|
of the
FLORIDA STATE MUSEUM
Biological Sciences
VOLUME 29 1983 NUMBER 2
TAXONOMY AND ECOLOGY OF THE GENUS
CHASMODES (PISCES: BLENNIIDAE)
WITH A DISCUSSION OF ITS ZOOGEOGRAPHY
JEFFREY T. WILLIAMS
UNIVERSITY OF FLORIDA
GAINESVILLE
Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL
SCIENCES, are published at irregular intervals. Volumes contain about 300 pages and are not
necessarily completed in any one calendar year.
OLIVER L. AUSTIN, JR., Editor
RHODA J. BRYANT, Managing Editor
Consultants for this issue:
JAMES E. BOHLKE
VICTOR G. SPRINGER
Communications concerning purchase or exchange of the publications and all manuscripts
should be addressed to: Managing Editor, Bulletin; Florida State Museum; University of
Florida; Gainesville, FL 32611, U.S.A.
Copyright by the Florida State Museum of the University of Florida
This public document was promulgated at an annual cost of $1,894.06 or
$1.894 per copy. It makes available to libraries, scholars, and all inter-
ested persons the results of researches in the natural sciences, emphasiz-
ing the circum-Caribbean region.
Publication date:
Price: $1.90
TAXONOMY AND ECOLOGY OF THE GENUS
CHASMODES (PISCES: BLENNIIDAE)
WITH A DISCUSSION OF ITS ZOOGEOGRAPHY
JEFFREY T. WILLIAMS'
SYNOPSIS: The western Atlantic blenniid fish genus Chasmodes is represented by two species,
C. bosquianus and C. saburrae. C. saburrae is the more tropical of the species and is found on
Florida's east coast from Edgewater southward around the tip of Florida, northward and
westward to the Chandeleur Islands, Louisiana. C. bosