in Maryland
Copeia, 1953 (4): 225-232, figs. 1-^.
Snyder, John Otterbein
1908a. Relationships of the fish fauna of the lakes of southeastern Oregon. Bull. U.S. Bur. Fish., 27: 69-102,
figs. 1^, 1 map.
1908b. The fishes of the coastal streams of Oregon and northern California. Bull. U.S. Bur. Fish., 27:
153-189, figs. 1-5, 1 map.
Svetovidova, A. A.
1948. Otikhookeanskoi minoge Entosphenus tridentatus (Gairdner) v sovetskoi chasti Beringova morya (On
the Pacific lamprev Entosphenus tridentatus (Gairdner) in the Soviet part of the Bering Sea). Dokl.
Akad. NaukSSSR,61 (1); 151-152, fig. 1.
Vladykov, Vadim D.
1950. Larvae of eastern American lampreys. I. - Species with two dorsal fins. La Naturaliste Canadien,
77 (3-4): 73-95, figs. 1-13.
1960. Description of young ammocqetes belonging to two species ot lampreys: Petromyzon marinus and
Entosphenus lamotienii. J. Fish. Res. Bd. Canada, 17 (2): 267-288, pis. 1-13.
Vladykov, Vadim D., and W. I. Follett
1958. Redescription of Lampetra ayresii (Giinther) of western North America, a species of lamprey
(Petromyzontidae) distinct from Lampetra fluviatilis (Linnaeus) of Europe. J. Fish. Res. Bd. Canada,
15(1): 47-77, figs. 1-15.
1965. Lampetra richardsoni, a new nonparasitic species of lamprey (Petromyzonidae) from western North
America. J. Fish. Res. Bd. Canada, 22 (1): 139-158, figs. 1-9.
1967. The teeth of lampreys (Petromyzonidae): their terminology and use in a key to the Holarctic genera. J.
Fish. Res. Bd. Canada, 24 (5): 1067-1075, figs. 1^.
Zanandrea, Giuseppe
1951. Rilievi e confronti biometrici e biologice sul Petromyzon ( Lampetra) planeri. Bloch. Nelle acque della
marca trevigiana. Boll. Pesca, Piscicult. e Idriobiol., 27 (n.s. 6) (1): 53-78.
1954a. Corrispondenza tra forme parassite e non parassite nei generi Ichthyomyzon e Lampetra (probiemi di
speciazione). Boll. Zool., 21 (2): 461-466.
1954b. Note sulla ecologia e distribuzione in Italia della Lampedra di ruscello (Lampetra planeri Bloch). Boll.
Pesca, Piscicult. e Idrobiol., 29 (n.s. 8) (2), for 1953: 252-269.
1955. La corrispondenza tra forme parassite e non parassite nei generi Ichthyomyzon e Lampetra nei
confronti del problema della specizione. Atti Accad. Sci. Inst. Bologna (ser. 11)2: 1-16.
1956. Neotenia in Lampetra planeri zanandreai (Vladykov) e Tendocrinologia sperimentale dei Ciclostomi.
Boll. Zool., 23 (2): 413^27, pis. 1-3.
1957a. Neoteny in a lamprey. Nature, 179: 925-926.
1957b. Esame critico e comparativo delle Lampedre cotturate in Italia. Arch. Zool. Ital., 42: 249-307, figs.
1-5, pis. 1-3.
1958a. Allri casi di Lamprede neoteniche e il lore apparato naso-faringeo. Atti 1st. Veneto Sci., Lett, ed Art.,
116: 179-191, pis. 1-10.
1958b. Posizione sistematica e distribuzione geografica di Lampetra zanandreai Vladykov. Mem. Mus. Civ.
Storia Nat., Verona, 6: 207-237, figs. 1-3, pis. 1-3.
1959a. Speciation among lampreys. Nature, 184: 380.
1959b. Lamprede parassite e non parassite nei bacino del Danubio e la nuova entita sistematica Eu-
dontomyzon danfordi vladykovi. Arch. Zool. Ital.. 44: 215-250, pis. 1-2.
1959c. Recenti ricerche sulle forme "appaiate" di Lamprede dell' Italia e del Danubio. Boll. Zool., 26 (2):
545-554.
1961. Studies of European lampreys. Evolution, 15 (4): 523-534, figs. 1-3.
1962a. Rapporti tra continent! e isole nella biogeografia delle Lamprede in Italia. Boll. Zool., 28, for 1961:
529-544.
1962b. Ulteriori rilievi biometrici su Lampetra zanandreai Wndykov . Boll. Zool., 28, for 1961: 703 715.
Scripps Institution of Oceanography, University of California, San Diego; La Jolla,
California 92037.
^^ ^jc.c^ MUS. COMP. ZOOL.
LIBRARY.
JUN 41971
HARVARD
UNiVSRSlTYi
RECENT OSTRACODES FROM CLIPPERTON ISLAND
EASTERN TROPICAL PACIFIC
EDWIN C. ALLISON AND JOHN C. HOLDEN
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 7 14 MAY 1971
-1
RECENT OSTRACODES FROM CLIPPERTON ISLAND
EASTERN TROPICAL PACIFIC
EDWIN C. ALLISONt AND JOHN C. HOLDEN
ABSTRACT. — The Recent ostracode fauna of Clipperton Island is derived from several biogeo-
graphic regions. It includes: (1) new species of Eticythenira, Neocaiidites, Paradoxostoma. and
Semicytherura; (2) Cytherelloidea praecipiia, Occ ultocytliereis angiista, Paracytheridea tschoppi
and species of Triehelina and Bairdia representing a distinct Caribbean aspect; (3) Xestolehcris
gracilis, Triehelina sertata, Sclerochilus sp. nov.. and Bairdia ritugerda clippertonensis subsp.
nov. forming a weak Indopacific link; (4) a restricted west American aspect represented by
Bairdia semuvillosa and Mini Ins convergens: and, (5) a cosmopolitan aspect provided by the
circumtropical species Pseiidocythere caiidata.
Clipperton Lagoon, open to the sea about 130 years ago, now supports a unique freshwater
ostracode fauna consisting of new species of Potainocypris, Cypridopsis and Liinnocythere.
Dominant species in the marine samples are Paracytheridea tschoppi and Mutihis convergens.
Common associates of these are species of Semicytherura, Paradoxostoma and Xestolehcris in
near-shore reef flat areas, and species of Macrocxprina. Neocaiidites, and Cytherelloidea farther
from shore. Members of the Family Loxoconchidae. characteristic of comparable Indopacific
habitats, are conspicuously absent at Clipperton Island.
RESUMEN. — La fauna de Ostracodos recientes de la isla Clipperton procede de varias regiones
biogeograficas. Ahi aparacen los siguientes: 1 ) Especies nuevas de Eiicythenira. Neocaiidites, Para-
doxostoma y Semicytherura: 2 ) Cytherelloidea praecipiia, Occiiltocythereis angiista, Paracythe-
ridea tschoppi y especies de Triehelina y Bairdia que presentan un distintivo aspecto Caribe;
3 ) Xestolehcris gracilis, Triehelina sertata, Sclerochilus sp. nov., y Bairdia ritugerda clipper-
tonensis subsp. nov., que constituyen un debil eslabon Indo-Pacifico; 4) Bairdia semuvillosa y
Miitilus convergens como representantes de las especies restringidas al oeste americano; y 5 ) la
especie tropical Pseiidocythere caiidata como representante cosmopolita.
La comunicacion de la laguna Clipperton con el Pacifico se abrio hace unos 130 anos, y
actualmente contiene una fauna excepcional de Ostracodos dulceacuicolas, como son las especies
nuevas de Potamocypris, Cypridopsis y Limnocythere.
Las especies dominantes en las muestras marinas son: Paracytheridea tschoppi y Mutiliis
convergens. Con estas se encuentran comunmente asociadas, especies de Semicytherura, Para-
doxostoma y Xestolehcris en los arrecifes llanos proximos a la costa, y especies de Macrocyprina,
Neocaiidites y Cytherelloidea en regiones mas alejadas. Es notable observar que los miembros de
la Familia Loxoconchidae, caracteristicos de habitats similares del Pacifico e Indico, estan ausentes
de la isla Clipperton.
INTRODUCTION
Clipperton Island, the easternmost Pacific atoll at latitude 10°I8' N, longitude 109°
13' W (Figure 1), occupies a critical place in the scheme of tropical biogeography. It offers
the only existing terrestial. littoral, or sublittoral habitats along the Clipperton Fracture
, Zone (Menard and Fisher. 1958) or within the great tropical oceanic area known as the
! East Pacific Barrier (Ekman. 1953) that separates Polynesian and west North American
shallow marine environments.
The atoll is oval in outline, about 3 by 4 km, and consists of a thin but unbroken ring
of both loose and lithified coral debris with a single remnant of the igneous basement,
j Clipperton Rock (29 m high), at the atoll's southeastern edge. A deep and completely
tDeceased, 3 January 1971
SAN DIEGO SOC. NAT. HIST., TRANS. 16(7): 165-214. 14 MAY 1971
166
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
rkec/cc CA
iqUA TOT^AL eOUNUIl CmijENT ,
J
ce&n
SOUTH WATO[(IAL CUK^NT
GAL4P4G0S
ISLANDS
Figure I. C'lipperlon Island and adjacent areas. Surface currents lor February adapted from Sverdrup. Johnson
and Fleming (1*^42). Previous published works dealing with podocopid and platycopid ostracodes in the east
Pacific are numbered within a circle in the approximate area of study. These are ( I ) Benson, 1959, (2) Benson and
Kaesler, 1963 (3) Brady, 1 S80, (4) Crouch, 1949, (5) Hartmann, 1953, 1957a, 1957b, 1959a, 1959b, (6) Holden,
1967, (7) Juday. 1907,'(S) LeRoy, 1943, 1945, (9) Rothwell, 1948a, 1948b, (10) Skogsberg, 1928, 1950, (11)
Swam, 1967, (12) Swain and Gilby, 1964, ( 1 3) Swain and Gunther, 1969, and ( 14) Triebel, 1954, 1956, 1957.
landlocked lagoon i,s fresh and generally palatable above 20 m but abruptly saline and
stagnant below that depth (Sachet, 1962c). Early historical accounts of ocean connections
(Sachet, 1963; Belcher, 1843) and in situ marine fossils, with a 370±I00 year radiometric
age (Fergusson and Libby, 1962), indicate that the lagoon was at least periodically marine
until recently.
The geologic age of Clipperton Island is unknown. But the low incidence of endemism
among the marine invertebrates does not support an old age for the faunas.
On the other hand, the strong Caribbean character of the ostracode fauna supports the
hypothesis that the Island has maintained a shallow water biota since the early Pliocene.
Prior to that time a seaway extending through middle America linking the east Pacific with
the Caribbean (Lloyd, 1963) would have allowed the North Atlantic Equatorial Current to
sweep from east to west over the Colombian Basin into the Pacific at the latitude of
Clipperton Island and could readily account for the Caribbean ostracode species now living
there.
The marine invertebrate fauna is an impoverished one in terms of diversity. It is
composed principally o'i central Pacific (Indopacific) and tropical west American (Pan-
amic) species. Many of these species are known to have tloating larval stages of long
duration or to be potentially subject to dispersal by rafting. Indopacific and Panamic
elements are almost equally represented in the inshore faunas, although the ratio of species
1971 ALLISON AND HOLDEN: CLIPPERTONOSTRACODA 167
representative of these provinces varies somewhat from group to group. Mixtures of
Indopacific and Panamic species in the shallow marine faunas of Clipperton Island mark a
blending of these two biogeographic provinces which otherwise are clearly distinct
(Hertlein and Emerson, 1953; Emerson, 1967). No modern Panamic species is known to
have dispersed farther westward than Clipperton Island. A small group of Indopacific
species which have crossed the East Pacific Barrier (Hertlein, 1937; Briggs, 1961;
Emerson, 1967) is almost completely represented in Clipperton Island faunas, thus
suggesting the islands role as a stepping stone. The failure of many other species to effect
westward or eastward dispersals once having reached Clipperton Island is one of the great
problems presented by that island and its faunas. Shifting Pacific North Equatorial
(westward) and Equatorial Counter (eastward) surface currents (Figure 1) as well as
subjacent currents, cross the eastern Pacific at the latitude of Clipperton Island, providing
possibilities for faunal dispersal in both directions (Wyrtki, 1965; U. S. Navy Hydro-
graphic Office, 1947, 1950, 1966).
The biogeographical importance of Clipperton Island, as well as the attraction of a
remote and scarcely known island, inspired brief visits by biologists before 1956. Two
expeditions with more ambitious aims were made possible in October-November 1956 and
August-September 1958 through the participation of the University of California Scripps
Institution of Oceanography in programs of the International Geophysical Year. The
research vessel Spencer F. Baird, commanded by Captain Alan W. Phinney, provided
transportation in both instances. The late Conrad Limbaugh served as scientific party chief
for both expeditions. A single dredge haul from a subsequent S.I.O. cruise, local-
ity B-8558, provided the only additional biological materials to which we have had
access. Samples and field notes on which the present account of Clipperton Island
ostracodes is based are the work of Allison who accompanied both the 1956 and 1958
expeditions. Sediment and algae samples which were the source of the ostracodes dealt
with here, were collected by free and SCUBA diving by Allison and Limbaugh except for
the dredge sample at station B-8558. The most comprehensive descriptions of the history,
geography, geology, and biology of Clipperton Island are to be found in published works of
Marie-Helene Sachet (1960, 1962a, 1962b, 1963), who was one of the participants of the
1958 expedition.
FAUNAL CHARACTERISTICS OF THE OSTRACODA
The marine ostracode fauna of Clipperton Island, like those of the other marine
invertebrates there, is impoverished but shows diverse biogeographic affinities. Nine
species are described as new and are considered here as endemics. These may, in fact,
refiect our poor knowledge of Pacific ostracodes. Areas from which eastern Pacific
podocopid ostracodes have been described are shown in Figure 1 .
The ostracode samples forming the basis of this account represent freshwater lagoon
and various marine reef and off-reef habitats. Species distributions are outlined tentatively
on the basis of six samples collected according to field evaluations of physical environmen-
tal factors and associated larger organisms. Species abundances, living-nonliving and
distributional relationships are shown in Table 1 .
Freshwater species. — Cypridupsis uceanus sp. nov., Limnocythere viaticum sp. nov.,
and Potamocypris insularis sp. nov., are abundant in Clipperton Lagoon. All presumably
were introduced within the last 130 years after the last sea connections were blocked and
marine conditions were replaced by the existing freshwater (Belcher, 1843). Only the
unlikely possibility of prior introduction to, or evolution in, permanent ponds along the rim
of the atoll, between the sea and the formerly marine lagoon, could account for a
168 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
freshwater ostracode history dating earlier than 130 years ago. It is unlikely that such
ponds ever existed on the narrow rimed atoll. The ostracodes probably were introduced by
marine birds which frequent the island during their migratory llights.
Disarticulated valves of several marine species occur in the lagoonal samples and
probably represent former marine conditions there. Bairdia semuvillosa appears to
represent former marine conditions in the lagoon. It does not occur in existing marine
habitats around the island, though it is reported living in a wide diversity of west American
habitats (Benson, 1959; Swain, 1967) and would appear to tolerate a wide range of
conditions.
Water in the lagoon varies in surface salinities from less than 0.1% to greater
than 5.0%, depending on seasonal variations in rainfall (Sachet, 1962b). Below about 20
m the water is saline with abundant sulfides and without evidence of an invertebrate fauna.
Marine species. — Known distributions of the ostracodes which occupy the marine
'linbitats give no clear indication of a dominant biogeographic relationship. Eucytherura
binocula. Mutilus convergens clippertonensis, Paradoxostoma limhaughi. and Semi-
cytherura qiiadraplana apparently represent an indigenous aspect of the Clipperton Island
ostracode faunas.
Five species have Caribbean affinities, Bairdia sp., Triebelina rugosa (not T. bradyi in
the sense of Puri, 1960), Paracytheridea tschoppi, Occultocythereis angusta, and Cytherel-
loidea praecipua. Paracytheridea tschoppi first appears in Miocene rocks of Trinidad, and
is found living in the Caribbean and tropical eastern Pacific (Panamic province). Species of
the genus Occultocythereis are common in early Tertiary deposits of North America and
Europe (Morkhoven, 1963:197) and now occur in the Mediterranean (Muller. 1894), off
the coast of Africa (Brady, 1911), and in the Caribbean. Occultocythereis angusta,
described originally from Madeira Island, northwest Africa (Brady, 1911: "cythere
deformis") also occurs in the Caribbean as far back as Miocene (Bold, 1963). Apart from
its discovery at Clipperton Island, the genus Occultocythereis is unknown elsewhere in the
Pacific. Bairdia sp. appears closely related to the undescribed Caribbean species Bairdia cf.
B. tuherculata of Puri (1960). Triebelina rugosa and Cytherelloidea praecipua occur only
in the modern Caribbean. Neocaudites is likewise a characteristic Caribbean genus
(McKenzie. 1967), though we are aware of one species living off Dakar, Africa (unpub-
lished), and two others (one fossil and one Recent) in the Hawaiian Islands (Holden, 1967).
The Clipperton form, N. pacifica pacifica is considered subspecifically distinct from the
living Hawaiian form, N.p. minima.
Indopacific and Panamic faunal aspects, clearly evident among associated Clipperton
Island marine invertebrates, are weakly represented. Xestoleberis gracilis, Sclerochilus
sp., and Triebelina serata may be Indopacific taxa, as might also Bairdia ritugerda
clippertonensis subsp. nov. The absence of the Loxoconchidae is striking because one or
more species of Loxoconcha and Lo.xoconchella are commonly represented in island
faunas of the Indopacific. Bairdia semuvillosa, probably restricted to the extinct marine
fauna of Clipperton Lagoon, provides the only evidence of a direct Panamic-Clipperton
Island link. Paracytheridea tschoppi occurs in the Panamic Province but probably has its
origin in the Caribbean.
Pseudocythere caudata is possibly a true cosmopolitan species. Other widely dis-
tributed species seem to be restricted to 2 or 3 provinces, as defined by other marine
invertebrate groups.
Two species, Mutilus convergens and Paracytheridea tschoppi, dominate all of the
marine samples, accounting for at least 50 "^"f of the individuals in each.
Living specimens of Paradoxostoma limbaughi and Sclerochilus sp. occur only on
1971 ALl ISON AND HOI DEN: CLIPPHRTON OSTRACODA 169
intcrtidal and slightly subtidal (locality B-4241) areas of the reef fiat. They belong to
genera known to live on marine plants. Xestuleberis gracilis also seems to prefer littoral
conditions but is represented by one living specimen and by several dead valves in deeper
water. Brady (1890) described that species as living in reef and shore pools of the tropical
Pacific.
Living specimens of Seniicytherura quadraplana occur only in sample B-6100, just
beyond the outer edge of the Clipperton reef flat, but associated species in the intermediate
area between reef (Tat (B-4241) and deeper outer slope (B-6120) samples range variously
shoreward and seaward.
Deeper habitats on the outer slope, beyond the outer edge of the ''ten-fathom terrace"
appear to be faunally distinguished by Neocaudites pacifica and Cytherelloidea praecipua
living in association with abundant Bairdia teeteri and with the ubiquitous Para-
cytheridea tschoppi and Mutilus convergent clippertonensis. The deepest Clipperton
sample. B-8538. at a depth of 92 m, lacks living ostracodes, although it contains numerous
valves of species found living in shallower samples.
METHODS
Detailed descriptions are presented for (1) all new species, (2) those that have been
inadequately described elsewhere, and (3) those of the Clipperton population that differ
somewhat from other populations. The term "aff." is used here to indicate a close
relationship between the Clipperton species and the species named. Whether they are
conspecific or not is impossible to determine based on the available information. The use of
"cf." denotes only a comparison to the species named and the two are probably distinct
species.
Most primary (holotypes) and some secondary types (paratypes and hypotypes) are
reposited in the collections of the U. S. National Museum (USNM), Washington, D. C,
and some are in the collections of the San Diego Society of Natural History at the
Museum of Natural History, San Diego, California (SDNH).
Measured specimens are adult instars unless otherwise indicated. Statistical measure-
ments are computed at the 95 per cent confidence limits (± two standard deviations). All
measurements are in microns (/t).
Clipperton Island Ostracode Localities
All samples (fig. 2) are assigned University of California Museum of Paleontology
locality numbers. Most of the material, except ostracode types and minor parts of the
samples, will be stored at the Edwin C. Allison Center for the Study of Pacific Faunas, San
Diego State College.
B-4244 - West side freshwater lagoon; on fossil reefs and in surrounding calcareous
sands; depth approximately 4 m.
B-4247 — West side freshwater lagoon; in sediment on steep slope off lagoon shelf;
depth 8-10 m.
B-4241 Reef flat off north side of island inshore from weakly developed algal ridge;
on algae and in calcareous sediment between widely spaced coral heads
I Porites and PocillopuraK depth intertidal to I- '2 m (in channels).
B-6100 Approximately 100 m off outer edge of reef flat on north side of island; in
sediment from broad sand patches near remains of sunken ships; depth 6-8
m.
B-6I()1 Approximately 100 m off outer edge of reef flat on north side of island,
opposite U.S.N.H.O. marker, about 30 m inshore from outer edge of most
170
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 2. Clippcrlon Island station locations. .Areas within circles indicate approximate station positions.
prominent submarine terrace, northwest of major sandy areas (B-6I00); in
small sediment pockets between and beneath massive living corals (mostly
Pavona, Poriies, and Pocillopora) which cover bottom; depth 10-12 m.
B-6 1 20 - Steep slope off north side of island opposite west end of near breach in atoll
margin (formed by waves during period between 1956 and 1958 expedi-
tions), below slope break at outer edge of principal submarine terrace; in
sediment between blocks of dead coral and sparse cover of living herma-
typic coral; depth 40-45 m.
B-8558 - (CARR 11 8 D)— Dredged living (ahermatypic) and dead coral debris and
calcareous sand from slope olT south-eastern side of Clipperton Island
(10°19'N, 109°12"W); depth 92 m. Scripps Institution o'i Oceanography
expedition CARROUStL (R/V Spencer F. Baird). 1 1 August 1964.^
\91\
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
171
^~~~~~^^^^ AND
SPECIES ^-^™
(B-4244, B-4247)
LAGOON
(B-4241)
REEF
(B-6100)
6-8ni.
(B-6I0I) ■:;•
I0-I2m.
^nV (B-6120)
■ ■:^40-45m.
(0
eo-
30
(B-e558) 40-
92 m. •"'■
so
Bairdia teeter i
2
2
1 -'
2
B. ritugerda clippertonensis
1
1
10
B. simuvlllosa
7
8. sp.
2
Triebelina r-ugosa
3
1
T. sertata
2
6
l^acrocyprina vargata
■^^7«^^
148
Potamocypns insularis
Cypridopsis oceanus
Pontocypris (?) sp.
1
2
Pseudocytttere caudata
1
Eucytt)erura binocula
7
3
Paracyttieridea tschoppi
9
Semicytherura quadaplana
1
1
Mutilus cor)vergens clippertonensis
5
22
■■'■^250^""^
^^193^"^
^■^^207^^"
^■^"321""""™
Limnocythere viaticum
Paradoxosfoma limbaughi
—^13
1
Scieroctiilus sp.
^^^3b^^
5
8
Neocaudites pacifica pacifica
3
Occultocythereis angusta
2
1
Xestoleberis gracilis
16
3
1
X off. X. eulitoralis
3
1
'Cythere" cf "C," caudata
1
Cytherelloidea praecipua
5
3
1
i— ^2Si^—
TOTAL OSTRACODES
ill 323
26. j
343
742
1
53
Table L Ostracode species-localit> check list. Heav\ lines indicate that some or all the individuals contained
soft parts and are therefore inferred to have been livina at that localil\.
Order Podocopida Muller. 1894
Suborder Podocopina Sars, 1866
Superfamily Bairdiacea Sars, 1888
Family Bairdiidae Sars, 1888
Genus Bairdia McCoy, 1844
Bairdia simuvillosa Swain, 1967
Fieure 3
172
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Bairdia simuvillosa Swain, 1967:34. pi, 1. tigs. 2a-l', 8; lc\l tigs. 30c-d, 32. 43a; not Bairdia sinnivillusa: McKenzie
and Swain. 1967:2S3. pi. 30, tig. i. ^
Bairdia sp. all". B. verdcsensis: Benson. 1959:42, pi. 1, tig. 6; pi. 8. tig. 16.
Diagnosis. — Elongate Bairdia. posteriorly tapered in side view, with straight venter;
greatest height in anterior third, greatest width just anterior to midlength; postercdorsum
slightly concave up due to brief hump on caudal process.
Description. — In side view: anteroventer evenly rounded; venter straight or slightly
concave; posteroventer gently rounded to pointed posterior; posterodorsum slightly convex
anterior to brief hump on caudal process; dorsum and anterodorsum almost straight,
divided by a rounded anterocardinal angle. Left valve overlapping right valve along all
margins except at extreme posterior ventral part of pointed caudal process. In dorsal view:
carapace roughly diamond-shaped; greatest width just anterior midlength. Surface of
valves smooth, marginal denticles absent, even in younger individuals.
Duplicature moderately wide; anterior and posterior vestibules large. Fused part of
duplicature transected b\ abundant simple radial pore canals, about 50 anteriorly, fewer
posteriorly. Normal pores abundant, small, relatively few in center of carapace.
Adductor muscle scars tending to fuse, pattern of an elongate scar above two larger
irregular scars which in turn top two smaller oval scars. Dimorphism not observed.
1 igurc 3. Bairdia siiiiuvilLisa Swam. l'Hi7. a-b. h\pot\pc. LJSNM 128066: a. righl \al\c \icw ofadult carapace:
b. dorsal \iew oT adult carapace, c. h\pol\pe. IjSNM 128067; interior otadidt right \al\c.
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
173
866
510
396
850
465
187
787
449
346
800
443
345
801
463
362
Dimensions. — Length Height Width
H\ polype, USNM 128066. Adult carapace, sta. B-4244, 47
Hypotype, USNM 128067. Adult right valve, sta. B-4244, 47
Hypotype, SDN H 04 1 89. Adult carapace, sta. B-4244, 47
Hypotype, SDN H 04 1 90. Adult carapace, sta. B-4244, 47
Hypotype, SDN H 04 1 9 1 . Adult carapace, sta. B-4244, 47
Discussion. Seven specimens were found only at station B-4244-47 in the brackish-
freshwater lagoon and are apparently relics from a past marine condition.
This species is identical to a species found in the Gulf of California and on the Pacific
side of the peninsula at Todos Santos Bay. Another form from Scammon Lagoon
(McKenzie and Swain, 1967) is not considered conspecific because it has a more rounded
dorsum and posterior, and relatively fewer normal pores
Figure 4. Bairdia ritugerda clippertonensis subsp. nov. a-c, holotype. SDNH 04192; a. lateral view of adult left
vaKe; b. interior viev\ of adult left vahe: e. dorsal \ievv of adult left valve.
174 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Bairdia ritugerda clippertonensis subsp. nov.
Figure 4
Diagnosis. — Centrally inflated Bairdia with greatest height at anterocardinal angle in
anterior third; caudal process humped, slightly pointed at posterior-most part; duplicatures
vestibulate; young with posteroventral marginal serrations. Subspecies B. r. clip-
pertonensis is smaller (600-750 /O than B. ritugerda sensu stricto; less accuminate
posteriorly.
Description. — Carapace small for genus, adult length 600-750 /a; surface of valves smooth
to inconspicuously pitted by large but shallow depressions. In side view: left valve much
higher than right valve along dorsum and at inturned area; dorsal margin broadly arched,
flattened in anterior third, sometimes flattened at midlength, slightly concave in posterior
above humped caudal process; greatest carapace height in anterior third of length;
anteroventeral margin smooth in adults, serate in young. In dorsal view: carapace inflated
at midlength or just anterior to midlength; posterior and anterior extremities pointed.
Duplicature wide, heavy; narrow vestibules present; straight or bifurcating radial pore
canals numerous, up to 50 anteriorly, most false; normal pores small, numerous except
around adductor muscle scar area. Eight adductor muscle scars in tight cluster near center
of valve.
Dimensions. — Length Height Width
Holotype. SDNH 04192. Adult left valve, sta. B-6101 595 413 176
ParatypcUSNM 128089. Adult left valve, sta. B-61 20 759 449 196
Paratype,USNM 128090. Adult right valve, sta. B-6 120 755 413 150
ParatypcUSNM 128091. Adult right valve, sta. B-6120 680 370 137
Discussion. — The species is much smaller at Clipperton Island than that at Hawaii, where
it reaches lengths of 1000/-1 and more (Holden 1967: 13). The size difference, together with
ditlerences in shape of the carapace distinguish the two populations as separate subspecies.
Its habitat preference is unknown as no living individuals were found. Ten specimens
were found off the submerged terrace at 40-45 m, whereas only two specimens were found
in shallower water, perhaps indicating a preference for moderately deep water.
Bairdia teeteri sp. nov.
Figures 5, 6
Diagnosis. Bairdia with upturned pointed caudal process; valves heavily pitted; antero
and posterolateral surfaces with horizontal ridges giving carapace a terminally blunt aspect
as seen from above.
Description. In side view: venter straight to slightly concave downward, anteroven-
ter and posteroventer about equal in length and convexity; posterodorsum and anterodor-
sum about equal in length and inclination from horizontal, each slightly concave up;
dorsum straight to slightly rounded. Left valve strongly over-reaching and over-lapping
right valve in dorsal region, with low keel along highest points of dorsum; horizontal
anterolateral ridge developed at midheight; horizontal posterolateral ridge extending along
pointed, upturned caudal process. Possible sexual dimorphism expressed by relatively
lower form (cf?) with height/length ratio =0.54 compared to (9?) 0.60.
In dorsal views: anteromost and posteromosl parts of horizontal lateral marginal
ridges sometimes knob-like giving carapace terminally blunt appearance; centrolateral
region inllated. compressed near margins; width/length ratio about 0.40; surfaces densely
pitted.
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
175
Hinge of " Bairdiopillata"-iype with small toothlets near posterodorsal and antero-
dorsal extremities in right valve and corresponding tiny sockets in left valve. Duplicature
wide, heavy, traversed by sparse simple radial pore canals numbering about 15 anteriorly
and posteriorly, tending to occur in pairs. Vestibules shallow. Adductor muscle scar
pattern with eight equant scars — a center scar with seven surrounding it; three smaller
mandibular scars just anteroventral to adductor group.
Dimensions. —
Holotype. USNM 128093. Adult carapace, sta. B-6120
Paratype, SDNH 04193. Adult left valve, sta. B-6101
Paratype. SDNH 04193. Adult right valve, sta. B-6101
Paratype, SDNH 04194. Adult carapace, sta. B-6120
Paratype, SDNH 04195. Adult left valve, sta. B-6101
Paratype, USNM 128092. Penultimate carapace, sta. B-6120
Paratype, SDNH 04196. Penultimate carapace, sta. B-6120
ength
Height
mdth
800
483
333
750
435
190
749
388
117
792
461
313
695
404
165
659
367
253
612
345
229
500
400
I-
X
300
200
o
VIII • .
•
•
•
•
VI
VII
• • •
V
>-
> •
400
500
600 700
LENGTH u
800
900
Figure 5. Length-height plot of tive growth stages of Bairdia leeleri sp. nov. from stations B-6120 (o). B-6101
(o). and B-4241 (o). The group isolated by dashed lines are thought to be males showing higher length-height
ratios (Kornicker, 1961 ). All measurements taken from entire carapaces or the larger left valves.
Discussion. Bairdia teeteri is closely related to B. attenuata Brady. 1880, from the
Indopacitic and possibly from otT the coast of west Africa (Egger, 1901 ) in general shape,
ornamentation, adductor muscle scar pattern and duplicature. Holden (1967: 14) described
the internal features of B. attenuata to which the present species can be compared. The
important difference between the two species is the presence of horizontal ridges on the
antero and posterolateral surfaces of B. teeteri which are lacking on B. attenuata. The
species might be confused with B. bradyi Bold, 1957, which has similar ornamentation and
somewhat the same shape in side view, but is much wider and diamond shaped in dorsal
176
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
view, not laterally compressed as B. teeteri. Another species belonging to the B. attenuata
group and closely related to the present species is Bairdia sp. c of Bold (1966) from Coco
Solo, Panama. It appears to have a poorly developed horizontal ridge on the posterolateral
surface. According to Bold (personal comm.) the species occurs on the Pacific side of Costa
Rica in rocks of "Young Neogene" age.
Figure 6. Bainlia leelcri sp. nov. a-b. holotypc, USNM 1 28093; a, right valve view of adult carapace; b, dorsal
view of entire carapace, c-d, paratype, SDNH 04193; c. interior view of adult left valve, d, interior view of adult
riizht valve.
At Clipperton Island sizes of individuals ditTer consistently between the stations B-
6120 and B-6101 (see text-fig. 5). The adductor muscle scar pattern and the "Bairdiopil-
laia" -lype dentition seem to be consistent as are other features and size dilTerences
apparently are not ta.xonomically significant.
The species is named for James Wallis Teeter, who in 1966 recognized its uniqueness
during a study of British Honduras ostracodes.
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
177
Bairdia sp. indet.
Figure 7
Description. — In dorsal view: carapace elongate, cylindrical, densely pitted, dark amber
colored; dorsal margin arched, parallel with arched venteral margin; posterodorsal margin
straight, angled ~ 45° from horizontal; anterior margin bluntly rounded beneath sharply
angled anterocardinal angle. In dorsal view: carapace width about equal height along mid
4/5 o{ length; terminally blunt; anterior and posterior valve junctures with small lip-like
ridge.
Anterior duplicature wide with large vestibule; posterior vestibule moderately wide
with outer marginal half fused. Radial pore canals simple, straight, many occupying
marginal denticles, alternating with interspaced false radial pore canals. Normal pores
small, open type, interconnecting internal pit to external. Muscle scars not observed.
Dimensions. — Length Height Width
Specimen, SDNH 04197. Adult right valve, sta. B-6 120 664 289 136
Specimen, USNM 128068. Penultimate? Left valve, sta. B-6120 471 232 99
Figure 7. Bairdia sp. a-c, specimen, SDNH 04197; a, lateral view of adult right valve, b, dorsal view of adult
right valve; c, interior view of adult right valve.
Discussion. — Only two specimens, of which one was an adult, were found at station B-
6120. The good condition of the adult carapace, including original coloration, suggests that
the species is living close by, perhaps in shallower water. The inflated cylindrical carapace
is indicative of a group of bairdiids including Bairdia acanthigera Brady from Cape Verde
at 1020 1 150 fms, B. tuherculata Brady from the Admiralty Islands at 16-25 fms, and B.
hanaumaensis Holden from the Hawaiian Islands at about 5 fms. The general shape alone
of these species would seemingly justify their assignment to a new genus.
178
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
The species is closely related and possibly conspecific with a Caribbean species listed
as Bairdia cf. B. tuherculata by Puri (1960), but ditTers primarily by being more elongate
and having a higher anterior margin as viewed from the side.
Figure 8. Triebelina sertata Tricbcl. 1948. a-c. hypotype, USNM 128069; a. right valve view of adult carapace;
b, dorsal view of adult carapace; c. interit)r view of adult right valve.
Genus Triebelina Bold, 1946
Triebelina sertata Tr\ehe\, 1948
Figure 8
Triebelina indopacifica van den Bold, 1946: 74, Fig. 7 in part .
Triebelina sertata i:nthc\. 1948: 29. pi. 19, figs. la-h. 2a-d; Key. 1953: 158. pi. 1. tig. 5; Puri. 1960: 132. figs. 3. 4;
(luha. 1968: 59, pi. 5. tig. 1.
Triebelina sp. cf. T. cubensis Kingma, 1948: 69. pi. 7. fig. 4.
Diagnosis. — Carapace robust, pitted, widest at two large swellings on each valve along
midlength; strong dorsal ridge curving downward in posterior part of left valve, confined to
dorsum in right valve.
Description. See Triebel (1948) for a complete description of the species.
Dimensions. Length Height Width
Hypotype, USNM 128069. Adult left valve, sta. B-61 20 572 310 170
Hypotype, USNM 128069. Adult right valve, sta. B-61 20 570 283 146
1971 ALLISON AND HOLDEN: CLIPPERTON OSTRACODA 179
Hypotype, SDNH 04199. Penultimate left valve, sta. B-6 120 484 244 142
Hypotype, SDNH 04200. 6th instar, left valve, sta. B-6 120 409 213 125
Discussion — Triebelina sertata and T. indopacifica are closely related (Triebel, 1948).
The most conspicuous differences between the two are the lack of swellings in the
dorsolateral areas of both valves and the reduction of the long ventrolateral ridge into two
broad nodes on each valve along the midlength in T. sertata.
According to Key (1953), Bold (1946) had a specimen of what was described as
Triebelina sertata in his collection of T. indopacifica from Ceram, West Indies. Key also
noted that the single valve of Kingma's (1948) Triebelina cf. T. cubensis, from the lower
Pliocene of Sumatra, is conspecific to T. sertata. One notices that the computed length-
height ratio from Kingma's data agrees well with those of other specimens of T. sertata but
does not agree with his illustrations, which must be distorted.
The species appears to be a shallow water inhabitant. At Clipperton Island it is found
from six to 45 meters (none living). One of us (Holden) collected it along beaches at
Vanuambalavu, Fiji; Puri found it on reefs in the Florida Keys; and, Triebel reported it
from shallow water in the Red Sea. Key's material consisted of one valve each at five
stations in the East Indies ranging in depth from 372 to 3221 meters probably representing
redeposition.
Triebelina rugosa sp. nov.
Figure 9
Triebelina bradyi : Puri. 1960: 1 32. pi. 6. figs. 7 8.
Diagnosis. — Carapace small, length less than 500/i, relatively elongate, L/H ratio about
2.0, valves nearly equal in height; carapace compressed with parallel sides; lateral surfaces
with small prominent tubercles in posterior and anterior lateral areas, two distinct
tubercles one above the other beneath posterior cardinal angle.
Description. — In side view: carapace elongate, L/H ratio about 2.0; dorsal margin
straight, subparallel with slightly concave downward ventral margin; posterodorsal margin
deeply concave upward above serrate caudal process terminating at midheight; anterior
margin denticulate beneath flattened anterodorsal margin. Valves unequally ornamented:
left valve with more strongly developed short tuberculate vertical posterior ridge than right
valve; right valve with two narrow horizontal, sometimes discontinuous, ridges inter-
connecting anterior lateral tubercles with posterior vertical ridge; both valves tuberculate
in anterolateral areas. In dorsal view: carapace compressed, L/W ratio about 2.8; sides
flattened, parallel; caudal region compressed behind vertical posterior ridges of right and
left valves.
Duplicature wide, heavy, shallow vestibules present with straight, thin radial pore
canals. Eight elongate, inclined adductor scars near midheight of valve interior.
Dimensions. — Length Height Width
Holotype, USNM 128094. Adult left valve, sta. B-61 20 478 237 167
Holotype, USNM 128094. Adult right valve, sta. B-6120 477 221 167
Paratype, SDNH 04198. Adult carapace, sta. B-6120 466 224 158
Discussion. — The specimens from Clipperton Island are conspecific with a species
identified incorrectly as Triebelina bradyi by Puri (1960) from the west coast of Florida,
and also known to occur in shallow waters of the British Honduran carbonate shelf
(Teeter, 1966). This Caribbean-Clipperton species is clearly distinct from the Indopacific
T. bradyi which is larger (more than 500/'), higher and has a few broad swellings for
180
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
ornamentation. Triehelina hradyi. in addition, lacks the heavily denticulate, broadly,
evenly rounded posteroventral margin of T. rugosa.
Triehelina rugosa may have a remote ancestor in Triehelina sp/498 of Kollmann
(1963) of Triassic (Rhaetic) age from the European Alps. They are strikingly similar in
outline and both have ubiquitous elongate pits for ornamentation. They differ in tubercle
and swelling arrangement on the lateral surfaces and size of carapace with T. rugosa being
less than half the size of T. sp/498.
t
o. ..■-
Figure 9. Triehelina rii^usu sp. no\ . a-c, holotype, USNM I28U94; a, left valve view of adult carapace; b, dorsal
view; c, internal view.
Puri did not give the depth distribution of the species in the Caribbean; however, we
presume it is a shallow water form. In the Caribbean it is found at Molasses Reef, off
Tavernier, in the Florida Keys (Puri, 1960). At Clipperton Island the species occurs no
shallower than 40 meters at station B-6120 on the rubble slope beneath the principal
submarine terrace of the island. One valve was found at 92 m at station B-8558.
Superfamily Cypridacea Baird, 1849
Family Cyprididae Baird, 1849
Subfamily Macrocypridinae Miiller, 1912
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
181
Genus MacrocyprinaTr'\ehe\, 1960
Macrocyprina vargata sp. nov.
Figures 10, 1 1
Diagnosis. — Carapace strongly arched, angled at highest point at mid-dorsum; posterior
bluntly pointed; light brown color pattern in live specimens distinctive with broad
somewhat inclined bands extending halfway down shell from cardinal angles, large
circular light brown spot surrounding muscle scar area, and at dorsum.
Description. — Carapace heavy, large, length 900-940 /x, light brown color pattern in live
specimens consisting of two somewhat oblique broad bands extending half way down
carapace from cardinal angles, large circular spot at center of shell corresponding with
adductor muscle scar pattern, large spot at mid-dorsum of carapace tending to elongate
and merge with central color spot. In side view: carapace reinform, dorsum highly arched,
somewhat angled at midlength; ventral margin broadly concave downward; anterior
margin evenly rounded, posterior margin bluntly pointed; right valve overlapping left valve
in anterodorsum, posterodorsum, along venter. In dorsal view: carapace bluntly pointed at
posterior and anterior; greatest width at midlength. Both sexes present; sexual dimor-
phism not evident in carapace.
Duplicatures wide, with irregular vestibules intruding into fused zone sometimes as
little pockets from which one or two true or false radial pores extend; radial pore canals
sparse for genus, some paired. Normal pores small, sieve type, about 40-50 in ventral half,
sparse in dorsal half. Hinge of right valve of finely crenulate bar terminating posteriorly
and anteriorly with small crenulate projecting cusps grading into terminal crenulate
grooves about 1 10 /i in length. Ten adductor muscle scars located beneath midheight and
just anterior to midlength; two mandibular scars located anteroventral to adductor group.
500
I-
X
l£ ■
ai
X
200
VIII
«:
1
VI
•
»
VI
.H
%.
V
•
IV
•
III
II
•
•.
•
600
LENGTH
900
r
Figure 10. Length-height plot of seven growth sti'ges of Macrocyprina vargata sp. nov. from stations B-6120
and 8-6101.
182
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Length Height
Dimensions. —
Holotype. USNM 128095. Adult carapace, sta. B-6101 926 410
Paratype, USNM 128096. Adult right valve, sta. B-6101 919 420
Paratype, USNM 128096. Adult left valve, sta. B-6101 925 422
Paratype, SDNH 04201. Penultimate carapace, sta. B-61 20 798 360
Paratype, USNM 128097. Penultimate carapace, sta. B-6120 821 354
Paratype, SDNH 04202. Adult carapace, sta. B-6120 937 430
Paratype, USN M 1 28098. 6th instar carapace, sta. B-6 1 20 550 225
Paratype, SDN H 04203. 6th instar carapace, sta. B-6 1 20 538 229
Discussion. — The type species of the genus, Macrocyprina propinqua Triebel ( 1
more evenly rounded dorsum, is more terminally pointed in dorsal view, and
larger (950-1008 i«.) than the Clipperton species. The color pattern is similar.
Width
307
316
316
254
260
314
179
194
960) has a
is slightly
though of
Figure 11. Mucrocvprina variiatii sp. nov. a-b. holotype, USNM 128095; a. lateral left valve view of adult
carapace; b, dorsal view of adult carapace, c-e. paratype, USNM 12S(»6; c, interior view of adult left valve; d,
dorsal view of adult Icl't valve; d, dorsal view of adult right valve, f, ejaculalory duct, f, third thoracic leg.
1971 ALLISON AND HOLDEN:CLIPPERTON OSTRACODA 183
greater relative size.and does not tend to form bands but rather spots in M. propinqua.
Macrocyprina vargata is also like the southern hemisphere species M. decora (Brady,
1866) in general shape though more terminally blunt, as seen from above, and smaller with
adult lengths ranging from 900-940 /a as opposed to 1005-1010 ju. as cited by Brady (1866,
1880). In addition, the color markings between the two species differ considerably
(compare with Brady, 1880, pi. 6, figs. 8a-b).
The species also resembles Macrocypris succinea MUller, 1894, from the Gulf of
Naples in general shape but, again, is more bluntly pointed in dorsal view. In these two
species the central muscle scar patterns are comparable each with the same number of
scars in approximately the same relative positions. The two small frontal scars shown on
Muller's pi. 13, fig. 25 do not appear on M. vargata, however. The male ejaculatory
apparatus (Zenker's organ) in the two species has the same characteristics, i.e., a central
spiny shaft terminating posteriorly in a smooth bulb-like structure and the same complexly
twisted tubing. In M. vargata, however, the posterior bulb-like structure is much smaller
and the central shaft and tubing are much narrower. Also, the central shaft possesses more
and longer spines. Zenker's organ of A/, propinqua and M. vargata appear very similar.
The specific name denotes the broad vertical color stripes shown in living individuals,
vargatus (L.), "striped."
Subfamily Cypridopsinae Kaufmann, 1900
Genus Potamocypris Brady, 1870
Potamocypris insularis sp. nov.
Figure 12
Diagnosis. — Smooth, highly unequivalved species of Potamocypris with posterior flange
of left valve overreaching right valve. As seen from above, anterior terminating in sharp
point canted slightly to the left.
Description. — In side view: carapace high, length/height ratio about 1.6; length of adult
600-700 /x; outline subtriangular. highest point just anterior to midlength at highly angled
dorsum; ventral margin straight to slightly concave; posterior margin of right valve steeply
truncate; bluntly pointed near venter in left valve; right valve larger than left valve,
overreaching left valve along dorsum where it is considerably higher and along venter and
anterior; left valve overreaching right valve posteriorly as a caudal flange. In dorsal view:
length/width ratio from 2.5 to 2.9; outline irregularly lenticular; greatest width near
midlength. anterior sharply pointed, posterior bluntly pointed.
Calcified duplicature poorly developed, present only in left valve anterior. Radial pore
canals short, simple. Normal pores numerous, small, open type. Hinge adont. Adductor
muscle scar pattern composed of five scars, top scar elongate, second and third an oblong
pair, fourth scar elongate, fifth scar small, circular.
Dimensions. —
USNM, Holotype 128099. Adult carapace, sta. B-4244, 47
USNM, Paratype 128100. Adult right valve, sta. B-4244. 47
SDNH, Paratype 04204. Adult carapace, sta. B-4244, 47
SDNH. Paratype 04205. Adult carapace, sta. B-4244, 47
Discussion. — Potamocypris insularis has only five scars in the adductor pattern, unlike
most species of the genus which have six or seven. There is an apparent reduction occurring
in the ventral part of the pattern.
The closest living Potamocypris to Clipperton Island is P. islagrandensis which occurs
Length
Height
Width
692
4r6
250
612
333
113
701
412
258
677
392
234
184
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
in Lake Nicaragua, Central America. Potamocypris insularis is relatively higher, has a
pointed posterior and has a different adductor muscle scar pattern than P. islagrandensis
(Swain and Gilby, 1964).
100 M
100 M
Figure 12, Potaniocvpris insularis sp. nov. a-b. hololype, USNM 128099; a. lateral left valve view of adult
carapace; b, dorsal view of adult carapace, c-d. paratype. USNM 12S1(K); c, interior view of adult right \alve; d.
interior view of adult left \alve.
Genus Cypridopsis Brady, 1868
Cypridopsis oceanus sp. nov.
Figure 13
Diagnosis. — Carapace small, 580 /a in length; smooth; moderately inflated (length/width
= 1.60); greatest height and width near midlength.
Description. — Carapace thin, transparent, smooth; living specimens covered with sparse
short hairs; width slightly greater than height, length 1.60 times width. In side view: dorsal
margin sloping off straight posteriorly and anteriorly from angled high point at carapace
midlength; posterior and anterior margins similarly shaped, broadly rounded; ventral
margin straight to slightly concave; valves somewhat unequal, left valve slightly over-
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
185
reaching right valve anteriorly, being barely overreached by right valve posteriorly; left
valve strongly overlapping right valve at ventral inturned area. In dorsal view: carapace
ovolenticular, greatest width behind midlength, width slightly greater than height.
Figure 13. Cypridopsis oceanus sp. nov. a-b, hoiot\pe. USNM 128101; a, lateral view of adult right valve; b,
dorsal vie\\ ofadult right valve, c. paratype, USNM 128102; interior view of adult left valve.
Anterior duplicature wide, fused zone narrow with many small simple radial pore
canals; posterior duplicature half as wide as anterior. Adductor pattern of five equant scars
in central field with sixth small scar in posteroventral part of field; antennal scars large,
oblong, beneath and in front of adductor muscle scar pattern. Normal pores minute,
sparse, evenly distributed.
Dimensions. —
Holotype, USNM 128101. Adult carapace, sta. B-4244, 47
Paratvpe, USNM 128102. Adult left valve, sta. B-4244. 47
Paratype, USNM 128102. Adult right valve, sta. B-4244, 47
Paratype, SDNH 04206. Adult carapace, sta. B-4244, 47
Paratype, SDNH 04207. Adult carapace, sta. B-4244, 47
Paratype, SDNH 04208. Adult carapace, sta. B-4244, 47
Paratype. SDNH 04209. Adult carapace, sta. B-4244, 47
ength
Height
Width
579
328
355
544
319
326
545
316
325
562
354
356
587
344
366
548
325
350
563
339
359
186
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Discussion. — This species bears some resemblance to Cypridopsis vidua (O. F. Miiller,
1776) but is much smaller, unpitled, and has a blunter posterior viewed from the side. Also,
there are six adductor scars as in C. vidua but their relative positions differ (compare with
Morkhoven, 1963, p. 48). The size of Cypridopsis oceanus is consistently less than 600
microns compared with 700 microns for C vidua (Wagner, 1957).
The actual salinity range of the lagoon when the species was collected is not known;
however, it was palatable. Allison noted when diving in the lagoon that the salinity
increased with depth. Breakers will occasionally reach the lagoon during storms. Consid-
ering these factors, Cypridopsis oceanus probably has a much higher salinity tolerance
than C. vidua which apparently cannot survive marine salinities greater than 0.8 % (Wagner,
1957:1 10; Reyment, 1964:75).
Figure 14. Pontocypris'^ sp. a-c, specimen, USNM 128070; a, interior view of adult left valve; b, dorsal view of
adult left valve; c, lateral view of adult left valve.
Subfamily Pontocypridinae Muller, 1894
Genus Pontocypris Sars, 1866
Pontocypris? sp.
Figure 14
Description. - Carapace accuminate posteriorly, terminating in a sharply pointed pos-
terior in both dorsal and side views. In side view: greatest height at sharply angled point in
anterior third; anterodorsum and posterodorsum sloping away from the highest point at
angles of about 30° from the horizontal; posterodorsal margin almost straight, terminating
ength
Height
Width
740
347
140
742
342
228
726
317
119
512
227
158
524
212
90
411
170
142
325
134
117
1971 ALLISON AND HOLDEN: CLIPPERTON OSTRACODA 187
in pointed ventral posterium; ventral margin straight except for slight convexity at inturned
area. In dorsal view: carapace compressed, greatest width in anterior quarter, right valve
overlapping left valve posterior to greatest carapace height in anterior third.
Duplicature wide in both posterior and anterior parts of valve. Fused zones narrow,
containing several straight, simple radial pore canals. About six oblong adductor scars
located in region above inturned area.
Dimensions. —
Specimen USNM 128070. Adult left valve, sta. B-6101
Specimen SDNH 04210. Adult carapace, sta. B-6120
Specimen SDNH 0421 1. Adult left valve, sta. B-6120
Specimen USNM 128071. 6th instar carapace, sta. B-6120
Specimen SDNH 04212. 6th instar right valve, sta. B-6120
Specimen SDNH 04213. 5th instar carapace, sta. B-6120
Specimen USNM 128072. 4th instar carapace, sta. B-6120
Discussion. — The carapace, as seen in side view, has the triangular shape of Pontocypris
but the muscle scar pattern suggests the genus Propontocypris. The two genera originally
were established on the basis of soft parts not preserved in the Clipperton collection.
Pontocypris'!' sp. is best compared to P. accuminata Muller, 1894, from the Gulf of
Naples. The Clipperton species has, however, a straighter dorsal margin in the posterior
two thirds, is more posteriorly accuminate and internally it has a less extensive duplicature
and lacks the typical Pontocypris muscle scar pattern.
Superfamily Cytheracea Baird, 1850
Family Bythocytheridae Sars, 1926
G^nxxs Pseudocythere Sdx?,, 1866
Pseudocythere caudata Sars, 1866
Figure 15
Pseudocythere caudata Sdvs. 1866:88: Brady. 1868:453, pi. 34, tigs. 49-52: Brady, 1880:144, pi. l,figs. 6a-d: Muller,
1894:285, pi. 16, figs. 5, 10, 30-36: Tressler, 1941:102, pi. 19, fig. 15; Wagner, 1957:35, pi. 12: Benson, 1964:13,
pi. 1, fig. 8; text-fig. 7.
Pseudocythere I A Maddocks, 1966:62, text fig. 46, no. 2.
Diagnosis. — Because there is little agreement on what the salient characteristics are that
detine this species, a diagnosis is not presented here.
Description. — Side view: dorsal margin almost straight from top of high truncate
caudal process to anterodorsal cardinal angle; anterior margin broadly rounded; ventral
margin concave downward at centrally located inturned area; posteroventral margin
formed by broad compressed marginal flange. Valves ornamented by continuous, discon-
tinuous, occasionally merging, narrow horizontal ridges everywhere except on most of
caudal process and on posteroventral flange which are smooth. In dorsal view: valve evenly
intlated along length excluding laterally compressed caudal process; width of carapace
would measure one-half length in entire specimen.
Duplicatures broad with large vestibules occupying one-half of duplicature width.
Radial pore canals straight, some with enlargements near line of concrescences, sparse,
about 10 anteriorly, relatively abundant in ventral half, about eight posteriorly. Normal
pores not observed. Hinge weakly developed with elongate bar and subjacent groove.
Adductor muscle scar pattern of three horizontally elongate scars in vertical row, bottom
scar possibly two fused scars. Oval frontal scar anterior to topmost adductor scar. Soft
parts not preserved.
188
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Dimensions. — Length Height Width
Hypotype, USNM 128073. Adult right valve, sta. B-6120 285 15^8 69
Discussion. — Benson (1964:14) pointed out the improbability that all the reports of
Pseudocythere caudata are referable to one species. However, no serious attempt has been
made to separate this geographically widespread group into species or even subspecies. The
single specimen found at station B-6120 is identified as P. caudata because it falls within
the range of variation of other known populations and insufficient material does not allow
a more critical analysis of it here.
Further studies may show that more important differences occur between warm water
and cold water forms, irrespective of depth of water, than between forms separated by
great distances of longitude. This relationship is suggested by a close resemblance between
the Clipperton Island specimen and another shallow water reef form from northern
Madagascar (Maddocks, 1966). In side view, specimens from both areas lack the
posteroventral spine, at least in the right valve, and are more quadrate, with almost parallel
ventral and dorsal margins, than the subtriangular, spined forms reported from cold or
deep water areas. Future taxonomists should pay particular attention to the number of
Figure 15. Pseudocythere ccnidata Sars, 1866. a-c, hypotype, USNM 128073: a, lateral \ ieu of adult right valve:
b, dorsal view of adult right valve: c, interior view of adult right valve.
1971 ALLISON AND HOLDEN: CLIPPERTON OSTRACODA 189
adductor muscle scars present. Some authors find five scars in the pattern, others only
four. Possibly there is a reduction in the number of adductor scars in warmer water forms;
indeed, the specimen from Clipperton Island approaches a condition of only three
adductor scars with the bottom two scars almost fused (see Figure 15c). A form illustrated
by Wagner (1957, pi. 12) from the Quaternary of the Pays Basin closely resembles the
Clipperton Island and Madagascar forms in those features discussed above and also has
only four adductor scars, but its ecology is unknown.
Pseudocythere caudata at Clipperton Island is considerably smaller than elsewhere,
being only 285 microns long. The specimen is well developed internally and must be
assumed to be an adult.
Family Cytheruridae G. W. MuUer, 1 894
Genus Eucytherura Muller, 1894
Eucytherura binocula sp. nov.
Figure 16
Diagnosis. — Small Eucytherura, length 258-290/'., very wide in posteroventer; surfaces
entirely reticulate, with swellings, tubercles and spines developed to various degrees; eye
tubercles and internal occular sinuses large, distinct duplicature vestibulate.
Description. — Carapace small, size variable, length 258-290 /»., males somewat smaller
than females. In side view: dorsal margin generally straight, parallel with ventral margin;
anterior margin flattened in dorsal half, strongly denticulate in rounded ventral half with
four to five denticles and spines; caudal process blunt, near dorsum; posterior margin
straight, obliquely angled at 45° beneath caudal process; surface of male valve usually with
three large swellings; an interior subcentral swelling, posterodorsal swelling, and pos-
teroventral swelling representing greatest width of shell, females without midswellings,
more inflated; large smooth eye tubercle located just behind sharply angled anterocardinal
angle in each valve; surfaces with deep reticulae, and variously developed, and variously
spaced spines and tubercles. In dorsal view: carapace lanceolate (cT ) to sublenticular ( 9 ),
greatest width always in posterior half at posteroventral swelling; caudal process com-
pressed and pointed; median sulcus poorly developed.
Posterior and anterior duplicatures of moderate width, each with small deep vesti-
bulae tending to dip into the few, straight radial pore canals. Normal pores numerous,
tending to occur in groups of up to three within the outlines of reticulae, usually
accompanied by tiny conical projections deep within the reticulae, the number of conical
projections approximates that of the pores. Hinge typical for genus: small entire terminal
teeth of right valve separated by finely crenulate groove. Muscle scar pattern and soft parts
not preserved.
Dimensions. —
Holotype, USNM 128103. Adult carapace, sta. B-6120
Paratype, USNM 128104. Adult left valve, sta. B-6120
Paratype, SDNH 04214. Adult right valve, sta. B-6120
Paratype, SDNH 042 1 5. Adult carapace, sta. B-6120
Paratype, SDNH 04216. Adult left valve, sta. B-8558
Paratype. SDNH 04217. Adult carapace, sta. B-6120
Discussion. — Two basic forms are present probably reflecting sexual dimorphism. The
males are compressed dorsally and swollen at the subcentral and posteroventral areas, as
shown in text figure 16a-b. These tend to be arrow-shaped in dorsal view as a result of the
pronounced posteroventral swellings. The presumed females are more abundant and more
ength
Height
Width
258
126
188
287
156
92
277
151
75
266
152
166
191
164
100
285
167
177
190
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
inflated laterally, but are no wider, and tend to be lenticular in dorsal view. Holden
(1964:413) noted a similar kind of dimorphism in Eucytherura spinata from the Upper
Cretaceous of California. The typical type of dimorphism in Eucytherura results in lower
and longer males (Morkhoven, 1963:357).
Ornamentation is variably developed. In the inflated females, an arcuate row of about
five or six tubercles runs from the eye tubercle to the posteroventral swelling via the
subcentral area and then up to the posterodorsum (Figure 16g). In the males the tubercles
are mostly lost at the expense of the various swellings.
One of the most prominent features is the large eye tubercles. The species appears to
be related to Eucytherura gihhera Miiller, 1894, which has a similar type of ornamentation
Figure 16. Eiicvtherura hinucula sp. nov. a-b, holotype, USNM 12S103; a, lateral left valve view of adult
carapace; b, dorsal view of adult carapace, c-d. paralype. USNM 128104; c, interior view of adult left valve; d.
dorsal view of adult left valve, e, parat>pc. SDNH 04214; dorsal view of adult right valve, f, normal pores within
reticulae as seen with transmitted light, g, generalized sketch sht)wing tubercle arrangement on the female
carapace, reticulations not drawn in.
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
191
and large eye tubercles. According to Bold (pers. comm.) the species is similar but not
identical to species living in the Caribbean.
The species is named with reference to its very large eye tubercles.
Genus Paracytheridea Miiller, 1894
Paracytheridea tschoppi Bold, 1946
Figures 17, 18, 19
Paracytheridea tschoppi van den Bold, 1946:85. pi. 16, figs. 6-7; van den Bold. 1957:245, pi. 4, fig. 7: Benson and
Coleman, 1963:33. pi. 6, figs. 7. 9, 10, 20.
Paracytheridea granti Swain, 1967:70 (in part), pi. 4, tigs. 10. 1 la. b, pi. 5, figs. 2a. b, 4a-c, 5, text fig. 47a.
Diagnosis. — Sharply and prominently caudate Paracytheridea with posterodorsal swelling
supporting 3-4 flange-like oblique ridges, horizontal alar ridge continuous to anterior
margin; posterior toothlet complex in hinge of right valve well developed; projecting
anterior toothlet complex poorly developed and not projecting.
Description. — In side view: outline of dorsum and venter parallel due to posteroventer
massive ala: dorsal and ventral margins actually highly and posteriorly accuminate,
terminating in well developed pointed caudal process at posterior midheight; anterior
margin of right valve broadly rounded, obliquely rounded in left valve due to extended
anterocardinal wing. In dorsal view: greatest carapace width in posterior third, height/
length ratio of 0.65 to 0.75. Valves deeply sulcate at midlength in dorsal three-quarters
dividing subcentral tubercle and highly inflated posterodorsal swelling. Ornamentation
principally of flange-like ridges characteristically arranged as discussed further on.
Duplicature wide, nonvestibulate duplicatures transected by sparse radial pore canals;
radial pore canals mostly false, about 12 anteriorly, 3 posteriorly, one of which occupies
conspicuous subcaudal dentical. Normal pores sieve type, sparse, sieve plate usually a
horseshoe shaped structure with about 25 perforations. Hinge lobodont, right valve with
prominent posterior element of five distinct toothlets, anterior element of five poorly
1-
UJ
X
iOO
1
VII
:^r ^"'
III
•
IV
>
V
VI
• •
•
1
n
100
200
300
LENGT
400
500
600
H U
Figure 17. Length-height plot of six growth stages of Paracytheridea tschoppi Bold from stations B-6101 and B-
6120. In all cases measurements were taken on entire carapace or the larger left valves.
192 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
developed toothlets, wavy median groove with about 20-25 notches. Five adductor muscle
scars on posterior side of well developed circular subcentral depression, second and third
scars up may be a divided scar preserving the fundamental pattern of four scars for the
adductor group. Frontal scars located on anterior side of subcentral depression numbering
six in two pairs of three, one group above the other.
Sexual dimorphism not observed.
Dimensions.-^ The following information was determined from a collection of 35 adult
carapaces: L = 520 ±20. 6 p.; H = 261 ± 18.4 /x. Nineteen adult carapaces gave a mean width
of 380 /J. with a range from 348 /x to 405 p..
Dimensions. —
Hypotype, USNM 128074. Adult right valve, sta. B-6101
Hypotype, USNM 128074. Adult left valve, sta. B-6101
Hypotype, SDNH 04218. Adult carapace, sta. B-6101
Hypotype, SDNH 04219. Adult carapace, sta. B-6101
Hypotype, SDNH 04220. Penultimate carapace, sta. B-6101
Hypotype, USNM 128075. Penultimate left valve, sta. B-6101
Hypotype, USNM 128076. 6th instar carapace, sta. B-6101
Hypotype, USNM 128077. 5th instar carapace, sta. B-6101
Hypotype, USNM 128078. 4th instar carapace, sta. B-6100
Discussion. — Paracytheridea tschoppi has not previously been reported from the Pacific
region though it is known to be widespread in the Caribbean and parts of the Gulf of
Mexico (Bold, 1946, 1957; Benson and Coleman, 1963). We believe that minor differences
in shell morphology are not sufficient evidence to separate the closely related populations
of P. tschoppi in the Gulf of California and Clipperton Island from those in the Caribbean
and Gulf of Mexico.
Terminology is introduced in Figure 18 for the ridge arrangement of Paracytheridea.
It is assumed that the positions, if not the degree of development, of ridges ornamenting
the valves of this genus are genetically controlled.
ength
Height
Width
514
241
192
517
267
192
524
251
360
530
275
367
449
209
300
430
204
150
364
167
237
300
142
203
203
102
143
Figure 18. Schematic diagram oi Paracytheridea ornamentation. A, L, P, and V represent the anterior, lateral,
posterior, and ventral ridges, respectively.
The Pliocene to Holocene Paracytheridea granti Le Roy, 1943 of California and Baja
California has been confused with P. tschoppi. Paracytheridea granti lacks the pronounced
posterodorsal swelling and possesses a more prominent posterodorsal cardinal angle than
P. tschoppi. In P. tschoppi a P2or P3 extends into the posterocardinal region. Ridge
ornamentation in P. granti is distinctive with a P2 or P3 running continuously into L|
which joins A 2 and which is the only horizontal ridge reaching the anterior margin. In P.
tschoppi. both At and V, reach the anterior margin. In P. granti V\ ultimately joins V3
1971
ALLISON AND HOLDEN; CLIPPERTON OSTRACODA
193
and merges with A^.
Paracytheridea tschoppi is characterized by a ridge arrangement as follows: P2 is well
developed and bifurcates near the median sulcus and can be traced, or extrapolated, across
the sulcus to L| and Lt respectively. P4 is interrupted medially and is traceable to L3. L]
and L3 merge in the anterior part of the subcentral tubercle and join A-, which continues to
the anterior margin. A strongly developed V| is continuous from the posterior end of the
alae to the anterior margin and is subparallel with L3-A2 in the anterior half of the shell.
V[ and Vt are equally developed.
Figure 19. Paracvtheridea tschoppi Bold, 1946. a, hypotype, SDNH 04219; lateral right valve view of adult
carapace, b, hypotype. SDNH 04218; dorsal view ol" adult carapace, c-f, hypotype, USNM 128074; c, interior
view of adult left valve; d, interior view of adult right valve; e, f, dorsal view of adult right and left valves
respectively, normal pore greatly enlarged as seen with transmitted light.
194
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Genus Semicytherura Wagner, 1957
Semicytherura quadraplana sp. nov.
Figure 20
Diagnosis. — Small Semicytherura with high pointed caudal process and ridge ornamenta-
tion resulting in minutely pitted posteroventral, anteroventral, and central fields when
viewed from the side; alate as seen from above.
Description. — Carapace heavy, small, 260 to 290 /x long. In side view: dorsal margin
nearly straight, parallel with straight ventral margin: venter very wide and flat; anterior
margin obliquely rounded, ventral half with four stubby marginal knobs; posterior margin
truncate beneath high, pointed caudal process; periphery of valves with continuous smooth
ridge, doubled along anterior margin and complex along dorsal margin; smooth lateral
ridge departing at right angle from anterior ridge at midheight, swinging down to venter
along the edge of wide alar process, then swinging irregularly back up to posterocardinal
angle thus creating two nearly equal fields in anteroventer and posteroventer with larger
central field between; compressed caudal area a fourth field; right valve somewhat higher,
overreaching left valve along dorsum. In dorsal view: carapace compressed in dorsal half;
greatest width along ventral midlength on well developed alar process; anterior blunt due
to doubled marginal ridge system; posterior compressed, pointed at caudal process.
Duplicatures broad; posterior duplicature greatly extended inward, almost to middle
of valve; posterior radial pore canals mostly false, some passing through marginal spine at
posteroventer, at least one running full length of caudal process; anterior duplicature wide,
with 15 to 20 irregular, enlarged, sometimes dividing radial pore canals; no vestibules.
Normal pores numerous, tiny, in small clusters of one to 18, each cluster apparently
narrowing to small external pit. Hinge elements of right valve consist of smooth anterior
tooth, fiange-like posterior tooth, and crenulate median groove. Four oblong adductor
muscle scars form vertical row in lower half of valve; elongate single frontal scar anterior
to topmost adductor scar.
Figure 20. Semicytherura quadraplana sp. nov. a, paratype, USNM 128106; internal view of adult right valve,
b, holotype, USNM 128105; external left valve view of adult carapace, c, paratype, SDNH 04221; dorsal view of
left valve, e, enlarged view of normal pore cluster as seen with transmitted light.
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
195
289
132
137
275
133
70
276
129
73
267
129
140
276
136
137
277
133
134
269
129
134
Dimensions. — Length Height Width
Holotype, USNM 128105. Adult carapace, sta. B-6100
Paratype, USNM 128106. Adult right valve, sta. B-6100
Paratype, SDNH 04221. Adult left valve, sta. B-6100
Paratype, SDNH 04222. Adult carapace, sta. B-6100
Paratype, USNM 128107. Adult carapace, sta. B-6100
Paratype, SDNH 04223. Adult carapace, sta. B-6100
Paratype, SDNH 04224. Adult carapace, sta. B-6100
Discussion. — The ridge arrangement of Semicytherura quadraplana is somewhat similar
to that found on 5". quadrata (Hanai, 1957:20) from Japan, though these species differ in
other aspects. The strongly developed alae set this new species apart from any known
Semicytherura. The unique ridge arrangement is a result of the singular lateral ridge
following each ala to the venter from anterior and posterior midheights.
Family Hemicytheridae Puri, 1953
Genus Mutihis Neviani, 1928
Mutilus convergens clippertonensis subsp. nov.
Figure 21, 22
Aurila convergens Swain, 1967:79, pi. 8, fig. 8; Gunther, 1967:97, pi. 1, fig. 8.
Diagnosis. — A species of Mutihis with highly arched dorsum, well developed poste-
rodorsal tubercle, prominent ornamental ridge and furrow from posterodorsum to
anteroventer across dorsolateral-anterolateral areas.
Description. — In side view: margins rounded except at small pointed caudal process near
400
300
I-
I
19
UJ
I
200
100
VII
•
VIII
•
- r
•• ••
VI
V
L •
• • • •
V
1
•
IV
•
•
!•.
100
200
300
400
500
600
LENGTH
f
Figure 21. Length-height plot of six growth stages of Mutilus convergens clippertonensis subsp. nov. from
stations B-4241, B-6101, and B-6100. All measurements taken on complete carapaces or the larger left valves.
Labeling of the instars assumes that the species has eight growth stages.
196
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
posteroventer beneath slightly concave posterium; dorsum gently rounded, continuous with
obliquely rounded anterior margin; ventral margin sinuous, slightly concave downward at
inturned area; right valve somewhat larger than left valve, overreaching left valve along
posterior, dorsum, and part of anterior margins; prominent angled posterodorsal tubercle at
juncture of ornamental ridges. Ornamentation of six horizontally trending ridges with
large reticulations in intermediate furrows; two parallel sinuous ridges extend from
posteroventer to anteroventer; prominent ridge and furrow from posterodorsal tubercle to
anteroventer via dorsolateral-anterolateral areas. In dorsal view: carapace lenticular,
greatest width at midlength; anterior and posterior blunt. Eye tubercles small, on heavy
marginal rim system. Males present but shell dimorphism not apparent.
Duplicature about 50 /x wide, continuous along venter. Radial pore canals abundant,
a
Figure 22. Mutilus convergens clippertonensis subsp. nov. a-b, paratype, USNM 128109; a, externa! right valve
view of adult carapace; b, dorsal view. c. holotype, USNM l2Si08; internal view of adult left valve, d, paratype,
SDNH 04226; dorsal view of adull left valve, e, paratype, SDNH 04225; dorsal view of adult right valve, f, 1st
antenna (incomplete), g, male 2nd antenna with long spineret bristle, h, female, 2nd antenna in part showing
reduced spineret bristle, i, mandible and maxilla, j. 1st thoracic lee.
ength
558
Height
346
Widt
150
550
345
296
537
316
142
542
335
150
456
281
218
380
234
183
302
195
133
242
158
103
1971 ALLISON AND HOLDEN: CLIPPERTON OSTRACODA 197
evenly spaced, straight, unpaired, each with small midswelling; vestibules shallow. Normal
pores large, sparse, sieve type. Hinge amphidont; anterior tooth of right valve stepped;
posterior tooth of right valve bifed in ventral part; median bar and tooth of left valve
smooth. Four adductor muscle scars, second scar from top distinctly divided into two equal
smaller scars; oblique row of three mandibular scars located anterior to top two adductor
scars; conspicuous oblong scar directly above adductor group in dorsal half of valve.
Dimensions. — The dimensions of the adults, based on the analysis of 50 carapaces and
larger left valves, are: L = 544.0 ± 19.2^1, H = 331.5 ± 18.8 ju; W = 264.0 ± 18.8 ^i. The
arithmetic mean widths of the instars VII, VI, V, and IV are 215 i^l, 169 /i, 136ju., and 103 ju,
respectively. Lengths and heights of the young are shown in figure 2 1 .
Holotype, USNM 128108. Adult left valve, sta. B-6120
Paratype, USNM 128109. Adult carapace, sta. B-6120
Paratype, SDNH 04225. Adult right valve, sta. B-6120
Paratype, SDNH 04226. Adult left valve, sta. B-6120
Paratype, USNM 1281 10. 7th instar carapace, sta. B-6101
Paratype, USNM 128111. 6th instar carapace, sta. B-6101
Paratype, SDNH 04227. 5th instar carapace, sta. B-6100
Paratype, SDNH 04228. 4th instar carapace, sta. B-6100
Discussion. — Almost 1000 specimens were counted and examined. The species is by far
the most abundant ostracode living in the shallow marine environments around Clipperton
Island from shoreline to depths of 40-45 m. It is still relatively abundant at 92 m, however,
this may be an artifact of redeposition as no living individuals were found at that depth.
The few specimens from Sta. B-4244-47 were all dead and we assume that they may have
lived there at a time prior to the enclosing and freshening of the inner lagoon.
The Clipperton Island specimens are assigned to a species occurring in the Gulf of
California (Swain, 1967) and the Gulf of Panama (Giinther, 1967). The most distinctive,
and apparently unique, feature of the valve is an ornamental furrow running from the
posterodorsum to the anteroventer; this is highly developed in the Clipperton Island
subspecies. Mutilus convergens is closely related to M. palosensis LeRoy (1943) from
California and the west coast of Baja California (Benson, 1959) and to the fossil Hawaiian
Island M. oahuensis Holden (1967). This group is characterized by a well developed
posterodorsal ridge juncture, a sinuous ventral margin paralleled by one or two ventrola-
teral ridges, and a tendency for the lateral ridges to converge anteroventrally. All of these
ornamental and morphological conditions are more prominently developed in the Clip-
perton species than in any other.
Family Limnocytheridae Klie, 1938
Genus Limnocythere Brady, 1868
Limnocythere viaticum sp. nov.
Figure 23
Diagnosis. — Carapace fragile, small, less than 400/1 long; lightly reticulate and punctate;
reniform-shaped as seen in side view; large dorsolateral swelling in front of median sulcus,
smaller swelling below, at center of valve; anterior wedge-shaped and sharply pointed as
seen from above.
Description. — In side view: shell reniform, ventral margin broadly concave, dorsal
margin straight to slightly arched; anterior and posterior margins broadly rounded; surface
of valves lightly reticulate in posterolateral and ventrolateral areas, lightly pitted in
198
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
anterior and on swellings; large swelling above smaller one between two dorsolateral sulci;
anteromost sulcus irregular, poorly developed; posteromost sulcus well developed, vertical;
third dorsolateral swelling behind posteromost sulcus poorly developed; small fourth
swelling on posterior part of ventrolateral inflation; left valve slightly larger, and
overreaching right valve anteriorly and posteriorly. In dorsal view: greatest width in
posterior half at ventrolateral swelling; posterior half of carapace inflated; anterior half
wedged shaped, pointed.
Duplicature narrow, traversed by sparse, evenly spaced radial pore canals, about 15
posteriorly and anteriorly. Four oblong adductor scars in vertical row in ventral half of
valve; single mandibular scar ventral and anterior to adductor group; single frontal scar
anterior and dorsal to adductor group. Hinge weak, left valve with terminal depressions
(sockets) near cardinal angles.
Dimensions. —
Holotype, USN M 128112. Adult carapace, sta. B-4244, 47
Paratype, USNM 1281 13. Adult right valve, sta. B-4244, 47
Paratype, SDNH 04229. Adult carapace, sta. B-4244, 47
Paratype, SDNH 04230. Adult carapace, sta. B-4244, 47
Paratype, SDNH 0423 1 . Adult carapace, sta. B-4244, 47
Paratype, SDNH 04232. Adult right valve, sta. B-4244, 47
Discussion. — Limnocythere viaticum is one of three freshwater species found in Clip-
perton lagoon. The taxon cannot be identified with any known species, though the
ength
Height
Width
ill
213
184
366
203
75
358
203
166
363
212
182
375
216
179
363
216
175
Figure 23. Limnocvlhere viaticum sp. nov. a-b, holotype, USNM 1281 12; a, lateral right valve view of adult
carapace: b, dorsal view of adult carapace, c, paratype, USNM 1281 13; interior view of adult right valve.
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
199
freshwater ostracode faunas of Central America, where one might expect these to have
originated, are very poorly known.
As discussed elsewhere, the freshwater lagoon is a relatively recent phenomenon. The
specific name alludes to the species, or its ancestors, trip to the island: viaticum (L.)
"voyager."
Family Paradoxostomatidae Brady and Norman, 1889
Genus Paradoxostoma Fischer, 1855
Paradoxostoma limbaughi sp. nov.
Figure 24
Diagnosis. — Elongate Paradoxostoma posteriorly terminating at midheight in blunt point;
greatest carapace height in posterior half; dorsal view of carapace lenticular and symmetrical
except for bluntly pointed anterior.
Description. — Shell fragile, transparent; relatively small for genus, length about 340 ix. In
side view: carapace elongate, length IVi times height; highest point of carapace just
posterior to midlength at broadly arched dorsum; posterodorsal margin flattened; pos-
terior margin bluntly pointed at midheight; ventral margin broadly concave downward at
inturned area in anterior half, broadly rounded in posterior 2/3 of valve. In dorsal view:
t
a
'S^-^'i^^i^^j^^S;^}^!^^'^
Figure 24. Paradoxostoma limbaughi sp. nov. a-b, holotype, USNM 1281 14; a, lateral right valve view of adult
carapace: b, dorsal view of adult carapace, c. paratype, USNM 128! 15: interior view of adult right valve.
ength
Height
Width
341
\11
90
328
129
50
326
134
92
334
130
87
338
135
92
340
137
93
200 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
outline symmetrically lenticular except for bluntly pointed anterior; greatest width at
midlength. Internal features not discernible.
Dimensions. —
Holotype, USNM 1281 14. Adult carapace, sta. B-4241
Paratype, USNM 1281 15. Adult right valve, sta. B-4241
Paratype, USNM 1281 16. Adult carapace, sta. B-4241
Paratype, SDNH 04233. Adult carapace, sta. B-4241
Paratype, SDNH 04234. Adult carapace, sta. B-4241
Paratype, SDNH 04235. Adult carapace, sta. B-4241
Discussion. — The species apparently belongs in the genus Paradoxostoma based on
general morphology; however, it is possible that it could be placed in Xiphicilus which is
usually more pointed at both ends, or Cytherois which is less bluntly pointed.
The species is similar to Paradoxostoma artum Bold, 1966, from the Caribbean and
Xiphicilus sp. cf. X. arenatus Brady from New Caledonia in the sense of Apostolescu,
1967.
The species is named for the late Conrad Limbaugh who helped collect the Clipperton
Island samples.
Genus Sclerochilus Sars, 1866
Sclerochihis sp.
Figure 25
Sclerochilus contnrtus: Muller, 1894: 282, pi. 16. fig. 2.
Sclerochilus sp. B. Holden, 1967: 39, text figs. 30a-c.
Description. — In side view: shell reniform, with broadly and evenly arched dorsum:
ventral margin sinuous, greatly rounded in posterior two-thirds, concave downward in
anterior half; carapace relatively high, length/height ratio = 2.0, posterior bluntly pointed
at midheight or broadly rounded. In dorsal view: carapace lenticular, compressed, length/
width ratio = 2.7; greatest width at midlength, posterior and anterior pointed. Dimorphism
not observed.
Duplicature wide; vestibules large; fused zone narrow, with continuous width of about
15 /.I. Radial pore canals simple, numbering 20 to 30 throughout duplicature. Normal pores
open, small, sparse. Five adductor muscle scars in oblong oblique pattern at midheight of
valve just anterior to midlength.
Dimensions. —
Specimen, USNM 128079. Adult right valve, sta. B-6I20
Specimen, USN M 1 28080. Adult left valve, sta. B-6 1 20
Specimen, SDNH 04236. Adult left valve, sta. B-6120
Specimen, SDNH 04237. Adult right valve, sta. B-6120
Specimen, SDNH 04238. Penultimate carapace, sta. B-6120
Discussion. — These specimens are identical to Sclerochilus sp. B (Holden, 1967) from late
Cenozoic drowned terraces in the Hawaiian Islands, and to a form from the Mediterranean
identified by Muller (1894) as S. contortus (Norman). Muller's illustrations (pi. 16, figs.
1-2) of this form show distinct sexual dimorphism, the males being the lower and relatively
more elongate of the two. Sclerochilus sp. is similar to the female, illustrated by Muller,
but not to the male. These specimens, including Muller's are considered distinct from 5".
contortus (a North Atlantic species) based on difTerences in the morphology of the shell.
Whether only females have been found at Clipperton Island or whether the population
ength
Height
Width
421
210
71
408
208
73
383
190
75
398
193
65
350
176
135
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
201
there shows no sexual dimorphism is unknown. Unfortunately the soft parts were not
preserved.
■^'t^.x
:-x>s
^:^
^:^
'■^k
H
Figure 25. Sclerochilus sp. a-c, specimen, USNM 128079; a, lateral view of adult right valve; b, dorsal view of
adult right valve; c, interior view of adult riaht valve.
Family Trachyleberididae Sylvester-Bradley, 1948
Genus Neocaudites Purl, 1960
Neocaudites pacifica paciHca sp. nov.
Figure 26
Diagnosis. — Moderate size Neocaudites. length to 559 /-i, ornamented with larged shallow
reticulations; distinctive, isolated, denticulate, submarginal ridge, paralleling anterior
margin; valves asymmetric with dorsal and lateral ridge juncture at posterodorsum more
posteriorly extended in right than left valve. Frontal scar v-shaped, three adductor scars.
Description. — In side view: carapace subquadrate. dorsal margin irregular to straight,
subparallel with gently concave ventral margin: anterior margin broadly rounded, finely
and evenly denticulate in ventral half; posterior subtruncate. with low, bluntly pointed,
caudal process; left valve overlapping right valve at postero- and anterocardinal angles.
Ornamentation of large shallow reticulations; marginal rim continuous from anterocar-
dinal angle around anterior, along venter, around posterior; lateral field with smooth,
straight centrolateral ridge extending from posterodorsal area to low, inconspicuous
subcentral tubercle; prominent, narrow submarginal ridge in anterolateral area, paral-
202 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
leling anterior margin; broad, shallow vertical sulcus anterior to midlength; eye tubercles
small, prominent. In dorsal view: carapace compressed, width/length ratio =0.30,
carapace of equal width from subcentral region to posterocardinal region; caudal and
anterior parts compressed; valves asymmetric: right valve with more posteriorly extended
ridge juncture.
Duplicature moderately broad, shallow vestibule irregularly shaped. Radial pores
sometimes branched, commonly with midswellings, about 30 in anterior, 25-30 in
posterior. Normal pores small, sieve type. Hinge holamphidont; left valve with entire,
projecting, stepped anterior tooth; entire reniform posterior tooth. Smooth median bar of
left valve with low, smooth anterior tooth. Three oblong adductor muscle scars on
posterior side of subcentral depression; bottom-most scar apparently a fused pair. Large
V-shaped frontal scar anterior to top-most adductor scar on side of subcentral depression.
Single circular mandibular scar beneath frontal scar.
Dimensions. — Length Height Width
Holotype, USNM 128117. Adult right valve, sta. B-6120 524 260 75
Holotype, USNM 128117. Adult left valve, sta. B-6120 524 270 89
Paratype, SDNH 04239. Adult right valve, sta. B-6120 523 253 81
Paratype, SDNH 04239. Adult left valve, sta. B-6120 533 266 92
Paratype, SDNH 04240. Adult carapace, sta. B-8558 600 318 184
Paratype, USNM 128118. Adult carapace, sta. B-8558 508 312 191
Paratype, USNM 128119. Adult carapace, sta. B-8558 559 302 175
Paratype, SDNH 04241. 6th instar, sta. B-6120 414 224 141
Paratype, SDNH 04242. 5th instar, sta. B-6120 350 183 146
Neocaudites paciflca minima subsp. nov.
Figure 26
Diagnosis. — Small, length about 450 i-i, ornamented with various sized reticulations and
pits; small, isolated, denticulate, submarginal ridge paralleling anterior margin; valves
asymmetric with dorsal and lateral ridge juncture at posterodorsum more posteriorly
extended in right valve; frontal scar s-shaped, four adductor scars.
Description. — Except for the differences stated in the diagnosis above, all other morpholo-
gical details of N. pacifica pacifica are the same as those of this subspecies.
Dimensions. — Length Height Width
Paratype, USNM 128120. Adult carapace, Hanauma Bay,
Hawaiian Islands 424
Paratype, SDNH 04243. Adult left valve, Hanauma Bay 458
Paratype, SDNH 04243. Adult right valve, Hanauma Bay 458
Discussion. — Neocaudites pacifica minima from Hanauma Bay, Oahu, Hawaii, is
believed to be subspecifically related to A', pacifica pacifica from Clipperton Island and is
diagnosed here for comparative purposes. The most apparent ditTerence between the two is
that of size. A', pacifica minima being much smaller (length 450 jx) than that of N . pacifica
pacifica (length = 525 /x). The Clipperton Island form occurs in deeper waters than the
Hawaiian Island form (10 m). At Clipperton, it was collected alive at locality B-6120
(40^5 meters) and dead at locality B-8558 (92 meters).
The genus Neocaudites has been characterized as a Caribbean taxon (McKenzie,
1967: 232). Previously, only one species had been reported from the Pacific basin, N. terryi
from off the Hawaiian Islands on submarine terraces. Although N. terryi is generally
217
144
216
100
209
95
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
203
similar to N. pacifica. its surface ornamentation is smooth rather than reticulate or pitted.
a
Figure 26. S'eocaudites pacifica sp. nov. a-c, holotype, USNM 128 117; a, external right valve view of adult
male carapace; b, dorsal view; c, internal view of right valve, d-e, paratype, SDNH 04239; d, dorsal view of adult
male left valve; e. dorsal view of right valve, \eocaudites pacifica minima subsp. nov. f, holotype, USNM
128120; external right valve view of adult female carapace; g-i. paratype, SDNH 04243; g, internal view of adult
male left valve; h. dorsal view of left valve; i, dorsal view of right valve.
Genus Occultocythereis Howe, 1951
Occultocythereis angusta Bold, 1963
Figure 27
Cythereis deformis Brady, 19 II: 397, pi. 20, figs. 7-8; not Cythereis deformis Baird, 1850: 256, pi. 18. figs. 4-6.
Occultocythereis angusta Bold, 1963: 391, pi. 9, tigs, la-c, pi. 12. fig. 6 new name for Cythereis deformis Brady.
Diagnosis. — Occultocythereis with posterodorsal tubercle and posteroventral marginal
rim heavy: dorsal rim weakly developed; lateral surface ornamentation very weakly
developed; dorsal margin concave as seen from the side.
204
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Description. — In side view: carapace small, length 450 ^i; length/height ratio =2.0; highest
point at anterocardinal angle in anterior third at midlength; dorsal margin straight
between elevated cardinal angles; ventral margin straight or slightly irregular; anterior
margin evenly and broadly rounded, with several well-developed denticles; larger left valve
over-reaching right valve along venter, posterior, and at anterocardinal hinge angle; left
valve asymmetric with elongate flange extending beneath valve along posteroventer
margin; broad, flattened anterior marginal rim continuous from poorly developed eye
tubercle to ventral inturned area; posteroventral area with complex massive tubercles;
posterocardinal angle occupied by large dimpled tubercle; valves conspicuously sulcate at
midlength; surfaces generally smooth between narrow, inconspicuous, serpentine ridges in
lateral areas. In dorsal view: greatest width in posterior third at ridge juncture terminating
in small, posteriorly pointing, lateral spine, anterior bluntly pointed due to thick antero-
marginal rim; posterior compressed behind posterocardinal tubercles.
Figure 27. Occultocythereis angusla Bold, 1963. a-d, hypotype. USNM 128081; a. lateral right valve view of
carapace; b, dorsal view of adult left valve; c, dorsal view of adult right valve; d, adductor muscle scar pattern of
left valve.
Duplicature of moderate width; vestibulae well developed, line of concrescence
irregular and forming pockets into fused zone from which emanate straight, simple,
abundant radial pore canals, about 30 in anterior; normal pores large, sparse, sieve type.
Muscle scar pattern as shown in Figure 27d.
Dimensions. — Length Height Width
Hypotype, USNM 128081. Adult carapace, sta. B-6120 461 229 183
Discussion. — Occultocythereis angusta is distinctive by the combination of features noted
in the diagnosis. The dorsal ridce and lateral surface ornamentation is subdued, like that of
1971 ALLISON AND HOLDENiCLIPPERTONOSTRACODA 205
O. lineata (Miiller, 1894) from the Mediterranean, in contrast to many of the early Tertiary
species (Hinte, 1964; Triebel, 1961; Howe and Law, 1936; etc.). These two Recent species
differ, O. angusta being smaller, relatively longer, and having a concave upward dorsal
margin instead of a slightly convex one as shown by Muller ( 1 894, pi. 29, fig. 2 1 ).
The Clipperton Island specimens more closely resemble the Caribbean form of O.
angusta illustrated by Bold (1963b, pi. 9, fig. 1) and Teeter (1966, pi. 6, figs. 20-21) than
the recent form from Madeira (Brady, 191 1, pi. 20, figs. 7 8); however, the dissimilarities
are slight and they appear to be conspecific.
Genus Xestolebris Sars, 1 866
Xestoleberis gracilis Brady, 1890
Figure 28
Xestoleberis gracilis Brady, 1890: 508. pi. 3, figs. 9-10.
Diagnosis. — A dorsoventrally compressed species o{ Xestoleberis with a broadly rounded
dorsal margin and straight flat venter.
Description. — In side view: males similar in profile to females; carapace low, dorsoven-
trally compressed, length/height ratio =2.5; ventral margin straight, flat; dorsal margin
evenly and broadly rounded; anterior margin low, sharply rounded but not pointed; surface
of valves smooth. In dorsal view: males lenticular, greatest width near midlength; females
posteriorly inflated, greatest width in posterior quarter.
Posterior duplicature narrow; anterior duplicature of moderate width with narrow
fused zone containing few (10 to 12) simple, straight, radial pore canals concentrated in
ventral part. Normal pores large, especially abundant in anteroventer. Hinge typical for
genus, terminal elements of right valve projecting crenulate plates, about 30 /x in length,
separated by a smooth arcuate groove. Four large, oblong adductor scars in oblique row in
anterior half at shell midheight; two frontal scars, one an arcuate bar, the other a spot
anterodorsal to it, directly in front of top two adductor scars. Wide, arcuate, highly
inclined, xestoleberid scar directly above adductor group near dorsum.
Dimensions. —
Hypotype, USNM 128082. Adult carapace, sta. B-6101
Hypotype, USNM 128083. Adult carapace, sta. B-6120
Hypotype, SDNH 04245. Adult left valve, sta. B-4241
Hypotype, SDNH 04246. Adult right valve, sta. B-4241
Hypotype, SDNH 04246. Adult left valve, sta. B-4241
Hypotype, SDNH 04247. Penultimate carapace, sta. B-4241
Hypotype, USNM 128084. Penultimate carapace, sta. B-4241
Discussion. — The species was originally described from Samoa living on reefs and
intertidal pools (Brady, 1890). At Clipperton it is most common on the reef flat but one
living specimen was found at 10 12 meters on the submerged terrace.
The species is somewhat similar to Xestoleberis humilis Klie, 1940, living in the
"algalzone" along the west coast of Africa.
Xestoleberis sp. aff. X. eulitoralis Hartmann, 1959
Figure 29
Xestoleberis eulitoralis Hdrim-dnn. 1959b: 224, pi. 42, figs. 134-136: pi. 43, figs. 137, 138, 140. 141.
Xestoleberis cf. .X. eulitoralis: McKenzie and Swain. 1967: 303. te.xt fig. 34.
Description. — Carapace moderately compressed, surface of valves smooth; sexual di-
ength
Height
Width
316
132
187
307
126
162
308
129
123
309
138
84
324
151
93
249
125
146
260
126
151
206
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
ioo/<
Figure 28. Xestoleheris i^racilis Brady, 1890. a-b, hypotype, USNM 128082; a, left valve view of adult female
carapace; b, dorsal view of adult female carapace, c-e, hypotype, USNM 128083; c, left valve view of adult male
carapace; d, dorsal view of adult male carapace; e, anterior view of adult male carapace, f-h, hypotype, SDNH
04245; f, interior view of adult female right valve; g, dorsal view of adult female left valve; h, dorsal view of adult
female right valve.
morphism not observed. In side view: valves suboblong, broadly rounded in outline, dorsal
margin sloping slightly anteriorly; ventral margin straight; anterior and posterior margins
bluntly rounded; carapace moderately compressed laterally, greatest inflation in ventral
third; surfaces smooth. In dorsal view; carapace oblong, anterior and posterior bluntly
rounded; greatest width at midlength.
Posterior duplicature narrow, entirely fused; anterior duplicature of moderate width,
vestibulatc. Radial pore canals simple, straight, equally spaced, numbering 20 in anterior,
about 20 in posterior. Hinge typical for genus: smooth median bar of left valve almost
straight as seen from above. Adductor muscle scar pattern a small vertical row of four
elongate scars; single frontal scar directly anterior to topmost adductor scar.
1971
ALLISON AND HOLDLN; CLIPPERTON OSTRACODA
207
Dimensions. — Length Height Width
Specimen, USNM 128085. Adult left valve, sta. B-6 120 302 164 86
Specimen, SDNH 04244. Adult carapace, sta. B-6120 305 169 1 14
Specimen, USNM 128086. Adult left valve, sta. B-6120 300 164 82
Discussion. — The Clipperton Island specimens resemble Xestoleberis sp. cf. X. eulitoralis
from Scammons Lagoon, Baja California, Mexico, more than they do the species from El
Salvador which has no vestibule and has relatively complex radial pore canals. More and
better preserved material would probably show this species to be conspecific to at least
those from Scammons Lagoon.
At El Salvador Xestoleberis eulitoralis was found in the intertidal zone of Mejanguera
Island, Gulf of Fonseca among rocks with corals, encrusting algae, barnacles, and oysters.
McKenzie and Swain report their species occurring throughout Scammons Lagoon from 4
to 75 feet. At Clipperton Island three disarticulated valves were found at B-6120 (40^5
meters) and one at B-8558 (92 meters).
,i:Usfeu%yi^!i-kJ»^5-=-^
lOO
^
Figure 29. Xestoleberis sp. alT. X. eulitoralis Hartmann, 1959. a-d, specimen, USNM 128085; a, dorsal view of
adult left valve; b, anterior view of adult left valve; c, lateral view of adult left valve; d, interior view of adult left
valve.
Genus Uncertain
"Cy there" cf. "C." caudata Brady, 1890
Figure 30
Description. — In side view: carapace elongate, length/height ratio = 2.5; dorsal margin
parallel to ventral margin throughout most of length; anterior margin broadly rounded;
posterior with large compressed, bluntly pointed caudal process most of which lies beneath
midheight. Shell ornamented with about 10 continuous and discontinuous glassy horizon-
208
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
tal ridges tending to parallel anterior margin. Caudal process and adjoining compressed
parts of carapace smooth. In dorsal view: anterior sharply pointed; carapace midhalf
parallel sided; posterior convex to highly compressed; caudal process of extreme posterior.
Internal features not observed.
Figure 30. "Cythere" sp. cf. "C" caudata Brady, 1890. a-b, specimen, USNM 128087; a, right valve view of
adult(?) carapace; b, dorsal view.
Dimensions. — Length Height Width
Specimen, USNM 128087. Entire specimen, sta. 8558 540 230 180
Discussion. — Only one specimen was collected by the carrousel dredge at 92 m. The
species closely resembles Cythere caudata Brady, 1890 from Sava, Sava Bay, Fiji, and
"Cythere" caudata from Manila (Keij, 1954) and Hawaii (Holden, 1967). The single entire
carapace from Clipperton Island is larger than those mentioned above. Brady's and Keij's
species are 460 /x and 450 /x respectively. The ones from Hawaii come from two
populations, one fossil with an individual 410 ;u. in length and one from Hanauma Bay with
very small adult individuals only 350 ii long. The Clipperton Island form is distinctive with
well developed horizontal ridge ornamentation, the elongate reticulae characteristic of the
other related forms being defined between ridges. At the present time it is not possible to
determine the specific relationships between the Clipperton forms and those described by
other authors. All of the above species belong to an undescribed genus.
Suborder Platycopina Sars, 1866
Family Cytherelloidea Sars, 1866
Genus Cytherelloidea Alexander, 1929
1971
ALLISON AND HOLDEN: CLIPPERTON OSTRACODA
209
Cytherelloidea praecipua Bold, 1963
Figure 31
Cytherelloidea praecipua van den Bold, 1963: 75, pi. 1, tigs. 1-7.
Diagnosis. — Carapace reticulate, becoming smooth centrally; valves with poorly devel-
oped horizontal ridges; left valve with strong dorsal tooth fitting into large socket of right
valve; large dorsal flange of right valve overlapping left valve at midlength.
Description. — In side view: carapace subquadrate; dorsal margin slightly rounded,
somewhat irregular centrally at articulation; posterior margin truncate, anterior margin
broadly rounded. Surfaces reticulate except in middle of valves where ornamentation is
reduced to small pits or absent; reticulation pattern parallel to anterior margin becoming
Figure 31. Cytherelloidea praecipua Bold, 1963. a-g, hypotype, USNM 128088; a, left valve view oF adult
female carapace; b, right valve view of adult female carapace; c, dorsal view of adult female carapace; d, dorsal
view of adult female left valve; e, dorsal view of adult female right valve; f, interior view of adult female left valve;
g, interior view of adult female right valve.
210 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
pronounced and deep behind strong anterior rim in left valve; broad undulating sulci in
dorsolateral and posterolateral areas, resulting in three ridge-like swellings along dorsum,
venter, and from posterior cardinal angle to center of valve beneath prominent circular
dorsocentral depression corresponding to internal adductor muscle scar swelling. Two
circular swellings in posterior quarter of female carapace. Right valve larger, overlapping
left valve around all margins. In dorsal view: female carapace lanceolate with greatest
width at truncate posterior; right valve strongly overlapping left valve just anterior to
midlength with large flange-like external tooth.
Anterior duplicature broad for genus, about 40 /j. at widest point, fused; 15-18 evenly
spaced, simple, anterior radial pore canals passing through marginal denticles. Hinge of
left valve with large flattened tooth just posterior to midlength; right valve with corre-
sponding "socket." About 1 1 oblong adductor muscle scars in typical Cytherelloidea
pattern on broad swelling in dorsal half of carapace at midlength.
Dimensions. — The collection consists of eight adult specimens of which six were entire.
Lengths range from 522 to 550 /x with an average of 535 /x; heights range from 297 to 3 14 ju,
with an average of 308 ju,; widths range from 1 89 to 228 /x with an average of 2 1 1 p..
Hypotype USNM 128088. Adult left valve, sta. B-6120
Hypotype, USNM 128088. Adult right valve, sta. B-6120
Hypotype, SDNH 04248. Adult carapace, sta. B-6120
Hypotype, SDNH 04249. Adult carapace, sta. B-6120
Hypotype, SDNH 04250. Adult carapace, sta. B-6120
Hypotype, SDNH 04251. Penultimate carapace, sta. B-6120
Hypotype, SDNH 04252. Penultimate carapace, sta. B-6120
Hypotype, SDNH 04253. Penultimate carapace, sta. B-6120
Discussion. — Small difl'erences can be noted between the Clipperton Island forms of
Cytherelloidea praecipua and those described by Bold (1963) from Tobago and Trinidad.
Bold's illustrations of the species show a more arched dorsum and concave downward
venter. In addition, the left valve hinge tooth appears smaller. In all other aspects the
Clipperton Island forms seem identical to those from the Caribbean.
ACKNOWLEDGEMENTS
We would like to thank W. A. van den Bold, Louisiana State University, for information concerning the
relationship of our fauna to those of middle America, and J. Teeter, University of Akron, who made unpublished
information available. Arnold Ross and J. R. Jehl, Jr., San Diego Natural History Museum, read the final
manuscript and offered many helpful suggestions. The laboratory work and manuscript preparation, largely
completed at San Diego State College during the summer of 1968, have received partial support from the
National Science Foundation through Grant GA-1403. This paper is a contribution from the Scripps Institution
of Oceanography, University of California, San Diego, California.
LITERATURECITED
Alexander, C. I.
1929. Ostracoda of the Cretaceous of North Texas: Texas Univ. Bull. 2907, 1 37 p. 10 pis.
Apostolescu, V.
1967. De'termination des ostracodes de la Mission Singer-Polignac en Nouvelle-Caledonie: Editions de la
Found. Singer-Polignac, Paris, p. 121-125, 2 pi.
Baird, W.
1849. Arrangement of the British Entomostraca, with a list of species, particularly noticing those which have
as yet been discovered within the bounds of the Club: Berwickshire Nat. Club (Hist.) Proc.(I842-
1849), 2: 145-158.
ength
Height
Width
538
300
189
538
304
189
540
309
215
533
297
209
550
313
228
477
266
156
482
275
152
484
273
158
1971 ALLISON AND HOLDEN: CLIPPERTON OSTRACODA 21 1
1850. The natural history of the British Entomostraca. Roy Soc. London, p. i-viii, 1-366, 36 pis.
Belcher, E.
1843. Narrative of a voyage round the world, performed in Her Majesty's Ship "Sulphur" during the years
1836-1842... .Vol. l.Coiburn, London, 387 (vol. 1 ) and 474 (vol. 2) pp.
Benson, R. H.
1959. Ecology of Recent ostracodes of the Todos Santos Bay region, Baja California, Mexico: Univ. Kansas
Paleont, Contr., Arthropoda. art. 1:1-80, 11 pi., 20 figs.
1964. Recent marine Podocopid and Platycopid ostracodes of the Pacific: Pubbl. Staz. zool. Napoli 33,
suppl.: 387-420.
Benson, R. H. and G. L. Coleman, II
1963. Recent marine ostracodes from the eastern Gulf of Mexico: Univ. Kansas Paleont. Contr.,
Arthropoda, art. 2: 1-52.
Benson, R. H. and R. L. Kaesler
1963. Recent marine and lagoonal ostracodes from the Estero de Tastiota region, Sonora, Mexico
(Northeastern Gulf of California): Univ. Kansas Paleont. Contr., Arthropoda, art. 3: 1-34.
Bold, W. A. v. d.
1946. Contribution to the study of Ostracoda with special reference to the Tertiary and Cretaceous
microfauna of the Caribbean region. Diss. Univ. Utrecht, Amsterdam, 167 p., 18 pi.
1957. Oligo-Miocene Ostracoda from southern Trinidad. Micropaleontology 3: 231-254, 4 pi., 2 figs.
1963. Upper Miocene and Pliocene Ostracoda of Trinidad. Micropaleontology 9: 361-424, 12 pi.
1966. Ostracoda from Colon Harbour, Panama. Caribbean Jour. Sci. 6(1-2): 43-64, 5 pi.
Brady, G. S.
1866. On new or imperfectly known species of marine Ostracoda. Zool. Soc. London, Trans. 5: 359-391, pis.
57-62.
1868. A monograph of the Recent British Ostracoda. Trans. Linn. Soc, London 26: 353-495, pis. 23-41.
1870. Notes on Entomostraca taken chiefly in the Northumberland and Durham District (1869) Trans. Soc.
Nat. Hist. Northumberland Durham 3: 361-373.
1880. Report on the Ostracoda dredged by the H.M.S. Challenger during the years 1873-1876. Rept.
Voyage Challenger. Zool. 1 (3): 1-184, pis. 1-44.
1890. On the Ostracoda collected by G. S. Brady in the South Sea Islands. Trans. Roy. Soc. Edinburgh 35:
489-525, 4 pi.
1911. Notes on marine Ostracoda from Madeira. Proc. Zool. Soc. London 2: 595-601, pis. 20-22.
Brady, G. S. and A. M. Norman
1889. A monograph of the marine and freshwater Ostracoda of the North Atlantic and northwestern
Europe: Sect. 1. Podocopida. Trans. Roy. Dublin Soc. Sci., ser. 2, 4: 63-270, pis. 8-23.
Briggs, J. C.
1961. The East Pacific Barrier and the distribution of marine shore fishes. Evolution 15: 545-554, 3 text fig.
Crouch, R. W.
1949. Pliocene Ostracoda from Southern California. J. Paleon. 23: 594-599, pi. 96.
Durham, J. W. and E. C. Allison
1960. The geologic history of Baja California and its marine faunas. Syst. Zool. 9 (2): 47-91, 7 text figs., 9
tables.
Egger, J. G.
1901. Ostracoden aus Meeresgrund-Proben gelothet von 1874-1876 von S.M.S. Gazelle. Abh. Math-Phys.
Kl. K. Bayerische Akad. Wiss. 21: 413-478, 8 pi.
Ekman, S. P.
1953. Zoogeography of the Sea (translated from Swedish by Elizabeth Parker). Sedgwick and Jackson,
London, XIV, 417 p.
Emerson, W. K.
1967. Indo-Pacific faunal elements in the tropical eastern Pacific, with special reference to the mollusks.
Venus 25: 85-93 text fig. 1
Fergusson, G. J. and W. F. Libby
1962. UCLA radiocarbon dates I. Radiocarbon 4: 109-1 14.
Fischer, S.
1855. Beitr'aige zur kenntnis der Ostracoden. Abh. Math.-Phys. Kl. K. Bayerische Akad. 7(3): 637-666, pis.
19-20.
Gunther, F. J.
1967. Ostracoda of the Gulf of Panama and Bahia San Miguel. Univ. Minnesota M.S. Thesis, 204 p., 8 pi.
Guha, D. K.
1968. On the Ostracoda from Neogene of Andaman Islands. J. Geol. Soc. India 9: 58-66, pis. 4-5.
212 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Hanai, T.
1957. Studies in the Ostracoda from Japan: 3, subfamilies Cytherurinae G. W. M'uller (emend. G. O. Sars,
1925) and Cytheroplerinae n. subfam. J. Tokyo Univ. Fac. Sci., sec. 2(1 1): 1 1-36, pis. 2-4, 9 figs.
Hartmann, G.
1953. Iliocythere meyer-abichi nov. spec, ein ostracodes des Schlickwattes von San Salvador. Zool. Anz.
151:310-316.
1956. Zur kenntnis des Mangrove-Estero-Gebeites von El Salvador und seiner Ostracoden-fauna. Kieler
Meeresforsch. 12:219-248.
1957a. Zur kenntnis des Mangrove-Estero-Gebeites von El Salvador und seiner Ostracoden-fauna. II.
Kieler Meeresforsch. 13: 134-159, pis. 39-50.
1957b. Contribucion al conocimiento de la region de estero y manglar de el Salvador y sua fauna de
ostracodos, la Parte Ecologia, 2a. Parte sistematica. Sobretiro Rev. Comun. Inst. Trop. Invest. Cient.
Univ. El Salvador 6: 47-108, 6 pi.
1959a. Beitrage zur kenntnis des Nicaragua-Sees unter besonderer Berucksichtigung seiner Ostracoden (mit
Beschreibung von 5 neuen Arten). Zool. Anz. 162: 269-294, 9 figs.
1959b. Zur kenntnis der lotischen Lebenbereiche der pazifischen Kuste von El Salvador unter besonder er
Berucksichtigung seiner Ostracoden fauna, 3, Beitragzur Fauna el Salvador. Kieler Meeresforsch. 15:
187-241, pis. 27-48.
Hertlein, Leo G.
1937. A note on some species of marine moUusks occurring in both Polynesia and the Western Americas.
Amer. Philos. Soc. Proc. 78 (2): 303-312, pi. 1, map.
Hertlein, Leo G. and William K. Emerson
1953. Mollusks from Clipperton Island (eastern Pacific) with the description of a new species of gastropod.
San Diego Soc. Nat. Hist., Trans. 11(13): 345-364, pi. 26-27.
Hinte, J. E. Van
1964. A new Occultocylhereis species of the Austrian Eocene. K. Nederlandse Akad. Wetensch., ser. B, 62:
108-115, 5 figs. '
Holden, J.C.
1964. Upper Cretaceous ostracods from California: Paleontology 7: 393-429, 28 figs.
1967. Late Cenozoic ostracodes from the drowned terraces in the Hawaiian Islands: Pacific Sci. 21: 1-50, 36
figs.
Howe, H. V.
1951. New Tertiary ostracode fauna from Levy County, Florida. Florida Geol. Surv. Bull. 34, 43 p., 5 pis.
Howe, H. V. and John Law
1936. Louisiana Vicksburg Oligocene Ostracoda. Dept. Conserv., Louisiana Geol. Bull. 7, 96 p., 6 pis.
Juday, C.
1907. Ostracoda of the San Diego region. II. Littoral forms. Univ. Calif. Pubis., Zool. 3: 135-358, pis. 18-
20.
Kaufmann. A.
1900. Zur Systematik der Cypriden. Mitt. Naturforsch. Ges. Bern.: 103-109.
Key, A. J.
1953. Preliminary note on the Recent Ostracoda of the Snellius Expedition: Koninkl. Nederlandse Akad.
Wetensch. Amsterdam, ser. B, 56: 1 55-168, 2 pis.
Kingma, J. T.
1948. Contributions to the knowledge of the Young-Caenozoic Ostracoda from the Malayan region. Diss.
Univ. Utrech, Kemink Printers, Utrecht, 1 19 p., 1 1 pis.
Klie, W.
1938. Ostracoda, Muschelkresse. In, DahFs Die Tierwelt Deutschland und der Angrenzenden Meeresteile,
34, 230 p.
1940. Beitrage zur Fauna der Eulitorals von Deutsch-Sudwest-Afrikas. Kieler Meeresforschungen 3: 404-
448.
Kollmann, K.
1963. Ostracoden ausder alpinen Trias II. Weitere Bairdiidae. Jahrb. Geol. 106: 121-203, 1 1 pis.
Kornicker, L. S.
1961. Ecology and taxonomy of recent Bairdiinae (Ostracoda). Micropaleontology 7 (1): 55-70.
LeRoy, L. W.
1943. Pleistocene and Pliocene Ostracoda of the coastal region of Southern California. J. Paleon. 17: 354-
373, pis. 58-62.
1945. A contribution to ostracodal ontogeny. J. Paleon. 19: 81-86, pis. 1-25, figs. 1-2.
1971 ALLISON AND HOLDEN:CLIPPERTONOSTRACODA 213
Lloyd, J.
1963. Tectonic history of the south Central American orogen. In, Backbone of Americas. Amer. Assoc.
Petrol. Geol. Mem. 2: 88-100.
Maddocks, R. F.
1966. Distribution patterns of living and subfossil podocopid ostracodes in the Nosy Be area. Northern
Madagascar. Univ. Kansas Paleont. Contr., Paper 12, 72 p. 63 text figs.
McCoy, F.
1844. Synopsis of the characters of the Carboniferous fossils of Ireland (Crustacea). Dublin Univ. Press,
Dublin: 159-168.
McKenzie, K. G.
1967. The distribution of caenozoic marine Ostracoda from the Gulf of Mexico to Australia. In, Adams,
C. G., and Ager. D. V., (eds). Aspects of Tethyan biogeography; a symposium. Syst. Assoc. Publ.
No. 7:219-238.
McKenzie, K. G. and Frederick M. Swain
1967. Recent ostracoda from Scammon Lagoon, Baja California. J. Paleon. 41 : 281-305, pis. 29-30.
Morkhoven, F. P. C. M., van
1963. Post-Paleozoic Ostracoda: their morphology, taxonomy, and economic use; vol. 2. Generic descrip-
tions. Elsevier Publ. Co., 478 p.
Menard. H. W., and R. L. Fisher
1958. Clipperton Fracture Zone in the northeast equatorial Pacific. J. Geol. 66: 239-253, 1 pi., 8 text figs.
Muller, G. W.
1894. Die Ostracoden des Golfes von Neapel und der Angrenzenden Meeres-Abschnitte. In, Fauna and
Flora des Golfes von Neapel und der Angrenzenden Meeres-Abschnitte, Monogr. 21: 404. 40 pis.
1912. Ostracoda. In, Das Tierreich. Eine zusammenstellung und kennzeichnung der rezenten Tierformen Im
Auftrage der K. Preuss. Akad. Berlin 31:1 -434. 92 figs.
Muller. O. F.
1776. Zoolgiae Danicae prodromus, seu animalium Daniae et Norvegiae indigenarum charateres, nomina
et synonyma imprimis popularium. Hallager, Copenhagen: 1-282.
Neviani, A.
1928. Ostracodi fossili dTtalia, I. Vallebiaja (Calabrino). Mem. Pont. Sci., Nuovi Lincei, ser. 2, 11, 120 p.,
21 pis.
Puri, H.S.
1953. The ostracode genus Hemicythere and its allies. Washington Acad. Sci., J. 43(6): 169-179, pis. 1-2.
I960. Recent Ostracoda from the west coast of Florida. Gulf Coast Assoc. Geol. Soc, Trans. 10: 107-149,6
pis., 46 figs.
Reyment, R. A.
1964. Notes on an upper salinity tolerance level for Cvpridopsis (Ostracoda). Crustaceana 7: lb-11 .
Rothwell, \V. T.
1948a. Distribution of living ostracodes, Newport Bay. California. Geol. Soc. Amer.. Bull. 59: 1380-1381
(abst.).
1948b. Paleoecological interpretations from ostracodes: Geol. Soc. Amer., Bull. 59: 1381 (abst.).
Sachet, M.
1960. Histoire de Tile Clipperton. Cahiers du Pacifique 2: 3-32, 1 pi., 1 text fig.
1962a. Flora and vegetation of Clipperton Island. California Acad. Sci., Proc, ser. 4, 31 (10): 249-307, 12
text figs.. 1 table, 1 map.
1962b. Geography and land ecology of Clipperton Island. Atoll Res. Bull. 86 (III), 115 p., 4 text figs., 5
tables.
1962c. Monographic physique et biologique d Tile Clipperton. Ann. Inst. Oceanogr. (Paris), 40 (1): 107 p.,
12 pis., 3 text figs., 3 tables.
1963. History of change in the biota of Clipperton Island. In, Pacific Basin biogeography (symposium, ed. J.
L. Gressit). Bernice P. Bishop Mus. Press: 525-534.
Sars, G. O.
1866. Oversight of Norges marine ostracoder. Forhandl. Vidensk. Selsk. Christiana 7: 1-130.
1888. Nye Bidrag til Kundskaben om Middelhavets Invertebratfauna: IV' Ostracoda Mediterranea. Archiv.
Math. Naturw. 12: 173-324, pis. 1-20.
Sars. GO.
1926. An account of the Crustacea of Norway, Cytheridae (cont.). Bergen Mus., Norway, pts. 13-14: 209-
240, pis. 97-112.
Skogsberg. T.
1928. Studies on marine ostracodes. Part. 2. External morphology of the genus Cylhereis and descriptions
214 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
of twenty-one new species. Occas. Papers, California Acad. Sci. 15: 1-144.
1950. Two new species of marine Ostracoda (Podocopa) from California: Proc. California Acad. Sci., ser. 4,
26: 483-505, pis. 27-30.
Sverdrup, H. V., Martin W. Johnson, and Richard H. Fleming
1942. The oceans, their physics, chemistry, and general biology. Prentice-Hall, Inc., x -i- 1027 p.
Swain, M.
1967. Ostracoda from the Gulf of California. Geol. Soc. America, Memoir 101, 139 p.
Swain, M. and J. M. Gilby
1964. Ecology and taxonomy of Ostracoda and an alga from Lake Nicaragua. Pubbl. Staz. zool. Napoli 33
(suppl.): 361-386, 4 pis.
Swain, W. and F. J. Gunther
1969. Recent ostracoda from San Juan del Sur, Nicaragua, and their relationship to other ostracode
populations of western Central America. Inst. Centroamericano Invest. Tech. Indust. (Guatemala, C.
A.), Trabcijos Technicos Presentados en la segundo Reunion de Geologos de America Central Publ.
Geol. I. C.A.I. T.I. 2: 54-55, 1 table (abst.).
Sylvester-Bradley, P. C.
1948. The ostracode genus C>'//?e/-m. J. Paleont. 22(6): 792-797, pi. 122.
Teeter, W. T.
1966. The distribution of Recent marine ostracodes from British Honduras. Ph.D. thesis, Rice University,
Houston, Texas, 212 p.. 19 pis.
Tressler, W. L.
1941. Ostracoda, part 4, q/ Geology and Biology of North Atlantic deep-sea cores between Newfoundland
and Ireland. U.S. Geol. Surv. Prof. Paper 196-C: 95-106, pis. 18-19.
Triebel, E.
1948. Zur Kenntnis der Ostracoden-Gattung Triebelina. Senckenbergiana 29: 17-22.
1954. Loxoconchella n. g. (Crust., ostr.): Senkendergiana Lethaea 35: 17-21,2 pis.
1956. Brackwasser-ostracoden ven den Galapagos-Inseln: Senckenbergiana Biol. 37: 447-467, pis. 54-58.
1957. Neue Ostracoden aus dcm Pleistozan von Kalifornien: Senckenbergiana Lethaea 38: 291-309, pis. 1-5.
1960. Die taxonomische Stellung und die Gattungen der Unterfamilie Macrocypridinae (Ostracoda):
Senckenbergiana Biol. 44: 109-124, pis. 13-20.
1961. Geschlechts-dimorphismus und asymmetric der klappen bei der ostracodengattung Occultocythereis:
Senkenbergiana Lethaea 42: 205-225, 5 pis.
U. S. Navy Hydrographic Office
1947. Atlas of surface currents, northeastern Pacific Ocean. Hydrographic Office Publ. 570, 12 sheets.
1 950. Atlas of surface currents, northwestern Pacific Ocean. Hydrographic Office Publ. 569, 1 2 sheets.
1966. Atlas of pilot charts. South Pacific and Indian Oceans, Third Edition. Hydrographic Publ. 107.
Wagner, C.W.
1957. Sur les ostracodes du quaternaire recent des Pas-Bas et leur utilisation dans Tetude geologique des
deposts holocenes. Mouton and Co., The Hague, 1 58 p. 40 pis.
Wyrlki, K.
1965. Oceanography of the eastern equatorial Pacific Ocean. Oceanogr. Mar. Biol. Ann. Rev. 4: 33-68.
Department of Geology, San Diego State College, San Diego, California 921 15, and
National Oceanic and Atmospheric Administration, Atlantic Oceanographic and Mete-
orological Laboratories, 901 South Miami Avenue. Miami. Florida 33130
j-^^J MUS. COMP. ZOOL.
LIBRARY
JUN 41971
HARVARD
UNIVERSITY
STUDIES ON THE TETRACLITIDAE
(CIRRIPEDIA: THORACICA)
A NEW TETRACLITELLAN FROM INDIA
ARNOLD ROSS
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 8 21 MAY 1971
STUDIES ON THE TETRACLITIDAE
(CIRRIPEDIA: THORACICA)
A NEW TETRACLITELLAN FROM INDIA
ARNOLD ROSS
ABSTRACT — Tetraclitella contains eight species, including T. karandei n. sp. from Mad'h Island, India,
all of which are restricted largely to the Indo-West Pacific faunal province. Two groups may be recognized in
this genus on the basis of opercular morphology. One species in each of these groups has radii that are
elevated well above the surface of the parietes. In T. danvini the elevated radii serve to strengthen the shell in
the absence of sutural ridges and denticulae; in T. karandei they probably create water turbulence and thus
enhance the fishing capabilities of the cirral net.
Tetraclitella comprises eight, relatively small, patelliform, balanomorph barnacles
that occupy habitats low in the intertidal zone. They are confined largely to the Indo-West
Pacific faunal province, contrary to the statement by Utinomi (1970: 349) that they are
"mostly circumtropical." All of the species occur predominantly on continental islands but
there are a few scattered mainland records. Exceptions to this distribution pattern are T.
purpurascens, which ranges from Australia to India, and T. divisa which is the only species
that occurs circumtropically (Ross, 1968: 14).
The barnacle fauna of India and adjacent areas is relatively well known through the
work of Annandale, Nilsson-Cantell, Karande (1966) and several contemporary Indian
workers. Therefore, it is surprising to note the presence of a new tetraclitellan from Mad'h
Island on the Bombay coast of India (Fig. 1). This new species is similar in many ways to
the widely occurring T. purpurascens, and records for that species should be reevaluated in
the light of the present discovery.
Dr. A. A. Karande, who collected the specimens reported on here, informed me that it
occurs on the under surface of rocks, low in the intertidal zone, where it normally remains
moist during periods of low tide. The shells commonly are covered with a dense mat of
brownish-green, finely particulate, organic matter. The associated animals include the
ubiquitous Planaxis sulcatus Born and a species of Acmaea. The ecological conditions
under which this species lives and the few animals with which it is associated do not differ
appreciably from those of other species of Tetraclitella.
Family Tetraclitidae Gruvel, 1903
Genus Tetraclitella Hiro, 1939
Definition. — Shell generally less than 20 mm in rostro-carinal diameter, patelliform,
ribbed; compartments discrete; parietes with 2 or more rows of tubes; radii broad, flush
with or raised above parietal surface, summits horizontal, tubiferous, lacking teeth or
denticles on articular surface; alae non-tubiferous; basis membranous, calcareous
peripherally or wholly calcareous; scutum transversely elongated or higher than wide,
commonly ornamented externally, lacking crests for depresser muscles; mandible with 5
teeth and spine-like lower angle; maxilla I with 6-8 major spines below subapical notch.
SAN DIEGO SOC. NAT. HIST., TRANS. 16 (8): 215-224, 21 MAY 1971
216
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 1 . Map showing position of Mad'h Island relative to other islands along the Bombay coast of India.
1971 ROSS: STUDIES ON THE TETRACLITIDAE 217
Type species. — Lepas purpurascens Wood (1815: 55), Recent, Australia, by original
designation of Hiro ( 1 939: 273).
Remarks. — Hiro (1939: 273) established Tetraclitella as a subgenus of Tetraclita.
Recently, in reevaluating the tetraclitids I raised the subfamily to familial status (Ross,
1968: 6), and accordingly the subgenera of Tetraclita were raised to genera to better retlect
relationships within the family (Ross, 1969: 237; Ross, 1970: 3). Utinomi (1970: 349)
independently also accorded Tetraclitella generic rank.
Species referable to Tetraclitella include: T. purpurascens (Wood, 1815: 55), T.
costata (Darwin, 1854: 339), T. chinensis (Nilsson-Cantell, 1921: 359), T. divisa (Nilsson-
Cantell, 1921: 362), T. darwini (Pilsbry, 1928: 314), T. multicostata (Nilsson-Cantell,
1930: 2) and T. pilsbryi (Utinomi, 1962: 234). Tetraclita squamosa depressa (Kolosvary,
1941: 42) from southern Australia, Tetraclita purpurascens darwini {Kolosvary, 1942: 140)
from Port Jackson, New South Wales, Australia, and Tetraclita radiata wagneri (Kolos-
vary, in Kolosvary and Wagner, 1941: 11) from Tasmania, on the basis of morphology and
biogeography, are apparently conspecific with T. purpurascens purpurascens.
KEY JOTHESPECIESOF TETRACLITELLA
1. Radii elevated above surface of parietes 2
1 . Radii flush with or sunken below surface of parietes 3
2. Scutum higher than wide; intermediate segments of cirrus
VI with 4 pairs of setae (Japan, Formosa) T. darwini
2. Scutum wider than high; intermediate segments of cirrus
VI with 3 pairs of setae (India) T. karandei
3. Scutum higher than wide 4
3. Scutum wider than high 5
4. Scutum with a row of small longitudinal pits; intermediate
segments of cirrus VI with 4 pairs of setae; basis calcareous
(Lesser Sunda Islands, Sulu and Philippine Archipelagos) T. costata
4. Scutum with 5 rows of longitudinal pits; intermediate segments of
cirrus VI with 3 pairs of setae; basis membranous (Japan) T. pilsbryi
5. Tergal spur essentially confluent with scutal margin 6
5. Tergal spur well separated from scutal margin (intermediate
segments of cirrus VI with 3 pairs of setae; circumtropical) T. divisa
6. Parietal plates without hollows 7
6. Parietal plates pierced by hollows (intermediate segments or
cirrus VI with 4, rarely 3 pairs of setae; southern China,
Formosa, Japan) T. chinensis
1. Shell with 14 or fewer primary longitudinal ribs; cuticle
persistent (West Irian, New Guinea) T. multicostata
1 . Shell with 20 or more primary longitudinal ribs; cuticle not
persistent (intermediate segments of cirrus VI with 2 pairs of setae;
New Zealand, Tasmania, Australia, Malay Archipelago, India) . . T. purpurascens
Tetraclitella karandei n. sp.
Diagnosis. — Radii transversely ridged, the apical 3-4 ridges extending like fingers out
and over adjoining plate; scutum transversely elongated, externally ornamented with
prominent nodes where longitudinal ridges cross growth lines; intermediate articles of
posterior cirri armed with 3 pairs of setae.
218 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Description. — Shell white or grayish white, patelliform, ovate in outline, covered with
persistent, hirsute, chitinous cuticle; parietes ornamented with prominent growth ridges,
and high, primary, longitudinal ribs intercalated with lower secondary and tertiary ribs;
ribs square or T-shaped in section, extending from orifice to or beyond basal edge of shell,
occasionally bifurcate basally (Fig. 2a, b); orifice diamond-shaped; radii broad, horizon-
tally ridged from base to apex, the ridges becoming progressively higher, produced and free
from the surface (Fig. 2a); articular margin and finger like projections tubiferous, the
apertural margins of the tubes being crenate; on the adjoining plate a narrow, longitudinal
ridge occurs on parietal surface where the radius butts against the plate (Fig. 2b); alae
broad, summits horizontal and crenate; sheath less than '/2 height of wall, basal margin not
depending; basis calcareous peripherally.
Scutum wider than high; external surface deeply sulcate (Fig. 2e); where the growth
ridges are crossed by longitudinal ridges prominent nodes are formed thus rendering a
scabrous appearance; articular ridge straight, about 2/3 length of tergal margin; adductor
ridge low, not undercut, apically fused with articular ridge, terminating basally at basi-
occludent angle; adductor muscle depression ovate, shallow, borders poorly delimited;
depression for lateral depressor muscle shallow, poorly defined; depression for rostral de-
pressor muscle lacking; apical portion of plate with weak ridges (Fig. 2d).
Tergum higher than wide; external longitudinal furrow open, broad, shallow, extend-
ing to base of spur; spur evenly rounded basally, confluent with scutul margin, width about
'/2 that of basal margin (Fig. 2g); articular ridge inclined; articular furrow wide and
shallow; 6-7 crests for depressor muscle, low, short, inclined; apical portion of valve ridged
or roughened (Fig. 20-
Measurements of the holotype are as follows (in mm): rostro-carinal diameter 10.1;
height 3.5; rostro-carinal diameter of orifice 3.6; height of scutum 1.5; width of scutum 2.1;
height of tergum 1.5; width of tergum 1.0. The mean rostro-carinal diameter of five
paratypes is 13.5 mm and the height is 3.8 mm.
Labrum with shallow, broad, medial depression; crest thick, heavily chitinized,armed
with short, fine bristles but rarely with teeth (Fig. 3a). Palps long, broad, distal end broadly
rounded; superior margin straight, basal margin convex; proximal setae on superior
margin short, stout, coarsely bipectinate; distal setae on margin long, slender, finely
bipinnate; basal portion of lateral surface covered with ctenae. Mandible with 5 unequally
spaced teeth; teeth 2 and 3 commonly with 1-2 subsidiary cusps; tooth 4 with 3-5 subsidiary
cusps; comb between tooth 5 and inferior angle containing 8-12 teeth; inferior angle
commonly with 1 long, slender and 1 short, stout tooth (Fig. 4). Maxilla I with 2 long, stout
and 1-2 shorter spines above sub-apical notch; 2-3 short, slender spines in notch; 6-8 stout
spines medially; 8-12 short, slender spines in basal cluster (Fig. 3c). Maxilla II bilobate;
setae along apical margin long; bipinnate, setae becoming progressively shorter toward the
notch; setae on basal lobe coarse, bipectinate.
Posterior ramus of cirrus I about 3/5 length of anterior ramus; intermediate articles
of both rami broader than high; segments of anterior ramus normal, but those of posterior
ramus protuberant; distal articles of both rami clothed with finely bipinnate setae (Fig. 3g).
Rami of cirrus II essentially equal in length, and slightly longer than posterior ramus of
cirrus I; medial segments of both rami protuberant; distal two segments of both rami
armed with bipectinate setae, proximal segments with bipinnate setae (Fig. 3h). Rami of
cirrus III of equal length, and same length as rami of cirrus II; medial segments of both
rami protuberant; distal 2 or 3 segments of anterior ramus and all segments of posterior
ramus armed with bipectinate setae (Fig. 3i). Cirri IV-VI essentially equal in length with
equal rami; 1-2 stout spines and 2-3 long, slender setae at each articulation along greater
1971
ROSS: STUDIES ON THE TETRACLITIDAE
219
M
i^ /
^A -^
^-^
LJ
■?.'
f
Figure 2. Shell and opercular plates of Tetraclitella karandei n. sp. a, apertural view of shell, x6; alar margin of
lateral compartment, x7; c, basal view of carina, x8: d, e, internal and external views, respectively, of scutum, x30;
f, g, internal and external views, respectively, of tergum, x30. Holotype (4000), a, c-g; paratype (400 1/c), b.
220
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
c
0.05mm
e
0,01
f 0,2 5
ad
gh i
0,1
0,25
Figure 3. Trophi and cirri of Tetraclitella karandei n. sp. a, iabrum and palp; b, enlarged view of labrum; c,
maxilla 1; d, maxilla II; e, intermediate segments of outer ramus of cirrus VI; f, penis; g, cirrus I; h, cirrus II; i,
cirrus III. Holotype (4000), c, f; paratypes, a-b, d-e, g-i (a, d = 4001/c;b = 400i/d;e, g-i = 4001/b).
1971 ROSS: STUDIES ON THE TETRACLITIDAE 221
curvature of intermediate segments; 1 or 2 rows of ctenae present on lateral face of
intermediate segments immediately below articulation; setation ctenopod, with 3 pairs of
setae on each intermediate segment, and commonly a single proximal, short seta; at base of
and between each major pair of setae there is a cluster of 3-5 short, slender setae (Fig. 3e).
Cirral counts for specimens in the type lot are summarized in Table 1 .
Table 1. Summary of data on cirral counts: range (R) and mean (X) value for number of segments in anterior
(a) and posterior (p) rami
I
II
III
IV
V
VI
a
P
a
P
a
P
a p
a
P
a p
N
10
10
12
12
11
11
7 7
6
6
4 4
R
lO-U
6-7
7-8
6-8
6-8
6-9
11-14 11-15
14-16
15-17
15-16 15-18
X
10.6
6.3
7.3
7.1
7.1
6.2
12.8 14.0
15.0
15.2
15.5 16.7
Intromittent organ annulated throughout its length, and sparsely covered with short,
slender bristles; distal extremity with 4 clusters of 1 1-14 setules (Fig. 30-
Remarks. — Of the presently recognized tetraclitellans, T. karandei may be distin-
guished by its radii, which have raised digitiform processes that extend out and over the
adjoining plates. The shape of the tergum in karandei is similar to that found in
multicostata, purpurascens and chinensis, but the scutum of this species has a scabrous or
nodose ornamentation externally rather than only simple growth ridges. The mandible
of karandei appears to be more variable than any other species in the degree of development
and number of subsidiary cusps on the second, third, and fourth teeth (see Fig. 4). The
crest of the labrum is commonly devoid of teeth as it is in purpurascens, costata and
darwini. But when teeth are developed, they appear as simple, low, rounded knobs, that
are few in number. The mouth parts and appendages have not been described for pilsbryi.
Disposition of types. — The holotype and four paratypes are housed in the collections
of the San Diego Society of Natural History, Marine Invertebrate catalogue numbers 4000
and 4001, respectively. Two paratypes are in the collections of the Zoological Survey of
India, Calcutta. The remaining specimens have been retained by the author.
Type locality. — Mad'h Island, Bombay Coast, India, approximately 19°8'N.,
72°47'22"E.; A. A. Karande coll. 1969; 10 specimens.
Comparative material. — I have examined specimens of the following species:
T. divisa: western side of Panto Hole Bay, east of town of Marigot, Dominica;
approximately 15°32'21"N., 61°17'31"W., intertidal on Tetraclita stalactifera (Lamarck);
E. Kirsteuer and K. Rutzler coll., 1-10 May 1966 (see Ross, 1968:13).
T. chinensis: Suo (Suao), Taiwan, approximately 24°35'45"N., 121°51'10"E., "... on
sheltered undersurface of stones in the littoral"; F. Hiro coll., 30 May 1938 (see Hiro,
1939:277).
T. purpurascens: Eddystone Point, Tasmania, approximately 40°59'30"S., 148°20'E.;
I. Bennett coll., 20 June 1964. The Nobbies, Phillip Island, Victoria, Australia, approx-
imately 38°30'S., 145°16'E.; E. C. Pope coll.. May 1949. Little Papanui, Otago Peninsula,
South Island, New Zealand, approximately 45°50'S., 170°43'E.; C. Hand coll., 4 Novem-
ber 1959.
T. darwini: Isle Hatake-zima, Tanabe Bay, Wakayma Prefecture, Japan, approx-
imately 33°43' N., 125°21'30" E.; F. Hiro coll., 3 April 1928.
Etymology. — The specific epithet honors Dr. Ashok A. Karande, Senior Science
Officer, Naval Chemical and Metallurgical Laboratory, Bombay, India, who collected the
specimens.
■>■>!
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 4. Mandibles of Tetraclilella karandei n. sp. and related tetraclitellans. A, E, right and left, respectively,
hololype (4000); B, F, right and left, respectively, paratype 400 1/b; C, G, right and left, respectively, paratype
4001/c; D, H, right and left, respectively, paratype 4001/d; I, K., right and left, respectively, paratype 4001/e; J.
right, paratype, Zool. Survey India; L, after Utinomi, 1962; M, from Dominica; N, from Taiwan; O, from
Tasmania; P, from Japan.
1971 ROSS: STUDIES ON THE TETRACLITIDAE 223
DISCUSSION
There are two groups in Tetraclitella based primarily on the morphology of the
operculum. In the first, consisting o{ costata, pilsbryi, and darwini, the scutum is higher
than wide and externally ornamented with one or more longitudinal rows of pits, and the
tergum composes about one-half or more of the bulk of the operculum. In the bala-
nomorphs a tall, narrow scutum generally correlates with a relatively tall shell; in these
three species the shell is relatively tall. In the second group, consisting oi purpurascens,
multicostata, divisa, chinensis, and karandei n. sp., the scutum is transversely elongated
and lacks the longitudinal rows of pits, and the tergum composes less than one-half of the
bulk of the operculum. I consider the costata group to be the phylogenetically more
primitive on the basis of the opercular valves, which are characteristic of geologically early
balanomorphs.
One species in each of the above groups develops radii that are elevated well above the
surface of the parietes {darwini and karandei). Radii develop essentially normal to the
parietes and function to enlarge and strengthen the shell. Similar functions are served by
the alae (Darwin, 1954: 36, 45-48), which are always non-tubiferous, contrary to the
statements of Pilsbry (1928: 316), and Hiro (1939: 273). The sutural surface of the radius
abuts against and fits into a furrow in the opposed compartment, the outer edge of which
may be raised to form a lip, as in darwini and karandei. In darwini this lip, an extension of
the parietes, is tubiferous. I infer that the elevated radii in darwini serve primarily as a
means of developing a larger sutural surface for strengthening the shell, especially in the
absence of sutural ridges and denticulae. Attempts to manually separate the plates in this
species are rarely if ever successful. Conversely, the plates in karandei are easily separated
from one another. However, in karandei the development of a prominent lip on the
adjoining compartments (Fig. 2b) may serve to strengthen the articulation of the plates.
Because the parietal plates of karandei are weakly articulated, and because karandei
occupies a protected habitat low in the intertidal zone, it is reasonable to assume that the
finger-like projections on the exposed radial surfaces serve a different function than the
raised radii of darwini. I believe that these projections function primarily to scatter the
initial energy of the incident currents into numerous smaller components. This would
create turbulence or change the water flow pattern over the shell, and consequently
enhance the fishing capabilities of the cirral net.
The mode of growth of the shell in T. chinensis sets it apart from all other
tetraclitellans. In the adult or large specimens the compartments are pierced by hollows,
one in each of the laterals and two in both the rostrum and the carina. Hiro (1939: 274)
considered them to be parietal tubes formed by the corrosion of the parietal wall, but it is
evident from the young stages that initially these hollows are external to the parietes, and
consequently they cannot be parietal tubes. The hollows result from the initial development
of distally flaring extensions from the shell, the lateral tips of which subsequently meet and
fuse in a manner somewhat analogous to the fusion of the terminally flanged radial
buttresses in whale barnacles. I believe that this method of shell growth enables chinensis
to rapidly develop a broad base of attachment in a high energy environment. Support for
this inference comes from the fact that in the few adult specimens I have seen the shell is
essentially circular in outline, lacks well preserved longitudinal ribs, and the peritreme is
eroded. Alternatively, this method of shell growth may be a means to prevent over-
crowding, but observations to support this suggestion are lacking.
ACKNOWLEDGMENTS
I thank A. A. Karande, Naval Chemical and Metallurgical Laboratory, India, Elizabeth C. Pope, the
224 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Australian Museum, Huzio Utinomi, Seto Marine Biological Laboratory, Brian Foster, Auckland University,
William A. Newman, Scripps Institution of Oceanography and Ernst Kirsteuer, American Museum of Natural
History, for their help in providing me with comparative material. Anne Acevedo prepared figures 1 and 2.
LITERATURE CITED
Darwin, C. R
1854. A monograph on the sub-class Cirripedia. The Baianidae, the Verrucidae, etc. London, Ray Society,
684 p.
Hiro, F.
1939. Studies on the cirripedian fauna of Japan. IV. Cirripeds of Formosa (Taiwan) with some geogra-
phical and ecological remarks on the littoral forms. Mem. Coll. Sci. Kyoto Imp. Univ. ser. B, 15(2):
245-284.
Karande, A. A.
1966. The sessile barnacles (Cirripedia) of the Bombay Coast. J. Bombay Nat. Hist. Soc. 63(1): 139-151.
Kolosvary, G.
1941. Balaniden-Studien. Zool. Anz. 135(1/2): 41-45.
1942. Studien an cirripedian. Zool. Anz. 137(7/8): 138-150.
Kolosvary, G. and J. Wagner
1941. Tengerbiologiai tanulmany a kacslabuak^puhatestuek es korallok tarsulasarol. A Tenger 31(1-3):
1-16.
Nilsson-Cantell, C. A.
1921 . Cirripeden-Studien. Zur Kenntnis der Biologic, Anatomic und Systematik dicscr Gruppc. Zool. Bidr.
Uppsala 7: 75-395.
1930. Diagnoses of some new cirripedes from the Netherlands Indies collected by the Expedition of His
Royal Highness the Prince Leopold of Belgium in 1929. Bull. Mus. Royal Hist. Nat. Belgique 6(4):
1-2.
Pilsbry, H.A.
1928. Littoral barnacles of the Hawaiian Islands and Japan. Proc. Acad. Nat. Sci., Philadelphia 79: 305-
317, pis. 24-26.
Ross, A.
1968. Bredin-Archbold-Smithsonian biological survey of Dominica. 8. The intertidal balanomorph Cirri-
pedia. Proc. U. S. Natl. Mus. 125(3663): 1-23.
1969. Studies on the Tetraclitidae (Cirripedia: Thoracica): Revision of Tetraclita. San Diego Soc. Nat.
Hist.. Trans. 15(15): 237-251.
1970. Studies on the Tetraclitidae (Cirripedia: Thoracica): A proposed new genus for the austral species
Tetraclita purpurascens breviscutum. San Diego Soc. Nat. Hist., Trans. 16(1): 1-12.
Utinomi, H.
1962. Studies on the cirripedian fauna of Japan. VIII. Thoracic cirripeds from western Kyusyu. Publ. Seto
Mar. Biol. Lab. 10(2): 21 1-239.
1970. Studies on the cirripedian fauna of Japan. IX. Distributional survey of thoracic cirripeds in the
southeastern part of the Japan Sea. Publ. Seto Mar. Biol. Lab 1 7(5): 339-372.
Wood, W.
1815. General conchology; or a description of shells arranged according to the Linnean System. London,
246 p.
Department of Invertebrate Paleontology, Natural History Museum, P. O.
Box 1390. San Diego, California 92112.
LIBRARy
JUL «3 1971
HARVARD
STRATIGRAPHY OF THE POWAY AREA,
SOUTHWESTERN CALIFORNIA
GARY L. PETERSON
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 9 9 JULY 1971
STRATIGRAPHY OF THE POWAY AREA,
SOUTHWESTERN CALIFORNIA
GARY L. PETERSON
ABSTRACT.— Post-batholithic sedimentary rocks near Poway, California, which were previously mapped as
"Poway Conglomerate," consist of three distinctively different formations. The Lusardi Formation (Late Cre-
taceous) consists predominantly of very poorly sorted conglomerate containing an assemblage of locally de-
rived clasts (Peninsular Ranges suite), which range in size from granules to boulders three meters in diameter.
This formation strongly resembles the type Lusardi near Rancho Santa Fe and "pre-Poway fanglomerates"
reported from several localities near Lakeside and Alpine to the southeast. Unconformably overlying the Lu-
sardi Formation are two late Eocene formations: 1) the Friars Formation (La Jolla Group), consisting pre-
dominantly of sandstone and shale, and 2) the Stadium Conglomerate (Poway Group). Conglomerates from
either of the Eocene units are moderately well sorted and consist mostly of exotic cobble-sized metavolcanic,
volcanoclastic, and quartzite clasts (Poway suite).
The Lusardi Formation fills a long narrow channel cut subsequent to mid-Cretaceous orogenesis during
a time of rugged topography. After Lusardi deposition, a more subdued erosion surface was developed on the
Lusardi Formation and on the basement rocks, and the terrain underwent severe weathering. The Eocene
formations were deposited on this deeply weathered erosion surface and received little locally derived coarse
detritus. The Poway suite of clasts accumulated in southwestern California after having been derived from the
east and transported across the low-lying erosion surface.
RESUMEN.— Las rocas sedimentarias post-batoliticas cercanas a Poway, California, mapeadas previamente
como "Conglomerado Poway," estan constituidas por tres formacions bien diferentes. La Formacion Lusardi
(Cretaceo Superior) esta constituida predominantemente por conglomerados de muy pobre seleccion, cuyos
clastos tienen una procedencia local (Peninsular Ranges suite) y varian en tamario desde granulos hasta pe-
riascos de tres metros de diametro. Esta formacion tiene un fuerte parecido con la Formacion Lusardi-tipo
cerca al Rancho Santa Fe y con los conglomerados de abanicos aluviales, pre-Poway, reportados en varias
localidades cerca de Lakeside y hacia el sureste en Alpine. Dos formaciones del Eoceno superior se super-
ponen inconformablemente sobre la Formacion Lusardi: 1) la Formacion Friars (del Grupo La Jolla) com-
puesta principalmente de areniscas y lutitas, y 2) el Conglomerado Stadium (del Grupo Poway). Los
conglomerados de eualquiera de las unidades del Eocene, estan moderadamente bien seleccionados y con-
sisten principalmente de bloques exoticos (de 5 a 15 cm en tamaiio) de rocas metavolcanicas, volcanoclasticas
y cuarcitas ( Powan suite).
La Formacion Lusardi rellena un canal largo y estrecho de la arrugada superficie producida despues de
la orogenesis del Cretaceo medio. Al terminar la depositacion del Lusardi y bajo la influencia de una severa
meteorizacion se desarrollo una superficie de erosion mas suave, tanto en las rocas del basemento como en la
Formacion Lusardi. Las Formaciones del Eoceno se depositaron sobre esta, profundamente meteorizada,
superficie de erosion recibiendo poco aporte local de detritos gruesos. La serie de clastos del Poway se derivo
aparentemente del este, siendo transportados a travez de la yaciente superficie de erosion y acumulados en la
parte sur-occidental de California.
INTRODUCTION
The geology of the Poway area, CaUfornia, consists of an igneous and metamorphic
basement complex overlain by about 150 to 200 meters of nearly flat lying sedimentary
rocks. On previously published geologic maps including the Poway area, all of the sedimen-
tary deposits were included under the designation "Poway Conglomerate" (Ellis and Lee,
1919; Hanna, 1926a). As implied by the name, the Poway area was regarded as the type
locality for that stratal unit. Ellis and Lee considered the "Poway Conglomerate" to be
Pliocene in age because it was composed in large part of the same type of materials present
in the coarser parts of the Pliocene San Diego Formation. Hanna (1926a; 1926b) and all
subsequent investigators recognized the "Poway Conglomerate" to be late Eocene in age on
the basis of fossils. The coarser materials in the San Diego Formation which strongly re-
semble those in the "Poway Conglomerate" are, in fact, the same; the source for much of
the San Diego Formation was the "Poway Conglomerate."
SAN DIEGO see. NAT. HIST, TRANS. 16(9): 225-236, 9 JULY 1971
226
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
RANCHO SANTA FE
LA JOLLA
0
L.
10
KM
POWAY
LAKESIDE
N
ALPINE
PT. LOMA
SAN
DIEGO
Figure \. Index map of southwestern California showing localities mentiimed in the text.
About five years ago several students practicing geologic mapping in the vicinity of
Poway reported a conglomerate that difiered significantly from typical exposures of the
"Poway Conglomerate." This anomalous conglomerate came to mind again subsequent to
detailed mapping in the Rancho Santa Fe area ( Fig. 1 ), where a thick boulder conglomerate
(now recognized as the Upper Cretaceous Lusardi Formation) that differs markedly from
the Eocene conglomerates in texture and clast content lies uncomformably beneath the
Eocene strata (Nordstrom. 1970; Peterson and Nordstrom. 1970).
Re-examination of the sedimentary rocks cropping out in the vicinity of Poway in-
dicated the presence of not one but three distinct and easily mappable formations. The first,
or lowest, rock unit is a very coarse, severely weathered conglomerate correlated with the
Upper Cretaceous Lu.sardi Formation of the Rancho Santa Fe area (Nordstrom. 1970). The
middle unit is predominantly sandstone and shale and is here regarded as an extension of
the newly recognized late Eocene Friars Formation (La Jolla Group) of Kennedy and
Moore ( 197 1 ). The upper unit is a thick and widespread cobble conglomerate and is corre-
lated with the late Eocene Stadium Conglomerate (Poway Group) of Kennedy and Moore
(1971).
The purpose of this paper is to describe and outline the principal distinctions between
the three formations, to show their distributions on a geologic map, to correlate the forma-
tions with those mapped and described in the San Diego to the west-southwest and with
formations recognized to the east-southeast in the Alpine-Lakeside area, and to indicate a
local and regional sequence of events implied by these observations and interpretations.
1971
PETERSON: STRATIGRAPHY OF POWAY AREA
227
The geology of the Poway area is depicted in figure 2. My principal interest was in the
sedimentary rocks: thus the basement complex was not subdivided. The basement rocks
consist predominantly of the mid-Cretaceous Southern California batholith (Larsen, 1948;
Bushee et ai, 1963) and a few small patches of the pre-batholithic Santiago Peak Volcanics
of Late Jurassic age (Fife et ai, 1967). An erosion surface having in excess of 300 meters of
local relief was developed on the basement complex and the sedimentary formations rest on
this irregular surface. The relief on the basement rocks exceeds the total thickness of the
younger flat-lying sedimentary deposits by over 100 meters; thus the area is characterized
by hills of relatively ancient basement rocks locally protruding through and standing well
above the younger sedimentary deposits.
Figure 2. Geologic map of the Powav area.
LUSARDI FORMATION
The Lusardi Formation crops out in a narrow, six kilometer long belt extending east-
northeast from Poway (Fig. 2). Apparently these deposits must have filled a former channel,
although the present topography is reversed, and the deposits cap a long narrow ridge with
the modern drainage incised deeply into the basement complex on either side. The ridge is
utilized as a roadway and excellent Lusardi exposures can be observed in some of the cuts
(Fig. 3).
The long exposure varies in altitude from about 550 meters near the eastern limit to
less than 200 meters at Poway. Thickness of the Lusardi ranges from about 20 to 30 meters.
The westernmost Lusardi exposure appears to fill a northeast-trending, broadly V-
shaped channel which terminates at Poway Valley. The channel probably extends farther to
the west beneath the alluvium and the Eocene formations.
The Lusardi conglomerate consists of extremely poorly sorted clasts ranging in size
from granules to boulders exceeding three meters in diameter. The clasts range in shape
from angular to spheroidal; most are at least partially rounded. The conglomerate matrix is
a poorly sorted, t^ner grained clastic assemblage dominantly derived from grus. The con-
glomerate has been severely weathered and many of the plutonic clasts have decomposed
to grus; in such cases they are difficult to distinguish from the matrix. Also, many of the
clasts are physically disintegrated in situ and yield sharp angular fragments giving the unit
228
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
the appearance of a coarse sedimentary breccia. The various clast types differ markedly in
resistance to weathering. Therefore, where an outcrop on the surface is inspected, it is domi-
nated by the more resistant clasts, whereas in a deep road cut the entire assemblage is ex-
posed.
kiT"»*»- ■-^'■^■■•■S -.T-JTt . -4 .'*■*• ^A"- '.*.«^ "**• 1
Figure 3. Outcrop of Lusardi Formation in road cut along Poway to Lakeside highway about two miles east of
Poway. Note the typical texture of this formation. Striped bar used for scale in photograph is a Jacobs staff (five
feet in length).
Some of the largest clasts are coarse-grained diorite and quartz diorite boulders
slightly more than three meters in diameter. Some of the diorite clasts are severely weath-
ered and contain finer grained resistant xenoliths, giving the appearance that the xenoliths
are "cobbles'' in a matrix of grus. Another abundant plutonic clast type is a resistant, me-
dium-grained granodiorite containing quartz phenocrysts. The plutonic rock types listed
above, together with minor amounts of aplite and vein quartz, constitute an estimated 60
per cent of the clasts in the conglomerate. All are common rock types found in the Southern
California batholith in the Poway area and the region to the east.
Other clast types include a number of fine to very fine grained, light brown, greenish
gray, and medium to dark gray metamorphosed tuff's. One of the most distinctive and com-
mon types is a medium gray weathering welded tuff containing finely crenulated flow band-
ing; where broken to reveal the inner appearance, the rock appears very fine grained, dark,
and without internal structure. Other clast types include very fine grained black hornfelsic
rocks, and minor amounts of fine-grained volcanic breccia. All of these clast types resemble
some of the more metamorphosed portions of the Santiago Peak Volcanics. The total as-
semblage of clasts found in the Lusardi is referred to as the Peninsular Ranges suite and is
characteristic of Cretaceous formations in the San Diego area (Peterson er a/.. 1968; Peter-
son, 1970a; Peterson and Nordstrom, 1970).
The Lusardi Formation was not recognized on previously published geologic maps of
the Poway area, and it was included with the basement complex (Ellis and Lee. 1919;
Hanna, 1926a). Likewise in the Rancho Santa Fe area, the type Lusardi appeared on pre-
1971 PETERSON: STRATIGRAPHY OF POWAY AREA 229
vious geologic maps labeled as everything from "weathered basement rocks" to "Quater-
nary terrace deposits." The easternmost portion of the elongate Lusardi outcrop in the
Poway area appeared on a map designated as "pre-Poway fanglomerate" (Gastil and
Bushee, 1961).
FRIARS FORMATION
In previous studies including the Poway area, all of the Eocene strata were mapped as
the "Poway Conglomerate" (Ellis and Lee, 1919; Hanna. 1926a), although Hanna clearly
recognized that his "Poway Conglomerate" could be locally subdivided into smaller stra-
tigraphic units, some of which were not conglomerate. In the Poway area, which serves as a
loosely defined type area for the "Poway Conglomerate." I have subdivided the Eocene sec-
tion and have recognized two widespread mappable formations. The lower formation con-
sists predominantly of sandstone and shale; the upper is dominated by conglomerate.
In the first general revision of the Eocene stratigraphic nomenclature since Hanna's
description of the La JoUa quadrangle, Kennedy and Moore (1971) recognized two groups
of Eocene formations: the La Jolla Group, which approximately coincides with the La Jolla
Formation as mapped by Hanna. and the Poway Group, which approximates the previous
"Poway Conglomerate." It is clear tYom the map of the San Diego area presented by Ken-
nedy and Moore (1971, figure 1), which overlaps the western margin of the geologic map
accompanying this report (Fig. 2), that they consider my lower sandstone and shale unit to
belong to the La Jolla Group and my upper conglomeratic unit to belong to the Poway
Group. In this report, I follow the stratigraphic nomenclature presented by Kennedy and
Moore and further suggest that my lower formation is equivalent to their newly defined
Friars Formation and that my upper formation is equivalent to their newly defined Sta-
dium Conglomerate.
The Friars Formation (La Jolla Group) of the Poway area lies unconformably on the
basement complex and is gradational with the overlying Stadium Conglomerate (Poway
Group). It crops out on the lower slopes of the hills along the southern margin of Poway
Valley and the area northwest of Poway (Fig. 2). The Friars Formation reaches a maximum
thickness of about 30 meters. In general, this formation is poorly exposed because of low
relief and a cover of coUuvium derived from the overlying Stadium Conglomerate. The best
exposures are in artificial cuts.
Despite the generally poor exposures, the Friars Formation is fairly easy to map. The
contact with the Stadium Conglomerate coincides closely with the 200 meter contour line
throughout the area and is marked by a slight change in slope and a change in vegetation.
The Friars Formation is generally easily separated from the basement rocks, except in a few
places where the basement rocks are severely weathered. In such places the basement rocks,
although superficially appearing very similar to the Friars Formation, contain at least a few
relict features such as joint planes, foliation, or small quartz veins by which they may be
recognized.
The Friars Formation is dominated by green to light brown, generally thickly bedded
sandstone and shale (some geologists might prefer to use the term "mudstone" for the as-
semblage). Grain sizes range from that of clay and silt to coarse sand. Typically the rocks
are neither well sorted nor well stratified. In addition, the Friars Formation locally contains
some thin beds of conglomerate and a few fairly sizeable lenses of conglomerate character-
ized by the Poway suite of clasts (more fully noted in the following section).
No fossils were found in any of the Friars outcrops examined in the Poway area. Ken-
nedy and Moore (1971) reported the age of the formation to be middle and late Eocene, as
based on fossil evidence and stratigraphic position at the type section of the formation
230
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
along the northern margin of Mission Valley.
STADIUM CONGLOMERATE
The Stadium Conglomerate (Poway Group) is the thickest, most extensive, and best
exposed stratigraphic unit in the Poway area. Along the southern side of Poway Valley, it is
up to 150 meters thick. In the area to the northwest of Poway, it is up to about 50 meters
thick. An excellent section of the Stadium Conglomerate is well exposed along Pomerado
Road in the southwest corner of the map area (Fig. 2).
Figure 4. Outcrop i>f Stadium Conglomerate along Pomerado Road about one-half mile oil' the southwestern
corner of the map (Fig. 2). Note the l\pical texture of this conglomerate and the presence of sandstone lenses.
Scale is .lacobs staff.
Although the Stadium Conglomerate is dominated by conglomerate, as implied by the
name, beds and lenses of sandstone and shale are locally evident, especially in the lower
portion. The finer grained sediments in the lower part closely resemble those in the Friars
Formation, whereas higher in the section the beds and lenses of sandstone and shale and
the matrix of the conglomerate are cleaner, better sorted, and white to light brown. Many of
the prominent sandstone lenses within the conglomerate probably represent sand bars de-
veloped in a river .system.
1971 PETERSON: STRATIGRAPHY OF POWAY AREA 231
The Stadium Conglomerate is typically fairly well sorted, at least relative to the Lu-
sardi conglomerate (compare Fig. 3 and 4). Clast sizes range from granules to boulders 60
centimeters in diameter, but clasts over 30 centimeters in diameter are rare. Typically, the
clasts are subrounded to rounded cobbles and small boulders. The texture of the Stadium
Conglomerate and of other Eocene conglomerates in the San Diego area differs so
markedly from that of the Lusardi that the two may be readily recognized on that basis
alone (Peterson. 1970a).
The types of clasts found in the Stadium Conglomerate and in other Eocene and post-
Eocene formations in the San Diego area are highly distinctive and easily recognized. The
clasts consist predominantly of slightly metamorphosed rhyolitic to dacitic volcanic and
volcanoclastic rocks, with a smaller but significant amount of quartzite. The various clast
types (together referred to as the Poway suite of clasts) and their proportions are well de-
scribed by Bellemin and Merriam ( 1958), De Lisle el al. (1965), and Woodford et al. ( 1968).
No fossils were found in the Stadium Conglomerate of the Poway area. However, in
other parts of the San Diego area, the Poway Group including the Stadium Conglomerate
has yielded a variety of marine and non-marine late Eocene fossils (Hanna, 1926b: Dusen-
bury, 1932; Cushman and Dusenbury, 1934; Stock, 1937, 1938; Kennedy and Moore.
1971). Evidently the Stadium Conglomerate was deposited very near sea level and repre-
sents a variety of fluvial, estuarine. and nearshore-marine depositional environments.
RELATIONSHIP BETWEEN CRETACEOUS AND EOCENE FORMATIONS
The westernmost outcrop of the Cretaceous Lusardi Formation is capped by a thin
patch of the Stadium Conglomerate. This is the only locality withm the map area where the
Eocene rocks are in contact with the Lusardi Formation. The Stadium Conglomerate at this
locality consists of a thin cobble conglomerate (approximately 7 meters thick), capping the
crest of the hill, underlain by a thin (about 7 meters) unit of greenish sandstone and shale.
Underlving this latter unit is the bouldery Lusardi Formation. Exposures of the three units
are poor except for several road cuts. The conglomerate cap is identical in all respects to the
Stadium Conglomerate as mapped throughout the area (figure 2). The sandstone and shale
unit is lithically identical to the Friars Formation and to the lenses of sandstone and shale in
the lower portion of the Stadium Conglomerate.
The Lusardi Formation of the Poway area is interpreted as a deposit formed by a very
fast flowine. turbulent river. The character of the deposits susaests that the Lusardi filled a
long narrow, fairly steep-walled canyon, although this topography is no longer evident. The
Lusardi channel (or canyon) fill extended from the northeast into the Poway area, and
probably continued far to the west of Poway.
After deposition of the Lusardi conglomerate, the area underwent erosion. A much
wider, north to south sloping valley was cut across the Poway area. During, or following,
this erosional episode the terrane (consisting of the Lusardi Formation and the basement
rocks) underwent severe weathering. When deposition began again, the Eocene formations
were deposited on the deeply weathered surface, filling in the low areas first and then lap-
ping over onto the adjacent highlands.
The initial deposits, the Friars Formation, are both mineralogically and texturally im-
mature. In gross character, the coarser grained sandstones of this formation strongly re-
semble grus. The finer grained portions were not studied in detail, but casual observation
suggests that they are dominated by weathered products derived from the deeply weath-
ered basement complex and probably from the Lusardi Formation as well. Much of the
finer sediments of the Friars Formation and the lower portion of the Stadium Con-
glomerate were evidently derived locally. The conglomerates with the exotic Poway suite of
232
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
clasts, however, had to be transported into the area from a considerable distance.
REGIONAL IMPLICATIONS
The sequence of post-bathohthic rock units in the Rancho Santa Fe area (Fig. 5) in-
cludes the Lusardi Formation, which is unconformably overlain by the Eocene La Jolla
Group. The unconformity between the two units can be demonstrated to have about 130
meters of relief, although both rock units are essentially flat lying ( Peterson and Nordstrom,
1970). The identical sequence of rock units, unconformable relationship between them, and
the implied sequence of events is evident in the Poway area except that the Eocene rock
units are far more conglomeratic and appear to be predominantly non-marine.
POway TERRACE
BALLENO GRAVELS
'V';.' '■"-'----/: cdvJ"''*^''
,£BN
ILPINE- LAKESIDE
LA JOLLA- PT LOMA
Figure 5. Correlation diagram emphasizing relationships between Cretaceous and Eocene rock units in the
northern and eastern part o^ the San Diego area. Thickness of rock units is not to scale, but is roughly
proportional.
East and southeast of the Poway area, a number of small and widely scattered outcrops
of "pre-Poway fanglomerate deposits" were reported and briefly described in a field trip
guidebook (Gastil, 1961; Gastil and Bushee. 1961). Most of the outcrops are within 15 ki-
lometers of the communities of Alpine and Lakeside (Fig. 1). I revisited several of the local-
ities for comparative purposes. In gross character, these deposits strikingly resemble the
Lusardi Formation of the Poway area and the type Lusardi near Rancho Santa Fe. The
deposits are very poorly sorted and contain a wide variety of clasts derived from the local
basement complex (Peninsular Ranges suite). Some of the clasts are exceptionally large (di-
ameter greater than three meters) and many clasts are deeply weathered. Locally the con-
glomerate has a reddish matrix; elsewhere the matrix is light brown, green, or gray. Clast
types characteristic of the Eocene formations (Poway suite) are absent.
Where field relations are evident (see especially the southwest and central portions of
geologic map number 2 of Gastil and Bushee. 1961), the "pre-Poway" conglomerate ap-
pears to be deeply channeled into the basement complex and is overlain by the "Poway
Conglomerate" (or the equivalent "Ballena Gravels"). On the basis of stratigraphic posi-
tion and lithic similarity, I regard all the outcrops of "pre-Poway fanglomerates" reported
1971 PETERSON: STRATIGRAPHY OF POWAY AREA 233
by Gastil and Bushee (1961) as further exposures of the Late Cretaceous Lusardi Forma-
tion.
No trace of fossils could be found in any of the Lusardi outcrops in any of the areas
discussed, nor does it appear likely that any will be found. The extremely coarse texture and
the deeply weathered nature of these deposits provides a very unfavorable environment for
preservation of fossils. In spite of the absence of fossils, a reasonable argument can be pre-
sented to indicate that the Lusardi is Late Cretaceous in age. Correlation of the units from
area to area on a physical basis is shown in Figure 5. When the Lusardi is traced to the west,
it extends beneath the Point Loma Formation in the Carlsbad area and in the subsurface of
the La JoUa-Point Loma area where it constitutes the lowest formation in the Late Cre-
taceous (Campanian and Maestrichtian) Rosario Group (Kennedy and Moore, 1971). In
addition, the clast content of the Lusardi is similar, but not identical, to conglomerates
found in the upper part of the Rosario Group (Cabrillo Formation) and is unlike other sedi-
mentary deposits (Eocene. Pliocene, Pleistocene, or Recent) of the San Diego area (Peter-
son, 1970a).
The Lusardi Formation is apparently of much greater extent than was previously rec-
ognized (Nordstrom, 1970; Peterson and Nordstrom, 1970). It would not be surprising to
find still further outcrops now that it has been recognized as a separate, distinct, and wide-
spread stratigraphic unit. Evidently, it was deposited over a large part of the San Diego
region following the emplacement of the Southern California batholith and the subsequent
uplift necessary to expose those deep-seated rocks. During Late Cretaceous time the San
Diego region was probably topographically very rugged and undergoing rapid erosion. The
high-relief topography was partially filled with debris derived from the batholithic and pre-
batholithic rocks and representing a very high energy depositional environment. Very
coarse stream deposits, alluvial fan deposits, and mudflow deposits were spread over the
area to unknown but highly variable depths. In the western part of the San Diego area, the
marine Point Loma and Cabrillo Formations were deposited over the Lusardi con-
glomerates (Fig. 5).
Following the Late Cretaceous depositional episode, the region underwent uplift, pos-
sibly slight deformation, and a widespread erosion surface of low to moderate relief was
produced across the Cretaceous sedimentary rocks and the basement rocks. This surface
was referred to as the "sub-La Jolla unconformity" in the coastal portion of the San Diego
area (Peterson and Nordstrom, 1970) and the "old erosion surface" to the east in the region
around Alpine and Lakeside (Gastil, 1961; Gastil and Bushee, 1961;Minch, 1970). During
this supra-baselevel episode, most of the Lusardi deposits were dissected and erosionally
removed, particularly in their eastern extent. The remaining remnants were left in low lying
areas (such as near Rancho Santa Fe), in buried canyons (such as at Poway), in areas that
were probably distant from major drainages during the time of erosion, or in areas where
the formation was covered by the Point Loma and Cabrillo Formations.
The "old erosion surface" had from several hundred to several thousand feet of local
relief from place to place. During or after the development of the erosion surface, the region
apparently underwent an episode of deep and severe weathering. The results of this weath-
ering are evident in all Lusardi outcrops. A similar observation was recorded from an area
several kilometers south of Tijuana, Baja California. Flynn ( 1970: 1793) described the pres-
ence of a widespread deeply weathered zone (paleosol) developed on the Cretaceous Re-
donda (probably equals Lusardi) and Rosario Formations and on the basement rocks. The
soil ranged up to nearly 15 meters thick and was overlain by the Eocene Delicias and
Buenos Aires Formations (equivalents to the La Jolla Group of the San Diego area).
The Eocene sedimentary rocks of the San Diego area were deposited on "the old ero-
234 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
sion surface" without totally destroying the relief. Thus the unconformity beneath the Eo-
cene rocks has several hundred to several thousand feet of relief, and hills composed of pre-
Eocene rocks locally protrude through the Eocene cover (Hanna, 1926a; Peterson and
Nordstrom. 1970).
With the beginning of Eocene deposition, the Poway suite of clasts first arrived in the
San Diego area. The clasts were evidently transported a considerable distance, at least 100
kilometers and probably more, for there is no known local source area. Potential distant
source areas from the Mojave Desert to Sonora, Mexico have been proposed (De Lisle et ai,
1965; Merriam, 1968; Woodford e/ a/., 1968; Minch, personal communication, 1970). The
intent here is not to resolve the Poway clast provenance problem. It is sufficient to say that
these exotic clasts did not arrive in the San Diego area until mid-Eocene time and thus they
have a chronologic significance. Regardless of where they came from, they are abundant in
all Eocene conglomerates of the San Diego area and are absent in pre-Eocene units.
In marked contrast to the Cretaceous conglomerates, the Eocene conglomerates con-
tain a very low proportion of clast types that might have been derived from the local base-
ment rocks. Outcrops containing more than 10 per cent clasts resembling the pre-
batholithic and batholithic rocks are rare. It should be pointed out that, although some
clasts resemble the local basement rocks, they too may have been transported into the San
Diego area along with the exotic Poway clasts (Minch, 1970). The paucity of locally derived
coarse detritus suggests that the San Diego region was relatively low lying during Eocene
deposition. Some of the fine detritus found in the Eocene formations, such as that of the
Friars Formation and the lower portion of the Stadium Conglomerate near Poway, was
probably locally derived and contributed by minor streams. The Poway suite of clasts, how-
ever, had to be transported across the low lying ''old erosion surface" to the site of accumu-
lation in the San Diego area.
Following Eocene deposition, the San Diego area has undergone several uplifts in-
cluding thecuttingof three widespread coastal terraces (Ellis and Lee, 1919; Hanna, 1926a;
Peterson, 1970b). This post-Eocene uplift and erosion has resulted in extensive dissection of
the Eocene deposits, locally revealing the underlying fragments of the Lusardi Formation.
ACKNOWLEDGMENTS
I would like to thank W. J. Elliott, A. H. James, and R. T. La Borde for pointing out the seemingly anomalous
conglomerate in the Poway area. In addition, I would like to thank J. A. Minch, C. E. Nordstrom, M. P. Kennedy
and R. G. Gastil tor many stimulating discussions concerning the conglomerates of the San Diego area. Thanks are
due to Armando Estrada for translation of the abstract into Spanish.
LITERATURE CITED
Bellemin, G. J., and R. H. Merriam
1958. Petrology and origin of the Poway Conglomerate, San Diego County, California. Geol. Soc. Amer.
Bull. 69: 199-220.
Bushee, J., J. Hoiden. B. Geyer, and G. Gastil
1963. Lead-alpha dates for some basement rocks of southwestern California. Geol. Soc. Amer. Bull. 74:803-
806.
Cushman, J. A., and A. N. Dusenbury, Jr.
1934. Eocene foraminifera of the Poway Conglomerate of California. Contr. Cushman Lab. Foram. Res.
10:51-65.
De Lisle, M., J. R. Morgan, J. Heldenbrand, and G. Gastil
1965. Lead-alpha ages and possible sources of metavolcanic rock clasts in the Poway Conglomerate, south-
west California. Geol. Soc. Amer. Bull. 76; 1069- 1074.
1971 PETERSON: STRATIGRAPHY OF POWAY AREA 235
Dusenbury, A. N., Jr.
1932. A faunule from the Poway Conglomerate, upper middle Eocene of San Diego County, California. Mi-
cropaleontology 3:84-95.
Ellis. A. J., and C. H. Lee
1919. Geology and ground waters of the western part of San Diego County, California. U.S. Geol. Survey
Water-Supply Paper 446: 1-321.
Fife, D. L., J. A. Minch, and P. J. Crampton
1967. Late Jurassic age of the Santiago Peak Volcanics, California. Geol. Soc. Amer. Bull. 78:299-304.
Flynn. C. J.
1970. Post-batholithic geology of the La Gioria-Presa Rodriguez area, Baja California, Mexico. Geol. Soc.
Amer. Bull. 81:1789-1806.
Gastil. G.
1961. The elevated erosion surfaces. In, Field trip guidebook. San Diego County. Geol. Soc. Amer.
(Cordilleran Section) 57th Ann. Mtg., 1-4.
Gastil, G., and J. Bushee
1961. Geology and geomorphology of eastern San Diego County. In, Field trip guidebook. San Diego
Countv. Geol. Soc. Amer. (Cordilleran Section) 57th Ann. Mtg.. 8-22.
Fianna, M. A.
1926a. Geology of the La Jolla quadrangle, California. Univ. California Publ. Geol. Sci. 16: 187-246.
1926b. An Eocene invertebrate fauna from the La Jolla quadrangle. California. Univ. California Publ. Geol.
Sci. 16:247-398.
Kennedy. M. P., and G. W. Moore
1971. Stratigraphic relations of Upper Cretaceous and Eocene formations, San Diego Coastal area, Califor-
nia. Amer. Assoc. Petrol. Geol. Bull. 55:709-722.
Larsen, E. S., Jr.
1948. Batholith and associated rocks of Corona. Elsinore. and San Luis Rey quadrangles, southern Califor-
nia. Geol. Soc. Amer. Mem. 29: 1 - 1 82.
Merriam, R. H.
1968. Geologic reconnaissance of northwest Sonora. In, Proceedings of conference on geologic problems of
San Andreas fault system. Stanford Univ. Publ. Geol. Sci. 1 1 :287.
Minch, J. A.
1970. Early Tertiarv paleogeographv of a portion of the northern Peninsular Range. //;, Pacific slope
geology of northern Baja California and adjacent Aha California. Amer. Assoc. Petrol. Geol. (Pacific
Section) Fall Field Trip Guidebook, 4-9.
Nordstrom, C. E.
1970. Lusardi Formation: a post-batholithic Cretaceous conglomerate north of San Diego, California. Geol.
Soc. Amer. Bull. 81:601-606.
Peterson, G. L.
1970a. Distinctions between Cretaceous and Eocene conglomerates in the San Diego area, southwestern Cal-
ifornia. In, Pacific slope geology of northern Baja California and adjacent Alta California. Amer.
Assoc. Petrol. Geol. (Pacific Section) Fall Field Trip Guidebook. 90-98."
1970b. Quaternary deformation of the San Diego area, southwestern California. ///, Pacific slope geology of
northern Baja California and adjacent Alta California. Amer. Assoc. Petrol. Geol. (Pacific Section)
Fall Field Trip Guidebook, 120-126.
Peterson, G. L., R. G. Gastil, J. A. Minch, and C. E. Nordstrom
1968. Clast suites in the late Mesozoic-Cenozoic succession of the western Peninsular Ranges province,
southwestern California and northwestern Baja California (abst.). Geol. Soc. Amer. Spec. Paper
115:177.
Peterson. G. L.. and C. E. Nordstrom
1970. Sub- La Jolla unconformity in vicinity of San Diego, California. Amer. Assoc. Petrol. Geol. Bull.
54:265-274.
Stock. C.
1937. An Eocene titanothere from San Diego County. California, with remarks on the age of the Poway
Conglomerate. Natl. Acad. Sci. Proc. 23:48-53.
1938. A tarsiid primate and a mi.xodectid from the Poway Eocene, California. Natl. Acad. Sci. Proc 24288-
293.
236 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
Woodford. A. O.. E. E. Welday, and R. H. Merriam
1968. Siliceous tuff clasts in the upper Paleogene of southern California. Geol. Soc. Amer. Bull. 79:1461-
1486.
Department of Geology, San Diego State College, San Diego, California 92115.
:3 - IN-H^ O;L0oP U'^OJ
<^c
MUS, COMP. ZOOL.
LIBRARY
SEP 10197]
I^ARVARD
UNfViftSlTY
HERPETOFAUNA OF THE PACIFIC COAST
OF NORTH CENTRAL BAJA CALIFORNIA, MEXICO,
WITH A DESCRIPTION OF A NEW SUBSPECIES
OF PHYLLODACTYLUS XANTI
DENNIS L BOSTIC
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 10 25 AUGUST 1971
HERPETOFAUNA OF THE PACIFIC COAST
OF NORTH CENTRAL BAJA CALIFORNIA, MEXICO,
WITH A DESCRIPTION OF A NEW SUBSPECIES
OF PHYLLODACTYLUS XANTI
DENNIS L. BOSTIC
ABSTRACT.— Three species of lizards are recorded from the Pacific slope of Baja California, Mexico, for
the first time: a leaf-toed gecko, Phvllodactylus xanti sloani n. subsp., Crotaphvtus collaris and Sauromalus
australis. The distribution, pattern, and scutellation of Gerrhonotus multicarinatus spp. indicates a south-
ward expansion of its range since glacial maxima via the cool, moist, coastal corridor, and its possible in-
tegradation with G. paucicarinatus. A southward coastal corridor diff"usion may also be true for Tantilla
planiceps eiseni, Coleon vx variegalus abbot ti and Lichanura roseofusca gracia. Of the 29 species of amphibi-
ans (2), lizards ( 16), and snakes (11) collected, only one lizard, Cnemidophorous labialis, is considered en-
demic to the Central Desert of Baja California. Homogeneity of habitats, the moderate climate and the extir-
pation of the Peninsular desert herpetofauna during glacial maxima probably have been important factors
in reducing or limiting species diversity and endemism.
INTRODUCTION
The coastal deserts of North America, of which more than 2000 miles are confined to
Baja California and Sonora, Mexico, remain biologically unknown because of their relative
inaccessibility, lack of potable water, and rugged terrain. Wiggins (1960a) identified these
and other regions in Baja California as in need of more careful biological exploration. One
area he mentioned was the Pacific coastal region between El Rosario (30°N) and the south-
ern boundary of the state of Baja California (28°N; Fig. 1). Excluding the immediate areas
of El Rosario and Rosarito, where the main road approaches within ten miles of the ocean,
this region has not been explored herpetologically.
In the spring and summer of 1969, 1 made several trips (Table 1) into the area in order
to: 1) better ascertain the distributional limits of the herpetofauna, 2) gather ecological
data; and 3) collect specimens for studies of geographical variation and evolution.
METHODS AND MATERIALS
A Taylor sling psychrometer and a Dwyer wind gauge (0-60 mph) were used to meas-
ure relative humidity and wind speed. A Taylor soil thermometer (0-50°C) and a multi-
channel tele-thermometer unit were used to record soil (approximately 0.5cm beneath
surface) and air temperatures (approximately 0.5cm above surface). Time (Standard) is ex-
pressed in 24 hour fashion. Throughout this paper when counts or measurements are pre-
sented in the following manner: 1 1 ±1.3(10-12)18, the first figure refers to the arithmetic
mean, the second figure to the standard error of the mean, the figures in parentheses to the
range, and the last figure to the number of observations. Occasionally, the standard error of
the mean is omitted, but the order, with this exception, remains the same. Standard devia-
tion is indicated by S.D.
All snout-vent measurements have been rounded off to the nearest whole number and
other measurements to the nearest tenth.
Within each major systematic grouping the species are arranged alphabetically by
genus. I have not been consistent in the treatment of subspecies and have omitted available
trinomials where geographical variation is poorly known. All material collected has been
deposited in the collections of the San Diego Society of Natural History.
SAN DIEGO see. NAT. HIST., TRANS. 16 ( 10): 237-264, 25 AUGUST I97I
238
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
J JEL ROSARIO iis° \ \
'"^y^jTX / //
^f** ,.- SorFernondooL- ^T^-^-' / / .' \l
'-
hir^^Sf--'-" \V^/ ^;?-^ Vk
o
^^
^ 1
VI '^MESA."DE / / \ /
P---^
^^^,san;<^carlos / \ J
\ ^^
o
Puerto de ^\'--,v^-";.:;;w<. ''-- cfSofito /cotarina V
V
""~^"~"^^^ //^o
San Carlos V--'? '<: ^/ /
\ f o o»Vw /
V''--o>-'/ 1 /
Puerto de \/^ / /
Sonto CatorinoV / /
\ 1 toBohio Sor
Luis Gonzaga
. \ >,>J\
CL
\
\
%
/ V^- \^ J
/ \\'""" Vllaguna Chopala
/
.•
^9° / VVV_-o'-^''^ ^^ Y 1
/
) \
">---.. ^ y Los Polomas 1 /
\ iP 1
% ^^\-^ (A
U(^
/ iv' Ca''doil-~^.y y^^-'^^h^^
) '^^'^
^ k\
PuntQ Prieta -.^
/
>
\\/
2Tk^^^^ I jK. /i 1
1 ^1
^/r^ S^/ /
""~~~~~~--^ \^L
A^ C^- -'-'7--- 1
/
/''*s^\r^^-\y
/ Bahio Sonta Rosolillitr^) *-oy<lso"W:::i^
\
' JBotiio de ^^--^
O //^ i!£f-i / V /
s
C Los Angeles
<j^ (resort; ,,^„
Son
Regis
? /
mill 1 1 1 rniln Millers Landing 1 y J
1 /
5 10 20 30 Iff ^/
1 /
1 1 1 1 1 kilometers
J\\
I /
5 10 20 40
SCALE .0 y 1 \
y. ^J^ / /Kjesu^M 0 r io
\ /
{ /
lA
/
V^
^~^'X^ ^^^^^^>^^3\b
/ \~i^
\r
/^xT~/
I r< X^ Guerrero Negro ^^^^^i^^^^ ^-W
''""/ f
V ^^--^ / //V°'^^?%s^ '
k-^^
/ \
A (.xWVl^
^"'^ - El Arc
o)
<s —
/ /
\ ^\uA r
^~'— '
\ \ \^TS \
1 ^^^"^^^^^
/ I
Figure 1. Map of the central region of Baja California, Mexico. Modified from Gerhard and Gulick, 1966.
1971
BOSTIC: BAJA CALIFORNIA HERPETOFAUNA
239
DESCRIPTION OF THE AREA
GENERAL GEOGRAPHY
The Pacific coast of Baja California between 30°N and 28°N is notably irregular, with
many small embayments. The coastal strand is hilly to mountainous, and is frequently in-
terrupted with valleys, coastal plains and marine terraces. The only well-defined mountain
range is the Sierra Colombia, with summits near 762 meters.
The area lies within the North American Desert Province (Shreve, 1942), but thus far
there has been no agreement as to a name for that part of the province in the middle of the
peninsula roughly delimited in the north by the southern tip of the Sierra de San Pedro
Table 1. Herpetological Collecting Stations in the Central Desert of Baja California del Norte, Mexico.
Station Date
Locality
1
30 March
6.6 miles
2
2-4 August
3
3 April
11.8 miles
4
3 April
19 miles
5
30-31 March
23.5 miles
6
1,3 April
24.5 miles
7
2 April
25 miles
8
3 April
9
1-2 April
1.8 miles
10
26-27 August
10.9 miles
11
26 August
10.3 miles
12
26 August
6.7 miles
13
26 August
4.3 miles
14
26 August
1.0 miles
15
24 June
1.7 miles
16
26 August
2.9 miles
17
26 August
3.6 miles
18
26 August
4.4 miles
19
26 August
6.5 miles
20
24-25 August
21
24-25 August
3.2 miles
22
25 August
0.5 miles
23
25 August
24
24-25 June
25
25 June
5.4 miles
26
26-29 June
27
30 June
15.5 miles
28
30 June
16.6 miles
29
30 June
14.4 miles
30
30 June
10.6 miles
31
1-5 July
32
5 July
16.6 miles
33
5-8 July
34
8 July
11.5 miles
35
8 July
9.7 miles
36
9-12 July
37
12 July
10.8 miles
38
12-15 July
0.5 miles
39
15 July
40
15 July
4.4 miles
41
16-19 July
10 miles
SE EI Rosario; 30°0rN, 1 15°38'W
Punta Baja; 29°58'N, 115°49'W
SE El Rosario; 29°58'N, 1 1 5°33'W
SE El Rosario; (RanchoSan Vicentito); 29°52'N, 115°33'W
SE El Rosario; 29°48'N, 1 1 5°33'W
SSE EI Rosario (Arroyo de San Fernando) ; 29°47'N, 1 15°33'W
SSE EI Rosario (Arroyo de San Fernando) ; 29°47'N, 1 15°33'W
San Felipe Springs (in Arroyo de San Fernando) ca. 9 miles NE
of the arroyo-coastal road junction; 29°52'N, 1 15°26'W
NW Puerto de San Carlos; 29°40'N, 1 15°29'W
NE Santa Catarina Ranch; 29°53'N, 1 15°04'W
NE Santa Catarina Ranch; 29°52'N, 1 15°04'W
NE Santa Catarina Ranch; 29°49'N, 1 15°05'W
NE Santa Catarina Ranch; 29°47'N, 1 15°06'W
NE Santa Catarina Ranch; 29°44TSr, 1 15°08'W
S Santa Catarina Ranch; 29°43'N, 1 15°08'W
S Santa Catarina Ranch; 29°41'N, 1 15°09'W
SW Santa Catarina Ranch; 29°40T^, 1 15°09'W
SW Santa Catarina Ranch; 29°40'N, 1 15°09'W
SW Santa Catarina Ranch; 29°39'N, 1 15°1 1'W
Mesa de San Carlos (SE); 29°38'N, 115°15'W
NE Puerto de Santa Catarina; 29=35^^1, 1 15°14'W
NE Puerto de Santa Catarina; 29°33'N, 115°16'W
Puerto de Santa Catarina; 29°32'N, 1 15°16'W
Punta Canoas; 29°26'N, 115°11'W
NE Punta Canoas; 29°26'N, 115°06'W
Arroyo San Jose; 29° 19'N, 115°51'W
S Arroyo de San Jose; 29°09'N, 1 14°42'W
NW Las Palomas; 29°14'N, 114°46'W
NW Las Palomas; 29°13'N, 114=46^
NW Las Palomas; 29°09'N, 114°40'W
Las Palomas; 29°08'N, 114°33'W
SE Las Palomas; 28°57'N, 114°29'W
EICardon;28°56'N, 114°29'W
SE El Cardon; 28°50'N, 114°28'W
S Punta Prieta; 28°49'N, 1 14°10'W
Arroyo Santo Dominguito (2.8 miles NE Santa Rosalillita);
28°42'N, 114°15'W
SE Santa Rosalillita; 28°37'N, 1 14°12'W
N San Javier; 28°32'N, 1 14°05'W
Miller's Landing; 28°28'N. 114°05'W
S Miller's Landing; 28°25'N, 1 14°04'W
S Jesus Maria; 28°13'N, 1 14°02'W
240
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Martir, in the east by the peninsular divide, and in the south by the northern and north-
western borders of the great lava plateau, but with a Pacific strip extending further south-
ward to the vicinity of Comondu. This semi-arid region was first named the Vizcaino Desert
District by Nelson (1921) who made a biological survey of the peninsula in 1905-1906.
Since then it has been called the San Borja Desert (Sauer and Meigs, 1927), the Vizcaino-
Magdalena Desert (Jaeger, 1957), the Central Desert (Aschmann, 1959), the Peninsular
Desert (Savage, 1960), and the Vizcaino Region (Shreve and Wiggins, 1964).
Sauer and Meigs' (1927) "San Borja Desert," based on a socio-economic division of the
mission era, implies too restricted a geographical area, and Savage's (1960) "Peninsular
Desert," is too inclusive. Jaeger's (1957) "Vizcaino-Magdalena Desert" is misleading. The
Magdalena Plain, farther south on the Pacific drainage of the peninsula differs floristically
from the Vizcaino Region (Shreve and Wiggins, 1964). I have chosen to call the area the
"Central Desert" as suggested by Aschmann ( 1959) because it seems desirable to restrict the
Vizcaino Desert, in accordance with local practice, to the dry coastal plain west of San Ig-
nacio.
CLIMATE
Until 1963 few chmatological data were available for Baja California, most of which
were qualitative. Important additions to these data were presented by Hastings and Turner
(1964, 1965a) and Hastings and Humphrey (1969).
Climatologically, the Central Desert may be classified on the origin of its climate (caus-
ally) and on the nature of its temperature (thermally), particularly in the winter. Causally, it
is a cool coastal phase of a subtropical desert, the Sonoran; thermally, it may be classified as
temperate.
The survey area is included within Meigs' (1966) "fog type" of temperate desert, and in
the system of notation used in the UNESCO arid homoclimatic maps (Meigs, 1953) would
be classified as Ac23— a desert climate with winter precipitation, the coldest month being
between 10°C and 20°C (50°-68°F) mean temperature, and the warmest month between
20°-30°C (68°-86°F) mean temperature.
Table 2. Irregular observations (n) of wind velocity and direction.
Wind Velocity (mph)
Direction
0700
1100
1500
0700
1100
1500
June
1.4-5.2
3.2-7.2
4.8-9.8
WNW(2)
WNW(5)
WNW(5)
(5)
(5)
(5)
ENE(3)
July
0.6-3.2
3.0-8.2
3.7-10.4
W(5)
W(14)
W(12)
(15)
(18)
(18)
NW(2)
WSW( 1 )
NW(2)
SW(2)
August
0.0-2.2
1.9-9.8
1.3-8.6
W(l)
W(3)
W(2)
(5)
(8)
(7)
SW(2)
SW(47
SW(2)
S(2)
The climate of the western coastal fringe of the Central Desert is greatly influenced by
the cold California Current of the Pacific Ocean and the prevailing westerly winds (Table 2)
which move layers of cool, moist air inland beneath dry descending air, producing consid-
erable fog and cloudiness, but no precipitation, and very mild conditions.
In our survey, onshore movement of the moist marine air, often as fog or low clouds.
1971
BOSTIC: BAJA CALIFORNIA HERPETOFAUNA
241
72.8(62-78)5
64.8(58-76)9
65.5(46-85)6
53.4(41-65)5
48.8(26-62)5
cr 45H
40
44.0(22-62)19
41.4(29-65)7
49.0(27-66)19 \ 49.3(31-66)3
1 1 1 — I 1 1 I
0700 MOO I500| 0700 1100 1500
JUNE JULY AUGUST
Figure 2. Relative humidity recorded in the Pacific coastal strand of the Central Desert of Baja California del
Norte, Mexico, during June, July and August, 1969. The first figure refers to the arithmetic mean, the figures in
parentheses to the ranges, and the last figure to the number of measurements.
generally began in mid-afternoon when wind velocities were greatest (Table 2). The relative
humidity increased in mid-afternoon and dropped substantially in mid-morning when the
fog and cloud cover dissipated (Fig. 2). The frequency and extent of the fog or cloud cover
diminished rapidly with distance from the ocean. Although Arnold (1957) reported
frequent fogs in the Chapala Basin, about 25 miles from the Pacific, during the spring and
summer of 1949 and 1950, fog was seldom observed during this survey more than five miles
from the ocean.
By late evening, along the coastal strand, visible drops of condensation formed on
those objects that had cooled most rapidly after sunset, and by early morning substantial
amounts of water, often 100 ml. or more, were present frequently in the depressions of rocks
and in the axils of the basal leaves of Agave. Similarly, in the sandy soil beneath woody
shrubs the extent of the plant drip was noticeable and, as recorded by Wiggins (1969) for
shrubs of the Vizcaino Desert, often the subsoil was dampened to a depth of 4-5mm. Dr.
James R. Hastings (pers. comm.) noted that on foggy mornings in the Vizcaino Region the
ground was visibly more moist under Opuntia cholla and Machaerocereus gummosus than
in open spaces, and that rivulets of condensate were observed running down the upper, con-
cave surface of the leaves of Yucca valida dind Agave sp., being funneled toward the caudex.
Hastings and Turner (1965b) suggest that some plants of the Vizcaino Region may utilize
the fog drip as a major source of water. Certainly the common epiphte Tillandsia recurvata,
which grows on woody shrubs and succulents, is dependent upon dew, as are many of the
lichens of the coast such as Ramalina reticulata. Distribution of these moisture-dependent
242
SAN DIEGO SOCIETY OF NATURAL HISTORY
Vol. 16
34.9(28.3-42.0)5
^33.3(30'.0-40.0)5,
41.5(30.6-50.4)19
40.3(30.0-48.8)8
39.2(25.8-50.0)19
37.1/26.6-48.8)7
33^(25.0-44.4)8
30.1(21.6-37.0)19
3y3.:^t23.3-42.2)7
26.8(2_3.3 31.6)5/273^22.4-33.9)19 , ^
26'nr23.3-3 1.4)57 / '^6.0(19.4-35.0) 7
24^.8(20.0-32.2)19 43,,(,9.4-32.2)7
•^21.9(18.6-27.7)19
9.8(17.2-21.1)5
-OPEN
-SHADE
0700 1 100 ISOOI 0700 MOO 1 500 1 0700 MOO ^00
JUNE
JULY
AUGUST
Figure 3. Soil temperatures recorded at various collecting stations in the Central Desert of Baja California,
Mexico, during June, July and August. 1969. See Fig. 2 for explanation of figures.
plants may serve to delimit the coastal area of the Central Desert.
As noted by McGinnies et al. ( 1968), evaporation retards heating of the soil and vege-
tation, and may eliminate or reduce excessive heat loads, or it may keep the plant tempera-
ture below that required for optimum growth. This factor, concomitant with the
temperature stabilizing effect of the ocean itself and the prevalence of strong, prevailing
onshore winds, may be important in maintaining a distinction between the east and west
coast floras.
The Pacitic coast of Baja California as far south as Bahi'a Magdalena, with a mean Jan-
uary temperature above 18°C (64.4°F) and a subtropical climate, receives its maximum
precipitation in winter (December-February), with the Central Desert receiving a winter
average of 56mm (Hastings and Turner, 1965a). Winter storms generally cover a large area,
are relatively gentle and may persist for days. But only in that area north of the Central
Desert, the approximate southern limit of the Sierra San Pedro Martir, do surface and
ground water occur with any regularity. Near the coastal strand, the only surface waters
encountered that were readily accessible to wildlife were the springs of San Felipe (Fig. 5)
and Las Palomas. a small stream in San Javier Arroyo (Fig. 6), and numerous "tinajas" or
tanks in the gulches and small canyons of the foothills.
Precipitation in fall (September-November), spring (March-May) and summer (June-
August), in that order, is progressively less. In summer, when relative humidity tends to be
low, rainfall is limited to thunder storms which are localized, relatively intense and of short
duration. Rainfall from such storms was experienced on 24 June and 24 August.
1971
BOSTIC: BAJA CALIFORNIA HERPETOFAUNA
243
45-
40-
^35H
CL
LlI
30-
<
25-
<
UJ
36.7(29.0-45.0)8
34.2(26.0-42.0)19 34.0(2^:^2.0)7
32.4(2B.p-38.5)8
30.6(27.0-34.0)6 /, . ^ .
/3I. 5(25.0-38.4)19 , , ,
, , , , /28.0(23.8-36.0)l9/?^-2(22.5-40.0)7
28.6(27.4-31.0)5/ .- //
25.0(23.3-27.2)5/27:8(22.0-34.0)19/^,52(19.0-38.0)7
^2/3(22.8-28.4)5 /a^.8(l8.4-29.0)l9 /22.9(l9.0-28.5)7
20-j/^9.3(l7.6-24.4)5 '20.7(18.0-24.0)19
''18.5(17.6-20.6)5
-OPEN
-SHADE
0700 MOO 1500
JUNE
0700 1100 1 500 1 0700 1100 1500
JULY AUGUST
Figure 4. The mean air temperatures recorded at various collecting stations in the Central Desert of Baia Cali-
fornia, Mexico, during June, July and August, 1969. See Fig. 2 for explanation of figures.
Combined mean monthly temperature data (taken from Hastings and Humphrey,
1969) from five coastal strand localities (El Rosario, Rosarito, Vizcaino, Bahi'a Tortuga and
Punta Abreojos) within the Central Desert and less than ten miles from the ocean show that
the highest mean monthly temperatures occur in August and September (ca. 24.2°C) and
the lowest in January and February (ca. I5.7°C), but less than 8.5 °C separate the mean
temperature of winter and summer.
Diurnal fluctuations in the summer soil and air temperatures recorded during the sur-
vey are shown in Figures 3 and 4. In general, soil and air temperatures rose rapidly in the
morning with the dissipation of coastal cloud or fog cover, peaked near mid-afternoon, and
thereafter showed a gradual decrease. Shade temperature decreased less rapidly than tem-
peratures in the open, and rose gradually from 1 100 through 1500 in June and July.
VEGETATION
The survey area falls within Wiggins' (1960b) Central Desert phytogeographic area,
specifically in the district of the Vizcaino Desert Subflora.
Characteristically, vegetation of the open coastal strand is stunted, seldom over one
meter high, widely spaced, and lacking in species diversity. According to Wiggins (1960b)
and Aschmann ( 1959) these characteristics are partially the result of strong, almost contin-
uous onshore winds that release very little moisture in their passage. In areas protected
from the direct effects of prevailing winds but still within reach of the fog and moist sea air,
in sandy arroyo floors where the water table is near the surface, and in areas where runoff"
244 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
concentrates at the base of slopes, the vegetation is fairly abundant. The most conspicuous
perennial plants of the Central Desert listed by Wiggins (1960b), Shreve and Wiggins
(1964), and Aschmann (1959), included the following: Agave spp.. Ambrosia chenopodii-
folia. Ambrosia magdalenae. Yucca valida, Idria columnaris, Machaerocereus gummosus,
Larrea divaricata, Lvcium californicum, A triplex polvcarpa, Viguiera deltoidea, Dudleva
spp., Encelia spp.. Euphorbia spp., Opuntia choUa, Viscainoa geniculata, Pachycereus
pringlei, Lophocereus schottii, Echinocereus brandegeei, Fouquieria splendens, Prosopis spp.,
Cercidium spp., Solanum spp., and Pachvcormus discolor.
Also common in appropriate coastal strand habitats were Frankenia palmeri, A triplex
canescens and A triplex julacea. '
SPECIES ACCOUNT
AMPHIBIANS
Bufo punctatus
Each evening at San Javier Arroyo (Fig. 6) trilling choruses of toads were audible. On
the evening of 14 July several pairs were observed in amplexus along a 100 meter stretch of
a stagnant stream. The tadpoles, with well-developed hindlimbs, were collected from a
small, shallow, algae-covered pool. A total of 83 specimens (65 adults, 18 tadpoles) were
taken at Station 38.
Hyla regilla deserticola
Ten adults and two tadpoles of this race (see Jameson et al, 1966, for distribution and
characters) were collected at San Felipe Springs (Sta. 8; Fig. 5), a small perennial spring
marked by luxuriant vegetation consisting of cottonwoods, willows, fan palms, cirio, pita-
haya and tules. The adults were found beneath rocks and in the grass and tules surrounding
the main body of water, a shallow pond about five meters wide. The tadpoles were collected
in another small pool.
Twenty-five adults were collected from rock crevices and from beneath rocks flanking
the stagnant San Javier stream (Sta. 38, Fig. 6). Adults called late into the evening.
LIZARDS
Callisaurus draconoides crinitus
These lizards were confined to a coastal (Sta. 40; 8 specimens) and inland (Sta. 41; 28
specimens) sand dune habitat. Many individuals were approached within several feet and
collected by stunning them with six-inch rubberbands.
Nine of the 18 females collected between 15-17 July had yolk-laden ovarian ova
greater than 3mm in diameter, and four of the 18 contained oviducal ova. The mean length
and width of ovarian ova in the left and right ovaries were 5.6(3.3-9.5)1 1 by 4.9(2.8-8.7)1 1
and 6.2(4.7-9.0)9 by 5.7(4.5-7.8)9, respectively. Oviducal ova in the left oviduct measured
15.2( 14.0-16.7)3 by 8.0(7.4-8.7)3 and in the right oviduct 15. 1( 13.7-17.8)7 by 9.2(7.3-1 1.0)7.
Male crinitus average longer than females; mean lengths for 18 males and 18 females
being 68.4mm (range 47-82mm) and 59.9mm (range 46-68mm). The mean testis size of the
series collected in mid-July was 4.6mm(range 3. 2-5. 8mm) by 3. lmm(range 2.4-4. 0mm). The
right testis was anterior to the left in all males examined.
Measurements and counts of crinitus are summarized in Table 3. The distance between
the anterior edge of the most anterior ventral tail bar and the posterior margin of the anus, a
'A list of other dominant plants representative of the major herpetological collecting stations in this area, is filed with the Na-
tional Auxiliary' Publication Service of the American Society for Information Service, and may be obtained by ordering NAPS
Document 01 547 from ASIS National Auxiliary Publication Service, CCM Information Corp.. 909 Third Ave.. New York, N.Y.
10022, remitting $5 per photocopy or $2 per microfiche copy.
1971
BOSTIC: BAJA CALIFORNIA HERPETOFAUNA
245
measurement suggested by Dr. Benjamin Banta, readily separates crinitus from rhodos-
tictus and appears to be much more reliable than the diagnostic characters used previously
to separate these races (see Tevis, 1944). The number of oblique lateral bars was variable
among the 35 crinitus examined; nine had none; three had two, two had four, and 21 had
the three characteristic of the race.
Except for snout-vent length, the only important sex difference was the hindlimb
length: males, 63.7(47-75)18, females, 55.2(45-64)18. Also, females tended to lack the ob-
lique body bars more frequently than males; eight females out of the 18 had none, whereas
only one of 17 males lacked the bars entirely.
Callisaurus draconoides rhodostictus
Individuals of rhodostictus were generally confined to sandy washes and broad, sandy
arroyos. At the most inland collecting stations, where the soil is largely decomposed granite,
Figure 5. Station 8, San Felipe Springs (in arroyo de San Fernando), about nine miles NE of the arroyo-
coastal road junction. Several adult Hvla regilla deserticola were collected beneath rocks and in the grass and
tules surrounding the main body of water (see arrow): tadpoles were collected in another small pool of water not
visible in photograph.
246 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
Table 3. Measurements and counts of Callisaurus draconoides crinitus and C. d. rhodostictus.
crinitus rhodostictus
Snout-vent length 64.1(46-82)36 68.9(38-85)29
S.D. = 9.9 S.D. = 9.8
Ratio, tail: total length 5.7(5-6) 19 5.8(4-6) 14
S.D. = 0.4 S.D. = 0.6
Distance between anus to most 15.9(7-26)34 3.4(1-7)29
anterior ventral tail bar S.D. = 5.2 S.D. = 1.5
Hindlimb length 59.4(45-75)36 64.5(32-84)17
S.D. = 8.7 S.D. = 8.9
Ventral body bars 2.2(0-4)35 1.8(0-2)29
S.D. = 1.3 S.D. =: 0.6
Ventral tail bars 6.8(0-10)20 7.6(4-10)30
S.D. = 1.8 S.D. = 1.4
Femoral pores' 18.2(14-22)35 13.8(11-18)27
S.D. = 1.6 S.D. = 1.6
'Femoral pores counted on one side only.
rhodosticus were often observed basking during mid-day atop small rocks.
Only four of the 17 female rhodostictus collected had enlarged yolk-laden ovarian ova,
[mean size: 5.3(3.2-6.9)12 by 4.8(3.2-6.0)12], and none had oviducal eggs. The specimens
with enlarged ovarian ova were collected in late August, whereas the crinitus, four of which
had oviducal ova, were collected in mid-July. This suggests that egg laying among crinitus
and rhodostictus ceases between the end of July and August in the Central Desert.
Males tend to be larger than females; males had a mean snout-vent length of 75. 3(61-
85)12 and females a mean snout-vent length of 64.3(38-75)17. The testes of eight rhodos-
tictus averaged 4.8(3.5-6.8) 15 by 3.3(2.0-4.5)15. The right testes was anterior to the left in all
males examined, and was usually slightly larger.
Other counts and measurements for rhodostictus are presented in Table 3. The mean
hindlimb length, 76.1(61-84)12 for males and 64.5(32-72)17 for females, and the mean
snout-vent length, previously mentioned, were the only apparent quantitative differences
between the sexes. Specimens were collected at Stations 1()(4 specimens), 11 (1), 12 (2), 13
(2), 14(3), 15(1), 16(1), 17(3), 18(4), 19(1), 26 (4), 31 (3).
Cnemidophorus hyperythrus schmidti
Walker and Taylor (1968) in their prehminary treatment of the geographical variation
among the ''hvpervthrus-like'' populations of Baja California lacked sufficient material
from Central Baja California to determine the variation and distribution of schmidti. The
specimens collected in this study possess a single mid-dorsal line, forked anteriorly, which is
characteristic of schmidti (Lindsdale, 1932; Murray, 1955). Data concerning scutellation
and pattern of those specimens are summarized and compared to similar data for hype-
rythrus and beldingi in Table 4.
Murray (1955) indicated that schmidti could readily be distinguished from hyperythrus,
the southern race, by the arrangement of the mid-dorsal lines. Separation of schmidti from
beldingi, the northern population, is based presently on the number of supraoculars sepa-
rated from the frontal by granules and less consistently by the presence of two mid-dorsal
stripes (Table 4). My data concerning the degree to which the supraoculars are separated
from the frontal by granules show that this character is of little diagnostic value when con-
sidered alone (Table 4). An apparent diagnostic difference among the three populations is
the number of granules around mid-body, intermediate in schmidti (Table 4).
1971
BOSTIC: BAJA CALIFORNIA HERPETOFAUNA
247
Murray (1955) stated that intergradation between schmidti and beldingi probably oc-
curs in the vicinity of El Marmol [about 45 miles NW of Laguna Chapala (Fig. 1)], because
individuals suggestive of intergradation have been recorded from Laguna Chapala and
Catavina (about 30 miles NW of Laguna Chapala). Murray's primary criterion was the par-
tial or complete separation of the second supraoculars by granules. The Catavina specimen
(see Lindsdale, 1932) was reported by Murray to be the only one from this part of the penin-
sula in which the second supraoculars were entirely separated by granules. I collected 15
individuals from Stations 2 through 38 (Table 1 and Fig. 1) that show this same condition.
One of these specimens (45554) from Station 38, about 125 miles to the south of El Marmcl,
also has two mid-dorsal lines, more suggestive of beldingi than either of the specimens dis-
cussed by Murray (1955). Specimens were collected at Stations 5(1), 6 (2), 7 (1), 8 (3) 15(1),
26 (15), 31 (16), 34(1), 36 (10), 38 (3).
Cnemidophorus labialis
Specimens of C. labialis from the localities below fill the distributional gap of 185 miles
between Miller's Landing and El Consuelo. Station 41, 30 miles south of Miller's Landing,
is the southernmost collecting locality, and probably is near the species southern hmit.
Table 4. Variation in scutellation and patterns among Baja California races of
Cnemidophorus hyperythrus.
beldingi
schmidti
hyperythrus
Granules around midbody
Granules separating
dorsolateral stripes
72.8±0.8(66-79)17' 75.2±0.6(66-83)54
S.D. = 3.3 S.D. = 4.1
25.4±0.4(23-30) 17' 24.2±0.3(21-29)53
S.D. = 1.5 S.D. = 1.5
Femoral pores (combined count) 31.9±0.5(29-37) 17' 31.5±0.4(26-39)52
S.D. = 2.2 S.D. = 2.6
Supraoculars (left-right) 3-4(2)',4-4(15)' 3-3(4),4-3(l),4-4(53)
Anteriormost supraoculars
separated from the frontal
by granules:
Part of third
5*
10=^19
Third
48' 14
Part of second
25*
3M6
Second
36*
15
Frontoparietal
Single
17
55
Partially divided
5
Divided
Number of mid-dorsal Unes
Three
95
Two
104'
2=3
One forked anteriorly
46' (extent of forking,
if present, not
More than one-third
stated.)
length
22= 10
Less than one-third
length
37M6
77.6±0.8(69-90)45'
S.D. = 5.2
33.6=h0.4(29-41)44'
S.D. = 2.6
3-3(16),3-4(3)
4-4(26),5-4(l)^
28'
443
4*
47'
48=
15^
'Data from Walker and Taylor (1968)
=Data from Lindsdale (1932)
'Combined data from Murray ( 1955) and Lindsdale (1932)
'Data from Van Denburgh (1922)
■Data from Burt (1931)
248
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
The specimens, all adults, showed a daily activity cycle and occupied habitats similar
to those previously recorded for the species (Bostic, 1968).
Scutellation and counts for Central Desert specimens were as follows: granules around
mid-body. 59.8±0.4(52-69)87; granules separating paravertebral stripes, 8.4±0.1(6-12)89;
femoral pore scales, left, 13.6±0.1( 1 1-16)87, right, 13.5±:0.1( 1 1-16)78. Specimens were
taken at Stations 2 (5), 26 (28), 28 (4), 3 1 (3), 33 (3), 36 ( 1 5), 37 (2), 4 1 (38).
Figure 6. Station 38, San Javier Arroyo. Several Bufo pimctatus tadpoles were collected from the small, shallow
algae-covered pool of water in the foreground. Adult B. pimctatus were particularly abundant, and many were
observed in amplexus. A Sauromalus australis was collected from within a crevice of the granite-strewn west
slope (see arrow). PhvUodactvlus xanti sloani were also collected beneath the exfoliating slabs of granite rock.
Cnemidophorus tighs multiscutatus
This species was relatively common throughout the survey area, but difficult to collect.
Individuals were most active during the mid-day hours, when they were frequently ob-
served foraging from shrub to shrub. They preferred the soft soil (sand and decomposed
granite) of the washes and arroyos to the compacted, rocky soil of the marine terraces.
Selected characters for the specimens collected are as follows: Postantebrachials gran-
ular in all but three individuals, which have these slightly enlarged; supraorbital semi-cir-
cles normal, except for two specimens in which they extend past the posterior margin of the
frontal; anterior nasal not in contact with the second supralabial in all but eight lizards;
fronto-parietal divided in all but one specimen; number of supraoculars 4-4, except for
seven specimens which have 5-5, 4-5, or 4-3 supraoculars; granules around body,
90.6±0.7(82-104)48; femoral pore scales (left), 20.0±0.2( 17-23)69, S.D. 1.5 1 ; femoral pore
scales (right), 19.9±0.2( 16-23)70, S.D. 1 .58. Specimens were obtained at Stations 2 ( 1 ), 5 (4),
8(11),9(2), 15(1), 16(1), 18(1), 19(1), 20(13), 21 (2), 26 (14), 31 (8), 33 (2), 36 (8), 38 (2).
1971 BOSTIC: BAJA CALIFORNIA HERPETOFAUNA 249
Coleonyx variegatus abbot ti
These specimens agree closely with Klauber's (1945) original description, confirm the
presence of the race in the Central Desert, and support Klauber's (1945) tentative assign-
ment of a damaged specimen at Calmalli, seven miles NW of El Arco, to this subspecies.
One individual was found beneath a small slab of shale on the SW slope of a clay-like foot-
hill (Sta. 1 ), and two were collected beneath the basal leaves of dead Agave at Station 38.
Crotaphytus collar is
One speciment from Station 20, an adult male, represents the first recorded occurrence
of Crotaphytus collaris west of the peninsular divide (Van Denburgh, 1922: 109; Smith and
Taylor, 1950:92).
The collecting station, Mesa de San Carlos, is a broad table-topped mountain near the
coast, which rises to an altitude of from 422 to 739 meters. The above individual was ob-
served foraging among large basaltic rocks on the edge of the mesa at approximately 1400
hrs. Another C. collaris was observed basking at 1730 hrs. on a small rock, part of a large
basaltic rock outcrop, surrounded by low shrubs on the mesa proper.
Crotaphytus wislizeni copeii
A single juvenile from Station 20 agrees in scutellation and pattern with Banta and
Tanner's (1968) account of the race. It was foraging in the late afternoon in a sandy wash
thickly overgrown with xeric vegetation.
Gerrhonotus multicarinatus ssp.
Table 5 shows that the Central Desert specimens agree closely with G. paucicarinatus in
degree of keeling and in some details of pigmentation. They appear more like G m. webbi in
numbers of longitudinal dorsal scale rows, and dorsal pattern; they have an intermediate
position between paucicarinatus and webbi in numbers of transverse dorsal scale rows, de-
gree of keeling and numbers of keeled temporal scale rows, lateral fold pigmentation and
ventral markings. Coloration and over-all pattern among the individuals show considerable
variation. Some xtSQmblQ paucicarinatus and others webbi. Individual counts and measure-
ments of the Central Desert specimens appear in Tables 6 and 7.
The above evidence suggests intergradation between G paucicarinatus and G multi-
carinatus. However, since a gap of about 250 miles separates these populations, it would be
premature to make a formal nomenclatural change at this time.
The Pacific coastal strand is suitable for the southern dispersal of G multicarinatus.
Similarly, G. paucicarinatus, once believed to occur only in the highland area of the Cape
Region, has now been recorded in the lowland area of the Cape (Richmond, 1965), and may
have dispersed farther northward along the Pacific Coast where cool, moist environments
suitable for anguids prevail.
Savage (1960) surmised that paucicarinatus separated from multicarinatus during a
Pleistocene glacial maximum, but whether this isolation has resulted in ecological and/or
reproductive isolation is unknown. The Vizcaino Desert is a possible barrier preventing
their contact. From here south through the Magdelena Plains region, coastal precipitation
is unpredictable, and often a summer phenomenon.
These specimens together with a specimen of G. multicarinatus ssp. from the Pacific
coast west of Punta Prieta(Bogert and Porter, 1967: 15) are Ihe first of Gerrhonotus from the
Central Desert. Six were collected beneath dead or partially dead Agave, and two beneath
pieces of tin at the abandoned settlement of Las Palomas. None of the examined females,
collected in July, had enlarged yolk-laden ova or oviducal ova, but the oviducts in all were
highly vascularized and convoluted.
Specimens were obtained at Stations 3 1 (1), 33 (3), 36 (3), and 38 (1). Additional speci-
250
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
mens examined— two G. multicarinatus ssp. [SDSNH 45016, 24 December 1969, Sta. 31;
American Museum of National History (AMNH) 75765, 22 April 1956, 5 mi. NE of Punta
Santa Rosalia]; 18 G. paucicarinatus from the Cape Region of Baja California (SDSNH
45006-45010,45033,45095-45098,45100-45101,45103-45106,53057-53058).
Table 5. Comparisons between adult Gerrhonotus paucicarinatus, G. multicarinatus ssp.,
and G. multicarinatus webbi.
Character
G. paucicarinatus
G. m. ssp.
G. m. webbi^
Average Snout-vent length
130
94.6
135(n = 35)
Scutellation
Dorsal Scale Rows
Transverse
50.7(50-51)10'
50.1(46-54)18
46.2(45-50)9
41.5(38-45)44
Longitudinal
15.4(14,16)10'
16.2(16,18)18
14
14
Ventral Scale Rows
Transverse
62.7(60-64)10-
64.1(60-67)18
64.6(62-68)9
63.2(60-66)44
Longitudinal
ir
12.1(12,14)18
12
12
Keeling
Temporals
None
Upper one to two
Upper two rows
rows faint or none.
or more.
Dorsal Rows
11.2(8-12)18
11.4(10-14)9
14
Upper Arm
None
None
Three rows or more.
Lower Arm
None
None
Average 2.8'
Tail
6-8
6-10
Eight plus several
lateral rows.
Pigmentation
Dorsal Head Spotting
Present (distinct)
Present
(distinct-faint)
Absent (normally)
Eye Color
Unknown to me.
Yellow
Yellow
Temporal Eye Stripe
Distinct
Distinct
Faint
Labials
Normally distinctly
Faintly to distinctly
Unicolor or the
banded with alternate
edged with black.
supralabials may be
black and white
faintly edged with
markings.
black.
Body Bands
ll.l(n= 10) when
11.1(10-14)9; Moder-
10.6(9-13)38; deeply
complete, but often
ately indented with
indented with distinct
bands are incomplete
white markings on
white markings on
dorsolaterally. White
fifth or sixth scale row
fifth scale row above
markings are reduced
above lateral fold. The
lateral fold and in
laterally and usually
white mid-dorsally is
middle of back.
absent mid-dorsally.
usually indistinct.
Lateral Fold
Ground color
Ground color
Ground color
predominates with
predominates with
predominates with
narrow black lines.
large whitish spots
scattered white spots
distinctly outlined and
but no black markings.
composed of groups
of white scales; black
markings when present
faintly diffuse.
Ventral Markings
Along middle of
Along middle of
Along middle of
longitudinal scale
longitudinal scale
longitudinal scale rows
rows forming distinct
rows forming distinct
forming faint longi-
longitudinal lines in
to faint longitudinal
tudinal lines in most.
most.
lines.
'From Fitch (1938) 'From Fitch (1938) except where noted. 'From Murray (1955)
1971
BOSTIC: BAJA CALIFORNIA HERPETOFAUNA
251
Table 6.
Desert.
Counts of body scales and cross bands in Gerrhonotus multicarinatus ssp. from the Central
Catalogue
Dorsal Scale Rows
Number
Keeled
Ventral Scale Rows
Cross
No.
Transverse
Longitudinal
Transverse
Longitudinal
Bands'
45992
45
14
12
64
12
10
45993
47
14
14
65
12
11
45994
45
14
12
66
12
10
45995
47
14
12
62
12
14
45996
45
14
10
64
12
10
45997
50
14
11
68
12
11
45998
47
14
10
63
12
10
45999
45
14
10
65
12
12
46016
45
14
12
64
12
12
'Partial bands not counted
Table 7. Measurements of Gerrhonotus multicarinatus ssp. from the Central Desert.
Catalogue
Sex
Collecting
Station'
Date
Snout
to vent'
Tail
Head
No.
Width
Length
Depth
45992
M
31
1 July 1969
98
Broken
14.0
20.0
9.6
46016
M
31
24 Dec. 1969
87
Regener-
ated
12.4'
19.3'
9.0'
45993
M
33
6 July 1969
104
Broken
16.5
21.0
12.0
45994
M
33
7 July 1969
92
Broken
13.3
19.0
9.0
45995
M
33
7 July 1969
100
150(79
whorls)
14.6
20.9
11.0
45996
F
36
9 July 1969
92
Regener-
ated
11.5
17.8
9.8
45997
F
36
10 July 1969
79
Regener-
ated
10.9
15.6
6.6
45998
M
36
10 July 1969
107
Regener-
ated
17.6
23.4
11.0
45999
F
38
14 July 1969
92
Broken
12.6
17.9
9.0
'Measurements before preservation
"See Table 1
Petrosaurus repens
The 10 specimens from Station 10 are the first known from the west coast of Baja CaH-
fornia del Norte. They were initially observed basking on huge granitic boulders between
1500 and 1830 hrs. When disturbed they usually sought refuge deep within rock fissures.
Two individuals were smoked out and hand captured.
None of the eight females collected on 27 August had enlarged (>3mm), yolk-laden
ovarian or oviducal ova. All the stomachs examined contained small black seeds similar in
appearance to those in the fruits of the barrel and fishhook cacti which occurred commonly
in the area. Many of the stomachs also contained small amounts of other nondescript vege-
tation and all contained the carapaces of small beetles.
Scutellation and measurements of the specimens collected are as follows: snout-vent
length,94.7±5.3(78-lll)6,S.D.= 13.0; head width, 16.7±0.8(14.2-19.6)7,S.D. = 2.0; femo-
ral pore scales (combined count), 24.8±0.4(22-26)8,S.D.= 1.2; dorsals 172.4±2.2(165-
182)7,S.D. = 5.8; head ventrals, 69.3±2.1(63-77)7,S.D. = 5.6; fourth toe lamellae,
27.5 ±0.3(26-28)8,S.D. = 0.7.
252
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Phrynosoma coronatum
Two active specimens were collected on a flat, sandy substrate sparsely covered with
low shrubs, and one was collected at 0650 hrs. by raking the sand beneath a hummock cov-
ered with ragweed; it was relatively sluggish and made no attempt to elude capture. These
specimens were taken at Stations 9 ( 1 ), 14 ( 1 ), and 41(1).
Phyllodactylus xanti sloani new subspecies
Holotype.-\d\x\i female (Fig. 7), SDSNH 45895, collected 23.5 miles SE of El Rosario
(29°48'N, 1 15°33'W), Baja Cahfornia del Norte, Mexico, from a crevice in a block of shale
by Thomas Cozens on 3 1 March 1969.
Paratypes.—A\\ seventeen paratypes collected are from Baja California del Norte,
Mexico: SDSNH 45896, Sta. 9; SDSNH 45897-45898, Sta. 25; SDSNH 45899-45900, Sta.
26; SDSNH 45901-45907, Sta. 3 1 ; SDSNH 45908, Sta. 33; SDSNH 45909-45912, Sta. 38.
Figure 7. Holotype (SDSNH 45895) of Phyllodactylus xanti sloani.
Diagnosis.— This race differs from all other races, except nocticolus, confined to south-
eastern California and the eastern desert regions of Baja California, and angulus, occurring
on Islas Salsipuedes and San Lorenzo Island, Gulf of California, by the absence of thigh
tubercles (see Dixon, 1969:79.1-79.2, for diagnostic accounts of the races of P. xanti); from
angulus in larger snout-vent length (51.1mm vs. 43.8mm), less numerous mid-orbital scales
( 1 8.0 vs. 20.5) and fewer paravertebral tubercles from axilla to groin (20.4 vs. 23.0) and from
rear of head to base of tail (37.8 vs. 40.0); from nocticolus in less numerous longitudinal
rows of ventral scales (27.2 vs. 35.2) and fewer tubercles in a paravertebral row between
axilla and groin (20.4 vs. 23.0).
Description of holotype.—Roslml twice as wide as high, its dorsal edge with two rec-
tangular internasals, their median edges in broad contact, bordered posteriorly by five
granules and postnasal on each side; nostril surrounded by rostral, internasal, labial, and
two postnasals; its ventral edge in contact with labial; slight depression between internasals
and in frontal region; 20 scales between eye and nostril; posterior dorsolateral loreals three
to four times larger than interorbital scales; 15 scales across snout between second labials,
1971 BOSTIC: BAJA CALIFORNIA HERPETOFAUNA 253
17 between third labials; 12 scales between anterior edge of orbits; 20 interorbital scales;
eye large, contained in snout length approximately one and one-half times; eyelid with two
rows of granules and one larger outer row of scales, the latter with seven posterior scales
bearing spines; diameter of ear contained in diameter of eye slightly less than two times; ear
opening not denticulate, anterior border with rounded and slightly pointed scales, posterior
margin with smaller rounded scales; top and rear of head granular, with faintly keeled,
larger, intermixed tubercles; 12 supralabials, seventh to center of eye; 1 1 infralabials, fifth
to center of eye; mental lyre-shaped, length and width equal; postmentals followed by a
transverse row of eight scales, followed by a second row of 12 smaller scales; postmentals
contacting first labial on right and left sides.
Dorsum with 12 longitudinal rows of enlarged, keeled, somewhat flattened tubercles,
1 1 rows reaching head, six at base of tail; 39 paravertebral tubercles, 24 between axilla and
groin; two median rows of enlarged tubercles separated from each other by two and three
rows of granules; each tubercle of enlarged dorsal series separated from proceeding tu-
bercle by one to three granules; three postanal tubercles on either side of anus, well differ-
entiated, rounded; 35 scales across venter, 72 from gular region to anus.
Ventral, antero-dorsal surfaces of limbs with large circular scales, postero-ventral sur-
faces granular; lower arm and leg granular, with scattered larger, keeled tubercles inter-
mixed; lamellae formula for left hand 7-9-10-1 1-8 (undivided 2-6-7-8-7), left foot 6-10-12-
13-1 1 (undivided 5-8-8-12-6); claws short, tip barely visible when viewed from below; ter-
minal pads rounded at tips; tail missing.
Measurements in mw.— Snout- vent length 53; headwidth 10.7; head length 13.9; axilla-
groin length 24.4.
Color in fl/co/?o/.— Mid-dorsum ground color pinkish-tan; dorso-lateral surfaces blue-
gray. Venter light pinkish-tan; dorsum with six reddish-brown broken crossbands, slightly
narrower than ground color interspaces; dorsal and lateral surfaces of head spotted with
light brown; area posterior of eye orbits, but anterior to first dorsal band, spotted with hght
brown on a tan ground color; dorsal surfaces of limbs with brown spots; tips of enlarged
dorsal tubercles cream, brown, or brown and cream.
Variation.— "Ho sexual dimorphism in size, color, or pattern is evident. Counts and
measurements are as follows: Snout-vent length 5 1.1± 0.6(32-6 1)17; enlarged series of dor-
sal tubercles, 1 1.8±0.3(9-14)17; postmental border scales, 7.3±0.2(6-10)18; nostril to eye
scales, 10.5±0.2(9-12)18; scales bordering internasals, 6. 8±0. 1(5-8)18; interorbital scales,
18.0±0.3( 15-20)18; third labial scales, 16.4±0.2( 15-18)18; lamellae beneath fourth toe,
12.5±0.2( 1 1-14)18; scales across venter, 27. 1±0. 9(21-35)17; number of paravertebral tu-
bercles, 37.8±0.7(32-40)17; axilla to groin, 20.4±0.3( 17-24)16. Of the 18 specimens, all but
one have the postmentals contacting two labials on each side. There are two postmentals in
all but two individuals, which have three. The number of interorbital scales is always equal
to or more than the number of scales across the snout between third labials. The color pat-
tern varies from incomplete, irregular bands, and spotting to complete bands on the dorsum
(Fig. 8). The ground color ranges from reddish-brown to gray-brown. The venter of all spec-
imens is immaculate.
Remarks.— ThtSQ specimens are the first of i^. xanti from the Pacific slope of the penin-
sula (see Dixon, 1966, Fig. 1). All individuals were collected beneath exfoliating slabs and
in fracture crevices of granite and shale, predominately the latter (see Fig. 6). This subspeci-
fic epithet honors Allan J. Sloan, Curator of Reptiles and Amphibians, San Diego Museum
of Natural History, whose assistance, enthusiasm and support were largely responsible for
making this survey a reality.
i?fl/2ge. -Known from 23.5 miles SE of El Rosario (29°48'N, 1 15°33'W) to San Javier
254
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
(28°32'N, 1 14°05'W) on the west coast of the peninsula.
Specimens examined.— Jht 18 specimens examined are listed under type and para-
types.
t^
4lfgS^
[:^i3'iio
HLOfltf
i.^^>
^
i^j
t.tl.'ttl
#
Figure 8. Dorsal variation among specimens of Phyllodactvlus xanti sloani collected along the Pacific coastal
strand of the Central Desert of Baja California del Norte, Mexico.
Sauromalus australis
An adult male was collected from deep within a crevice on the afternoon of 12 July
1969 approximately 15 yards up the steep granite-strewn west slope of Arroyo San Javier
(Station 38; see Fig. 6).
Despite two additional days working suitable habitats in and around the arroyo, no
other Sauromalus were observed. That this specimen was not a "waif," however, was in-
dicated by large amounts of dried fecal material, and by the observation of similar large
lizards by a resident rancher, Senor Lopez of "Mi Ranchita," who stated that chuckwallas
could be found about three miles to the east of his ranch.
1971 BOSTIC: BAJA CALIFORNIA HERPETOFAUNA 255
Undoubtedly more chuckwallas will be collected in this area during a more favorable
time of year, spring and early summer, when plant food is available and the daily tempera-
ture not so high. I suspect that most Sauromalus in this region undergo a state of estivation
deep within granitic fissures when the Vegetation is dormant. Plant food was also decreased
in and around Arroyo San Javier by domestic goats. Johnson (1965) noted that almost all
activity of a Mojave Desert population of Sauromalus obesus ceased by 1 August owing to a
lack of food and water.
This specimen, the first collected on the Pacific side of the Peninsula, fits the parame-
ters of scutellation and measurements established for the species by Shaw (1945).
The pattern and coloration of this species differ from those described for the type speci-
men by Shaw (1945). In pattern it is hke one (SDSNH 17708) he described from La Paz,
and in coloration it is similar to Sauromalus ater in being yellowish-black (and brown) in-
stead of the gray characteristic of 5". aust rails.
Sceloporus magister rufidorsum
Sceloporus magister, difficult to collect, were most frequently associated with impene-
trable thickets of thorn bush and pitahaya. At Station 33, a coastal sand dune habitat, they
inhabited hummocks covered with tree sunflower (Encelia ventorum). Of the adults (snout-
vent >93mm) collected, six were females and 13 were males, with 21 and 19 being juvenile
female and male, respectively. Two females with snout-vent lengths of 90mm and 93mm
contained a combined total of 13 oviducal eggs, the mean size of which was 18.0mm by
10.0mm (range 1 1.5mm-17. 1mm by 8.8mm-12.0mm). The mean number of oviducal ova in
the left and right oviducts was 2.5 and 4.0, respectively.
Phelan and Brattstrom (1955), in their analysis of the variation among 5". magister pop-
ulations, concluded that the basic diff'erences are those of coloration of the adult males, scu-
tellation characters being so variable that they were not significant.
Variations in scutellation and other measurements of the Central Desert specimens are
compared (Table 8) to data provided by Phelan and Brattstrom (1955). Excluding the cir-
cumorbital and femoral pore counts, these data fit the parameters established by Phelan
and Brattstrom for S. m. rufidorsum. The Central Desert specimens tend to have the circum-
orbital scales broken up into smaller units, which accounts for the greater range and mean.
There was little consistency in color pattern among the Central Desert specimens. Of the 13
adult males examined only one had a typical rufidorsum pattern, six had a basic rufidorsum
pattern but lacked side bars, five had a lineatus pattern, and one had no pattern (see Phelan
and Brattstrom, 1955, Fig. 1). Adult females showed a similar variation in pattern; juveniles
showed a much greater one.
In summary, dorsal patterns of adult males are so variable as to be of little diagnostic
value. Consequently, I question the reliability of subspecific recognition based primarily on
the dorsal pattern of adult males. Specimens were collected at Stations 2(2), 6(1), 12(1),
15(1), 16(1), 21(3), 26(10), 31(7), 33(5), 36(8), 37(2), 38(5), 40(1),41(21).
Sceloporus orcutti orcutti
Seven individuals of S. orcutti were associated with large granitic rock outcrops, and
one was collected from among the basal leaves of din Agave where it had traveled after it was
first discovered in a thicket of thorn scrub.
Two of three females collected on 4 and 13 July contained a total of 14 oviducal eggs,
the mean size of which was 15.4mm x 9.5mm (range 14.2- 16.5mm x 9.1-10. 1 mm).
Scutellation and measurements for the specimens, five females and three males, taken
at Stations 8( 1), 10(3), 26( 1 ), 3 1( 1 ), and 38(2) are as follows: snout-vent length 83.8±4.3(67-
I02)8,S.D.= 12.2; ratio, tail: snout-vent, 1.2±0.5(0.92-1.28)5,S.D. = 0.1; dorsal scales
31.I±0.3(30-32)8,S.D. = 0.8; femoral pores, 13.2±0.4(12-15)8,S.D.= 1.1; gular scales.
256
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Table 8. Scutellation and measurements of adult Sceloporus magister rufidorsum.
Phelan and Brattstrom
(1955)
Central Desert Specimens
Males
Females
Snout-vent length
Ratio, tail: snout-vent
Dorsal scales
Femoral pores
Gular scales
Supralabials
Infralabials
Supraoculars
Circumorbitals
Lamellae, fourth toe
Auricular lobules
Ventrals
131.0-maximum
1.4(1.2-1.5)6
29.8(29-31)8
17.9(15-20)14
15.9(15-18)8
4.4(4-5)13
6.2(5-7)13
5.3(5-6)13
5.1(3-6)14
110.0+2.1(97-119)12
S.D. = 7.4
1.2±0.03(1.0-1.4)12
S.D. = 0.1
29.0+0.2(28-30)11
S.D. = 0.7
18.6±0.4(16.5-20.0)12
S.D. = 1.5
18.4=1:0.4(17-20)12
S.D. = 1.2
4.1 + 0.1(4.0-4.5)12
S.D. = 0.2
6.2+0.1(6.0-6.5)12
S.D. = 0.2
5.5 + 0.2(5-6)5
S.D. = 0.4
6.9+0.4(6.0-10.5)12
S.D. = 1.3
22.7+0.4(20-25)12
S.D. = 1.4
5.5+0.2(4.5-6.0)12
S.D. = 0.7
39.1 ±0.8(35-43) 11
S.D. = 2.6
96.2+1.7(93-105)6
S.D. = 4.2
1.3+0.04(1.2-1.3)6
S.D. = 0.1
29.3±0.5(28-30)3
S.D. = 0.9
17.5+0.5(16-19)6
S.D. = 1.2
18.6±0.3(17-19)6
S.D. = 0.7
4.1+0.1(4.0-4.5)5
S.D. = 0.2
6.5 + 0.1(6.0-7.0)5
S.D. = 0.3
5.7+0.9(5-6)5
S.D. = 0.4
8.1+0.9(5-11)5
S.D. = 2.0
22.3+0.3(21.0-23.5)6
S.D. = 0.8
5.5+0.3(4.5-6.0)5
S.D. = 0.6
40.2+0.6(39-42)6
S.D. = 1.4
16.8±0.4(15-18)8,S.D.= 1.0; infralabials, 5.8±0.1(5.5-6.0)8,S.D. = 0.4; supraoculars,
5.0±0.2(4-6)8,S.D. =0.5; circumorbitals, 6.1 ±0.2(5.0-6.5)7,S.D. = 0.6; lamellae, fourth toe,
20.6±0.9(15-23)8,S.D. = 2.4; auricular lobules, 5.6±0.2(5-6)8,S.D. = 0.5; ventrals,
38.7±0.7(35-41)7,S.D.= 1.8.
Urosaurus microscutatus
All individuals were initially observed basking or foraging in rocky areas and when
approached generally retreated to rock crevices. The collected specimens, from Stations
10(2), 20(1), 33(1) and 38(3), represent over half of all Urosaurus observed during the
survey.
Uta stansburiana
Side-blotch lizards were the most frequently observed reptile in the Central Desert.
They occupied every conceivable habitat, and were generally the first and last reptiles ob-
served each day. Specimens were collected at Stations 2(14), 5(2), 6(1), 8(2), 9(6), 10(7),
14(1), 16(2), 18(0,20(8). 21(16), 23(1), 24(3), 25(1), 26(19), 28(5), 29(1), 30(1), 3 1(10), 33(7),
36(5), 37(7), 38(4), 39(4), 40(3), and 41(46).
Xantusia vigil is wigginsi
This species was most commonly found beneath the basal leaves of dead Agave and
less frequently in or under dead decaying stems of cirio and Yucca. Specimens were taken at
stations 2(7), 5(4), 26(4), 31(8), 33(8), and 36( 1 ).
These specimens fill the distributional gap of approximately 85 miles between the
northermost collecting locality, 23.5 miles north of Punta Prieta, Baja California del Norte,
recorded for this race (Savage, 1952), and a single specimen collected near El Rosario which
1971 BOSTIC: BAJA CALIFORNIA HERPETOFAUNA 257
Savage stated seemed "to be nearer wigginsi than to the northern form," {X. v. vigilis).
SNAKES
Chilomeriscus cinctus
An aduh male was collected at Station 38 by raking through the base of a small
hummock of sand. A Phrvnosoma coronatum and a Sceloporus magister were collected in
the same fashion, but beneath the sand of a larger hummock covered with ragweed.
Counts and measurements for this individual are as follows: ventrals, 124;
subcaudals, 25; dorsal body bands, 22; tail bands, 5; and dorsal scale rows, 15-15-13.
Crotalus enyo envo
A juvenile specimen was collected at Station 3 beneath a dead Agave. Scutellation
and pattern agree with Klauber's (1931b) account of the nominal race.
Crotalus ruber ruber
The number of body blotches and the scale counts of these specimens fall within the
parameters established for the race by Klauber ( 1964: Table 2:7).
Crotalus ruber occupied a diversity of macrohabitats; one was observed in a coiled
position about 10 yards above the high tide mark of a cobblestone beach and another in a
coiled position beneath an ocotillo in bloom in a sandy, dune-like environment. Speci-
mens were collected at Stations 2( 1 ), 3( 1 ), 6( 1 ), 7( 1 ), 8( 1 ), 2 1(1 ), 22( 1 ), 38( 1 ), 4 1 ( 1 ).
Crotalus viridis helleri
A juvenile specimen collected at Station 41 is distinctly light colored with a sharply
defined pattern. It was observed at 0920 hrs. coiled beneath a small, sparsely branched
ragweed shrub at the fringe of an isolated sandy dune area.
Details of pattern and scale counts agree with those summarized by Klauber (1964,
Table 2:7) for the race.
Hypsiglena torquata klauberi
An active immature female was collected at 1710 hours beneath a dead Agave in an
eroded, sandy-bottomed wash (Sta. 2). Scutellation, coloration and pattern are similar to
those reported by Tanner (1944) for the race.
Lichanura roseofusca gracia
Compendia dealing with North America reptiles list two species of Lichanura, trivir-
gata and roseofusca, the latter species represented by two races, roseofusca and gracia.
Klauber (1933) reported a specimen of rosy boa from Guaymas, Sonora, Mexico, that
agreed exactly with L. trivirgata in coloration and pattern but more closely approached L. r.
gracia in scutellation. He remarked that perhaps we might be dealing with three subspecies,
trivigata, gracia and roseofusca. However, he did not suggest uniting the two species before
additional material between Guaymas and southern Arizona and in central Baja California
demonstrated intergradation.
Since Klauber's (1933) remark, additional specimens have been collected from these
areas, but according to Gorman (1965) we still lack a basis for uniting the two species of
Lichanura in view of the great uniformity of pattern of trivirgata throughout its known
range, and the absence of obvious intergrades with gracia.
The Central Desert specimen from Station 38 is of particular interest since the locality
is the southernmost for Lichanura in Baja California del Norte, and is only 100 miles from
San Ignacio, the northernmost for L. trivirgata.
The Central Desert specimens appear to agree with trivirgata in dorsal and ventral
counts, but more closely agree with gracia in all other counts (Table 9). In pattern and color-
258
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
ation there is close agreement with Klauber's (1931a) description of gracia . I tentatively,
then consider these specimens to be gracia.
One boa, a mature male, was collected in a grain field shortly after it had been killed by
a rancher (Sta. 4), and the other, an active fem.ale, was collected in the late afternoon from
beneath the basal leaves of Q.n Agave (Sta. 38).
Table 9. Scale characters of Lichanura trivirgata and L. roseofusca.
L. trivirgata
L. roseo
fusca'-
Central
Speci
SDSNl
45957
Desert
mens
BNos.
Gorman*
Klauber^
gracia
roseofusca
45958
Dorsals
Ventrals
Caudals
Oculars
Supralabials
Infralabials
39.2(36-41)10
218.5(219-223)10
45.0(42-49)10
10.1(9-11)10
12.8(12-14)10
13.8(13-15)10
41.4(40-43)7
222.0(218-227)7
44.0(42-46)7
9.7(9-11)9
12.8(12-13)7
13.8(13-15)7
41.3(40-43)9
230.0(220-236)9
46.0(42-49)9
9.8(8-11)9
14.1(13-15)9
15.4(14-17)9
40.9(35-43)38
232.0(221-244)38
47.0(39-51)38
9.1(7-10)38
14.1(12-15)38
15.0(13-17)38
42
225
46
11-10
15
15
40
222
43
10
14
14-15
^Data from Gorman, 1965
'Data from Klauber, 1931
Masticophis flagelliim piceus
Specimens, all adult females, were taken at Stations 2(1), 9(1), and 28(1). One active
individual was collected from within the hollow, dead stalk of an Agave at 1655 hrs. An-
other was collected at 1500 hrs. from beneath a large shrub in a sandy, eroded arroyo. The
most active individual was first observed in early afternoon foraging on the leeward side of
a large inland sand dune.
Phyllorhynchus decurtatus decurtatus
An adult male was collected at approximately 2000 hrs. as it crossed a sandy stretch of
road at Station 35. Scutellation and counts are as follows: caudal blotches, 7; dorsal body
blotches, 39; ventrals, 168; caudals, 36; snout-vent length, 367; and tail length, 58.
Pituophis melanoleucus annectens
This adult female was killed by a farmer who saw it foraging in a grain field (Sta. 4).
Scutellation and other counts, except for the ratio of total length to tail length, fit the
parameters established by Klauber (1946) for the race. The aforementioned ratio is .107
(total length 2565mm/tail length 265mm) considerably less than the . 1 55 reported by Klau-
ber (1946) for female annectens, which he states is probably the longest tailed of all the go-
pher snakes.
Pituophis melanoleucus bimaris
One specimen, an adult male from Sta. 21, was observed at 0930 hrs. as it foraged in
a sandy area studded with pitahaya. It attempted to elude capture by retreating down a
mammal hole. The o\htr bimaris, an active, immature female from Station 33, was collected
from beneath the basal leaves of a dead Yucca on a coastal foothill. This individual, as in-
dicated by the bulge in its stomach, had recently fed on a small woodrat (Neotoma).
Salvadora hexalepis klauberi
One specimen was collected in late afternoon while basking on a dirt road (Sta. 5). The
other individual was collected at 1630 hrs. as it foraged in an open sandy area (Sta. 36).
Scutellation, pattern and counts generally fit Klauber's (1946) description of the race.
The exceptions are as follows: SDSNH 45953 has 241 ventrals, much lower than the
range of 253-257 given by Klauber, and SDSNH 45954 has a tail-to-total-length ratio of
1971 BOSTIC: BAJA CALIFORNIA HERPETOFAUNA 259
0. 168, higher than the 0. 140 reported by Klauber for the race.
Tantillaplaniceps eiseni
A specimen, found dead on a sandy-dirt road adjacent to a flat sparsely vegetated
sandy area (Sta. 5) is the fifth of T! p. eiseni from the peninsula (Tanner 1966) and the first of
Tantilla from the Pacific side of central Baja California.
Scutellation and measurements of the specimen, an adult female, are as follows: ven-
trals, 176; caudals, 62; ventral-caudal total, 238; total length, 211; tail length, 30; ratio of tail
to total length, 0.142. These counts and measurements, excluding tail to total length ratio,
fall within the range recorded by Tanner (1966) for female eiseni; but the tail to total length
ratio of 0. 142 is considerably less than the range of 0. 1 78-0.256 reported by Tanner. Pattern
and coloration of the specimen fit Tanner's (1966) description of the subspecies.
DISCUSSION
This report treats 29 species of amphibians and reptiles from the Pacific coastal strand
of Baja California del Norte's Central Desert, including elements from three Peninsular
faunal zones; the Californian, the Colorado Desert District, and the Cape Region.
Only one species, Cnemidophorus labialis, may be considered to be endemic to the
Central Desert, and only if one considers the coastal region between Arroyo Santo Tomas
and 20 miles north of El Rosario to be Sonaran Desert. This area, based on the dominant
forms and composition of the flora and fauna appears to be Sonoran Desert (Short and
Crossin, 1967; Bostic, 1968). Since Shreve (1936) referred to this area as the Chaparral-
Sonoran ecotone many workers have arbitrarily included it within the California faunal
region.
The relative absence of endemic forms and the lack of species diversity support, in
part, the theory that during periods of glacial maxima the deserts of the Peninsula were aU
but eliminated, and that reconstitution of the desert herpetofauna occurred during glacial
minima (Savage, 1960).
Homogeneity of habitats and the moderate chmate of the Pacific coastal strand have
also been important factors in reducing species diversity. Savage (I960) listed 32 species
of amphibians and reptiles comprising his central peninsular assemblage, including two
amphibians, 16 lizards and 14 snakes. To this list may be added Hvla regilla desert icola,
Cnemidophorus labialis, Gerrhonotus multicarinatus ssp., Lichanura roseofusca gracia,
Tantillaplaniceps eiseni and Petrosaurus repens.
As I have delimited the Central Desert, Savage's inclusion of Scaphiopus couchi and
Dipsosaiirus dorsalis should be considered marginal. Both genera in Baja California del
Norte show a decided preference for mesquite and creosote bush deserts. These plant com-
munities are rare and never extensive in the coastal strand region. Only inland and south of
El Arco (below 28°N. latitude), where they were prominent, did we observe Dipsosaurus
dorsalis.
The following snakes, included by Savage in his Peninsular Desert assemblage, were
not recorded in the survey: Leptotvphlops humilis, Lichanura trivirgata. Arizona elegans,
Masticophis lateralis, Sonora mosaueri and Crotalus mitchelli. As pointed out by Myers and
Rand (1969), snakes are a herpetofaunal segment that is difficult to sample adequately, ow-
ing in part to their lower population densities and their behavioral and structural adaptions
designed to avoid discovery.
None of the five species reported for the first time from the Pacific slopes of the Central
Desert appear to be recent arrivals. They were probably overlooked during previous years
of faunal exploration.
Sauromalus australis and Petrosaurus repens, based upon current knowledge of their
260 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
distribution, ecological associations, and tolerances, appear to be contiguous with the pen-
insular populations. The scarcity of favorable habitats within the area surveyed preclude
their occurrence elsewhere.
Analysis of the distribution of Gerrhonotus multicarinatus spp., contrary to Savage's
(1960) interpretation, indicates a southward expansion of its range since glacial maximum
via the cool, moist coastal corridor. The same may be true for other temperate-tolerant
types such as Tantilla planiceps eiseni, Coleonyx variegatus abbotti and Lichanura rose-
ofusca gracia. The ranges of other temperate adapted forms from the Cape refugium, such
as Gerrhonotus paucicarinatus may be expanding northward via the Pacific coastal corridor.
The Crotaphvtus collaris of Mesa de San Carlos appear to represent an isolated popu-
lation. The discontinuity of favorable habitat and climatic conditions within the survey
area, excluding the Sierra Colombia with summits near 762 meters, together with the ap-
parent distributional gap between this population and the peninsula's east coast popu-
lations seem to support this view.
San Carlos Mesa is about 15 miles long in a northwestern and southeastern direction
by six miles wide and rises to an altitude of from 422 to 739 meters. The basaltic rock out-
crops around the edge and on the top of the mesa provide suitable habitat for C. collaris.
The mesa proper is a favorable habitat for the species. In contrast to the surrounding low-
lands and foothills, it is subject to a greater duration and intensity of solar radiation, and
concomitantly less frequent and shorter durations of coastal cloud cover and fog.
Phyllodactylus xanti sloani probably represents a marginal population of the mainland
stock that recently immigrated to the Pacific slope via the foothills of the southern extremity
of the Sierra de San Pedro Martir occupying marginal, but suitable habitats to the north
and south.
ACKNOWLEDGEMENTS
I wish to express my sincere gratitude to the Belvedere Scientific Fund of San Francisco for their financial
support of this survey.
I am indebted to Thomas Cozens for his field and laboratory assistance, and to Margery Stinson for her labo-
ratory and office assistance.
To Glen Contreras and Dennis Roberts for their able field assistance, I am especially grateful.
For the generous help, interest and encouragement of Allan J. Sloan, Curator of Herpetology, San Diego
Natural History Museum, I am particularly indebted.
Special thanks are extended to Reid Moran, Curator of Botany at the SDNHM for his identification of the
plants; to Elsie Arena for technical assistance; Bill Hite, Victor Limon and John Waldrup for field assistance; to
Mike Langdon and DeDe Miller for their illustrations of Fig. 1 and Figs. 2,3 and 4, respectively; to Ted Karounos
for Figs. 7 and 8; to Mr. and Mrs. Robert Eckhart for office and laboratory assistance; to Richard P. PhiUips, for his
letters of introduction which proved to be invaluable assets; to Richard G. Zweifel, American Museum of Natural
History, for the loan of Gerrhonotus from Baja California; to Charles Coutts and Eugene Stevens, Palomar Col-
lege, for placing the facilities of the Life Sciences Division at my disposal; and to Leon Rector for his excellent
preparation and maintenance of the vehicles used in the survey.
I am particularly grateful to James R. Dixon, Richard Ethridge, Reid Moran, Clifford H. Pope, Allan J. Sloan
and Ira L. Wiggins for their sound editorial comments and criticisms of the manuscript.
My sincerest appreciation is extended to Mrs. Norrine Keesee and Mrs. Sophie Bartlett who typed the final
body of the manuscript, and the tables, respectively.
1 am also thankful for the help and cooperation of Dr. Rodolfo Hernandez Corzo, Director General de la
Fauna Silvestre de la Secretaria de Agricultura y Ganaderia, who issued the collector's permit.
LITERATURE CITED
Arnold, B. A.
1957. Late pleistocene and recent changes in land forms, climate, and archaeology in central Baja Cahfornia.
Univ. California Publ. Geogr. 10(4): 201-318.
1971 BOSTIC: BAJA CALIFORNIA HERPETOFAUNA 261
Aschmann, H.
1959. The central desert of Baja California: Demography and ecology. Ibero-Amer. 42. Univ. California
Press, Berkeley. 282 p.
Banta. B. H., and W. W. Tanner.
1968. The systematics of Crolaphvtus wislizeni. the leopard lizards (Sauria: Iguanidae). Part 11. A review of
the status of the Baja California peninsular populations and a description of a new subspecies from
Cedros Island. Great Basin Nat. 28(4): 183-194.
Bogert, C. M.. and A. P. Porter.
1967. A new species of /l/)roAj/a( Sauria. Anguidae) from the Sierra Madre del Surof Oaxaca, Mexico. Amer.
Mus. Novitates 2279 : 1 -2 1 .
Bostic, D. L.
1968. Thermal relations, distribution, and habitat of Cnemidophorus labialis (Sauria: Teiidae). San Diego
Soc. Nat. Hist., Trans. 15(3): 21-30.
Burt, C. E.
1931. A study of the teiid lizards of the genus Cnemidophorus with special reference to their phylogenetic
relationships. U.S. Natl. Mus. Bull. 154: 1-286.
Dixon, J. R.
1966. Speciation and systematics of the Gekkonid lizard genus PhvUodactvlus of the islands of the Gulf of
California. Cahfornia Acad. Sci. Proc. 33( 13): 415-452.
1969. PhvUodactvlus xanti. Cat. Amer. Amphibians Reptiles: 79. 1-79.2
Fitch. H. S.
1938. A systematic account of the alligator lizards ('Gerr/jo^onwJ in the western United States and Lower Cali-
fornia. Amer. Midland Nat. 20(2): 381-424.
Gorman, G. C.
1965. The distribution of Lichanura trivirgata and the status of the species. Herpetologica 21(4): 283-287.
Hastings, J. R. (Ed) and R. R. Humphrey r£J.j.
1969. Climatological data and statistics for Baja California. Technical reports on the meteorology and cli-
matology of arid regions, no. 18 Tucson: Univ. Arizona Atmos. Phys.
Hastings, J. R., and R. M. Turner.
1964. Climatological data for Baja California. Technical reports on the meteorology and chmatology of arid
regions, no. 14. Tucson: Univ. Arizona Atmos. Phys.
1965a. Seasonal precipitation regimes in Baja California, Mexico. Geografiska Annaler 47 Ser. A: 204-223.
1965b. The changing mile: an ecological study of vegetation change with time in the lower mile of an arid and
semi-arid region. Univ. Arizona Press, Tucson. 2 1 7 p.
Jaeger, E. C.
1957. The North American deserts. Stanford Univ. Press. 308 p.
Jameson, D. L., J. P. Mackey, and R. C. Richmond.
1966. The systematics of the Pacific tree frog, Hvlaregilla. California Acad. Sci., Proc. 33( 19): 551-620.
Johnson, S. R.
1965. An ecological study of the chuckwalla, Sauromahis obesus Baird, in the western Mojave Desert. Amer.
Midland Nat. 73(1): 1-29.
Klauber. L. M.
1931a. Anewsubspeciesof the California boa, with notes on the genus L/c/ianwra San Diego Soc. Nat. Hist.,
Trans. 6(20): 305-318.
193 lb. Crotaliis tigris and Crotalus envo. two little known rattlesnakes of the southwest. San Diego Soc. Nat.
Hist.. Trans. 6(24): 353-370.
1933. Notes on Lichanura. Copeia(4): 214-215.
1945. The geckos of the genus Coleonvx with descriptions of new subspecies. San Diego Soc. Nat. Hist.,
Trans. 10(11): 133-2''l6.
1946. The gopher snakes of Baja California, with descriptions of new subspecies of Pitiiophis catenifer. San
Diego Soc. Nat. Hist., Trans. 11(1): 1-40.
1964. Rattlesnakes. Their habits, life histories, and influences on mankind. Univ. Cahfornia Press. 2 vol.
Linsdale. J. M.
1932. Amphibians and reptiles from Lower California. Univ. California PubL Zool. 38(6): 345-386.
McGinnies, W. G. (Ed). B. J. Goldman (Ed), and P. Paylore (Ed).
262 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
1968. Deserts of the world. An appraisal of research into their physical and biological environments. Univ.
Arizona Press. 788 p.
Meigs, P.
1953. World distribution of arid and semiarid homoclimates. In Reviews of research on arid zone hydrology.
Unesco, Paris Arid Zone Prog. 1 : 202-210.
1966. Geography of coastal deserts. Unesco, Paris. Arid Zone Research 28: 140 p.
Murray, K. F.
1955. Herpetological collections from Baja California. Herpetologica 11:33-48.
Myers, C. W., and A. S. Rand.
1969. Checklist of amphibians and reptiles of Barrow Colorado Island, Panama, with comments on faunal
change and sampUng. Smithsonian Contrib. Zool. ( 10) : 1-11.
Nelson, E. W.
1921. Lower California and its natural resources. Natl. Acad. Sci. 16, First Memoir. 194 p.
Phelan, R. L. and B. H. Brattstrom.
1955. Geographical vansition in Sceloporus magister. Herpetologica 1 1(1): 1-14.
Richmond, N. D.
1965. Distribution oi Gerrhonotus paucicarinatus Fitch. Copeia (3): 375.
Sauer, C. and P. Meigs.
1927. Lower Califomian studies. I. Site and Culture at San Fernando de Velicata. Univ. California Publ.
Geog. 2(9): 271-302.
Savage, J. M.
1952. Studies on the lizard family Xantusiidae I. The systematic status of the Baja California night lizards
allied to Xantusia vigilis, with the description of a new subspecies. Amer. Midland Nat. 48(2): 467-479.
1960. Evolution of a peninsular herpetofauna. In symposium: The biogeography of Baja California and adja-
cent seas. Syst. Zool. 9(3-4): 184-212.
Shaw, C. E.
1945. The chuckwallas, genus Sawrowa/w.?. San Diego Soc. Nat. Hist., Trans. 10( 15): 269-306.
Short, L. L., Jr., and R. Crossin.
1967. Notes on the avifauna of northwestern Baja California. San Diego Soc. Nat. Hist., Trans. 14(20): 281-
300.
Shreve, F.
1936. The transition from desert to chaparral in Baja California. Madrono 3: 357-264.
1942. The desert vegetation of North America. Bot. Rev. 8(4): 195-246.
Shreve, F. and I. L. Wiggins.
1964. Vegetation and flora of the Sonoran Desert. Stanford, Stanford Univ. Press. I: 1-840.
Smith, H. M. and E. H. Taylor.
1950. An annotated checklist and key to the reptiles of Mexico exclusive of the snakes. U.S. Natl. Mus. Bull.
199: 1-253.
Tanner, W. W.
1944. A taxonomic study of the genus Hypsiglena. Great Basin Nat. 5(3-4): 25-92.
1966. The night snakes of Baja California. San Diego Soc. Nat. Hist., Trans. 14(15): 189-196.
Tevis, L.
1944. Herpetological notes from Lower Cahfomia. Copeia ( 1): 6- 18.
Van Denburgh, J.
1922. The reptiles of western North America. Occ. Papers Cahfomia Acad. Sci. (10): 1-1028.
Walker,;. M.
1966. Morphology, habitat and behavior of the teiid lizard. Cnemidophorus labialis. Copeia (4): 644-650.
Walker, J. M. and H. L. Taylor.
1968. Geographical variation in the Teiid lizard Cnemidophorus hyperythnis. I. The caeruleus-Mke subspecies.
Amer. Midland Nat. 80( ! ) : 1 -27.
Wiggins, I. L.
1960a. Investigationsin the natural history of Baja California. California Acad. Sci., Proc.30(l): 1-45.
1960b. The origin and relationships of land flora. In symposium: The biogeography of Baja California and
adjacent seas. Syst. Zool. 9(3-4): 148-165.
1971 BOSTIC: BAJA CALIFORNIA HERPETOFAUNA 263
1969. Observations on the Vizcaino Desert and its biota. California Acad. Sci., Proc, 4th series 36(1 !)• 317-
346.
Life Sciences Department, Palomar College, San Marcos, California 92069
i/lliis. CO.viK ZCCL.
LIBRARY
FEB 1 8 1972
HARVARD
UNIVERSITY
A NEW GENUS OF CHTHAMALIDAE
(CIRRIPEDIA) FROM THE SOUTHEASTERN
PACIFIC ISLAND OF SAN AMBROSIO
ARNOLD ROSS
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 1 1 26 OCTOBER 1971
A NEW GENUS OF CHTHAMALIDAE
(CIRRIPEDIA) FROM THE SOUTHEASTERN
PACIFIC ISLAND OF SAN AMBROSIO
ARNOLD ROSS
ABSTRACT .—Jehlius gilmorei n. gen., n. sp. is proposed for a chthamalid apparently endemic to Isla San
Ambrosio, a volcanic island about 800 km west of Chanaral, Chile. This new barnacle has a grade of shell
construction transitional between 6 and 4 plates.
From 15 May through 6 July 1970 the U.S. Antarctic Research Program (USARP)
trawler Hero cruised the southwest and central coasts of Chile, and visited Isla Robinson
Crusoe of the Juan Fernandez group, and Islas San Ambrosio and San Felix of the Des-
venturados group. The cruise objectives were to obtain data on marine mammals and birds.
But at my request barnacles were collected as opportunity permitted. Gilmore (1971: 10)
gave a preliminary report of this cruise.
The Islas de los Desventurados include the oceanic islands of San Ambrosio and San
Felix, together with a lesser rock, Gonzalez, at about 26° south and 80° west, or approx-
imately 800 km off the coast of Chanaral, Chile (Fig. 1). These volcanic islands rise some
4000m from the sea floor. The surface waters here have a salinity of about 34.5% and an
average surface temperature during February-March of 20°-21° C, and during July-Sep-
tember of 17°-18° C (Meteorological Office, 1956; Murphy, 1936: 104; Wyrtki, 1966: 40).
San Ambrosio, type locality for the new chthamalid described herein, is about 4 km long
and 1 km wide with an estimated maximum elevation of 480 m (Fig. 2).
PREVIOUS STUDIES ON THE BIOTA
The biota of the Desventurados islands remains poorly known owing to their relative
inaccessibility and the lack of good landing sites (Fig. 3; see Douglas, 1970: 345). On the
basis of a short visit, Bahamonde N. (1966) presented a popular, broad, and general account
of the biota.
Studies on the flora were published by Johnston (1935) and by Skottsberg (1937, 1952),
both of whom Usted references to earlier studies. The avifauna was treated by Murphy
(1936) and by Johnson (1965, 1967), who also cited earlier references. Allen (1899) dis-
cussed briefly the hunting and virtual extermination of fur seals {Arctocephalus) in rook-
eries on the two major islands (see also Gilmore, 1971: 10), and Kellogg (1943: 306) pres-
ented data on the size of the catch during the early years of American sealing in these wa-
ters. Other studies are those by Serafy (1971: 165) who described a new Clypeaster from San
Felix, and by McLean (1970: 362), who described two new fissurelUd gastropods.
The only mention of the crustacean fauna of the island with which I am familiar is by
Bahamonde N. (1966: 7) who stated "En la zona supramareal hay una franja muy nitida de
Cirripedos. En sus cercanias es posible capturar ejemplares de la 'jaiba corredora' {Lepto-
grapsus variegatus), designada por Philippi como Grapsus obscurus, por su coloracion. Alh
es muy abundante. Tambien se halla habitualmente en las pozas profundas ejemplares de
Rhynchocinetes balsii y en las areas en que predomina las algas de los generos Padina y
Corallina se obtuvieron individuos de Plagusia chabrus.''
CHTHAMALID COLONIZATION OF ISLA SAN AMBROSIO
Under the influence of the west wind drift, South Pacific Temperate Water flows east
SAN DIEGO SOC. NAT. HIST., TRANS. 16(11): 265-278, 26 OCTOBER 1971
266
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
1^
05
80° W
1^
55
Rocas Catedral de
Peterborough
San Felix
O Islo
te Gonzalez
San Ambrosio
Roca Conico
•Bahi'a Covadonga
/^ ' Roca Mas Afuere
Roca Bass
05
I
80 W
_|
SCALE 1250000
Figure \. Map showing position of Isla San Ambrosio relative to other islands in the Desventurados group,
Chile.
toward South America (Wyrtki, 1968: 131). Near Chile at about 50° S this water mass di-
vides, one branch turning south and eastward around the tip of South America, the other
flowing northward along the coast as the Peru Current ( = Humboldt Current) or the Peru-
Chile Current System. The offshore Desventurados Islands are under the influence of this
current system. The Peru Current extends as far north as Ecuador and then swings west just
south of the Equator to become part of the South Equatorial Current. The northward flow
of the Peru Current, generally at 25° S, is divided into two components, the Peru Coastal
Current and the Peru Oceanic Current, between which is a southward moving subsurface
current, the Peru Countercurrent, which carries equatorial subsurface water as far as 22° S
(Wyrtki, 1966: 59; 1968: 121).
The prevailing north-flowing currents argue for colonization of San Ambrosio from
the southeast, much as the biota of the Juan Fernandez Islands, in the main, also appears to
have been derived from South America. I have discounted a direct Australia-New Zealand
origin of the Desventurados chthamalid largely because of the vast distance separating the
two regions, the apparent absence of any living or extinct populations of chthamahds in the
region between, and because the temperate Southeast Pacific chthamalids have their great-
est affinity with the Tropical American fauna (ZuUo, 1966: 142). Elminius and Austroba-
lanus in the southeastern Pacific, although seemingly good indicators of biogeographical
affinities, are two groups that remain poorly known (the type species of Austrobalanus is
apparently a six-plated tetraclitid and the remainder true balanids; Elminius until recently
contained two species referable to the tetraclitid Epopella and the remaining two or three
widely separated species offer no clues as to their origin [Ross, 1970: 9]).
Based on morphological and hydrographic evidence, this new chthamalid probably
evolved from or shared a common ancestry with Chthamalus cirratus Darwin, 1854, which
1971
ROSS: A NEW GENUS OF CHTHAMALIDAE
267
Figure 2. View from the northwest of Isla San Ambrosio. Small prominence to the right of San Ambrosio is
Roca Conico. Photo by R. M. Gilmore.
occurs commonly along the west coast of South America from the Chonos Archipelago in
Chile (about 45° S) to Guayaquil, Ecuador (2° 13' S) (Pilsbry, 1916: 321; Nilsson-Cantell,
1957: 11).
Three possible modes of colonization are offered in what I believe to be increasing
probability, 1) introduction by or through an agency of man, 2) larval colonization, and 3)
adult colonization by natural drift or rafting. I have discounted the first because Chtha-
malus and its derivatives are essentially shore barnacles, although they are known to foul
marine structures. Also, the evolutionary state of this new species argues for colonization
prior to the origin of man in the new world. I also doubt that the islands were colonized by
larvae, because the nauplii of Chthamalus and other balanomorphs in general lack the long
tomentose flotation setae characteristic of pelagic species, and because the nauplii of inter-
tidal barnacles probably remain in the plankton less than two weeks, which is apparently
not long enough to reach San Ambrosio. The efficacy of natural rafting is well documented
in the Hterature, and it appears most probable that colonization of San Ambrosio was ef-
fected by rafting.
As Crisp and Southward concluded (1953: 209), even narrow seas pose a barrier to
animals that are predominantly intertidal. The relatively small size of the two major islands
in the Desventurados Group (San Ambrosio— 4 km long, 1 km wide; San Felix— 3 km long,
1 km wide), and their great distance from the South American mainland, would tend to
preclude repetitive colonization from the mainland. Many workers have remarked that in
order to estabUsh a viable population some minimum density is required. However, since
there is good evidence that many species of Chthamalus are readily capable of self fertiliza-
tion under certain conditions (Barnes and Barnes, 1958: 550), the initial propagule could
have been only a single individual.
268
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 3. View of landing site at Bahi'a Covadonga, Isla San Ambrosio. The two wooden shacks are used by
transient lobster fishermen. Photo by R. M. Gilmore.
1971 ROSS: A NEW GENUS OF CHTHAMALIDAE 269
SYSTEMATICS
Family Chthamalidae Darwin, 1854
Remarks —The new taxa described below are assigned to this family, which was diag-
nosed recently by Newman, Zullo, and Withers (1969: 283) and emended subsequently by
Newman and Ross (1971: 139). The assignment of genera to this family differs in several
details between that of the above workers and that proposed by Utinomi (1968: 36). Type
Genus— Chthamalus Ranzani, 1817 (for Lepas stellatus Poh, 1791, by original designation,
Recent, Bay of Naples, Italy).
KEY TO GENERA OF LIVING CHTHAMALIDAE
1. Shell composed of 8 parietal plates 2
1. Shell composed of 6 or 4 parietal plates 5
2. Shell with 2 or more whorls of basal plates 3
2. Shell without whorls of basal plates 4
3. Shell with 6-8 whorls of basal plates; caudal
appendages lacking (1 sp.) Catomerus
3. Shell with 2-6 whorls of basal plates; caudal
appendages present (2 spp.) Catophragmus
4. Shell in young individuals with eight plates, in older
individuals with 6 or 4; mandible quadridentoid;
cirrus III more like cirrus II than IV; caudal
appendages present (8 spp.) Pachylasma
4. Shell never with fewer than 8 plates, mandible
tridentoid; cirrus III more like cirrus IV than II,
caudal appendages lacking (3 spp.) Octomeris
5. Shell with a single whorl of basal plates (1 sp.) Chionelasmus
5. Shell without whorls of basal plates 6
6. Mandible tridentoid (12 spp.) Euraphia
6. Mandible quadridentoid 7
7. Shell with inflected basal rim (1 spp.) Tetrachthamalus
7. Shell without inflected basal rim 8
8. Shell of adult with 6 wall plates (13 spp.) Chthamalus
8. Shell of adult with 4 wall plates, or transitional
between 6 and 4 wall plates 9
9. Wall plates coalescing in juvenile stage; scutum with
adductor ridge; anterior cirri armed with grapple-like
spines (3 spp.) Chamaesipho
9. Wall plates coalescing in adult stage; scutum without
adductor ridge; anterior cirri lacking grapple-like
spines (1 sp.) Jehlius
Jehlius n. gen.
Definition.— SheW of adult in transitional stage between 6 and 4 plates; reduction in
number of plates by fusion rather than exclusion; in 4 plated stage wall plates not second-
arily coalesced; plates disposed asymmetrically or symmetrically; fusion pattern variable
throughout population; compartments lacking radii and inflected basal rim; basis mem-
branous; scutum with well defined depression for adductor muscle, but no adductor ridge;
cirrus III structurally and probably functionally more similar to cirri IV-VI than to cirrus II;
270
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
cirrus II lacking grapple-like spines; caudal appendages lacking; mandible with four teeth,
basal comb, and spine-like inferior angle.
Type species.— J ehlius gilmorei new species.
Remarks.— J ehlius is unique in that the parietal plates are not always symmetrically
disposed and that the pattern of fusion of the plates has not become fixed at least within the
few specimens available for study (Fig. 4). Also, shell development is obviously transitional
between six and four plates, and in the two specimens with only four plates there is no in-
dication whatsoever that these plates will secondarily coalesce.
The shell of Chamaesipho is fundamentally composed of six plates, rostrum, carina,
and paired rostrolaterals and laterals, but the genus is regarded as tetramerous (see Moore,
1944; Pope, 1965; Newman, ZuUo and Withers, 1969). The six sutures separating these
plates in C. columna are rarely seen in individuals beyond 2 mm in rostro-carinal diameter
and occasionally they are obliterated in individuals as small as 0.5 mm in diameter (Moore,
1944: 317). In C bninnea, on the other hand, the sutures delimiting the plates are no longer
visible by the time individuals reach 6 mm in diameter (Moore, 1944).
In both species o^ Chamaesipho mentioned above, the rostrolaterals are united with the
laterals. In specimen 1 ofJehlius (see Fig. 4), the arrangement of the wall plates appears to
be the same as in Chamaesipho columna. In specimen 3, the right lateral is fused and sec-
ondarily coalesced in part with the carina, but the left lateral is fused with the rostrolateral.
In specimen 2 the right lateral and rostrolateral are fused with the rostrum, and the left
Pachylasma Tetrachthamalus Chamaesipho
(^ O B O
Pac
hylasrra
Oc
tomens
Catt
phragmus'
I
X",
)
CHTHAMALIDAE
e) (J o
Figure 4. Plan views of wall construction in the Chthamalidae. Numerals at left indicate grades of decreasing
complexity. Shell outlines ofJehlius on right side are camera lucida drawings of the internal surface showing de-
gree of development of sutures, which are not readily discernible on external surface of shell (right side of shell
is on left side in drawing). Number in center of orifice refers to number in plan view below. Specimen No. 3 is
holotype, S.D.S.N.H. No. 4003/3.
1971
ROSS: A NEW GENUS OF CHTHAMALIDAE
271
rostrolateral and lateral remain separate. Specimen 4 is the most unusual of the lot. It has
one major suture, between the rostrum and the right rostrolateral, and all of the other plates
are partially coalesced (Fig. 5). Aside from the unusual arrangement of the wall plates there
is nothing to suggest that the shells are pathologically malformed.
Jehlius also differs from Chamaesipho in the articulation of the opercular plates. The
junction between the scutum and tergum on each side, when viewed internally, in Chamae-
sipho takes the form of the Greek letter omega, but in Jehlius it is simpler and only slightly
sinuous. Jehlius also differs in that cirri I-III lack the grapple-like spines and the scutum
lacks an adductor ridge but has a well defined deep pit for the insertion of the adductor
muscle (Fig. 6).
Jehlius obviously is derived from an Eastern Pacific stock of Chthamalus, whereas
Chamaesipho probably was derived from an Indo-Pacific stock. Furthermore, Chamaesipho
is restricted to the austral region and the probability of penetrating the East Pacific barrier
is remote.
Tetrachthamalus, also a genus with four plates that evolved from Chthamalus, differs
from Jehlius in that the rostrolaterals are fused with the rostrum to form a tripartite plate,
and during the ontogeny of individuals in this genus the four plates coalesce.
Etymology— Earned for Dr. Joseph R. Jehl, Jr., San Diego Museum of Natural His-
tory, longtime friend and colleague, and collector of the specimens reported on herein.
Jehlius gilmorei n. sp.
Diagnosis— Crest of labrum armed with 50-60 simple conical teeth; cutting edge of
maxilla II with 10-13 long spines in medial cluster; intermediate articles of posterior cirri,
which have rami of equal length, bear 5 pairs of setae; basal segment of anterior ramus of
cirrus I armed with stout spines.
Description.— Shell white or grayish-white, low conic, broadly ovate to subcircular in
outline; basal portion of compartments ribbed and periphery of shell irregular or strongly
toothed (Fig. 5); upper portion of external surface corroded, exfoliating; aperture relatively
large owing to corrosion; radii lacking; sheath less than Va height of compartments, basal
margin not depending; surface below sheath smooth. Basis membranous.
Figure 5. Jehlius gilmorei n. gen., n. sp. External and internal views, respectively, of shell. Paratype, S D.S.N.H.
No. 4004/4; actual rostro-carinal diameter, 8. 1 mm.
272
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Scutum transversely elongated (Fig. 6); length about '/s greater than height; external
surface poorly preserved, exfoliating; only last 3-4 newly formed growth ridges preserved
along basal margin of plate; articular ridge poorly differentiated from articular surface; ad-
ductor ridge absent; depression for adductor muscle deep, well delimited; depression for
lateral depressor muscle deep, well defined, crossed by 3-4 septa; depression for rostral
depressor muscle commonly shallow, poorly delimited; apical portion of plate lacking
ridges, crests, or pits.
Tergum higher than wide (Fig. 6); external surface poorly preserved, exfoliating; ex-
ternal longitudinal furrow apparently lacking; spur rounded or pointed distally, and not
distinctly separated from articular margin; articular ridge low, poorly developed; parallel
and immediately adjacent to articular ridge there is a row of shallow, oblong pits; there are
2 prominent and 1-2 lesser crests for the insertion of the lateral depressor muscle; apical
portion of plate either slightly pitted or roughened.
Measurements of the holotype (in mm) are as follows: rostro-carinal diameter 9.7. lat-
eral diameter 9.1, height 5.1, rostro-carinal diameter of orifice 5.2, height of scutum 3.0,
width of scutum 3.9, height of tergum 2.8, width of tergum 2.0. The range in rostro-carinal
diameter of the four specimens is 8.1-10.1 (x = 9.2), and the range in height is 2.6-5.1 (x =
3.3).
Figure 6. Opercular plates of Jehlius gilmorei n. gen., n. sp. External views of scutum and tergum. respectively
(top row), and internal views of scutum and tergum, respectively (bottom row). Paratype, S.D.S.N.H. No.
4004/2. Drawings by Anthony D'Attilio.
1971
ROSS: A NEW GENUS OF CHTHAMALIDAE
273
Figure 7. Jehlius gilmorei n. gen., n. sp. a, right mandible; b, left mandible: c, maxilla II; d, intermediate articles
of cirrus VI; e, crest of labrum; f, maxilla I. Holotype, S.D.S.N.H. no. 4003/3.
Crest of labrum thin, with broad U-shaped medial notch toothed its whole width; teeth
50-60, close spaced, simple, conical; bristles behind and parallel to teeth along crest short
and densely packed (Fig. 7). Palps elongate, rounded distally, the basal margin convex and
free of setae; superior margin densely clothed with coarsely bipinnate, long, slender setae;
setae on distal extremity longer than on proximal, and finely bipinnate. Cutting edge of
mandible armed with 4 teeth, basal comb, and spine-like inferior angle; teeth 2-4 bicuspate;
comb between tooth 4 and inferior angle with 50-60 acicular teeth (Fig. 7). Maxilla I with 2
long stout and 1-2 shorter stout spines above subapical notch, 4-5 short slender spines in
notch, 10-13 long stout spines medially, 14-20 slender spines in basal cluster (Fig. 7). Cut-
ting edge of maxilla II distinctly bilobate; setae along apical margin long, finely bipinnate,
setae progressively shorter toward notch; notch free of setae; setae on basal lobe finely bi-
pinnate (Fig. 7).
Anterior ramus of cirrus I about 1/5 longer than posterior ramus; intermediate articles
of both rami about twice as broad as high; proximal segment of anterior ramus armed with
5 or 6 short, stout spines along posterior border (Fig. 8); 1 row of coarse ctenae present on
lateral face of segments of each ramus immediately below articulation; ctenae better devel-
oped on posterior ramus than on anterior ramus; setae on both rami bipinnate. Rami of
cirrus II essentially equal in length and about same length as rami of cirrus I; 1 row of coarse
ctenae present on lateral face of segments of both rami immediately below articulation;
setae on both rami bipectinate. Cirri III-VI essentially equal in length and with equal
rami; 1-2 long slender, and 1-2 shorter slender setae at each articulation along greater cur-
274
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
0.^ nini
ANTERIOR RAMUS
T,''
y
Y'\
f
k/
n.e 5 5 5 6 8
I II III IV V VI
mean
no.
POSTERIOR RAMUS
;--
/
/
1 II III IV V
5
VI
Figure 8. Outline drawings of cirri I-VI (setae omitted; paratype, S.D.S.N.H. No. 4004/2) and summary of data
on cirral counts for the holotype and two paratypes.
vature of intermediate articles; 1 row of ctenae on lateral face of intermediate segments
below articulation; setation ctenopod, with 5 pairs of setae on each intermediate segment;
at base of each pair of setae there are 2-3 short, slender setae. Cirral counts for specimens in
the type lot are summarized in Figure 8.
Intromittent organ annulated throughout its length, and sparsely covered with short,
slender bristles; distal extremity bilobed and each lobe bearing about 15 or 16 short, slender
setae.
Type Locality— On northeast coast, just west of Bahia Covadonga, Isla San Ambrosio,
Islas de los Desventurados, Chile, approximately 26° 20' 15" S., 79° 15' 45" W., I. M. W. SG
17; intertidal on volcanic rock; J. R. Jehl, Jr. coll., 27 June 1970; USARP cruise 70-3.
Disposition of types.— The holotype and three paratypes are housed in the collections of
the San Diego Society of Natural History, Marine Invertebrate catalogue numbers 4003/3
and 4004/ 1, 4004/2 and 4004/4 respectively.
Etymology.— The specific epithet honors Dr. Raymond M. Gilmore, Research Associ-
ate, San Diego Natural History Museum, and chief scientist aboard the trawler Hero during
USARP cruise 70-3.
INTRAFAMILIAL RELATIONSHIPS
Four families are presently recognized within the suborder Balanomorpha, namely
1971 ROSS: A NEW GENUS OF CHTHAMALIDAE 275
Chthamalidae, Bathylasmatidae, Tetraclitidae, and Balanidae (see Newman and Ross,
1971: 137). Of these, the Chthamalidae are more generalized structurally and appear in
the fossil record before any of the others.
Within the Balanomorpha the evolutionary history has been one of reduction in the
number of compartments composing the shell (Pilsbry, 1916: 291; Withers, 1928: 46). In
the Chthamalidae this reduction has been accompanied further by structural modification
of the mouth parts and cirri for feeding (Zullo, 1963: 190).
Based on the probable mode of reduction in the number of shell elements, two Hne-
ages are evident in the Chthamalidae. In the first, consisting solely of Pachvlasma, the
shell initially contains 8 plates, including rostrum, carina, and paired rostrolaterals, later-
als, and carinolaterals. Subsequently, the rostrolaterals coalesce with the rostrum forming
a tripartite plate, and the carinolaterals may coalesce with the laterals yielding a shell of
only 4 plates (Fig. 4).
The second lineage {Octomeris-Chthamahis group) includes the remaining genera
(Fig. 4; see Newman and Ross, 1971: 141; cf. Utinomi, 1968: 36). Of these, Catophragmus
(including the subgenera Catomerus and Pachvdiadema) and Octomeris have the same
number and arrangement of the plates as does Pachvlasma. Early in the evolution of this
lineage, the number of shell elements was reduced through eUmination or exclusion since
Chthamalus has only six wall plates (rostrum, carina, and paired rostrolaterals and later-
als), the carino-laterals lacking. The small size of the carina and the presence of alae point
to reduction by exclusion. From Chthamalus a further reduction in the number of wall
plates, by fusion, is evident in Tetrachthamalus and Chamaesipho. In Tetrachthamahis the
rostrolaterals are fused with the rostrum, as shown by the size of the composite plate and
by the fact that it has radii, thus forming a tripartite plate essentially similar to that in
Pachvlasma and the bathylasmatid Tessarelasma. In Chamaesipho columna and C.
bnmnea the shell initially contains six plates; the rostrolaterals fuse with the laterals
(Withers, 1928: 45; Moore, 1944: 324) rather than with the rostrum as in Tetrachtha-
malus. By the time individuals of C. columna reach a rostrocarinal diameter of 2 mm and
individuals of C. bnmnea a diameter of 5-6 mm, all the plates coalesce secondarily, and
the sutures are obliterated. In adults of Tetrachthamalus oblitteratus, which reach a rostro-
carinal diameter probably not much greater than 6 mm, the sutures are commonly dis-
tinct; but then coalescence occurs, and remnants of these sutures can be observed in the
sheath (Newman, 1967: 427).
In all chthamalids with 6 or 4 plates fusion of shell elements and their subsequent
coalescence proceeds in a uniform manner. To judge from the specimens available, this
apparently is just the opposite of what takes place in Jehlius (Fig. 4). Although two speci-
mens of Jehlius have in part attained a grade of construction comparable with that found
in 4-plated individuals of Chamaesipho, two specimens are effectively intermediate be-
tween six and four plates. In the two specimens that have attained a 4-plated grade of
construction, there is no secondary coalescence and obliteration of the sutures uniting
these wall plates.
In his classification of the chthamalids Zullo (1963:190) stressed the modification in
mandibular and cirral structures attending the reduction in number of the wall plates. In
Octomeris, Chthamalus, Chamaesipho, Tetrachthamalus and Jehlius the mandible is char-
acteristically quadridentoid, but in Catophragmus, Catomerus, Chionelasmus, Euraphia
and Pachvlasma it is tridentoid. In the Octomeris-Chthamalus lineage the third cirrus is
relatively unmodified; but in the Pachvlasma lineage, feeding adaptations involve the
modification of cirrus III as a mouth appendage, such as is found in the balanids.
Pachvdiadema from the Cretaceous (U. Senon.) of Sweden is the oldest known
276 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
chthamalid with eight wall plates and at least two whorls of imbricating basal plates. The
number and arrangement of wall plates, simple opercular valves, caudal appendages, and
unmodified third cirrus all tend to link Pachvdiadema with the scalpellid lepadomorphs
(Newman, ZuUo, and Withers, 1969: R 269). '
Pachvdiadema is probably ancestral to Catomerus (Withers, 1935: 390; Pope, 1965:
15), which also possesses eight wall plates and several whorls of imbricating plates. Ca-
tophragamus also may have been derived from Pachvdiadema, or possibly from Cato-
merus. The presence of caudal appendages in Catophragamus suggests derivation from
Pachvdiadema rather than from Catomerus which lacks these appendages. Chionelasmus
with but six wall plates (carinolaterals lacking) and a single whorl of basal plates, and
with caudal appendages, is probably an off-shoot from Catophragmus.
Octomeris lacks the basal whorls of plates and caudal appendages, and hence is prob-
ably derived from Catomerus, which also lacks caudal appendages, and the articulation of
the opercular plates is simple rather than complex as it is in Catophragmus. From Octo-
meris it is a single step, through loss of the carinolaterals, to Chthamahts and Euraphia,
which probably share a common ancestry. However, Euraphia has retained the lepa-
domorph or early chthamalid tridentoid mandible whereas Chthamahts has evolved the
quadridentoid mandible with a basal comb.
Chamaesipho evolved from Chthamahts (Newman, 1967: 431), and probably rather
recently. Although young individuals of Chamaesipho brumiea and C. cohunna develop ros-
trolateral plates initially, these soon fuse with the laterals forming a shell with only four
plates; later the sutures coalesce, and are obliterated. In Euraphia the plates apparently
never coalesce but they do develop an inflected basal rim (see Newman, 1961). Tet-
rachthamahis is also an offshoot from Chthamalus (Newman, 1967: 431) but apparently of
greater antiquity than Chamaesipho. In Tetrachthamahts there is no evidence in the
ontogeny of a stage having six plates as in Chamaesipho, but as in Chamaesipho the plates
eventually coalesce. The wall plates in Tetrachthamahis, unlike those in Chamaesipho and
Jehlius, develop an inflected basal rim. Jehlius is apparently the most recent offshoot from
Chthamahis, and is most closely related to C cirratus.
ACKNOWLEDGMENTS
For the loan or gift of comparative materials I thank Elizabeth Pope, The Australian Museum, Brian Fos-
ter, University of Auckland, William A. Newman, Scripps Institution of Oceanography (S.I.O.), and Meredith
L. Jones, Smithsonian Institution. I thank my personal physician Dr. Wayne L. Heath and his assistant Mrs. Su-
san D. Dobbin for providing me with x-rays of the specimens. Mrs. Marguerette Schultz, S.I.O., brought to my
attention the recent studies by K. Wyrtki, and my wife Cecelia, S.I.O., helped me locate bathythermograph data
from the ships Yelcho, Anion Briinn and Esmeralda. For criticisms and comments on manuscript copy, or other
courtesies, I thank W. A. Newman, S.I.O., Joseph R. Jehl, Jr., Raymond M. Gilmore, Reid Moran, and Dwight
W. Taylor, San Diego Natural History Museum.
LITERATURE CITED
Allen, J. A.
1899. Fur-seal hunting in the southern hemisphere, p. 307-319. In. Jordan, D. S.. et al.. The Fur seals and
fur seal islands of the north Pacific Ocean. Part III. Special papers relating to the fur seal and to the
natural history of the Pribilof Islands. Washington, Gov't. Print. Off.
Bahomonde N., N.
1966. Islas Desventuradas. Mus. Nac. Hist. Nat. Chile, ser. Educ. 6: 1-15.
Barnes, H., and M. Barnes
1958. Further observations on self-fertilization in Chthamalus sp. Ecology 39(3): 550.
Crisp, D. J., and A. J. Southward
1953. Isolation of intertidal animals by sea barriers. Nature 172(4370): 208-209.
Douglas, G.
1970. Draft check list of Pacific oceanic islands (foreward by E. M. Nicholson). Micronesia 5(2): 327-463.
1971 ROSS: A NEW GENUS OF CHTHAMALIDAE 277
Gilmore, R. M.
1971. Observations on marine mammals and birds off the coast of southern Chile, early winter 1970. An-
tarctic J. United States 6(1): 10-11.
Johnson, A. W.
1965-1967. The birds of Chile and adjacent regions of Argentina, Bolivia and Peru. Vol. 1, 1965: vol. 2,
1967. Buenos Aires, Piatt Establicimientos Graficos S. A.
Johnston, I. M.
1935. The flora of San Felix Island. J. Arnold Arbor. 16(4): 440-447.
Kellogg, R.
1943. Past and present status of the marine mammals of South America and the West Indies. Ann. Rept.
Smithsonian Inst. 1942: 299-316.
McLean, J. H.
1970. Descriptions of a new genus and eight new species of Eastern Pacific Fissurellidae, with notes on
other species. Veliger, 12(3): 362-367.
Meteorological Office
1956. Monthly meteorological charts of the eastern Pacific Ocean. London, H.M.S.O., M.O. 518: 1-122 (not
seen).
Moore, L. B.
1944. Some intertidal sessile barnacles of New Zealand. Trans. Roy. Soc. New Zealand 73(4): 3 15-334.
Murphy, R. C.
1936. Oceanic birds of South America. Amer. Mus. Nat. Hist. Vol. 1, 640 p.
Newman, W. A.
1961. On the nature of the basis in certain species of the Hembeli section of Chthamalus (Cirripedia, Thora-
cica). Crustaceana 2(2): 142-150.
1967. A new genus of ChthamaHdae (Cirripedia, Balanomorpha) from the Red Sea and Indian Ocean. J.
Zool. London 153: 423-435.
Newman, W. A., and A. Ross
1971. Antarctic Cirripedia. Vol. 14. Antarctic Research Series, Amer. Geophys. Union. 257 p.
Newman, W. A., V. A. Zullo and T. H. Withers
1969. Cirripedia, p. 206-295. In, R. C. Moore (ed.), Treatise on Invertebrate Paleontology, Part R, Arthro-
poda 4.
Nilsson-Cantell, C. A.
1957. Thoracic cirripeds from Chile. Reports of the Lund University Chile Expedition 1948-49. Lunds
Univ. Arsskrift 53(9): 1-25.
Pilsbry, H. A.
1916. The sessile barnacles (Cirripedia) contained in the collections of the U.S. National Museum: in-
cluding a monograph of the American species. U.S. Nat. Mus. Bull. 93: 1-366.
Pope, E. C.
1965. A review of Australian and some Indomalayan Chthamalidae (Crustacea: Cirripedia). Proc. Linnean
Soc. New South Wales 90( 1): 10-77.
Ross, A.
1970. Studies on the Tetraclitidae (Cirripedia: Thoracica): a proposed new genus for the austral species
Tetraclita purpurascens breviscutum. San Diego Soc. Nat. Hist., Trans. 16(1): 1-12.
Serafy, D. K.
1971. A new species of Clypeaster (Echinodermata, Echinoidea) from San Fehx Island, with a key to the
Recent species of the Eastern Pacific Ocean. Pacific Sci. 25(2): 165-170.
Skottsberg, C.
1937. Die flora der Desventuradas-inseln (San Felix und San Ambrosio). Gotesborgs Kungl. Vetensk. Vit-
terh. samhalles Handl., ser. B. 5(6): 1-88.
1952. Weiter Beitrage Zur flora der Insel San Ambrosio . . . Arkiv fur Botanik, n.s., I (not seen).
Utinomi, H.
1968. A revision of the deep-sea barnacles Pachvlasma and Hexelasrna from Japan, with a proposal of new
classification of the Chthamalidae (Cirripedia, Thoracica). Publ. Seto Mar. Biol. Lab. 16(1): 21-39.
Withers, T. H.
1928. Catalogue of fossil Cirripedia in the Department of Geology. Vol. 1. Triassic and Jurassic. Brit. Mus.
(Nat. Hist.).
1935. Catalogue of fossil Cirripedia in the Department of Geology. Vol. 2. Cretaceous. Brit Mus. (Nat.
Hist.).
Wyrtki, K.
1966. Oceanography of the eastern Equatorial Pacific Ocean. Oceanogr. Mar. Biol. Ann. Rev. 4: 33-68.
278 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
1968. Circulation and water masses in the eastern Equatorial Pacific Ocean. Intl. J. Oceanol. Limnol. 1(2):
117-147.
Zullo, V. A.
1963. A classification and phylogeny of the Chthamalidae (Cirripedia: Thoracica). Proc. 16th Internatl.
Congr. Zool., Washington, 1: 190.
Department of Invertebrate Paleontology, Natural History Museum, P. O. Box 1390,
San Diego, California 92112
LIBRARY "*
FEB 1 8 1972
HARVARD
THE LARVAL AND PUPAL STAGES OF FOUR SPECIES OF
CAFIUS (COLEOPTERA: STAPHYLINIDAE) WITH NOTES
ON THEIR BIOLOGY AND ECOLOGY
GARY J. JAMES, IAN MOORE AND E.F. LEGNER
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 12 5 NOVEMBER 1971
THE LARVAL AND PUPAL STAGES OF FOUR SPECIES OF
CAFIUS (COLEOPTERA: STAPHYLINIDAE) WITH NOTES
ON THEIR BIOLOGY AND ECOLOGY
GARY J. JAMES, IAN MOORE, AND E. F. LEGNER
ABSTRACT.— Staphylinid beetles of the genus Cafiiis live in and under piles of decaying seaweed on
beaches in southern California. Seven species (seminitens. canescens. luteipennis. lithocharinus. decipiens,
opacus. sulcicollis) occur together in this habitat. Their food consists largely of fly (Fucellia) larvae and
pupae, although some were seen to prey upon amphipods and barnacles and scavenge on dead fish, others
were predaceous on their own larvae as well as those of other species of Cafius. In mating, end-to-end pos-
tures were observed, but more commonly males assumed a superior position. In the laboratory, eggs depos-
ited in sand about one inch below the surface hatch in about 6 days, pupation occurring about 27 days later
and adult eclosion on day 39. Early developmental stages are described and illustrated for luteipennis. lith-
ocharinus. canescens and seminitens.
A unique group of arthropods live on decaying seaweed on the beaches of southern
Cahfornia. The habitat consists chiefly of surf grass, four species of brown algae, and ten
species of red algae (Dawson, 1945, 1966). These plants are found together in clumps of
all sizes, extending from the strand to the high tide level of the beach. From the moment
that this vegetation appears on the shore, it is colonized by flies (Fucellia and Leptocera)
and amphipods tolerant of repeated wetting and occasional submersion in sea water.
Higher on the beach other accumulations of seaweed provide a habitat for additional
species of flies as well as arachnids, mites, isopods and a variety of Coleoptera.
Three species of Fucellia are probably most common in these habitats, while of the
coleopterans, the Staphylinidae are usually the most abundant (Moore. 1956). This paper
is concerned with Cafius. one of the more conspicuous genera of Staphylinidae. The four
commonly found species are Cafius seminitens Horn, C. canescens Maklin, C luteipennis
Horn, and C lithocharinus LeConte. Less common are C decipiens LeConte, C. opacus
LeConte, and C sulcicollis LeConte. The slim elongated bodies of these specie enable them
to move easily within the clumps of seaweed and to burrow into the upper layer of mixed
sand and seaweed.
METHODS
Observations and samples of this community were taken weekly from June 1966
through October 1967 on the beaches of San Clemente and Corona Del Mar in Orange
County: and of La Jolla. Ocean Beach, Sunset Cliff's, and Coronado, all in San Diego
County. All study sites were relatively undisturbed by beach cleaning machinery. Field
and laboratory studies were conducted on feeding and mating behavior, the eff"ects of
physical factors and competition (James, 1968).
RESULTS AND DISCUSSION
Habitat Observations.— V\! q noted that staphylinids always inhabited not only the sea-
weed but also the wet and slimy upper layer of sand beneath. If disturbed, they moved to
the tips of the drier seaweed, then flew to other nearby clumps.
Laboratory experiments showed that all species were attracted to the smallest sand
particles found in the beach habitat (James, 1968), and prefer a relative humidity of 95
percent. We also found that all species could survive without food for about a week, but
that further starvation was detrimental. Survival of individuals on the surface of seawater
SAN DIEGO SOC. NAT. HIST., TRANS. 16 (12); 279-290. 5 NOVEMBER 1971
280 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
in a plastic container ranged from 45 to 72 hours.
We observed one individual o^ Cafius lithocharinus which had the longest survival in
sea water, and found that it supported itself on its tibia on the water surface film for about
2': hours. This individual, when placed in a plastic enclosed container partly filled with
sea water, flew three times but finally resorted to merely floating on the surface as was
characteristic of the other three species studied. When pushed beneath the water, all four
species curled their abdomens up and back toward the thorax. The crook thus formed en-
trapped a bubble of air. which was carried beneath the surface. Upon release the beetle
floated back to the surface and extended its abdomen, then groomed the head and an-
tennae with the forelegs. Flotation ceased when the beetle dropped its abdomen below
the water surface, curved the abdominal tip back towards the head, and ceased leg move-
ment. The body then sank to the bottom and movement stopped.
Sea water thrown on beached seaweed caused beetles to come to the surface and fly
away. A thorough soaking of the deposit drove out all beetles. Beetle flight was always
away from the ocean, either up the beach or parallel to the surf, the ultimate goal being
undisturbed piles of seaweed nearby.
Fresh piles of seaweed were colonized by large numbers of adult staphylinids within
two weeks of deposition. Once 800 individuals of Cafiiis lithocharinus were captured,
marked with white paint, and released on the beach near their capture. None of the
marked individuals was ever recovered.
Predation.—Cafius canesceus and C. seminitens were voracious predators of both lar-
vae and pupae of Fucellia. while C. lithocharinus and C. luteipennis were only casual feed-
ers on larvae under experimental conditions (James. 1968). The adults of the four com-
mon species are chiefly predatory, although some were seen to scavenge on dead grun-
ion. Other known prey of Cafius includes amphipods and small barnacles; at times they
also preyed on their own larvae and pupae as well as those of related species.
Feeding Behavior.— Cafius seminitens and C. canescens upon encountering a fly larva
would grasp it with the mandibles, and break the larval intergument. Oozing body fluids
attracted other staphylinids. which joined in the consumption of the prey. We observed C.
canescens breaking the surface layers of seaweed with its mandibles to feed on fly larvae
within. This action attracted additional staphylinids which then shared the kill. On one
occasion seven beetles consumed a fly larva in nine minutes. Cafius seminitens and C. ca-
nescens were capable of excavating a hole in the puparium of a fly larva, and consuming
the oozing fluids. Pupae were rarely shared.
Mating Behavior.— Ahhough end-to-end mating postures were observed, usually the
male assumed a superior position. The males use their mandibles to grip the females on
the 2nd and 3rd abdominal segments below the elytra, in addition to using their legs to
hold the female in position. In C. canescens the male extended the adaeagus while curling
his abdomen around and downward to meet the female's upcurved abdomen. This posi-
ti(m was retained for as long as 77 seconds.
Immature Stages. — \n the laboratory, individually placed eggs of Cafius canescens
were deposited about 1 inch below the sand surface. Gestation was about 6 days at room
temperature. A newly hatched larva immediately excavated a burrow about 5 inches deep
in a sand-filled test tube. Pieces of cockroaches which were dropped into the test tube
were examined bv the staphylinid larva on the surface and finalK pulled into the sand
burrow.
The larva oriented itself with its head toward the sand surface. The burrow was en-
larged and extensions were made in succeeding larval stages. Pupation occurred about 27
days after egg laying at the 1/2 inch level in sand. Adult eclosion occurred on the 39th
1971
JAMES, MOORE AND LEGNER: CAFIUS
281
dav after the egg was laid.
Larvae of C canescens were first collected from the beach at Coronado on 15 March
1967. Overwintering apparently occurs in either the egg, pupal or adult stage. After 1
April, C. canescens larvae were collected regularly in small numbers at all study sites.
Cafius hiteipennis larvae were placed in a cage on 25 March and pupated 31 days later.
At^ter 4 May, larvae of C. luteipennis were collected in small numbers from the Coronado
site. On 8 Julv, larvae of C lithocharimis were first collected at Coronado. and the pupae
were formed 34 days after a second set of larvae was collected on 27 August.
The large larvae of Cafius seminiiens were first observed at Corona Del Mar on 18
September. Pupation occurred 18 days later and adults emerged on the 28th day. Only
Cafius seminitens larvae were seen to feed on seaweed fly larvae, devouring their prey in a
similar manner as the adults.
The various species of Cafius apparently breed at different times of the year, as in-
dicated by their appearance at different dates. Cafius canescens and C luteipennis appear
to breed in early spring, C lithocharinus in early summer, and C. seminitens in late sum-
mer.
Figure 1. Larva of Cafius canescens Maklin. a,
anterior margin of clvpeus; b, urogomphus and
pseudopodia; c. antenna; d, maxilla: e. dorsal
view of body.
^^^.^
V
y-
DESCRIPTIONS OF EARLY STAGES
LARVAE
The larva of the European Cafius sericeus Holme was described by Rey (1887) and
that of C xantholoma (Gravenhorst) by Rupertsberger ( 1880). Paulian (1941) redescribed
both of these species, using the name Remus sericeus for the former. Remus generally is
considered a subgenus of Cafius. Paulian gave no generic description for the larva of
Cafius. But a combination of characters from his key makes a good diagnosis of this
genus. It follows:
Cephalization accentuated: neck present; epicranial suture very long: gular sutures very
long: ocelli four, arranged in a compact group near bases of antennae: nasal present: max-
282 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
illarv palpus four-segmented; galea present, movable, with the aspect of a segment; lacina re-
duced at maximum to some local spines in the apical region of the stipes; prosternum
strongly chitinized.
The combination of these diagnostic characters and the seashore habitat, permit easy
recognition of Cafius larvae.
In his key PauHan used other characters which new material shows to be too variable
for a generic definition. Thus the middle teeth of the nasal do not differ from the lateral
teeth in all the Pacific Coast species. And although the urogomphus is two-segmented and
longer than the pseudopod in two of our species, in the other two it is shorter than the
pseudopod. In one of the latter the urogomphus is distinctly one-segmented and spheri-
cal.
KEY TO THE KNOWN LARVAE OF CAFIUS
1 A. Urogomphus longer than pseudopod.
2A. Second segment of urogomphus widest at base, tapered to apex.
3A. Composite macrosetae much more numerous than simple macrosetae. . . sericeus
3B. Simple macrosetae much more numerous than composite
macrosetae xantholoma
2B. Second segment of urogomphus long, slender, cylindrical.
4A. First segment of maxillary palpus one-half as long as second
segment luteipennis
4B. First segment of maxillary palpus about as long as second
segment lithocharinus
1 B. Urogomphus shorter than pseudopod.
5A. Urogomphus two-segmented, the segments subcylindrical canescens
5A. Urogomphus one-segmented, spherical seminitens
LARVA OF CAFIUS LUTEIPENNIS HORN
Co /or.— Pale, with head dark testaceus.
Head subquadrate. widest near basal angles, slightly narrowed to apical angles.
Neck about three-fourths as wide as head. Ocelli four, in a small cluster near apical an-
gles. Clypeal margin with nine teeth, the two outer teeth on each side smallest, the next
two on each side longer than wide, the central tooth little more than half as long as
those next to it (Fig. 3A). Antennae four-segmented, the first segment short, the second
and third about as wide as first and each about twice as long as wide, the third with a
small ovoid seta at apex, the fourth about half as wide and half as long as third with a
very small, round modified seta at apex (Fig. 3C). Maxilla (Fig. 3D) with the stipes al-
most as long as palpus; galea small elongate-ovoid; maxillary palpus four-segmented,
the first segment about as long as wide, the second as wide as and twice as long as first,
the third somewhat narrower and shorter than second, the fourth small, elongate-ovoid.
Ligula about as long as first segment of labial palpus, pubescent basally. Labial palpus
three-segmented, the first segment about twice as long as wide, the second a little nar-
rower and shorter than first, the third much narrower and shorter than second. Gular
sutures united in basal three-fifths, thence divergent to apex.
Thorax.— Pronolum a little wider than long, widest near basal angles, narrowed
slightly to apical angles, with a few scattered setae at sides and on disc. Mesonotum and
metanotum shorter and a little wider than pronotum. with sparse scattered setae.
Abdomen with parallel sides in basal half, thence slightly narrowed to apex, the
1971
JAMES. MOORE AND LEGNER: CAFIUS
283
segments of about equal length throughout, sparsely setose. Pseudopod about twice as
long as wide. Urogomphus two-segmented, longer than pseudopod, the segments very
slender, the second segment much narrower and somewhat shorter than first.
Length.— 1 mm.
Material examined.— Hold Del Coronado Beach, Coronado, San Diego Co., Cali-
fornia, April 1967, Gary James coll.
Notes.— This species can be distinguished by the combination of the very long slen-
der two-segmented urogomphus and the very short first segment of the maxillary pal-
pus.
pi
.^i^/AfJ/1-,
kT^"^
Figure 2. Larva of Cafius seminitens Horn, a,
anterior margin of clypheus: b, urogomphus and
pseudopodia: c, antenna; d, maxilla; e, dorsal
view of body.
LARVA OF CAFIUS LITHOCHARINUS LE CONTE
Co/or.— Head and thorax dark ferruginous, abdomen pale ferruginous.
Head subquadrate, widest near basal angles, slightly narrowed from base to apex.
Neck about four-fifths as wide as head. Ocelli small, dark, in a small cluster near apical
angles. Frontal suture joining epicranial suture at an obtuse angle near anterior third of
head. Clypeal margin with nine teeth, the central tooth and two outer teeth smallest (Fig.
4A). Antenna with first segment short, the second and third each about twice as long as
wide, the third with an ovoid modified seta at apex, the fourth much narrower and shorter
than third, with a small ovoid modified seta at apex (Fig. 4C). Maxilla (Fig. 4D) with stipes
as long as palpus; galea very small, ovoid; maxillary palpus four-segmented, the first seg-
ment about twice as long as wide, the second a little narrower and shorter than the first, the
third much narrower and somewhat shorter than second, the fourth small, ovoid. Ligula
shorter than first segment of labial palpus, pubescent. Labial palpus three-segmented, the
first segment almost twice as long as wide, the second narrower and shorter than first, the
third small, ovoid. Gular sutures united in basal two-thirds, divergent anteriorly.
Thorax .-Pronotum about as wide as long, widest near basal angles, narrower ante-
284
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
riorly. Mesonotum and metanotum much shorter and slightly wider than pronotum.
Abdomen gently tapered from base to apex, the first segment short, the others progres-
sively slightly longer, with scattered sparse setae throughout. Pseudopod nearly three times
as long as wide. Urogomphus longer than pseudopod, two-segmented, the first segment al-
most as long as pseudopod, the second long and very slender (Fig. 4B).
Length.— S-l 1 mm.
Material examined.— Ten specimens: Hotel Del Coronado Beach, Coronado, San
Diego Co., California, August 7, 1967, Gary James coll.
Notes.— This larva most closely resembles that of C luteipennis, from which it may be
distinguished by the relatively longer first segment of the maxillary palpus and by shorter
clypeal teeth.
Figure 3. Larva of Cafiiis luleipennis Horn. a.
anterior margin of clypeus; b, urogomphus and
pseudopodia; c, antenna; d. maxilla: e. dorsal
view of body.
Figure 4. Larva of Cafius lithocharinus Le-
Conte. a, anterior margin of clvpeus: b. uro-
gomphus and pseudopodia: c. antenna: d,
maxilla: e, dorsal view of body.
LARVA OF CAFIUS CANESCENS MAKLIN
Co/or.— Pale testaceous, with head ferruginous. Ocelli and base and apex of mandibles
dark.
//ertJsubquadrate, widest just before the rounded basal angles, gradually narrowed to
near the ocelli. Neck about five-sevenths as wide as head. Ocelli tour, in a close cluster near
the anterior angles. Frontal suture joining epicranial suture at an obtuse angle at about the
anterior third of the head. Clypeal margin with nine similar teeth, the penultimate outer
tooth on each side somewhat shorter than the others (Fig. 1 A). Antenna with first segment
widest, about as long as wide, the second segment about twice as long as first, the third
1971
JAMES, MOORE AND LEGNER: CAFIUS
285
about as long as second and with a small modified segment at apex, fourth segment much
narrower and shorter than third (Fig. IC). Maxilla (Fig. ID) with stipes almost as long as
palpus, about twice as long as wide. Galea small, ovoid: maxillary palpus with first two seg-
ments subequal, the third much shorter and narrower than second, the fourth minute. Li-
gula about as long as first segment of labial palpus, pubescent in basal half Labial palpus
three-segmented, the first segment about twice as long as wide, the second a little shorter
and distinctly narrower than first, the third narrower than second, very little longer than
wide. Gular sutures united in basal three-fifths, thence divergent to apex.
Thorax —VronoXuva subquadrate. a little wider than long, widest near middle, thence
narrowed slightly to base and to apex. Mesonotum and metanotum much shorter than and
about as wide as pronotum. Each segment with a row of setae at anterior, lateral and pos-
terior margins and a very few scattered setae on disc.
Abdomen widest at base, slightly tapered to apex; first segment shortest, the segments
increasing in length progressively to apex; a little more densely setose than thorax. Pseudo-
pod about twice as long as wide. Urogomphus two-segmented, shorter than pseudopod,
subcylindrical(Fig. IB).
Length.— 9 mm.
Material examined— ¥'\\Q specimens, Vi mile west of pier, San Clemente, Orange Co.,
California, 12 April 1967, Gary James coll.
Notes— T\\\?, larva differs from the other larvae of Cafius in having a two-segmented
urogomphus that is much shorter than the pseudopod.
Figure 5. Pupa of Cafius canescens Maklin. a,
ventral view; b, lateral view.
LARVA OF CAFIUS SEMINITENS HORN
Co/or.— Pale testaceous, with head and pronotum ferruginous, the base and apex of
mandibles darker. Head subquadrate, a little wider than long, widest near basal third.
286
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
thence slightly narrowed to apex. Neck about three-fourths as wide as head. Ocelli very
pale, difficult to detect. Frontal sutures joining epicranial suture at an obtuse angle at about
the apical third of head. Clypeal margin with nine teeth, the central tooth distinctly smaller
than the others (Fig. 2A). Antenna four-segmented, the first segment short, the second and
third as wide as and more than twice as long as first, the third with a rounded modified seta
at apex, the fourth small ovoid (Fig. 2C). Maxilla with stipes as long as palpus, about twice
as long as wide; galea small, oval; maxillary palpus (Fig. 2D) with a separate sclerotization
forming a very short ring at base in the form of an extra segment which may represent the
lacina; first segment of palpus about twice as long as wide, the second nearly as wide and as
long as first, the third a little narrower and shorter than second, the fourth small, ovoid.
Ligula about as long as first segment of labial palpus, pubescent at base. Labial palpus
three-segmented, the first segment about twice as long as wide, the second narrower and a
little shorter than first, the third much narrower and shorter than second. Gular sutures
united in basal three-fifths, thence divergent to apex.
Thorax .—Pvonolum transverse, the sides well rounded, widest at basal third, with a few
scattered setae on disc and at sides. Mesonotum and metanotum narrower and shorter than
pronotum, with a row of setae at base, sides and apex.
Abdomen slightly tapered from base to apex, the first few segments short, the apical
segments progressively narrower and longer. First four segments irregularly, and densely
set with short stout peg-like setae, the next five segments progressively more sparsely setose.
Pseudopod about twice as long as wide. Urogomphus one-segmented, spherical, much
shorter than pseudopod (Fig. 2B).
Lenglh.— \0-\4mm.
Material examined.— Five specimens. Corona Del Mar, Orange Co., California, 16
September 1967, Gary James collector.
Note.— This larva differs in several respects from other known larvae ofCaJius, but par-
Figure 6. Pupa of Cafius seminilens Horn, a,
ventral view; b, lateral view.
1971
JAMES. MOORE AND LEGNER: CAFIUS
287
ticularly in the spherical one-segmented urogomphus, the "extra segment" at the base of
the maxillary palpus, the very transverse pronotum and the densely setose first four abdom-
inal segments.
On the basis of larval characters this species might be placed in a separate genus or
even separate subfamily, but adult characters preclude such a course. Because of the great
similarity of their imagoes, this species and C canescens are usually placed by themselves
in the subgenus Bryonomus.
PUPAE
The pupa of the European species C sericeus Holme, was described and illustrated by
Paulian (1941). In his key to the genera of the pupae of the Staphilinoidea he diagnosed
sericeus as follows:
Pronotum with strong marginal setae, without discal setae: dorsum of abdomen fiat, epi-
pleurae prominent: with two long slender cerci which have whorls of fine setae apicallv.
The pupae of the Pacific Coast species show that some of these characters are specific
rather than generic.
Hinton ( 1958. 1963a. 1963b) called attention to the fact that among the few pupae of
Coleoptera studied the most apparent useful taxonomic characters are the number and ar-
rangement of tubercles, macrosetae and pubescence. The pupae of Cafius which have been
studied share the following characters:
Body without fine pubescence except dense fine pubescence at extreme tip of urogomphus.
Tubercles arranged in a single row at anterior margin of pronotum, two to four rows on middle
and posterior tibiae and one tubercle each at lateral men-gin of abdominal segments five and
six. Macrosetae restricted to pronotal and abdominal tubercles.
Figure 7. Pupa of Ccifius luieipennis Horn. a.
ventral view: b. lateral view.
288
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure 8. Pupa of Cafius lithocharinus Le-
Conte. a, ventral view; b, lateral view.
KEY TO THE KNOWN PUPAE OF THE AMERICAN SPECIES OF CAFIUS
1 A. Anterior margin of pronotum with a single row of
nine setigerous tubercles each side seminitens
1 B. Anterior margin of pronotum with a single row of
fewer than nine setigerous tubercles each side.
2A. Anterior margin of pronotum with a single row of
three setigerous tubercles each side canescens, lithocharinus
2B. Anterior margin of pronotum with a single row of
four setigerous tubercles each side hiteipennis
Characters have not been found for the separation of the pupae of C canescens and
lithocharinus.
ACKNOWLEDGMENTS
We are indebted to Dr. Lauren B. Anderson. Mrs. Patricia Farreli and Mrs. Arlene Sansevero of the Univer-
sity of California at Riverside and to Fred G. Andrews of the California Department of Agriculture at Sacramento
for aid in locating literature.
LITERATURE CITED
Dawson. E. Y.
1945. An annotated list of the marine algae and marine grasses of San Diego County, California. Occas.
Papers San Diego Soc. Nat. Hist. 7: 1-87.
Dawson, E. Y.
1966. Marine botany. An introduction. Holt Rinehart and Winston, Inc., N.Y. 371 p.
James, G. J.
1968. The biology and ecology of four species of the genus Cafius (Coleoptera: Staphylinidae). M.S.
Thesis, San Diego State College. 72 p.
1971 JAMES, MOORE AND LEGNER: CAFIUS 289
Moore, I.
1956. Notes on intertidal Coleoptera with descriptions of the early stages (Carabidae, Staphylinidae, Mala-
chiidae). Trans. San Diego Soc. Nat. Hist. 12: 207-230.
Paulian. R.
1941. Les Premier etats des Staphvlinoidea. Etude dc morphoioaie comparee. Mem. Mus. Hist. Nat. Paris,
n. ser.. 15: 1-361.
Rey, C.
1887. Essai d'etudes ser les larves des coleopteres. Ann. Soc. Linn. Lyon 33: 146-1480.
Rozen, J. G.. Jr.
1958. Systematic study of the pupae of the Oedemeridae (Coleoptera). Ann. Ent. Soc. Amer. 52: 299-303.
1963a. Preliminary systematic study of the pupae of the Nitidulidae (Coleoptera). Amer. Mus. Novitates
2124:1-13.
1963b. Two pupae of the primitive suborder Archostemata (Coleoptera). Proc. Ent. Soc. Washington 65:
307-310.
Rupertsberger. M.
1880. Biologic der Kafer Europas. Eine Ubersicht der biologischen Literatur. Donau. Linz. 295 p.
Orange Coast College. Costa Mesa, California 92626 (G.J.J. ), and Division of Biological
Control, Department of Entomology, University of California. Riverside, California 9 '")()''
(LM.andE.F.L.)
MUij. CUM p. ^oc;
LIBRARY
FEB 1 8 1972
HARVARD
UNIVERSJTY
THE COLOR PATTERNS OF DOWNY YOUNG
RATITES AND TINAMOUS
JOSEPH R. JEHU JR.
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 13 15 NOVEMBER 1971
THE COLOR PATTERNS OF DOWNY YOUNG
RATITES AND TINAMOUS
JOSEPH R. JEHL, JR.
ABSTRACT.— Plumage patterns of downy young ratites indicate that the Casuariidae and Dromiceiidae are
closely related, and they suggest that the Struthioniformes and Casuariiformes may be more closely related to
each other than either is to any other hving ratite taxon. Relationships of the Rheiformes and Apterygiformes
are not clarifed. The Tinamidae fall into two distinct groups of genera: 1) Tinamus, Nothocercus, and Cryplu-
rellus. and 2) Rhynochotus. Nothura, Nothoprocta, and Tinamotis; chicks of Taoniscus were not examined.
Chick plumages provide no evidence for close relationship between tinamous and any ratite taxon.
The ratites are large, flightless, running birds with an unkeeled sternum which, with
one minor' exception, are now restricted to the southern hemisphere. They include the ex-
tant families Struthionidae, Rheidae, Casuariidae, Dromiceiidae, and Apterygidae, and the
extinct Opisthodactylidae, Dromornithidae, Emeidae, Eleutherornithidae, and Aepyorni-
thidae (Brodkorb, 1963). Interrelationships among these famihes have long been among
the most controversial problems in avian systematics (Bock, 1963), as has the question of
whether these flightless birds share a common ancestor. (References to much of the relevant
literature are contained in Bock, 1963, de Beer, 1956, 1964, and Parkes and Clark, 1966).
Some recent authors (e.g.. Bock, 1963; Parkes and Clark, 1966) have argued that the ratites
are probably monophyletic, but ornithologists have yet to reach a consensus on this point.
The tinamous (Tinamidae) are ground-dwelling, chicken-like birds of the Neotropics.
Their possible close relationship to the ratites, and particularly to the rheas, has received
much attention, but relationships within the Tinamidae have been largely ignored.
Because chick color patterns have been used to elucidate relationships within certain
other taxa (e.g., Podicipedidae: Storer, 1967; Anatidae: Delacourand Mayr, 1945; Tetrao-
ninae: Short, 1967; Charadrii: Jehl, 1968) their importance in suggesting relationships wi-
thin the ratites and tinamous was investigated. The results provide limited evidence in
support of relationships that have been suggested among ratites, and they clarify the subdi-
visions of the tinamous.
In this study I was able to examine specimens or descriptions of most ratite species, as
well as living chicks of Struthio camelus, Dromiceius novaehoUandiae and Crypturellus soui
in the San Diego Zoo. Most tinamou genera were also available, but many species were not.
Studies in other groups have shown that an index to color pattern within a genus can usu-
ally be obtained from a few representative species. Thus, the general conclusions reached
here are unlikely to be aff'ected by the limited material. Nevertheless, further collecting is
desirable, particularly of species in the genera Nothoprocta, Crypturellus, and Taoniscus.
In the following section the major color patterns are described for each taxon. For each
species the number of specimens examined is given in parentheses. Species for which I have
examined only a description in the literature are denoted by an asterisk. In view of the lim-
ited material, no attempt has been made to provide descriptions adequate for species iden-
tification.
RATITES
STRUTHIONIFORMES: STRUTHIONIDAE
Struthio: camelus (2)
Ostrich chicks (Fig. lA) are covered on the back with a thick mat of tan and blackish
down feathers. Several barbs on each of these feathers are prolonged, flattened, and
SAN DIEGO SOC. NAT. HIST., TRANS. 16 ( 13): 291-302, 15 NOVEMBER 1971
292
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
\
3
Hf.*
Figure 1. Downy young ratites in dorsal and ventral view: (A): Siruthio camelus (B) Pterocnemia pennata, (C)
Dromiceius novaehoUandiae, (D) Casuarius bennetti.
1971
JEHL: RATITES AND TINAMOUS
293
twisted, and intertwine with those from adjacent feathers. The resuhing appearance is that
of a pile of straw and I cannot determine whether any underlying color pattern is present.
There is a definite though variable striped pattern on the neck. Because several of the
stripes are discontinuous, the configuration of this pattern is not as evident in flattened study
skins as it is in living chicks (see photos by Sauer and Sauer, 1966: Fig. 31, 32). It consists
(Fig. 2) of a mid-dorsal stripe (A) and one dorso-lateral (B) stripe on each side of the neck;
an interrupted stripe (C) on each side of the neck; an interrupted stripe (D) on the ventro-
lateral surface of the neck that starts near the base of the bill and continues to the upper
chest, and (E) a short interrupted stripe in the throat region. On the head a stripe extends
from the base of the upper mandible, dorsal to the eye, to the ear region; facial markings are
variable but usually include a dark spot posterior to the eye and a short hne from the rictus
that passes dorsally anterior to the ear.
Figure 2. The interrupted pattern of head and neck striping in Struthio camelus. The major stripes are indicated.
RHEIFORMES: RHEIDAE
Rhea: americana (6). Pterocnemia: pennata (4)
Color patterns of Pterocnemia (Fi^s. IB, 3C) and Rhea chicks are identical. A dark cen-
tral stripe extends from the crown to the rump, but broadens to a diamond-shaped figure on
the mid-back and sends branches along the dorsal surface of the wing. Lateral stripes ex-
tend from the rump to the mid-back, where they turn ventrally. When the chick's wings are
folded, the wing and lateral stripes appear to form a continuous stripe along the entire
length of the body. The chin and belly are whitish; the neck is dusky gray and this color-
ation extends onto the chest as a thin central line.
The ground color of Pterocnemia chicks is whitish and the patterned areas are choco-
late brown; minor pattern variations occur in the width of the striping. In Rhea the ground
color is tan, the patterning dark brown. Rhea chicks hatched in captivity show considerably
more color variation than wild chicks. This presumably results from inbreeding and selec-
tion for albinistic birds.
CASUARIIFORMES: CASUARIIDAE, DROMICEIIDAE
Casuarius: casuarius (3), unappendiculatus, bennetti (2)
Species limits in the cassowaries are not well known. Peters (1931) lists six species,
294
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
but Rand and Gilliard (1967) recognize only three.
Cassowary chicks are pale brown with well-marked longitudinal stripes on the back.
The head is chestnut or tan, and may be unmarked or dotted irregularly with dark brown
markings (Figs. ID, 3B). On the back three major dark brown stripes extend from the
shoulder region to the rump; within each of these stripes a light central stripe of varying
prominence is formed by the chestnut tips of the feathers. A dark stripe on the side is par-
alleled ventrally by an indistinct stripe that appears to be continuous with the leg stripe.
The belly and chin range from light tan to light brown and are unmarked. The neck and
chest are irregularly flecked with gray-brown markings; in most specimens the neck color-
ation continues on to the chest as a thin central line, similar to that found in the Rheidae
(cf. Figs. IB, D).
Figure 3. Diagrammatic color patterns of; (A) Dromiceius novaehoUandiae, (B) Casuarius casuarius, and (C)
Pterocenmia pennata.
Dromiceius: novaehoUandiae (5)
Emu chicks are boldly patterned (Figs. IC, 3 A, 4). The head markings show no con-
sistent arrangement but the neck and back markings are distinctive. Dorsally, a central
(A) and two lateral (B) stripes extend from the occiput to the rump; a stripe from the au-
ricular region broadens at the shoulder, where it acquires a buff"y central stripe, and con-
tinues to the flank; a buff'-centered stripe on the lateral surface of the thigh is bordered by
a thin (sometimes interrupted) black stripe on the antero-lateral surface. On the ventral
surface of the neck stripes run from the base of the bill to the sides of the chest (D); a
short central stripe (E) is present in the throat region. Pattern details are variable. In some
birds parts of stripes are interrupted or missing, and in the bird shown in Figure 4 the
posterior part of the central stripe has fused with a lateral stripe.
1971
JEHL: RATITES ANDTINAMOUS
295
Figure 4. Head and neck pattern of Dromiceius novaehollandiae. The major stripes are indicated.
APTERYGIFORMES: APTERYGIDAE
Apteryx: australis*, oweni, haasti
Newly-hatched Kiwis seem to be unpatterned. No trace of a color pattern could be
detected in a well-developed embryo o^ Apteryx australis preserved in alcohol at the Car-
negie Museum (Mary H. Clench, pers. comm.). Oliver (1957: 48) described the nestling of
Apteryx australis as follows: "Upper surface black streaked with brown mainly on the
shafts and bases of the feathers. On the head, breast and abdomen it is greyer."
TINAMOUS
TINAMIFORMES: TINAMIDAE
Tinamus: tao, soUtarius*, osgoodi, major (5), guttatus
De Schauensee's (1966) classification of tinanmous is followed in this paper.
The head pattern in T. major is complex (Figs. 5, 6); a grayish patch extends from the
base of the bill onto the forehead; posteriorly, a brown crown patch extends over the occi-
put and onto the neck; a gray-brown postorbital stripe runs from above and behind the
eye to the side of the neck; a brownish line of variable prominence extends from the base
of the bill to the anterior corner of the eye and continues posteriorly as a broad band
through the auriculars; the cheeks and throat are grayish, except for a short, dark malar
stripe. Feathers on the nape of the upper back are brown, lightly barred with gray,
whereas those on the midback appear uniformly brown. A broad, light brown or golden
band on the lower back extends to the rump and is bordered laterally by a thin line of
dark feathers (Fig. 5). The color pattern of T. solitarius is similar (Salvadori, 1895: 502).
Nothocercus: bonapartei (I), Julius (1), nigricapillus
The coloration of the two species o^ Nothocercus at hand differs slightly, but there are
no important differences in color pattern. In A'^. bonapartei (Figs. 5, 6) a light grayish patch
from the base of the bill extends onto the forehead, where it blends with a dark gray crown
that extends onto the occiput; the face and cheeks, including a broad supraorbital stripe,
are grayish, and an obscure dark line runs from the base of the bill to the anterior corner
296
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
of the eye; the throat is grayish-white. The back is uniformly brown, individual feathers
being thinly barred with black. In N. Julius the crown patch is grayish-white and is sharply
bordered laterally by a dark stripe; the facial area is orangish.
Figure 5. Downy young tinamous. Left to right: Tinamus major, Crypturellus boucardi, Crypturellus undulatus,
Crypturellus soui, Nothocercus bonapartei.
Crypturellus: cinereus, soui {!), ptaritepui, obsoletus, undulatus (1), brevirostris, bartletti,
variegatus (2), atrocapillus, noctivagus, duidae, cinnamomeus (1), transfasciatus, strigulosus,
casiaquiare, boucardi (2), saltuarius, kerriae, parvirostris, tataupa (2)
Color patterns in this large genus are variable (Figs. 5. 6). In C boucardi, variegatus,
cinnamomeus, tataupa and undulatus, the head pattern is similar to that of Tinamus except
that a light brown narrow median stripe is enclosed in the posterior part of the crown
Figure 6. Head patterns of tinamou chicks. Left to right: Tinamus major (2). Crypturellus undulatus, Cryptu-
rellus soui, Nothocercus bonapartei.
1971
JEHL: RATITES AND TINAMOUS
297
patch, and the auricular stripe is narrower and much less prominent (Fig. 6); in variegatus
(see Beebe. 1925; Fig. 22) the borders between the prominent head patches are less dis-
tinct; in sold the auricular patch is absent and the head patches blend into each other, as
in Nothocercus. The back patterns of boncardi, variegatus and cinnamomeus are similar to
that of Tinamus but lack a pale patch on the lower back; in undiilatiis and tataupa the
back is uniformly brown and lightly barred; in soid the feathers appear uniformly brown-
ish but are finely barred with black, as in Nothocercus.
Figure 7. Downy young tinamous. Left to right: Rhvnchotus rufescens, Nothura maculosa. Nothoprocta curvi-
rostris, Nothoprocta pentlandii, Nothoprocta perdicaria. Eudromia elegans.
Rhvnchotus: rufescens (2)
The color pattern o^ Rhvnchotus chicks (Figs. 7, 8) is unlike that of the preceding gen-
era. The back appears to be longitudinally streaked with dark and light feathers in no ob-
.,A^ • 1 1 iiV.
Figure 8. Head patterns of tinamou chicks. Lett to right: Rhvnchotus rufescens. Nothura maculosa. Nothoprocta
curvirostris, Nothoprocta perdicaria, Eudromia elegans.
298 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
vious pattern and to be overlain by a thin coat of stiff bristles. This streaked pattern
reflects the structure of the dorsal down feather, which consists of a rachidial main feather
and a prominent aftershaft of almost equal size. The rachis of the main feather bears
dark-colored barbs for about two-thirds of its length, then forks to form a pair of stiff
bristles; the aftershaft bears light colored barbs for almost its entire length. A similar
down structure is present in Nothura, Nothoprocta, and Eiidromia. In down feathers ofTin-
amiis. Nothocercus and Crypturelliis, a rachis is poorly developed and aftershafts seem to
be lacking or rudimentary.
The head pattern is well defined. A narrow blackish crown patch which extends from
above the eye to the occiput is bordered by a thin buffy stripe that begins at the base of
the bill but forks on the forehead to encircle the crown patch; the buffy stripes, in turn,
are bordered by a dark stripe that extends from the base of the bill to the nape. The face
is generally buffy, but with interrupted dark stripes in the post-orbital and auricular re-
gions; there is a thin malar stripe (Fig. 8).
Nothoprocta: taczanowskii, kalinowski, ornata (1); perdicaria (1), cinerascens, pentlandii
(2), curvirostris (3)
Chicks of Nothoprocta (Figs. 7, 8), Hke those of Rhynchotus have a streaked pattern.
In Nothoprocta, however, the bristle-like tips of the main feather are much less prominent
and usually are flanked by one or more additional barbs. The chick of TV. perdicaria, in
addition to being streaked dorsally, has a sHghtly barred appearance, because the dark
feathers are buffy at both the base and tip.
The head markings in perdicaria and ornata are similar to those of Rhynchotus, ex-
cept that the boundaries between the major stripes are less pronounced. In curvirostris
dind pentlandii the head is dotted irregularly with black, brown, and white, but the pattern
is a variation of that found in perdicaria.
Nothura: boraquira, minor, darwinii, maculosa (1), chacoensis
The chick of Nothura maculosa (Figs. 7, 8) also has a streaked pattern and is ex-
tremely similar to that oi Rhynchotus; the hairlike bristles of the dark back feathers, how-
ever, are less strongly developed than in that genus. The head pattern is hke that of
Rhynchotus, except that the borders between the major stripes are less clearly defined.
Nothura lacks a post-orbital stripe; auricular and malar stripes, though present, are incon-
spicuous.
Taoniscus: nanus
I have seen neither a chick nor a description of the downy plumage of this species.
Eudromia: elegans (5),formosa
The dorsal color pattern of Eudromia (Fig. 7) is similar to that of the other streaked
genera, although light-colored feathers are less abundant than in Rhynchotus, Noth-
oprocta and Nothura. The darker feathers are subterminally barred with blackish brown,
so that the chick, like that of Nothoprocta perdicaria, appears slightly barred. The bare
tips of these feathers are much shorter than in the genera hsted above.
The dorsal surface of the head and neck is flecked with gray and brown; from each
side of the biU an indistinct whitish line extends across the crown to the occiput; the face
and throat are generally buffy-white, but the lores are dark; posterior to the eye a brown
stripe extends through the auriculars to the side of the neck; there is prominent malar
stripe (Fig. 8).
Tinamotis: pentlandii (3), ingoufi
So little down remains on the three half-grown chicks of T. pentlandii that I have ex-
1971 JEHL: RATITES AND TINAMOUS 299
amined that no assessment of the body pattern is possible. The head, however, is boldly
striped. Two broad dark stripes, one on each side of the bill, pass dorsal to the eyes and
around the periphery of the crown to the nape; a small white-centered patch on the occi-
put extends onto the nape as a thin median line; a stripe from the lores passes through the
eye to the auricular region; and a malar stripe extends from the gape through the cheeks
and onto the side of the neck. At first glance the head markings of Tinamotis seem unique
but the pattern is clearly a variant of those found in genera with streaked chick plumages
and closely resembles that of Nothoprocta perdicaria.
DISCUSSION
RATITES.-In the following discussion I assume that similarities in complex patterns and
the potential for easy transformation of one pattern to another are evidence for close rela-
tionship. The sequence of pattern transformation cannot be determined in the absence of
information regarding the ancestral downy pattern. However, if one assumes that the ra-
tites are monophyletic, it is reasonable to infer that a striped pattern of some sort may
have been primitive, inasmuch as a striped pattern or presumed remnant thereof is pres-
ent in four of the five extant ratite families and is lacking only in the Kiwis, whose bur-
row-nesting habits are unusual in that group.
The downy young plumages provide limited evidence regarding relationships among
ratite families. The long-accepted close affinity of cassowaries and emus is confirmed by
the similar color patterns of their chicks. The transformation of a cassowary pattern to
that of an emu requires only a change in head pattern (variable in cassowaries) and the
introduction of a light central stripe to each of the major dorsal stripes. A hint of that
line— the light chestnut tips to the central feathers of each stripe— is present in the casso-
waries.
Ostrich chicks lack any discernible dorsal pattern, but the pattern of head and neck
striping is closely similar to that of an emu (cf Figs. 2, 4). This suggests, as Sibley (1960)
and Glenny (1965) have indicated, that the Struthioniformes and Casuariiformes may be
more closely related to each other than either is to any other hving ratite order. If so, the
neck pattern in ostriches might represent the remnant of a striped pattern that extended
over much of the body. One could speculate that this pattern was replaced by a uniform
pattern, and was complemented by a straw-like down structure, insuring crypticity in
areas of sparse vegetation as proto-ostriches became adapted to desert habitats.
Downy plumage patterns do not suggest an alliance between rheas and other ratites.
Although it would be possible to derive the striped rhea pattern from that of a cassowary,
for example, no easy transformation is evident. Similar chest patterns in rheas and casso-
waries are simple and could result from convergence. Thus, they provide no evidence for
relationship.
Kiwis are thought to be most closely related to the extinct moas and to the living cas-
sowaries and emus (Parkes and Clark, 1966). Young kiwis are unpatterned and their plu-
mage offers no evidence on their possible relationship to other ratites. The lack of a
distinct pattern may be a derived condition associated with the burrow-nesting habits of
these birds.
TINAMOUS.-Downy young tinamous fall into two distinct groups of genera: 1) Tinamus,
Nothocercus, and Crypturellus; 2) Rhynchotus, Nothura, Nothoprocta, Eudromia, and
Tinamotis. Chicks of Taoniscus (not available) presumably fall into the second group.
These groups correspond to the subfamilies Tinaminae and Nothurinae, respectively, of
Miranda-Ribiero (1938). The downy young provide no evidence for von Boetticher's
300 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
(1934) subdivision of the Nothurinae into two subfamilies, Rhynochotinae {Rhynochotus,
Nolhura, Nothoprocta, Taoniscus) and Eudromiinae {Eudromia, Tinamotis).
Because the Tinaminae are forest dwellers whereas the Nothurinae are birds of the
grasslands, pattern similarities within these groups might be attributable to convergence.
This seems unlikely because patterns in the Tinaminae are complex, and the Tinamus
pattern can be easily transformed into the more uniform pattern of Nothocercus through a
small series of steps such as are represented in existing species oT Crypturellus (Figs. 5, 6).
Chicks of the Nothurinae are united by similarities in back pattern and feature structure;
differences between color patterns of Rhynchotus and Eudromia are largely bridged by in-
termediate patterns within Nothoprocta (Figs. 7, 8).
Interrelationships of tinamou genera diagrammed by von Boetticher (1934; also re-
produced in Ward, 1957: 336) are largely supported by Ward's study of mallophagan par-
asites on tinamous. Ward's suggestion that Nothocercus is more closely related to
Crypturellus than to Tinamus, however, is also indicated by the similarity of chick color
patterns. In addition, chick plumages suggest that Nothoprocta may be somewhat more
closely allied to Eudromia (and Taoniscus) than von Boetticher postulated. In the species
that I have examined there are no close similarities between patterns in the Tinaminae
and Nothurinae.
The attempt to establish a phylogenetic hnk between the ratites and tinamous, and
particularly between the rheas and tinamous, dates at least to the early 19th century, and
the downy young have been used to give some support to that view. Salvadori (1895: 494)
stated that newly hatched tinamous "... are covered with down, and more or less closely
resemble the young of some of the Ratitae." While I agree with Salvadori's implicit
thesis— that the downy plumages of birds may be of great taxonomic value— I differ with
his conclusion. Nowhere among the tinamous that I have examined, nor in species whose
chicks are described by Salvadori, are there patterns resembling those of rheas or any
other ratite. This evidence, of course, cannot be used to refute the possibility that ti-
namous may be more closely related to rheas than to any other living taxon, but neither
does it provide any support for that hypothesis. Whether analyses of other taxa with pre-
cocial young might suggest alternate relationships for the Tinamiformes (e.g., Galli-
formes, see Verheyen, 1960; Chandler, 1916) is problematical but worthy of investigation.
ACKNOWLEDGMENTS
Specimens used in this study were borrowed from or examined at tiie American Museum of Natural His-
tory, Carnegie Museum, The University of Kansas Museum of Zoology, The University of Michigan Museum of
Zoology, Peabody Museum, Museum of Vertebrate Zoology, Field Museum of Natural History, Philadelphia
Academy of Sciences, Los Angeles County Museum, and San Diego Natural History Museum. I am indebted to
the curators of these collections for their assistance. I am also grateful to K. C. Lint and James Dolan for mak-
ing the facilities of the San Diego Zoo available to me.
K. C. Parks, R. W. Storer. J. Strauch, G. A. Clark, Jr., W. Bock, P. Devillers, and J. Cracraft provided use-
ful criticisms of the manuscript.
LITERATURE CITED
Beebe, W.
1925. The Variegated Tinamous Crypturus variegatus variegatus (Gmelin). Zoologica 6(2): 195-227.
Bock, W. J.
1963. The cranial evidence for ratite affinities. Proc. XIII Intern. Ornithol. Cong.: 39-54.
Boetticher, H. von.
1934. Beitrage zu einem phylogenetisch begrundeten naturlichen System der Steisshuhner (Tinami) auf
Grund einer taxonomisch verwarbaren Charaktere. Jenaische Zeits. fur Naturwiss. 69: 169-192.
Brodkorb, P.
1963. Catalogue of fossil birds. Bull. Florida State Mus. 7(4): 179-293.
1971 JEHL: RATITES ANDTINAMOUS 301
Chandler, A. C.
1916. A study of the structure of feathers, with reference to their taxonomic significance. Univ. California
Publ. Zool. 13(11): 243-446.
de Beer, G.
1956. The evolution of ratites. Bull. Brit. Mus. (Nat. Hist), Zool. 41: 59-70.
1964. Ratites, phytogeny of the, p. 681-685. In A. L. Thomson (ed.), A new dictionary of birds. McGraw-
Hill, New York.
Delacour, J., and E. Mayr.
1945. The family Anatidae. Wilson Bull. 57: 3-55.
de Schauensee, R. M.
1966. The species of birds of South America with their distribution. Livingston Publ. Co., Narberth, Penn-
sylvania, 577 p.
Glenny, F. H.
1965. Main cervical and thoracic arteries of some flightless birds. Ann. Zool. 5(1): 1-8.
Jehl, J.-R., Jr.
1968. Relationships in the Charadrii (shorebirds): a taxonomic study based on color patterns of the downy
young. San Diego Soc. Nat. Hist. Memoir 3.
Miranda-Ribiero, A. de.
1938. Notas ornithologicas (XIII). Tinamidae. Rev. do Mus. Paulista 23: 667-788.
Oliver, W. R. B.
1957. New Zealand birds. 2nd ed. A. H. and A. W. Reed, WeHington, N. Z. 661 p.
Parkes. K. C, and G. A. Clark, Jr.
1966. An additional character linking ratites and tinamous, and an interpretation of their monophylv. Con-
dor 68: 459-471.
Peters, J. L.
1931. Check-list of birds of the world. Vol. 1. Harvard University Press, Cambridge, Mass. 345 p.
Rand, A. L., and E. T. Gilliard.
1967. Handbook of New Guinea birds. Wiedenfeld and Nicholson, London. 612 p.
Salvadori, T.
1895. Catalogue of Chenomorphae (Palamedeae, Phoenicopteri. Anseres), Crypturi, and Ratite in the col-
lection of the British Museum. Brit. Mus. (Nat. Hist.), London.
Sauer, E. G. P., and E. M. Sauer.
1966. The behavior and ecology of the South African ostrich. Living Bird. Fifth Annual, p. 45-47.
Short, L. L., Jr.
1967. A review of the genera of grouse (Aves, Tetraoninae). Amer. Mus. Novitates 2289.
Sibley, C. G.
1960. The electrophoretic patterns of avian egg-white proteins as taxonomic characters. Ibis 102: 215-284.
Storer, R. W.
1967. The patterns of downy young grebes. Condor 68: 469-478.
Verheyen, R.
1960. Les Tinamous dans les systemes omithologiques. Bull. Inst. Roy. Sci. Nat. de Belgique 36( 1): 1-11.
Ward, R. A.
1957. A study of the host distribution and some relationships of mallophaga parasites on birds on the Order
Tinamiformes. Part I. Ann. Ent. Soc. Amer. 50: 335-353.
San Diego Natural History Museum, P. O. Box 1390, San Diego, California 92112
J/1U:d. CuiviP. ^.k^u;
LIBRARY
FEB 1 8 1972
HARVARD
UNIVERSITY
CENOZOIC CALCAREOUS NANNOFOSSILS
FROM THE PACIFIC OCEAN
DAVID BUKRY
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 14 7 DECEMBER 1971
CENOZOIC CALCAREOUS NANNOFOSSILS
FROM THE PACIFIC OCEAN
DAVID BUKRY
ABSTRACT.— The typical stratigraphic ranges of key Cenozoic calcareous nannofossil taxa in Pacific
Ocean cores are presented. Two new genera and 16 new species from Pacific Ocean cores are described;
these include: CoccoUihus magnicrassus. C. miopelagicus. Coccolithus? orangensis. Cyclicargolithus n. gen.,
Cvcloliihella kariana. Discoaster bifax. D. intercalaris, D. loehlichii. D. neorectus, Fasciciilithiis clinalus. He-
licoponiosphaera heezcnii, H. rhomba. Sphenolithus coniciis, S. obtiisus, S. spiniger, Strialococcolilhus n.
gen., Siriatococcolithus pacificanus, and Triquetrorhabdulus milowii.
INTRODUCTION
Calcareous nannofossils are microscopic calcite skeletal elements produced largely
by Coccolithophyceae— marine, planktonic, one-celled, golden-brown algae. These 1-50
micron skeletal elements, composed of many still smaller calcite crystallites, have been
preserved in marine strata since their earliest known occurrence in deposits of Early Ju-
rassic age. Owing to their great abundance and evolutionary structural diversification,
nannofossils can be used to subdivide marine strata into a sequence of biostratigraphic
zones. The planktonic life-style of fossil nannoplankton in the light-penetrated and there-
fore current-influenced layer of the ocean ensured rapid dispersal of new forms. This fac-
tor contributes to the utility of nannofossils in transoceanic stratigraphic correlation.
Light microscopes set at magnifications of 250-1000 X and electron microscopes at
1000-20,000 X are used in the identification of nannofossils. For rapid comparison of
many samples and for stratigraphic zonation utilizing assemblages, the light microscope is
most convenient. For delineation of detailed surficial crystallite patterns that aid in phy-
logenic and taxonomic studies, the electron microscope is useful. But the internal crystal-
lographic orientation of the individual crystallites provides important distinctions for
taxonomic discrimination, and this information comes only from cross-polarized light mi-
croscopy.
Approximately 3000 species of Cenozoic nannofossils have been described. The most
important forms for zonation are the star-shaped discoasters, the placoliths (shaped like
sewing-machine bobbins), and the cone-shaped sphenoliths (examples in Plate 1). Tax-
onomic distinctions within these groups are based for discoasters on the number and form
of the rays and on accessory ornamentation of the rays and central area as seen in plan
view; for placoliths on the crystallite crystallographic orientation, on relative proportions
and circularity of the central area and rims of the upper and lower shields, and on any
distinctive central-area ornamentation; for sphenoliths on the orientation of basal and
apical spines as seen in cross-polarized light at various angles to the polarization.
ZONATION
The potential of calcareous nannofossils for biostratigraphic zonation was first in-
dicated by Bramlette and Riedel (1954), and in 1967 the first general sequence of Ceno-
zoic calcareous nannofossil zones was published (Bramlette and Wilcoxon, 1967; Hay et
al., 1967). These zones were based on study of stratigraphic type stages in Europe, the Ci-
pero and Lengua Formations of Trinidad, long cored sequences from the JOIDES Blake
Plateau drilling, and numerous short cores taken on oceanographic expeditions. This
tYamework has provided useful guidelines for later studies based on the Deep Sea Drilling
Project cores and on restudy of type-stage sections using the ranges of many newly de-
scribed species. Recent studies furthering zonal refinement include: Gartner, 1969, 1971;
SAN DIEGO SOC. NAT. HIST.. TRANS. 16 (14): 303-328, 7 DECEMBER 1971
304 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
Bukry and Bramlette, 1970: Milow, 1970; Roth, 1970; Bukry, 1971; and Martini and
Worsley, 1971. Because nannofossils are small and occur in vast numbers in a given
sample, the whole assemblage can be scanned in a few minutes. Therefore, it is conven-
ient to base zonal identifications on the character of the whole assemblage. Boundaries of
zonal units (table 1) can usually be identified by closely spaced first and last occurrences
of several species. Some of the key species used to recognize zone and subzone boundaries
are indicated in tables 2 and 3. A full discussion of the character of these zonal units is
given in the report on nannofossil stratigraphy for Deep Sea Drilling Project Leg 16 that
explored the eastern equatorial Pacific Ocean (Bukry, in press).
NANNOFOSSIL DISSOLUTION
The diversity of species that compose nannoplankton assemblages is, to a large ex-
tent, controlled by selective dissolution of skeletal elements between the time of death in
surface water and the time of final burial below the ocean bottom. Skeletal elements that
bypass or survive ingestion by nannoplankton herbivores— microscopic, planktonic proto-
zoans such as foraminifers and radiolarians (Tappan, 1971)— owing to their calcite com-
position are subjected to increased inorganic dissolution rates in progressively more
calcite-undersaturated water at progressively greater ocean depth (Peterson, 1966). Al-
though Berger (1970) has estimated that about four-fifths of the calcite supphed to the
ocean floor is being redissolved, nannoplankton skeletons are more resistant to this effect
than are other calcite microplankton skeletons. Part of their resistance to dissolution may
be the result of incorporation of acid-resistant, fibrillar, cellulose-like polysaccharide ma-
terial with the skeletal calcite (Franke and Brown, 1971). Chave and Suess (1967) have
stated that organic coatings inhibit the precipitation of calcium carbonate on carbonate
surfaces. Such coatings that inhibit carbonate-sea water interactions probably also retard
dissolution of calcium carbonate particles in undersaturated water (Smith et al., 1968;
Pytokowicz, 1969).
In addition to the possible organic coatings, variation in nannoplankton skeletal
thickness relative to optic-axis orientation apparently accounts for some of the solution
resistance of nannoplankton. Some of the most resistant taxa have a similar relation be-
tween the exposed surfaces and optic-axis orientation of their calcite crystallites. Dis-
coaster and the upper shields of the placoliths Coccolithus and Cvclococcolithina are
typically the last remnants of a strongly dissolved fossil nannoplankton assemblage. In
cross-polarized light, all of these appear dark because of the vertical orientation of the
principal optic axis of their crystallites. Diff'erences in dissolution rates along dilTerent
crystallite axes, in conjunction with variation in crystallite thickness, could cause a signifi-
cant range of structural differences to explain selective solution along taxonomic group-
ings.
The most diverse assemblages, those from warm-water areas that are little affected
by calcite undersaturation, occur in deposits from the sublittoral shelf to the basal conti-
nental slope (approx. depths 50-2000 m). Such assemblages, which may contain common
pentagonally-shaped Braarudosphaera or Micrantholithus (JOIDES Blake Plateau cores,
for example), have been characterized as "nearshore'' (Bramlette and Martini, 1964). As
these nannoplankton are distinctively shaped, their general absence in deep-ocean (2000-
6000 m) sediment is easy to determine. Indeed, some of these presumed nearshore in-
dicator taxa have been reported in mid-ocean plankton and island samples, for example,
Braarudosphaera in North Atlantic water (Hulburt, 1962; Hulburt and Rodman, 1963)
and Braarudosphaera and Micranihohthus in shallow-water sediment from the Tonga Is-
lands (Bramlette, 1970), suggesting that they are not restricted to inshore areas by envi-
1971
BUKRY: CENOZOIC PACIFIC NANNOFOSSILS
305
ronmental factors while they are alive. Instead, these forms are probably poorly resistant
to solution. Their spotty and far-flung distribution suggests that they are preserved in
shallow (near-saturated) depositional areas and dissolved in deep (undersaturated) areas.
Table 1. Cenozoic calcareous nannoplankton zones and subzones. Approximate ages of series and subseries in
million years from Berggren ( 197 1 ).
SERIES OR
SUBSERIES
AGE
M. Y.
ZONE
SUBZONE
HOLOCENE
0.01
1.85
3.7
5.1
- 10.5
- 14
- 22.5
- 38
- 45
- 49
- 53.5
- 65
Errrlliania huxleyi
Gephyroaapsa oaeaniaa
PLEISTOCENE
Cocaolithus doroniooides
Gephyrocapsa aaribbeaniaa
Emiliania annula
UPPER
PLIOCENE
Disaoaster brouueri
Cyaloaoaaolithina maaintyrei
Disaoaster pentaradiatus
Disaoaster tamalis
Re tiouto fenestra pseudownbilioa
Disaoaster asyrnmetriaus
LOWER
Sphenolithus neoabies
PLIOCENE
Ceratotithus trioorniculatus
Ceratolithus rugosus
Ceratolithus amplifiaus
Triquetrorhabdulus rugosus
Disaoaster quinqueramus
Ceratolithus primus
UPPER
Disaoaster berqgrenii
MIOCENE
Disaoaster neohamatus
Disaoaster neoreatus
Disaoaster bellus
Disaoaster hamatus
MIDDLE
Catinaster aoatitus
MIOCENE
Disaoaster exitis
Disaoaster kugleri
Cocaolithus miopelagiaus
Sphenolithus heteromorphus
Heliaopontosphaera ampliaperta
LOWER
Sphenolithus belemnos
Triquetrorhabdulus aarinatus
Disaoaster druggii
MiOCENb
Disaoaster deflandrei
Diatyoaoaaites abisectus
Sphenolithus aiperoensis
OLIGOCENE
Sphenolithus distentus
Sphenolithus predistentus
Heliaopontosphaera retiaulata
Cyaloaoaaolithina formosa
Cocaolithus subdistichus
UPPER
Disaoaster barbadiensis
Cyaliaargolithus reticulatus
EOCENE
Disaoaster tani tani
Retiaulofenestra umbiliaa I
Nannotetrina quadrata
Cocaolithus staurion
MIDDLE
Chiasmolithus gigas
EOCENE
Disaoaster mirus
Disaoaster sublodoensis
Disaoasteroides kuepperi
Disaoaster lodoensis
LUwhK
Tribraahiatus orthostylus
EOCENE
Disaoaster diastypus
Disaoaster multiradiatus
Campy losphaera eodela
Chiasmolithus bidens
Disaoaster nobilis
PALEOCENE
Disaoaster mohleri
Heliolithus kleinpellii
Fasaiaulithus tympaniformis
Cruaiplaaolithus tenuis
1
. — J
306
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Comparison of fossil nannoplankton assemblages from Deep Sea Drilling Project
cores taken from many ocean depths shows that the most solution-resistant genera— Z)/5-
coaster, Coccolithiis, Cyclococcolithina, Reticulofenestra, and Dictyococcites—occ^xx in
nearly all samples, shallow and deep. In fact, these taxa persist in some deep-ocean red-
clay deposits after all other calcitic microfossils have been dissolved. On the contrary,
such taxa as Braanidosphaera, Micrantholithus, Transversopontis, Scyphosphaera, and He-
licopontosphaera are absent from red-clay deposits. By comparing many different coeval
Table 2.
dashed.
Typical ranges of some key late Cenozoic calcareous nannoplankton. Rare or sporadic occurrence
MIOCENE
PLIOCENE
PLEISTOCENE >
AND HOLOCENE M
C3'3<3Co&:Coejt)rit)t)t)i3o^<^<^<J3t)t)t3r^triQ:i£::i ct]
ft U5
ft
ft
TO
o
3
o
/s
ha
s-
s;
Co
3
O
h3
TO
ft
K
Co
S"
iyxj
o
Co
TO
s
d-
ft
^S
ft
ft.
r>.
ft
<M-
s;
Co
Emiliania huxleyi
Gephyrocapsa oceaniaa
G. aaribheaniaa
Ceratolithus cristatus
Cyctoaoccolithina maaintyrei
Ceratolithus rugosus
Discoaster brouweri
D. pentaradiatus
D. sur cuius
D. asymmetricus
D. tamalis
D. variabilis decorus
Reticulofenestra pseudoumbiliaa
Sphenolithus neoabies
Ceratolithus tricomiculatus
C. amplificus
C. primus
Triquetrorhabdulus rugosus
Discoaster quinqueramus
D. berggrenii
D. neorectus
D. neohamatus
D. bellus
D. hamatus
Catinaster coalitus
Discoaster exilis
Discoaster kugleri
Coccolithus miopelagicus
Sphenolithus heteromorphus
Cyclicargolithus floridanus
Helicopontosphaera ampliaperta
Sphenolithus belemnos
Discoaster druggii
Triquetrorhabdulus carinatus
Dictyococcites abisectus
1971
BUKRY: CENOZOIC PACIFIC NANNOFOSSILS
307
assemblages, a general order of selective solution can be determined that reflects the rela-
tive depth of ancient ocean areas. The following list ranks lower Cenozoic nannofossil
genera from those least common in deep-ocean sediment, at the beginning, to those most
characteristic of very deep sediment at the end: Transversopontis, Syracosphaera, Rhab-
dosphaera, Discolithina [perforate], MicranthoUthiis, Braarudosphaera, Lophodolil/ms,
Scvphosphaera, Helicopontosphaem, Discolithina [imperforate], Sphenolithus, Chiasmo-
lithus, Reticulofenestra, Dictyococcites, Cyclococcolithina, Coccolithus, Discoaster.
Coeval samples from two nearby Pacific Ocean sites of greatly differing water depth
are cited below as specific examples of taxonomically selective dissolution. Assemblages
from the deep-water site of the pair are always less diverse. The species common to both
the shallow site (DSDP 62: depth 2591 m, lat 1°52.2'N., long 141°56.0'E.) and the deep
site (DSDP 63: depth 4472 m, lat 0°50.2'N., long 147°53.5'E.), are excluded, and the taxa
listed below for each geologic subseries and zone are those solution-prone forms that oc-
Table 3. Typical ranges of some key early Cenozoic calcareous nannoplankton.
PALEOCENE
EOCENE
OLIGOCENE
AGE
TO
S
K
Co
•
1
ft
s
Co
•
TO
TO
1
S"
TO
>^
s
o
^.
Co
cy
ft.
TO
3
Co
pi
TO
O
TO
^~»
ft
a.
ft
Co
n-
Co
O
Co
s;
Co
t3
O
TO
S
Co
Co
?^
K
TO
t3
TO
"^
Co
S
C3-
^^
o
§^
TO
S
Co
Co
3
Co
Co
Co
ri-
ft
^.
O
to
c;)
ft
t3
ri-
ft
S
rf
ft
S
TO
C1-
Ci
K
ft
n-
Co
p
Co
s;
C3^
r^.
Co
<r1-
Ci
Co
pi
i
Co
a
Co
13
TO
C^
r^
Co
rl-
TO
n-
K
Co
Co
ft.
Co
TI-
TO
s
d-
Co
"C
TO
O
TO
S
Co
Co
Cfi tsi /
GOO /
ro » z /
O / o
Z / en Q
w / Tj n
/ wo
/ no
/ M f
/ MM
/ Crt H
/ ^
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Sphenolithus aiperoensis
Diatyoaoaaites abiseatus
Sphenolithus distentus
Discoaster tani tani
Sphenolithus predistentus
Reticulofenestra umbiliaa
Cyclococcolithina formosa
Coccolithus subdistichus
Cyclicargolithus reticulatus
Discoaster barbadiensis
Chiasmolithus grandis
Coccolithus staurion
Nannotetrina quadrata
Chiasmolithus gigas
Discoaster mirus
D. sublodoensis
Rhabdosphaera inflata
Coccolithus crassus
Discoasteroides kuepperi
Discoaster lodoensis
Tribrachiatus orthostylus
Discoaster diastypus
Campy losphaera eodela
Discoaster multiradiatus
D. nobilis
D. mohleri
Chiasmolithus bidens
Fasciculithus tympaniformis
Heliolithus kleinpellii
Cruaiplacolithus tenuis
308
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
c5
1971 BUKRY: CENOZOIC PACIFIC NANNOFOSSILS 309
cur at the shallow site alone.
Lower Pleistocene Coccolithus doronicoides Zone:
Discoaster perplexus, Helicopontosphaera sellii, Oolithotus antillarum, Rhabdosphaera
clavigera, Umbilicosphaera mirahilis.
Upper Pliocene Discoaster brouweri Zone:
Discoaster perplexus, Helicopontosphaera sellii, Oolithotus antillarum, Scyphosphaera
apsteinii, S. intermedia, Thoracosphaera spp.
Lower Pliocene Ret iculo fenestra pseudoumbilica Zone:
Discoaster perplexus, Discolithina japonica, Helicopontosphaera kamptneri, H. sellii,
Oolithotus antillarum, Scyphosphaera apsteinii, S. globulata, S. pulcherrima, Thora-
cosphaera spp.
Upper Miocene Discoaster quinqueramus Zone:
Discoaster perplexus, Discolithina japonica, Helicopontosphaera kamptneri, Scyphos-
phaera intermedia, Sphenolithus abies, Thoracosphaera spp.
Middle Miocene Discoaster hamatus Zone:
Discoaster perplexus, Helicopontosphaera kamptneri, Scyphosphaera sp. cf. 5". pulcher-
rima, Sphenolithus abies.
Establishing the relative order of nannofossil dissolution is important to provide in-
formation for interpretating the paleoecology of the assemblages (Douglas. 1971; Law-
rence, 1971), and to improve precision in stratigraphic zonation.
SYSTEMATIC PALEONTOLOGY
Genus Coccolithus Schwarz, 1894
Coccolithus magnicrassus n. sp.
PI. 2, figs. 1-5
Description.— This large, elliptic placolith is characterized by a small central area and
a broad finely striate rim. In light-microscope examination, the central area is prominent
and the rim faint, being at high and low relief with respect to the mounting medium
(n= 1.518). In cross-polarized light the central area is bright, forming a small elliptic col-
lar around an elliptic central opening: whereas the rim is faint, with diffused strongly
curving extinction bands. The upper rim has 55-80 radial crystallites, and is distinctly
larger than the lower rim.
Remarks.— Coccolithus magnicrassus is distinguished from other similar placoliths by
the combined characters of (1) large overall size; (2) small, high relief central area with
simple central opening; (3) broad upper rim, composed of many elements, that is only
moderately bright in cross-polarized light and has diffuse, strongly curving extinction
bands. Toweius craticulus Hay and Mohler is smaller with a narrower rim; Reti-
culofenestra hillae Bukry and Percival has a larger central opening and in cross-polarized
light a fully bright rim with broader less curved extinction bands; Coccolithus crassus
Bramlette and Sullivan is distinctly smaller, and the upper, larger rim is dark in cross-
polarized light. A comparison of C. crassus with C. magnicrassus is shown in PI. 2, fig. 2.
Plate I. Electronmicrographs of carbon-platinum replicas showing surface crystallite patterns of some typical
forms of Cenozoic calcareous nannofossils. 1. Svracosphaera pulchra Lohmann, Pleistocene, Shatsky Rise,
DSDP 47.0-1-4, 77-78 cm. 11,000 X. 2. Helicoponiosphaera kamptneri Hay and Mohler, Pleistocene, Shatsky
Rise. DSDP 47.0-1-4, 77-78 cm. 7,000 X. Diatom fragment at lower right corner. 3. Rhabdosphaera clavigera
Murray and Blackman, Pleistocene, Shatsky Rise, DSDP 47.0-1-4, 77-78 cm. 8,000 X. 4. Group of placoliths
and discoasters. Pliocene, Caroline Ridge. DSDP 57.2-1-6, 0-3 cm. 1,700 X. 5. Group of placoliths and a dis-
coaster. Miocene, Caroline Ridge, DSDP 55.0- 1 1-5, 78-80 cm. 4,000 X. 6. Group with placolith, discoaster, and
sphenolith. Eocene, Horizon Ridge, DSDP 44.0-3-5, 145- 150 cm. 3,000 X.
310 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Occurrence.— Coccolithus magnicrassus occurs in lower Eocene marine sediment
cored in the North Pacific and North Atlantic Ocean and in the Donzacq Marl of France.
It does not range as high as C. crassus, for it is recognized only from the Discoaster lo-
doensis Zone thus far.
^/ze. — 16-20 microns.
Holotvpe.-\JS^U 176883 (PI. 2. figs. 1-2). -
Paratvpe.-\J'S^M 176884-176886.
Type locality.-\:>^D? 47.2-7-3, 104-105 cm, Shatsky Rise, northwestern Pacific
Ocean.
Coccolithus miopelagicus n. sp.
PI. 2, figs. 6-9
Description.— This large placolith has a medium-sized central area and a broad dis-
tinctly striate rim. In light-microscope examination, both the rim and central area are
prominent. In cross-polarized hght the central area is bright with distinct extinction
bands; the lower (smaller) rim is bright, but the upper rim is dark. A small simple ellipti-
cal opening or sht in the central area is aligned with the long axis of the placolith.
Remarks.— The similarly constructed species, Coccolithus eopelagicus (Bramlette and
Riedel) is distinguished from Coccolithus miopelagicus by several criteria: (1) Rim counts
for C. eopelagicus are higher, 50 to 61 instead of 40 to 49; (2) Measurement often typical
specimens shows that the central area of C. eopelagicus occupies a greater percentage of
the long axis, 59+ 1 percent instead of 50 + 5 percent; (3) The central area also occupies a
greater percentage of the short axis, 49 + 2 percent instead of 42 + 3 percent. The general
distinction of C. miopelagicus from C. eopelagicus and large specimens of the younger C.
pelagicus (Wallich) s.s. is the distinctly smaller central area of C. miopelagicus with respect
to the rim area. Large specimens of C miopelagicus are 20 microns in major axis length,
but as indicated by Bramlette and Riedel (1954) these middle Tertiary forms, similar to C.
eopelagicus. are generally smaller.
Occurrence.— Coccolithus miopelagicus is most common in lower and middle Miocene
sediment from the Atlantic and Pacific Oceans and Caribbean Sea. The appearance of C.
miopelagicus populations near the Oligocene- Miocene boundary is probably a gradual
transition from C. eopelagicus resulting from increasing temperatures. Some tropical
middle Eocene C. eopelagicus populations have a fair percentage of associated C. sp. cf.
C. miopelagicus, whereas lower Oligocene (cooler temperatures) and high latitude middle
Eocene assemblages contain only C. eopelagicus. The disappearance of C. miopelagicus at
the Catinaster coalitus Zone is abrupt.
Size. — \2> to 18 microns.
Holotvpe.-\JS^U 176888 (PI. 2, figs. 7-8).
ParatYpe.-\]?>^U 176887, 176889.
Type /ofa//7v.— DSDP 63.0-3-4, 80-81 cm. East Caroline Basin, western equatorial
Pacific Ocean.
Plate 2. Photomicrographs: 2,000 X. 1-5. Coccolithus magnicrassus. n. sp. ( 1) holotype USNM 176883, DSDP
47.2-7-3, 104-105 cm; (2) holotype at left, Coccolithus crassus Bramlette and Sullivan at right, cross-polarized, (3)
USNM 176884, (4) USNM 176885, cross-polarized, (5) USNM 176886, DSDP 47.2-7-2, 100-101 cm, cross-polar-
ized. 6-9. Coccolithus miopelagicus, n. sp. (6) USNM 176887, DSDP 63.1-8-3, 80-81 cm. (7) holotype USNM
176888, DSDP 63.0-3-4, 80-8f cm, (8) holotype, cross-polarized, (9) USNM 176889, DSDP 70.0-3-3. 63-64
cm. 10-11. Coccolithus? orangensis, n. sp. (10) holotype USNM 176890, DSDP 55.0-13-1, 120-121 cm, (11)
holotype, cross-polarized; three small Cvclicargolithus sp. cf. C floridamts (Roth and Hay) with straight extinc-
tion lines are present below the holotype.
1971
BUKRY: CENOZOIC PACIFIC NANNOFOSSILS
311
312 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Coccolithus? orangensis n. sp.
PL 2. fig. 10; PI. 3, figs. 1-3
Description— This small elliptic placolith has exceptionally high relief with respect to
the mounting medium (n= 1.518). Individual crystallites are not discernible by light mi-
croscope. In cross-polarized light, the coccolith is bright except for a small central open-
ing. The extinction bands are strongly curved and become abruptly diffused in a narrow
margin at the outer perimeter that is slightly brighter than the rest of the nannofossil.
Remarks.— Coccolithus? orangensis is not readily assignable to any presently de-
scribed genus. What might be a side view of this species is shown at the right side of fig-
ure 1-2 of Plate 3. The exceptionally high relief, characteristically exceeding that of
discoasters, small size, and bright perimeter help to distinguish C? orangensis from other
elliptic nannofossils. There appears to be a narrow, high relief collar structure around the
small central opening, resulting in a ring-like depression between the perimeter and col-
lar. C? orangensis is distinguished from Cyclicargolithus fhridanus (Roth and Hay) by its
elliptic outline, high relief, and bright perimeter.
Occurrence.— Coccolithus? orangensis is never abundant but is a consistent low-fre-
quency member of upper Oligocene and lower Miocene assemblages assigned to the
Sphenolithus ciperoensis Zone and Triquetrorhahdulus carinatus Zone. It occurs in both
Pacific and Atlantic Ocean deep-sea cores.
Size.— A to 6 microns.
Holot\pe.-\JS^U 176890(P1. 2, figs. 10-11).
Paratypes.-\JSNU 176891-176892.
Type locality — DSDP 55.0-13-1, 120-121 cm, Caroline Rise, western equatorial Paci-
fic Ocean.
Cyclicargolithus n. gen.
Description.— CirculaT to subcircular placoliths constructed of two shields connected
by a central tube that may be closed or open. In plan view, the upper shield is bright in
cross-polarized light.
Type species.— Coccolithus floridanus Roth and Hay in Hay and others, 1967, Gulf
Coast Assoc. Geol. Socs. Trans., v. 17, p. 455, PI. 6, figs. 1-4.
Remarks.— J\\Q genus Cyclococcolithina Wilcoxon (1970) included circular to sub-
circular forms of two kinds— those with dark upper shields and those with bright upper
shields when viewed in cross-polarized light. The significant difference in the orientation
of the optic axis of the shield crystallites that accounts for this distinction is considered to
be of generic rank. Therefore Cyclococcolithina Wilcoxon is herein restricted to those
forms having dark upper shields such as Cyclococcolithina formosa and the type species
Cyclococcolithina leptopora. Forms with bright upper shields are transferred to Cycli-
cargolithus. The circular to subcircular outline of this genus distinguishes it from elliptical
Coccolithus Schwarz, which has a dark upper shield in cross-polarized light.
Cyclicargolithus floridanus (Roth and Hay) n. comb.
Coccolithus floridanus Roth and Hay in Hay and others, 1967. Gulf Coast Assoc. Geol. Socs. Trans. 17: 455. PI.
6, figs. 1-4.
Cyclococcolilhus neo^ammalion Bramlette and Wilcoxon, 1967. Tulane Studies Geol. 5: 104. PI. 1. figs. 1-3; PI. 4,
figs. 3-5.
Cvclococcolilhus floridanus (Roth and Hay), ot^ Mulier, 1970. Geologica Bavarica 63: 1 13, PI. 2, figs. 1-3.
Cyclococcolithus fhridanus {Roth and Hay), of Roth. 1970, Eclogae Geo!. Helv. 63: 854, PI. 5, fig. 6.
^£'m^//A:.v.— The original definition of Coccolithus floridanus Roth and Hay describes a
small (3.6 to 5 micron) eUiptic placolith. The original definition of Cyclococcolithus neo-
1971 BUKRY: CENOZOIC PACIFIC NANNOFOSSILS 313
gammation Bramlette and Wilcoxon describes a medium sized (6 to 12 micron) circular
placolith. According to Roth ( 1970) the electronmicrograph paratypes of these two species
show the same construction and number of rim elements. The slight difference in size and
shape noted in the written descriptions probably resulted from the description of end
members of the same species and from the use of different instrumentation.
CvcIicargoUthiis luminis (Sullivan) n. comb.
Cvclococcolithus luminis Sullivan. 1965, Univ. Calif. Puhi. Geol. Sci. 53: 33, PI. 3, figs. 9a, b.
CyclicargoUthus reticulatus (Gartner and Smith) n. comb.
Cvclococcolithus reticulatus Gartner and Smith, 1967, Univ. Kansas Paleont. Contr., Paper 20, p. 4, PI. 5, figs. I-
4.'
Genus Cvclolithella Loeblich and Tappan, 1963
Cyclolithella kariana n. sp.
PI. 3. figs. 4-5
Description.— J\\\s small circular coccolith has a small circular central opening that
occupies about a quarter to a third of the diameter. The thick upper shield is composed of
about 12 to 20 curving crystallites that are strongly imbricated, indicated by the spiralling
effect as focus is raised or lowered through the coccolith. The margin of the central open-
ing and outer perimeter is generally smooth, but the perimeters of a few etched(?) speci-
mens appear slightly scalloped. In cross-polarized light, no sharp black extinction bands
are seen; instead, four light-gray rays each occupy two or three crystallites.
Remarks.— Cyclolithella kariana is distinguished from other species of Cvclolithella by
the small central opening. It is further distinguished from the most similar species, Cvcl-
olithella pactilis Bukry and Percival, by curved, gray extinction bands instead of straight
black ones, when viewed in cross-polarized light.
Occurrence.— Cvclolithella kariana occurs commonly in lower Eocene sediment as-
signed to the Discoaster lodoensis Zone at DSDP 47.2 on the Shatsky Rise of the north-
western Pacific Ocean.
Size.— 6 to 9 microns.
Holotvpe.-VSl<\M 176893 (PI. 3. figs. 4-5).
Type' locality.-DSDP 47.2-7-3, 104-105 cm, Shatsky Rise, northwestern Pacific
Ocean.
Genus Discoaster Tan, 1927
Discoaster bifax n. sp.
PI. 3. figs. 6-11
Description.— This small species is constructed of 10 to 15 (typically 14) approx-
imately radial rays that are appressed and terminate in broad points. High central stems
extend from each side of the discoaster. The stem on one side is slender, occupying only
25 percent of the shield diameter, whereas the stem on the other side is consistently twice
as wide, occupying 50 percent of the shield diameter. No birefringence is seen in cross-
polarized light.
Remarks.— Discoaster bifax is distinguished from other compact discoasters, Dis-
coaster barbadiensis Tan, D. circularis Hoffmann, D. multiradiatus Bramlette and Riedel,
D. saipanensis Bramlette and Riedel, and D. salisburgensis Stradner, by tall central stems
314
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
1971 BUKRY: CENOZOIC PACIFIC NANNOFOSSILS 315
on both sides of the discoaster shield instead of only on one side. It is distinguished from
other double-stemmed forms such as D. diastypus Bramlette and Sullivan and D. bollii
Martini and Bramlette by stems of strongly contrasting width on opposite sides of the
same specimen.
Occurrence.— Discoaster bifax occurs commonly in Stetson 21 at 147-152 cm from the
Atlantic Ocean. Associated nannofossils in this sample such as Chiasmolithus grandis
(Bramlette and Riedel) and Reticulofenestra umbilica (Levin) indicate a middle Eocene
age. D. bifax is sparse in coeval Pacific Ocean sediment from the Reticulofenestra umbilica
Zone of the East Pacific Rise, DSDP 74.0-12-3, 24-25 cm.
Size.— 5 to 10 microns.
Holotvpe.-VSNM 176895 (PI. 3, figs. 7-8).
Paratvpes.-VSNM 176894. 176896-176897.
T}pe locality: Stetson 21, 147-152 cm, northwestern Atlantic Ocean.
Discoaster intercalaris n. sp.
PI. 3, fig. 12; PI. 4, figs. 1-2
Discoaster brouweri Tan, of Stradner and Papp, 1961, (partim). Jahrb. Geol. Bundesanst. [Wien], v. 7, p. 85, PI.
20, fig. 6.
Description.— This medium-sized, six-rayed species has a large central area and a cen-
tral stem. The symmetric radially arrayed rays show a distinct tapering and terminate in
simple rounded points. Some specimens have a small indentation at the tip.
Remarks.— Discoaster intercalaris is a simple form that is distinguished from the Dis-
coaster variabilis group by a single-pointed termination of the rays instead of a broadly
flaring bifurcation. It is distinguished from Discoaster brouweri brouweri Tan, emended,
by the wide central area and marked taper of the rays. It is distinguished from Discoaster
neorectus Bukry by its smaller size and the straight to slightly concave sides of the rays.
Occurrence.— Discoaster intercalaris is common in upper Miocene to upper Pliocene
marine sediment cored by the Deep Sea Drilling Project during Leg 5 at sites off" northern
California. The cool-water aspect of the associated nannofossil assemblages and the sim-
ilarity in form of D. intercalaris and D. variabilis variabilis suggest that D. intercalaris may
be a cool-water relative of D. variabilis variabilis that failed to develop bifurcations. This
possibility is indicated by the small size of the bifurcations of D. variabilis variabilis speci-
mens associated with D. intercalaris. More southerly populations of D. variabilis variabilis
have larger, more robust terminations (see Martini and Bramlette, 1963, PI. 104, figs. 4-8).
Size. — \Q to 16 microns.
Holotvpe.-\}^^M 176899(P1. 4. fig. 1).
Paraiypes.-\J'S>NU 176898, 176900.
Type /oca//7v.— DSDP 36-12-5. 77-78 cm, western flank of Gorda Rise, northeastern
Pacific Ocean.
Discoaster loeblichii n. sp.
PI. 4, figs. 3-5
Plate 3. Photomicrographs: 2.000 X. 1-3. Coccolithus? orangensis n. sp. (1) USNM 176891, DSDP 74.0-4-4,
63-64 cm, cross-polarized, plan view on left, (2) same, bright field, (3) USNM 176892, DSDP 77B-37-4, 65-66
cm. cross-polarized. 4-5. Cvclolithella kariana n. sp. (4) holotype USNM 176893, DSDP 'M. 1-1-7,. 104-105 cm,
(5) holotype, cross-polarized. 6-11. Discoaster bifax n. sp. (6) USNM 176894, STETSON 21, 147-152 cm, tilted,
(7) holotype USNM 176895, high focus, (8) holotype, low focus, (9) USNM 176896, tilted, (10) USNM 176897,
high focus, (11) same, low focus. 12. Discoaster intercalaris n. sp. (12) USNM 176898, DSDP 36-12-5, 77-78
cm.
316 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
Description.— This small- to medium-sized, six-rayed species has a central area oc-
cupying about a third of the total discoaster diameter. It has a small central knob, and the
rays are distinctly tapered, being widest near the central area. The tips of the rays have
distinctive unequal bifurcations that are bent slightly out of the plane of the rays. Both
limbs of the bifurcation taper to points, but all six sets show one Umb that is consistently
more than twice as long as the other. The sense of direction of the resulting asymmetric
bifurcation is always the same for all six rays of a given specimen.
Remarks.— Besides Discoaster loeblichii, the only other six-rayed discoaster with or-
dered, therefore primary, crescent-forming bifurcations is Discoaster calcaris Gartner,
which is a large form more comparable in size, proportions, and occurrence to five-rayed
D. hamatiis Martini and Bramlette. Discoaster loebichii is distinguished by a proportio-
nally larger central area, by shorter, broader, and more tapering rays, and by a smaller
average size (12 to 16 microns instead of 16 to 22 microns). It appears to have been de-
rived from the Discoaster variabilis group, and differs from D. variabilis variabilis Martini
and Bramlette by the unequal bifurcations that form an asymmetric crescent at the ray
tips.
Occurrence.— Discoaster loeblichii was a limited stratigraphic range in the early late
Miocene Discoaster neohamatus Zone, where it is most common in the middle to upper
part of that zone. Geographically, D. loeblichii is known from the tropical Pacific Ocean
areas cored during Deep Sea Drilling Project Legs 7 to 9.
Size. — \ 1 to 16 microns.
Holotvpe.-XJS^M 176902 (PI. 4, fig. 4).
Paraiype.-\}^^U 176901, 176903.
Type locality. —DS>D? 83A-15-6, 130-131 cm, Panama Basin, eastern equatorial Paci-
fic Ocean.
Discoaster neorectus n. sp.
PL 4, figs. 6-7
Description.— This gigantic six-rayed species has a small central stem but no sepa-
rately marked central area. The rays are long and symmetrically arranged, with sides that
are straight or slightly convex. The rays have a slight taper and terminate in simple sharp
points.
Remarks.— Discoaster neorectus has simple, pointed terminations that distinguish it
from Discoaster brouweri brouweri Tan, emended, which has rays bent like umbrella ribs,
and D. brouweri rutellus Gartner, which has blade-like wedges at the end of each ray. Dis-
coaster neorectus is distinguished from D. intercalaris by the narrower taper of the rays
and by the lack of a significant central area.
Occurrence.— \n nannofossil assemblages from the Pacific Ocean, Discoaster neorectus
is common in only a limited stratigraphic horizon of the upper Miocene, upper Discoaster
neohamatus Zone to lower Discoaster quinqueramus Zone. The unusually large size of this
species makes it a convenient guide. Slightly less robust specimens than those of the Paci-
fic, but equally large, are common in Core DSDP 3-10 from the Gulf of Mexico.
Size.—2Q to 38 microns.
Plate 4. Photomicrographs: 2,000 X. unless noted otherwise. 1-2. Discnaster intercalaris n. sp. (1) holotype
USNM 176899, DSDP 36-12-5, 77-78 cm, (2) USNM 176900. 3-5. Discoaster loeblichii n. sp. (3) USNM
176901, DSDP 83A-16-4, 64-65 cm, (4) holotype USNM 176902. DSDP 83A-15-6, 130-131 cm. (5) USNM
176903, DSDP 83A-16-4, 64-65 cm. 6-7. Discoaster neorectus n. sp. (6) USNM 176904, DSDP 72.0-3-4, 63-64
cm, magnification LOOO X, (7) holotype USNM 176905. 8-9. Fasciculithus clinatus n. sp. (8) holotype USNM
176906, DSDP 47.2-9-5, 77-78 cm, (9) holotype, cross-polarized.
1971
BUKRY: CENOZOIC PACIFIC NANNOFOSSILS
317
^: '*
%
-1
'i
1
318 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Ho/ofvpe.-VSNM 176905 (PI. 4, fig. 7).
Pararype.-VSNM 176904.
Type locality— DSDP 72.0-3-4, 63-64 cm, western flank East Pacific Rise, equatorial
Pacific Ocean.
Genus Fasciculithus Bramlette and Sullivan. 1961
Fasclculithus clinatus n. sp.
Pi. 4, figs. 8-9
Description.— This small simple species has a short conical form with a slightly
rounded top that produces an almost triangular outline in side view. The base line is es-
sentially straight in side view and is slightly longer than the upper sides, which are
straight to slightly convex. In cross-polarized light, a single median extinction band bi-
sects the triangular outline.
Remarks.— Fasciculithus clinatus is distinguished from other species of Fasciculithus
by its small size and almost triangular outline. The only comparable small form, F. tym-
paniformis Hay and Mohler, is cylindric, with parallel instead of inclined sides. Fascicu-
lithus ma^nus Bukry and Percival may have inclined sides for only half of its height and is
much larger than F. clinatus. Also, F. clinatus lacks the pit-and-ridge ornamentation de-
veloped in several other species, such as F. involutus Bramlette and Sullivan.
Occurrence.— Fasciculithus clinatus is common in upper Paleocene sediment of the
Shatsky Rise in the northwestern Pacific Ocean.
^'/ze.— height. 4 to 6 microns.
Holotype.-VSNM 176906 (PI. 4, figs. 8-9).
Type locality.— DSDP 47.2-9-5, 77-78 cm, Shatsky Rise, northwestern Pacific Ocean.
Genus Helicopontosphaera Hay and Mohler, 1967
Helicopontosphaera heezenii n. sp.
PI. 5, figs. 1-5
Description.— This large species has a long bar, aligned with the long axis of the nan-
nofossil. that dominates the central area. The length of the bar is 53 to 61 percent of the
total nannofossil length. The bar is rounded at the ends, and although the sides are nor-
mally smooth, some etched specimens show irregularities suggesting small perforations.
In cross-polarized light, the central bar is brightest when aligned with a polarization di-
rection.
Remarks.— Helicopontosphaera heezenii is distinguished from similar forms such as
Helicopontosphaera lophota (Bramlette and Sullivan) and H. papillata Bukry and Bram-
lette by the greater length and the axial alignment of the central bar. It is distinguished
from H. reticulata (Bramlette and Wilcoxon) by its non-rhomboid shape and non-diago-
nal central bar.
Occurrence.— Helicopontosphaera heezenii is common in the upper middle Eocene at
DSDP 44 on Horizon Ridge in the northwestern Pacific Ocean, where it was first recog-
nized. It also occurs in coeval sediment of Stetson 21, northwestern Atlantic Ocean.
Plate 5. Photomicrographs: 2,UUU X. 1-5. Helicopontosphaera heezenii n. sp. ( 1 ) USNM 176907. DSDP 44.0-4-
6, 145-150 cm. 45°, (2) holotype USNM 176908, 45°. (3) holotype. cross-polarized, 0°, (4) USNM 176909, 45°,
(5) same as hgure 1. cross-polarized 90°. 6-9 Helicopontosphaera rhomba n. sp. (6) USNM 176910, DSDP 54.0-2-
I, 67-68 cm, 45°, (7) holotype USNM 17691 1, DSDP 54.0-2-4, 81-82 cm, 45°, (8) holotype. cross-polarized. 90°.
(9) same as figure 6. cross-polarized, 90°. 10-12. Sphenoliihus conicus n. sp. ( 10) holotype USNM 176912. DSDP
80-5-2. 63-64 cm, 90°, (11) holotype, cross-polarized, 90°. ( 12) holotype, cross-polarized. 45°.
1971
BUKRY: CENOZOIC PACIFIC NANNOFOSSILS
319
320 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
Size. — \3 to 18 microns.
Holotvpe.-VSNM 176908 (PI. 5, figs. 2-3).
Paratypes.-VSNM 176907, 176909.
Tvpe localitv.—DSDP 44.0-4-6, 145-150 cm. Horizon Ridge, northwestern Pacific
Ocean.
Helicopontosphaera rhomba n. sp.
PL 5, figs. 6-9
Description.— This large species has a large elongate central opening that is bridged
by a diagonal bar oriented at about 45° to the long axis of the nannofossil. In cross-polar-
ized light, the diagonal bar is bright when the long axis of the nannofossil is aligned with
a polarization direction, but it is dark at 45°. The short-axis extinction bands are broad
and diffuse when the nannofossil is at 90°. The bar is not in optical continuity with the
rim.
Remarks.— Helicopontosphaera rhomba is distinguished from Helicopontosphaera loph-
ota (Bramlette and Wilcoxon) by its more elongate shape, larger central opening, and
more diffuse short-axis extinction bands. It is distinguished from H. intermedia (Martini)
by its larger, non-sigmoid bar; from H. parallela (Bramlette and Wilcoxon) by its more
open central area and diagonally aligned bar; and from H. bramlettei Mtiller ( = H. wilcox-
onii Gartner) by its more elongate outline and diagonal (approx. 45°) bar.
Occurrence.— Helicopontosphaera rhomba is presently known only from the Philip-
pine Sea in lower middle Miocene deposits cored during DSDP Leg 6.
Size. — \5 to 18 microns.
Holotvpe.-\]S^U 17691 1 (PI. 5, figs. 7-8).
ParatYpes.-\}^^U 176910.
Type locality. -r>S\:>? 54.0-2.4, 81-82 cm, Philippine Sea.
Genus Sphenolithus Deflandre, 1952
Sphenolithus conicus n. sp.
PI. 5, figs. 10-12
Description.— This large species is characterized by its tall triangular outhne in side
view. The several apical spines are partly coalesced to form the triangular to rounded
triangular upper section of the nannofossil. In cross-polarized light, the base is divided
into quadrants by the extinction bands when the long axis of the nannofossil is ahgned
with a polarization direction. The height of the lower quadrants is equal to or slightly
greater than the upper quadrants. The apical complex is bright when oriented at 45° to
the polarization directions.
Remarks.— Sphenolithus conicus could be mistakenly identified as a large Spheno-
lithus heteromorphus Deflandre but is distinguished by the greater proportion of the fossil
that is formed by the basal quadrants instead of by the apical complex. It is distinguished
from S. moriformis (Bronnimann and Stradner) by its triangular instead of hemispheric
outline.
Occurrence.— Sphenolithus conicus occurs in lower lower Miocene sediment of the
lower Triquetrorhahdulus carinatus Zone from the Pacific Ocean.
Size.—l to 12 microns.
//o/o/v/7e.-USNM 176912(P1. 5, figs. 10-12).
Type locality.-U^D? 80-5-2, 63-65 cm, East Pacific Rise, eastern equatorial Pacific
Ocean.
1971 BUKRY: CENOZOIC PACIFIC NANNOFOSSILS 321
Sphenolithus obtusus n. sp.
PI. 6, figs. 1-9
Description.— This species has a short cycle of small basal spines and a large tapering
apical spine constructed of two vertically matched halves. The apical spine halves are
flush and terminate together. Seen in side view and cross-polarized light, the contact be-
tween the two spine crystallites is planar, because (1) a black median line appears when
the median plane of the spine is aligned to a polarization direction, (2) a solid black or
white spine appears if a specimen is rolled so that the median plane is parallel to the mi-
croscope stage and thus perpendicular to the polarization directions, and (3) oblique ori-
entations produce an off-center black line. In bright field, with a single polarizer, the
apical spine is at low relief when aligned to a polarization direction, and the basal cycle is
at high relief When aligned perpendicular to a single polarization direction, the broad
base of the apical spine shows a round outline. In cross-polarized light this contributes to
the diagnostic obtuse angle made by the extinction line between the apical spine and the
basal spine cycle. The basal cycle is short and simple. No side-oriented spines lie between
the apical spine and the downward, proximally directed basal cycle.
Remarks.— Sphenolithus obtusus is distinguished from S. furcatoUthoides Locker by
the consistent obtuse angle formed between the apical spine and the extinction line be-
tween the basal spine in cross-polarized light. Sphenolithus furcatoUthoides has a single
straight extinction fine that is perpendicular to the nannofossil axis; it also has divergent
halves of the apical spine. Sphenolithus obtusus is distinguished from S. distentus (Mar-
tini) by the three-line extinction pattern of the basal spines, which are also longer than
those of 5'. distentus.
In its most typical orientation on prepared slides, S'. obtusus has the median plane of
the apical spine perpendicular to the slide surface. In cross-polarized light at 15° to 25°
one whole side of the nannofossil— the base and apical side— is dark. At 45°, the extinction
line is missing from the apical spine: instead a light blue line is present that marks the
trace of the median plane.
Occurrence.— Sphenolithus obtusus is common in upper middle Eocene sediment from
Horizon Ridge, northwestern Pacific Ocean. Sphenolithus furcatoUthoides occurs with S.
obtusus only in the lower part of the range of 5'. obtusus. This distribution, together with
the similarity in construction, suggest the derivation of 5'. obtusus from S. furcatoUthoides.
Size.— 6 to 12 microns.
Holotvpe.-VSNM 176913 (PI. 6, figs. 1-6).
Paratypes.-USNM 176914-176915.
Type locality.— DSDP 44.0-4-2, 145-150 cm, Horizon Ridge, northwestern Pacific
Ocean.
Sphenolithus spiniger n. sp.
PL 6, figs. 10-12; PI. 7, figs. 1-2
Description.— This small species is dominated by a basal ring of spines. In cross-polar-
ized light, the lower basal quadrants are about twice as tall as the upper basal quadrants.
The apical structure appears to be a single small spine that is bright at 45° and dark
black, when oriented parallel with a polarization direction. When the nannofossil axis is
oriented parallel with a polarization direction, the median extinction band is flared near
the base of the specimen, giving the large lower quadrants a rounded appearance. At 45°
to the polarization directions, the area occupied by the dark flare of 0° or 90° contains
two bright spines that form an inverted "v'' and that are outlined by black extinction
322
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
1971 BUKRY: CENOZOIC PACIFIC NANNOFOSSILS 323
bands.
Remarks.— Sphenolithus spiniger is distinguished from other species of Sphenolithus
by the unique optical pattern of the basal spines in cross-polarized light and also by the
small size and vestigial apical structure. It is distinguished from S. dissimilis Bukry and
Percival by the triangular outline, the smaller apical structure, and the smaller upper
quadrants of the basal structure.
Occurrence.— Sphenolithus spiniger is common in upper middle Eocene sediment of
Horizon Ridge in the northwestern Pacific Ocean.
^/ze.— width, 4 to 5 microns; height. 5 to 6 microns.
//o/o/i/?e.-USNM 176916 (PI. 6, figs. 10-12).
Paraiype.-XJ^^U 176917.
Type locaUty.—D'bD? 44.0-4-6, 145-150 cm, Horizon Ridge, northwestern Pacific
Ocean.
Striatococcolithus n. gen.
Description.— These circular and subcircular placoliths are composed of two simple
shields connected at the center by a small tube. Each shield is composed of a single cycle
of narrow essentially radial elements. In cross-polarized light either both the shields and
the small central area are dark to faintly visible, or the shields are dark, but a tiny central
area is bright. The lower shield is distinctly smaller than the upper shield.
Type species.— Striatococcolithus pacificanus n. sp.
Remarks.— Striatococcolithus is distinguished from other genera constructed of two
shields and having small simple central areas by the consistent crystallographic alignment
of its shield and central area crystallite elements that results in a typical dark appearance
of the entire placolith in cross-polarized light. Of the most similar genera, Cyclococcoli-
thina Wilcoxon is distinguished by the bright appearance of its smaller shield in cross-
polarized light. Markalius Bramlette and Martini has strongly inchned and imbricated
rim elements and a central area that is consistently bright in cross-polarized light.
Striatococcolithus pacificanus n. sp.
PI. 7, figs. 3-8
Description.— This circular to subcircular placolith has two distinct shields, each com-
posed of a single cycle of 40 to 60 narrow, radial crystallites. The diameter of the larger
shield is 1.6 to 1.7 times that of the smaller shield. The central area is small, occupying
only 15 percent or less of the diameter of the larger shield. In cross-polarized light both
shields and the central area are typically dark or only faintly visible. A few specimens
have a small, vestigial, elliptic, central area that is bright.
Remarks.— Striatococcolithus pacificanus differs from other circular to subcircular
placoliths by lacking birefringence in both shields and in the central area. The radial crys-
talhtes appear as prominent bands extending from the margins to the centers of the
shields. Specimens of S. pacificanus that have a small bright central area are distinguished
from Markalius inversus (Deflandre) by their thin, radial, slightly imbricate shield con-
struction.
Plate 6. Photomicrographs; 2,000 X. 1-9. Sphenolithus obtusus n. sp. ( 1) holotype USNM 176913, DSDP 44.0-
4-2, 145-150 cm, cross-polarized, 0°, (2) holotype, cross-polarized, 45°, (3) holotype, cross-polarized, 22°, (4)
holotype 0°, (5) holotype, 45°, (6) holotype. 90°, (7) USNM 176914, 45°, (8) cross-polarized, 20°, (9) USNM
176915, cross-polarized, 45°, median plane perpendicular to polarization directions. 10-12. Sphenolithus spin-
iger n. sp. (10) holotype USNM 176916, DSDP 44.0-4-6, 145-150 cm, 90°, (11) holotype, cross-polarized, 45°,
(12) holotype, cross-polarized, 0°.
324
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
^^^^^^^^^^^^^^L JI^^^^^^^^^^^^^^^^^Hl * 1
9
1971 BUKRY: CENOZOIC PACIFIC NANNOFOSSILS 325
Occurrence.— StriatococcoUthus pad ficanus occurs through lower Eocene sediment of
the Shatsky Rise in the Pacific Ocean.
Size.~\Q to 14 microns.
Holotype.-\JS^M 176919 (PI. 7, figs. 4-5).
Paratvpes.-XJS^M 176918, 176920-176921.
Type locality.— DSDP 47.2-7-3, 82-83 cm, Shatsky Rise, northwestern Pacific Ocean.
Genus Triquetrorhabdulus Martini, 1965
Triquetrorhabdulus milowii n. sp.
PI. 7, figs. 9-12
Description.— This small species is constructed of three blades joined symmetrically at
a common axis. Oriented in side view, the nannofossil outline, which is formed by two of
the blades, is elliptic to rounded rhomboid. In this same orientation, the third blade is
seen in edge view. Maximum relief above the mounting medium (n= 1.518) and max-
imum birefringence occur at 45° to the polarization directions. In cross-polarized light,
the typical color pattern ofyellow for the two blades in profile, red for the area next to the
third blade (edge view), and blue for the third blade, is a measure of the various thick-
nesses of the nannofossil at this orientation. Minimum relief and birefringence (dark) for
the whole nannofossil is parallel with the polarization directions. Typically the width of
the nannofossil is equal to one half or more of the length.
Remarks. — Triquetrorhabdulus milowii is distinctly shorter than any other species of
Triquetrorhabdulus. It is distinguished from T. inversus Bukry and Bramlette and T. ru-
g<95z/5 Bramlette and Wilcoxon by the orientation of the optic axis of the three blades. It is
distinguished from T. carinatus Martini, with which it shares the same optic-axis pattern,
by a shorter more "inflated" profile. T. milowii is typically one half or two thirds as wide
as long and blade margins tend to be curved, whereas T. carinatus is only one third or one
fourth as wide as long. Younger specimens of T. milowii tend to be shorter and more ellip-
tic in outline than older specimens, with one end slightly wider than the other. This nan-
nofossil has been recorded as T. carinatus [short] in Deep Sea Drilling Project reports on
the tropical Pacific Ocean. The stratigraphic utility of this species was suggested by Dean
Milow (pers. comm., 1969).
Occurrence. — Triquetrorhabdulus milowii is common to rare in lower Miocene sedi-
ments of the Pacific Ocean and Cipero section of Trinidad. Early forms of T. milowii over-
lap the upper range of T. carinatus in the Triquetrorhabdulus carinatus Zone, but T.
milowii persists upward into the Sphenolithus belemnos Zone and possibly into the lower
Helicopontosphaera ampliaperta Zone, which are above the range of T. carinatus.
Size.— 6 to 12 microns.
Holotvpe.-VS^M 176922 (PI. 7, figs. 9 and 12).
Paran/7e.-USNM 176923.
Type locality.— DSDP 74.0-4-4, 63-64 cm, western flank East Pacific Rise, equatorial
Pacific Ocean.
Plate 7. Photomicrographs: 2,000 X. 1-2. Sphenolithus spiniger n. sp. ( 1) USNM 176917. DSDP 44.0-4-6, 145-
146 cm, cross-polarized, 45°, (2) cross-polarized. 0°. 3-8. StriatococcoUthus pacificanus n. sp. (3) USNM
176918. DSDP 47.2-7-2, 100-101 cm. (4) holotype USNM 176919, DSDP 47.2-7-3, 82-83 cm, (5) holotvpe, cross-
polarized, (6) USNM 176920, DSDP 47.2-7-3, 104-105 cm, cross-polarized, (7) USNM 176921. DSDP 47.2-7-3,
82-83 cm, (8) cross-polarized. 9-12. Triquetrorhabdulus milowii n. sp. (9) holotype USNM 176922, DSDP 74.0-
4-4, 63-64 cm, 45°, ( 10) USNM 176923, 45°, ( 1 1) cross-polarized, 45°, ( 12) holotype, cross-polarized, 45°.
326
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
SAMPLE LOCALITIES
DSDP3 (23°OrN., 77°43'W.)
DSDP36 (40°59'N., I30°07'W.)
DSDP44.0(19°19'N., I69°00'W.)
DSDP47.0(32°27'N.. 157°43'E.)
DSDP47.2(32°27'N., 157°43'E.)
DSDP54.0(15°37'N.. 140°18'E.)
DSDP55.0( 9°18'N.. 142°33'E.)
DSDP57.2( 8°4rN.. 143°32'E.)
DSDP63.0( 0°50'N., 147°53'E.)
DSDP63T( 0°50'N., I47°53'E.)
DSDP70.0( 6°20'N.. I40°22'W.)
DSDP72.0( 0°26'N., 138°52'W.)
DSDP74.0( 6°14'S., 136°06'W.)
DSDP77B( 0°29'N.. 133°I4'W.)
DSDP80 ( 0°58'S., 12I°33'W.)
DSDP83A( 4°03'N., 95°44'W.)
STETSON 21, (38°58'N., 72°28'W.)
Sample provided by M. N. Bramlette.
ACKNOWLEDGMENTS
The majority of the samples examined in this study were provided by the National Science Foundation
through the Deep Sea Drilling Project.
I wish to thank M. N. Bramlette, Scripps Institution of Oceanography, R. G. Douglas, Case Western Re-
serve University, A. D. Warren, Mobil Oil Corporation, and G. W. Moore, U.S. Geological Survey for helpful
discussions on various aspects of this paper. Publication has been authorized by the Director of the U.S. Geolog-
ical Survey.
LITERATURE CITED
Berger. W. H.
1970. Biogenous deep-sea sediments; Fractionation by deep-sea circulation. Geol. Soc. America Bull. 81:
1385-1402.
Berggren, W. A.
1971. Tertiary boundaries and correlations. ///, Funnell, B. M., and W. R. Riedel (eds.) The micro-
paleontology of oceans. Cambridge University Press, p. 693-809.
Bramlette, M.N.
1970. Calcareous nannoplankton. U.S. Geol. Surv. Prof. Paper 640-A: 18.
Bramlette, M. N., and E. Martini
1964. The great change in calcareous nannoplankton fossils between the Maestrichtian and Dan-
ian. Micropaleontology 10: 291-322.
Bramlette, M. N.. and W. R. Riedel
1954. Stratigraphic value of discoasters and some other micro fossils related to recent coccolithophores. J.
Paleont. 28: 385-403.
Bramlette, M. N., and J. A. Wilcoxon
1967. Middle Tertiary calcareous nannoplankton of the Cipero section, Trinidad, W. 1. Tulane Studies
Geol. 5:93-131.
Bukry, D.
1971. Coccolith stratigraphy Leg 7. Deep Sea Drilling Project. Deep Sea Drilling Proj. Initial Repts. 7:
1513-1528.
Coccolith stratigraphy Leg 16, Deep Sea DriUing Project. Deep Sea Drilling Proj. Initial Repts. 16,
in press.
Bukry, D., and M. N. Bramlette
1970. Coccolith age determinations Leg 3, Deep Sea Drilling Project. Deep Sea Drilling Proj. Initial
Repts. 3: 589-611.
Chave, K. E., and E. Suess
1967. Suspended minerals in seawater. New York Acad. Sci. Trans. 29: 991-1000.
1971 BUKRY: CENOZOIC PACIFIC NANNOFOSSILS 327
Douglas, R. G.
1971. Cretaceous Foraminifera from the northwestern Pacific Ocean: Leg 6, Deep Sea Drilling Proj-
ect. Deep Sea Drilling Proj. Initial Repts. 6; 1027-1053.
Franke. W. W., and R. M. Brown, Jr.
1971. Scale formation in Chrysophycean algae. III. Negatively stained scales of the coccolithophorid Hy-
menomonas. Archiv Mikrobiologie 77: 12-19.
Gartner, S., Jr.
1969. Correlation of Neogene planktonic foraminifer and calcareous nannofossil zones. Gulf Coast As-
soc. Geol. Socs. Trans. 19: 585-599.
1971. Calcareous nannofossils from the JOIDES Blake Plateau cores, and revision of Paleogene nannofos-
sil zonation. Tulane Studies Geol. Paleont. 8: 101-121.
Hay, W. W., H. P. Mohler, P. H. Roth, R. R. Schmidt, and J. E. Boudreaux
1967. Calcareous nannoplankton zonation of the Cenozoic of the Gulf Coast and Caribbean-Antillean
area, and transoceanic correlation. Gulf Coast Assoc. Geol. Socs. Trans. 17: 428-480.
Hulburt, E. M.
1962. Phytoplankton in the southwestern Sargasso Sea and north equatorial current, February 1961. Lim-
noi. Oceanogr. 7: 307-315.
Hulburt, E. M.. and J. Rodman
1963. Distribution of phytoplankton species with respect to sahnity between the coast of southern New
England and Bermuda. Limnol. Oceanogr. 8: 263-269.
Lawrence, D. R.
1971. The nature and structure of paleoecology. J. Paleont. 45: 593-607.
Martini, E.. and M. N. Bramlelte
1963. Calcareous nannoplankton from the experimental Mohole drilling. J. Paleont. 37: 845-856.
Martini, E., and T. R. Worsley
1971. Tertiary calcareous nannoplankton from the western equatorial Pacific. Deep Sea Drilling Proj. In-
itial Repts. 7: 1471-1507.
Milow, E. D.
1970. Tentative nannofossil zones and subzones and their radiometric age, northeast Pacific. Deep Sea
Drilling Proj. Initial Repts. 5: 8-10.
Peterson, M. N. A.
1966. Calcite: rates of dissolution in a vertical profile in the central Pacific. Science 154: 1542.
Pytkowicz, R. M.
1969. Chemical solution of calcium carbonate in seawater. Amer. Zool. 9: 673-679.
Roth. P. H.
1970. Oligocene calcareous nannoplankton biostratigraphy. Eclogae Geol. Helvetiae 63: 799-881.
Smith, S. v., J. A. Dygas, and K. E. Chave
1968. Distribution of calcium carbonate in pelagic sediments. Mar. Geol. 6: 391-400.
Tappan, H.
1971. Microplankton, ecological succession and evolution. North Amer. Paleont. Convention Proc. H:
1058-1103.
Wilcoxon, J. A.
1970. Cvclococcolithina Wilcoxon nom. now (nom. subst. pro Cyclococcolithus Kampter, 1954). Tulane
Studies Geol. Paleont. 8: 82-83.
U.S. Geological Survey, Scripps Institution of Oceanography, LaJolla, California 92037.
MUS. COIVIP. ZUUi-
LIBRARY
FEB 1 8 197^
HARVARD
UNIVERSITY
AN UPPER PLEISTOCENE MARINE FAUNA
FROM MISSION BAY,
SAN DIEGO, CALIFORNIA
J. PHILIP KERN, TOM E. STUMP,
AND ROBERT J. DOWLEN
TRANSACTIONS
OF THE SAN DIEGO
SOCIETY OF
NATURAL HISTORY
VOL. 16, NO. 15 29 DECEMBER 1971
AN UPPER PLEISTOCENE MARINE FAUNA FROM MISSION BAY,
SAN DIEGO, CALIFORNIA
J. PHILIP KERN. TOM E. STUMP, AND ROBERT J. DOWLEN
ABSTRACT.— Sixty-nine invertebrate species and one chordate have been collected from the upper Pleisto-
cene Bay Point Formation on the northeast shore of Mission Bay in San Diego, California. This protected-
bay assemblage lived in water depths of 1 to 2 m. Rocky-shore species at the base of the section were re-
placed bv mudflat species as the initially deposited gravel and boulders were covered by sand and mud. The
fauna includes three or four southern extrahmital species; their paleoclimatic implications are not clear.
INTRODUCTION
The upper Pleistocene Bay Point Formation crops out in a number of small, isolated
exposures on the lowest well-developed, emergent marine terrace (Nestor terrace of Ellis,
1919: pi. 6; La Jolla terrace of Hanna, 1926: 194-195) and at corresponding elevations in
coastal embayments from Oceanside, California to northern Baja California. Marine fos-
sil assemblages are preserved in several outcrops of this formation in the area of Mission
Bay in northern San Diego (Fig. 1). Exposed-coast faunas occur at Pacific Beach (Valen-
tine, 1961: 359-361. tables 19. 20) and Sunset Chfts (Valentine and Meade. 1961: 11-13,
table 2). Fossils at Crown Point, the type locality of the Bay Point Formation, hved on or
near a barrier beach that protected the Pleistocene Mission Bay to the east from strong
wave action (Valentine. 1959: 687); the present-day barrier is a mile farther west. Two
small, sheltered-water faunas were reported by Stephens (1929: 253. 255). one from the
northeast shore of Mission Bay (the railroad cut locality) and one from the south shore.
Another sheltered-water fauna was described by Emerson and Chace (1959) from Teco-
lote Creek on the east shore. All the above localities are shown in Figure 1.
The sheltered-water fauna reported by Stephens (1929: 253) from the northeast
shore of Mission Bay (railroad cut locality) has been referred to subsequently by Valen-
tine (1959: 687: 1961: 359) and by Emerson and Chace (1959: 340). but this fauna has
never been adequately studied. The locality is near the base of the steep northeastern
slope of the present embayment, and the sediments here clearly were deposited close to
the eastern shore of the Pleistocene Mission Bay. The purpose of this paper is to describe
this fauna and discuss its paleoenvironmental implications.
San Diego State College locahty 2276 (Figs. 1-3) is 230,600 ft. north and 1.704,800 ft.
east in zone 6 of the California coordinate system (U.S. Geol. Surv. 7.5 minute La Jolla,
California quad.. 1967 ed.). It is in a low cut on the east side of the tracks of the Santa Fe
Railroad between Morena Boulevard and Interstate 5. Fossils should not be collected
from this locahty without the permission of officials of the Santa Fe Railroad.
STRATIGRAPHY
At this locahty the Bay Point Formation lies unconformably on Pliocene rocks of the
San Diego Formation. From the lower part of the exposed Bay Point Formation we col-
lected reworked Pliocene fossils including Astrangia sp.. Opalia varicostata Stearns, and
two unidentified species of the gastropod families Turridae and Thaididae.
The exposed Pleistocene section is approximately 2 m thick and is fossiliferous for a
lateral distance of about 30 m (Figs. 2 and 3). At the base of the section are 60 to 70 cm of
poorly sorted conglomerate containing sub-rounded boulders up to 60 cm in diameter.
The matrix of the conglomerate is poorly consolidated, poorly sorted, predominantly
coarse-grained, brown sand. The fossiliferous upper part of this bed is finer-grained, and
SAN DIEGO see. NAT. HIST.. TRANS. 16(15): 329-338. 29 DECEMBER 1971
330
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
Figure \. Map of the Mission Bay area showing the location of San Diego State College locality 2276 (Stephens'
railroad cut locality) and other Bay Point Formation fossil localities at Pacific Beach (1), Crown Point (2), Teco-
lote Creek (3), south Mission Bay (4), and Sunset Cliffs (5). Inferred late Pleistocene land areas are shaded. The
extension of the barrier beach south of Crown Point is based on the presence of sheltered-bay fossil faunas be-
hind the barrier and the typical development of such barriers today. Probably the barrier was alternately open
and closed.
1971
KERN. STUMP AND DOWLEN: PLEISTOCENE MARINE FAUNA
331
the conglomerate grades rather abruptly upward into somewhat better sorted, poorly con-
solidated, fine-grained, brown sand. There is a slight upward decrease in grain size
through the upper part of the section. Grain-size analyses for the beds described below
are shown in Table 1.
Table 1. Sediment grain-size analyses made by the dry sieve method described by Folk (1968:
Ranges are given because two or more samples from each bed were analyzed.
34-36).
coarser than sand
coarse sand
-l</>to 1</)
medium sand
l4>to 24>
fine sand
2(|)to4</)
silt and clay
>4</)
bed 3
1-4%
5-12%
12-22%
53-70%
9-12%
bed 2
2-6
4-7
22-31
54-60
4-12
bed 1
30-32
14-16
21-22
24-31
4-6
conglomerate
70
8
9
10
3
Fossils are distributed irregularly throughout the section above the lower part of the
conglomerate. Collections were made from three rather arbitrarily defined stratigraphic
intervals in order to evaluate temporal changes in the fauna. Bed 1 is a highly fossiliferous
stratum in the upper 15 to 30 cm of the conglomerate (Fig. 3). Bed 2 is a poorly defined
fossiliferous interval tYom 30 to 45 cm thick directly overlying bed 1. Bed 3 is an irregular
stratum 30 cm thick and 15 cm above the top of bed 2.
rf^^m^^ ^.
if^S
Figure 2. San Diego State College locality 2276 viewed from the southwest. The tracks of the Santa Fe Railroad
are in the foreground, and the houses in the background are east of Morena Boulevard. Exposure A is the exca-
vation in the bank directly below the street sign near the left edge of the view; exposure B is the smaller excava-
tion at the right edge of the view. The lower parts of both excavations were filled in after the collections were
made and before this photograph was taken. Fossils that have weathered out on the bank are visible between the
two exposures.
332 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL. 16
METHODS AND FAUNAL LIST
Each bed was sampled at two exposures, designated A (at the northern end of the
outcrop) and B ( 10 m to the south of A; Fig. 2). We collected 18 to 22 kg of sediment and
fossils in each sample except the one from 2A, which was twice that size. Each sample was
washed through a 10-mesh (2 mm) screen, and all identifiable fossil specimens were
picked from the material retained in the screen. Sediment passing through the screen was
examined for microfossils but none were found.
Sixty-seven species of mollusks, one chordate species, and borings of unidentified
species of Polydora, an annelid, and Cliona, a sponge, have been identified. The distribu-
tion and abundance in the three beds of those species collected during this study are
shown in Table 2. Also indicated in Table 2 are species present in earlier San Diego State
College collections from this locality and in the collections of the San Diego Natural His-
tory Museum. The latter include at least part of Stephens' ( 1929: 253) collection from this
locality, though we have not been able to find two of the species he listed, Macoma cal-
carea and Phacoides californica. His Macoma probably is Psammotreta viridotincta, which
is common in his collection, but that collection now contains no specimens likely to be
confused with Phacoides [ = Lucina] californica.
AGE OF THE FAUNA
Odostomia diegensis is the only species in this fauna not known to be living. Addicott
and Emerson (1959: 24) suggested that all essentially modern fossil faunas preserved at
the lowest emergent terrace level in southern California probably are correlative with the
regional type upper Pleistocene Palos Verdes Sand at San Pedro. The Bay Point Forma-
tion, which is deposited in part on this terrace, probably is correlative, then, with upper
Pleistocene deposits at San Pedro, Cayucos, and San Nicolas Island that have been dated
radiometrically at between 95,000 ± 15,000 and 140,000 ± 30,000 years (see discussion
in Kern, 1971: 812). These faunas may have lived during Sangamon interglacial time or
during an earlier interglacial episode.
ENVIRONMENT OF DEPOSITION
Environmental interpretations are based on distribution data for living species from
Berry (1922), Oldroyd (1927), Grant and Gale (1931), Burch (1944-1946), Hertlein and
Strong (1955), Morris (1966), Ricketts and Calvin (1968), McLean (1969), and Keen
(1971). The fossil fauna is characteristic of the protected-bay environment, consistent with
the probable paleogeographic setting of the locality (Fig. 1). However, the inferred coast-
line suggests that this shore was exposed to over a mile of open water behind the barrier
beach, and several moUuscan species also suggest that the fauna lived in a somewhat ex-
posed part of the bay. Littorina scutulata, a common species at this locality, is most abun-
dant today in the less protected parts of bays (Ricketts and Calvin, 1968: 237). Lottia
gigantea and Mopalia ciliata, rare species in this fauna, are characteristic of a protected
outer coast (Ricketts and Calvin, 1968: 26, 138), but they may occur rarely in the rela-
tively exposed parts of bays.
ORIGIN OF FAUNAL ASSEMBLAGE
The composition of the fossil fauna and the abundance of worn and broken shells
that are concentrated in irregular beds, lenses, and pockets reflect local reworking of sedi-
ment and mixing of shells from slightly different depth zones. Littorina planaxis, L. scutu-
lata. Assiminea translucens, Melampus olivaceus, and Heterodonax himaculatus live at or
1971
KERN. STUMP AND DOWLEN: PLEISTOCENE MARINE FAUNA
333
Table 2. Distribution and abundance of fossil species in beds 1. 2, and 3 in two exposures at San Diego State College locality 2276. Numbers
are pairs of bivalves and individual specimens of other fossils. The first two columns indicate species present m other collections at San Diego
State College ISDSC) and at the San Diego Natural History Museum (SDNHM). Data on abundance and on distribution within the outcrop are
not available for these collections.
Collections
Species
SDNHM SDSC lA IB
2A
2B 3A 3B
Porifera
Cliona sp.
Annelida
Polydora sp.
Polyplacophora
Mopalia ciliata (Sowerby, 1840)
Gastropoda
Acnwea insessa (Hinds, 1843)
Acmaea instabilis (Gould, 1846)
Acnwea scabra (Gould, 1846)
Lottia gigantea (Sowerby, 1833)
Lucapinella callomarginata (Dall, 1871)
Tegula gallina (Forbes, 1850)
Liotia fenestrata (Carpenter, 1864)
Epitonium indianorum (Carpenter, 1864)
Littorina planaxis (Philippi, 1847)
Littorim scutulata (Gould, 1849)
Lacuna sp.
Assiminea translucens (Carpenter, 1864)
?Solariorbis sp.
Alabina tenuisculpta (Carpenter, 1864)
Cerithiopsis carpenteri (Baitsch, 1911)
Cerithidea californica (Haldeman, 1840)
Hipponix lumens (Carpenter, 1864)
Creptdula perforans (Valenciennes, 1846)
Crepidula sp.
Crepipatella lingulala (Gould, 1846)
Eupleura muriciformts (Broderip, 1833)
Morula lugubris (C. B. Adams, 185 2)
Anachis coronata (Sowerby, 1832)
Mttretla carinata (Hmds, 1844)
Nassarius tegula (Reeve, 1853)
Olivella biplicata (Sowerby, 1825)
Conus californlcus (Hinds, 1844)
Odostomia diegensis (Dall & Bartsch, 1903)
Peristichia pedroana (Dall & Bartsch, 1909)
Pyramidella adamsi iCMpentei. 1864)
Turbonilla sp.
Acteoclna culcitella (Gould, 1853)
Acteon punctocoelata (Carpenter, 1864)
Melampus olivaceus (Carpenter, 1856)
Pedipes liratus (Binney. 1860)
Bivalvia
Nucula aff . A', exigua (Sowerby, 1833)
Anadara multicostata (Sowerby, 1833)
Septlfer bifurcatus (Conrad, 1837)
Ostrea lurida (Carpenter, 1864)
Argopecten circularis (Sowerby, 1835)
Leptopecten latiauratus (Conrad, 1837)
Lima sp.
Anomia peruviana (Orbigny, 1846)
Crassinella branneri (Arnold, 1903)
Lucina nuttallii (Conrad, 1837)
Here excavata (Carpenter, 1857)
Diplodonta sericata (Reeve, 1850)
Laevicardium substriatum (Conrad, 1837)
Pilar newcombianus (Gabb, 1865)
Chione californiensis (Brodenp, 1835)
Chione fluctifraga (Sowerby, 1853)
Chione gnidia (Broderip & Sowerby, 1829)
Callithaca staminea (Conrad, 1837)
Mactra californica (Conrad, 1837)
Spisula cf. 5. hemphilli (Dall, 1894)
Tellina bodegensis (Hinds, 1844)
Tellina meropsis (DaU, 1900)
Macoma nasuta (Coruad, 1837)
Psammotreta viridotincta (Carpenter, 1856)
Cumingia californica (Conrad, 1837)
Donax californlcus (Coniid, 1837)
Heterodonax bimaculatus (Linnaeus, 1758)
Tagelus calif ornianus {Qontzi, 1837)
Cryptomya californica (Conrad, 1837)
Corbula luteola (Carpenter, 1864)
Corbula sp.
Chorda ta
Myliobatis sp.
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10
31
48
12
35
1
1
27
12
61
13
26
25
25
70
66
2
3
24
1
428
112
25
28
1
4
13
5
1
3
2
10
2
10
10
5
9
2
17
15
2
1
5
3
4
2
6
10
2
1
1
1
1
10
1
1
26
1
6
133
7
7
2
3
5
190
415
75
42
38
35
2
89
9
15
1
70
2
3
43
3
9
2
13
15
2
8
1
42
110
22
18
25
28
1
1
1
110
45
2
2
65
1
18
1
334
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
near high tide level where few other marine species live, and Cehthidea californica is most
common on middle intertidal mud and sand flats where few other species are abundant.
Yet at this locality these species are part of a rather diverse assemblage of invertebrates,
most of which extend from the lower part of the intertidal zone into deeper water. Be-
cause of the relative turbulence of the intertidal zone, even in sheltered environments,
such mixing of species from diflerent intertidal levels is not unexpected.
■y
Figure 3. Exposure A at San Diego State
College locality 2276. The lower part of
the excavation has been covered since the
collections were made, and the basal con-
glomerate and bed I are no longer ex-
posed. The abundant fossils in the lower
part of the exposure are in bed 2, and the
small lens of fossils higher in the exposure
is in bed 3. The irregularity and dis-
continuity of these beds is evident. The
pencil is 13 cm long.
However, in spite of this evidence for turbulent conditions and local reworking of
sediment, there apparently was no large-scale sediment transport. With the few excep-
tions described below all the species in this fauna live in the intertidal zone of sheltered
bays; thus there was no mixing of shells from widely different environments. In situ pres-
ervation of the majority of species is suggested by the presence of approximately equal
numbers of right and left valves of several bivalve species and by high percentages of ar-
ticulated specimens of Lucina nuttalUi, Diplodonta sehcata, Chione californiensis, Psam-
motreta viridotincta, and Tagelus californianiis. The deeply burrowing T. californianus
commonly is preserved in life orientation in these beds.
WATER DEPTH
This fauna clearly lived in sediments deposited in or very near the intertidal zone.
The shoreline angle of the Nestor terrace in this area is not more than 2 or 3 m higher
1971 KERN, STUMP AND DOWLEN: PLEISTOCENE MARINE FAUNA 335
than the fossil-bearing beds. Littorina scutulata, Cerithidea californica, Melampus oli-
vaceus, Pedipes liratus, and Septifer bifurcatus are relatively abundant and they are re-
stricted today to the intertidal zone, as are the less abundant Lottia gigantea, Lucapinella
callomarginata, Littorina planaxis, Assiminea translucens, and Heterodonax bimaculatus.
All but four of the other species range from the intertidal zone into deeper water.
One of the four exceptions, Crassinella branneri is common in this fauna, and the
genus has a reported depth range of 2 to 40 m (Keen, 1963: 105). Psammotreta virido-
tincta, also common, occurs today "mostly offshore in depths to 14 fathoms" (Keen. 1971:
231). However, the depth significance of this species is not clear because its present-day
minimum depth is not known, and the change in its geographic range since late Pleisto-
cene time (Table 3) suggests that its environmental tolerance limits may have changed.
The collections also include four specimens each of Nucula aff. N. exigua and Pitar new-
combianus. The former is known to live today in depths of 1 1 to about 2000 m (Keen,
1971: 26) and the latter in depths of 9 m or more (McLean, 1969: 78). Though these
depth ranges are inconsistent with those described above, the few specimens of these two
species do not warrant substantial modification of the suggested depth interpretation, es-
pecially in light of the questionable identity of the Nucula. However, the presence of these
four apparently subtidal species in the fauna suggests that deposition may have occurred,
at least in part, slightly below the intertidal zone.
Bed 3 contains fewer species than beds 1 and 2, and most of the abundant species in
bed 3, including Littorina scutulata, Cerithidea californica, Nassarius tegula, Melampus
oUvaceus, Chione fluctifraga, and Tagelus californianus, today are restricted to or are most
abundant in the intertidal zone. The other common species in bed 3. Diplodonta sericata,
Protothaca staminea, and Tellina meropsis, are more abundant in bed 2, and these species
are not restricted to or most abundant in the intertidal zone today. The four possibly sub-
tidal species are uncommon in bed 3. Thus bed 3 apparently was deposited in the lower
part of the intertidal zone, perhaps 1 m below mean sea level, and bed 1 was deposited in
water perhaps 1 m deeper. Probably sea level was stable throughout the period of depo-
sition, and the change in water depth reflects the thickness of sediments, about 1 m, de-
posited from bed 1 through bed 3. The base of the outcrop is approximately 14 m above
present mean sea level, so sea level when these sediments were deposited probably was
about 16 m higher than it is today. The shoreline angle of the Nestor terrace in this area is
within 30 m to the east and no more than 2 or 3 m higher than locality 2276, suggesting
that at this level the sea was close to its maximum extent on the terrace. The approximate
position of the coastline at that sea level is shown in Figure 1.
The terrace deposits at Tecolote Creek (Fig. 1) are at a present elevation of 6 to 8 m
at the base of the section and 14 to 16 m at the top. The fauna (Emerson and Chace, 1959:
table 1) includes nearly all the intertidal species present at locality 2276, though their dis-
tribution within the section is now known. Possibly the lowest beds were deposited some-
what offshore in depths of 8 to 10 m and the highest beds, at about the same elevation as
the beds at locality 2276, were deposited in or near the intertidal zone after this part of
the basin had filled with sediments. Alternatively, sea level may have been rising during
deposition of these sediments.
Valentine (1959: 685, 687) suggested that maximum sea level during cutting of the
Nestor terrace at Crown Point, Pacific Beach, and Sunset Cliffs (Fig. 1) was between 60
and 70 feet (18 and 20 m) above present sea level and that the shallow-water assemblage
at Crown Point, at a present elevation of 5 m at the bottom of the section and 9 m at the
top, probably lived when the sea was well below its maximum extent on the terrace. This
apparent difference in the sea levels under which these two faunas lived suggests that they
336
SAN DIEGO SOCIETY OF NATURAL HISTORY
VOL. 16
may not have been contemporaneous.
SUBSTRATE
Stratigraphic faunal changes also reflect a temporal change in the substrate. Tegula
gallina, Littorina scutulalcL Septifer hifurcatus, Ostrea hirida, and Anomia peruviana live
only or chiefly on hard substrates, and all are very abundant in bed 1 and are either rare
or absent in beds 2 and 3. On the other hand, the characteristic mudflat species Tagelus
californianus, Cerithidea californica, and Melampus olivaceus are rare in bed 1 and in-
creasingly abundant in beds 2 and 3. Thus the stratigraphic change in faunal composition
reflects both the decrease in water depth and the change from a coarse gravel and boulder
substrate to mud.
CLIMATE
Species whose present-day geographic ranges end near or do not include San Diego
are listed in Table 3 with their ranges. Six species do not live today south of the San
Diego-Ensenada region, and a seventh lives only as far south as Bahia San Quintin. The
questionably identified Spisula hemphilli also lives only as far south as Ensenada. Seven
species live only as far north as the San Diego-Los Angeles region. The overlapping pres-
ent-day ranges of these 14 species suggest that the late Pleistocene shallow-water marine
climate in which they lived was similar to that at the same latitude today.
However, the assemblage also contains several southern extralimital species, species
that do not live today as far north as San Diego (Table 3). Anachis coronata, Psammotreta
vihdotincta, and Chione gnidia live only as far north as Laguna Scammon or Lsla de
Cedros (28° north latitude). Eupleura muriciformis also is reported to live today only as
far north as lsla de Cedros, though Hertlein and Strong (1955: 258) included in its synon-
ymy Ranella triquefra Reeve, 1844 from San Diego. Both Eupleura muriciformis and
Table 3. Species with geographic ranges that end near or do not include the San Diego area.
Species
Geographic Range
Acniaea instabilis
Hipponix tumens
Mopalia ciliata
Epitonium indianorum
Assiminea translucens
Callithaca staminea
Spisula cf. S. hemphilli
Cerithiopsis carpenteri
Alabina tenuisculpta
Nucula aff. A^. exigua
Atwdara cf. /I. multicostata
Morula lugubris
Pedipes liratus
Crassinella branneri
Eupleura muriciformis
Anachis coronata
Chione gnidia
Psammotreta viridotincta
Alaska to San Diego (Morris, 1959: 57)
Crescent City to San Diego (Oldroyd, 1927,2(3): 113-114)
Alaska to Bahia Todos Santos (Berry, 1922: 449-451)
Alaska to Bahia Todos Santos (Oldroyd, 1927, 2 (2): 58)
Vancouver to Punta Banda (McLean, 1969: 28)
Alaska to Bahia San Quintin (Grant and Gale, 1931: 329)
Santa Barbara to Ensenada (McLean, 1969: 82)
San Pedro to South Coronado Island (Oldroyd, 1927, 2 (2):
253)
San Pedro to Magdalena Bay (Oldroyd, 1927, 2 (3): 14)
Los Angeles to Ecuador (Burch, 1944-1946, no. 33: 7)
Newport Bay to Galapagos Islands (Keen, 1971: 48)
San Diego to Panama (Keen, 1971: 554)
San Diego to Golfo de California (Oldroyd, 1927, 2(1): 54)
San Diego to Panama (Oldroyd, 1927, 1: 110)
(San Diego ?) lsla de Cedros to Lobitos, Peru (Hertlein and
Strong, 1955: 258)
Laguna Scammon to Ecuador (San Diego Natural History
Museum Coll.)
lsla de Cedros to Peru (Keen, 1971: 188)
Laguna Scammon to Costa Rica (San Diego Natural History
Museum Coll.)
1971 KERN, STUMP AND DOWLEN; PLEISTOCENE MARINE FAUNA 337
Chione gnidia are rare in this assemblage, but the other two species are rather abundant.
Two additional uncommon species, Crassinella branneri and Nucula cf. A', exigua, have
been regarded in some studies as southern extralimital species, though there are conflict-
ing records on their geographic ranges; both species have been reported in the San Diego
area. Thus there apparently are three or four species in this fauna that do not live today
in the San Diego region or in the area of overlap of the present-day geographic ranges of
all the other species in the assemblage.
The paleoclimatic significance of extralimital species has been discussed by Emerson
(1956: 326-327). Valentine (1955: 465-468; 1961: 393-400), Kern (1971: 819-820; in
press), and others. Southern extralimital species commonly have been interpreted as in-
dicating that shallow-water marine climates have been substantially warmer in the past
than today, at least locally. However, the presence in this fauna of seven or eight mollus-
can species that do not live today south of the San Diego region suggests that the late
Pleistocene marine climate in this area was not as warm as the climate at Laguna Scam-
mon and Isla de Cedros today. The geographic ranges of these and the extralimital spe-
cies do not overlap today, and paleoclimatic interpretations based on assumed thermal
limitations of their ranges must involve more complex changes than simple warming or
cooling. It must also be recognized that some of these species may have changed physio-
logically and ecologically since late Pleistocene time, and some of them may be limited
geographically by factors other than water temperature (see discussion in Kern, in press).
ACKNOWLEDGEMENTS
The fossils on which this study are based were collected and prepared by members of a graduate class in
paleoecology at San Diego State College. In addition to the authors this class included William Cunningham,
Dennis Dowd, Rogers Hardy III, and Michael Hart. Sediment analyses were made by Dowd and Hardy. We
also are indebted to Willard Libby, Geological Survey of Western Australia, for reviewing the manuscript, and
to Arnold Ross and George Radwin, San Diego Natural History Museum, who granted access to the museum's
fossil and Recent invertebrate collections.
LITERATURE CITED
Addicott, W. O., and W. K. Emerson
1959. Late Pleistocene mvertebrates from Punta Cabras, Baja California, Mexico. Amer. Mus. Novitates
1925; 1-33.
Berry, S. S.
1922. Fossil chitons of western North America. California Acad. Sci., Proc. (ser. 4) 11: 399-526.
Burch, J. Q. [ed.]
1944-1946. Distributional list of the west American marine Mollusca from San Diego, California, to the Po-
lar Sea. Conch. Club S. California, Minutes.
Ellis, A. J.
1919. Physiography, p. 20-50. In A. J. Ellis and C. H. Lee, Geology and ground waters of the western part
of San Diego County, California. U.S. Geol. Surv., Water Supply Paper 446.
Emerson, W. K.
1956. Pleistocene invertebrates from Punta China, Baja California, Mexico. Amer. Mus. Nat. Hist , Bull.
Ill: 313-342.
Emerson, W. K., and E. P. Chace
1959. Pleistocene mollusks from Tecolote Creek, San Diego, California. San Diego Soc. Nat. Hist Trans.
12: 335-346.
Folk, R. L.
1968. Petrology of sedimentary rocks. Hemphill's, Austin, Texas. 170 p.
Grant, U. S., IV, and H. R. Gale
1931. Catalogue of the marine Pliocene and Pleistocene Mollusca of California and adjacent regions. San
Diego Soc. Nat. Hist., Mem. 1. 1036 p.
Hanna, M. A.
1926. Geology of the La Jolla quadrangle, California. Univ. Calif Pub. Geol. Sci. 16: 187-246.
338 SAN DIEGO SOCIETY OF NATURAL HISTORY VOL.16
Hertlein. L. G., and A. M. Strong
1955. Marine mollusks collected during the "Askoy" expedition to Panama, Colombia, and Ecuador in
1941. Amer. Mus. Nat. Hi.st.. Bull. 107: 159-318.
Keen, A. M.
1963. Marine molluscan genera of western North America. Stanford Univ. Press, Stanford, California. 126
P-
Keen, A. M.
1971. Sea shells of tropical west America. Second Edition, Stanford Univ. Press, Stanford, California. 1064
P-
Kern, J. P.
1971. Paleoenvironmental analysis of a late Pleistocene estuary in southern California. J. Paleont. 45: 810-
823.
Kern, J. P.
Early Pliocene marine climate and environment of the eastern Ventura basin, southern California.
Univ. Calif Pub. Geol. Sci., in press.
VlcLean. J. H.
1969. Marine shells of southern California. Los Angeles Co. Mus. Nat. Hist., Sci. Ser. 24, Zool. 11. 104 p.
Morris. P. A.
1966. A field guide to the shells of the Pacific coast and Hawaii. Houghton Mifflin, Boston. 297 p.
Oldroyd, I. S.
1927. The marine shells of the west coast of North America. Stanford Univ. Pub. Geol. Sci. 1, 2 (parts 1-3).
Ricketts, E. P., and J. Calvin
1968. Between Pacific tides. Fourth Edition, Stanford Univ. Press, Stanford. California. 614 p.
Stephens, F.
1929. Notes on the marine Pleistocene deposits of San Diego County, California. San Diego See. Nat.
Hist., Trans. 5: 247-255.
Valentine. J. W.
1955. Upwelling and thermally anomalous Pacific coast Pleistocene molluscan faunas. Amer. J. Sci. 253:
462-474.
Valentine, J. W.
1959. Pleistocene molluscan notes, I, The Bay Point Formation at its type locality. J. Paleont. 33: 685-688.
Valentine, J. W.
1960. Habitats and sources of Pleistocene mollusks at Torrey Pines Park, California. Ecology 41: 161-165.
Valentine, J. W.
1961. Paleoecologic molluscan geography of the Californian Pleistocene. Univ. California Publ. Geol. Sci.
34: 309-442.
Valentine, J. W., and R. F. Meade
196 1. Californian Pleistocene paleotemperatures. Univ. California Publ. Geol. Sci. 40: 1-46,
Department of Geology, San Diego State College, San Diego, California 92115
7 9Ub u92
Date Due
-^m^MM^r
ACME
BOOKBINOiNG CO.. INC.
MAR 8 - 1984
100 CAMDRCCE :.:r^EET
3 2044 093 361 319
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Books to Borrow
Open Library
TV News
Understanding 9/11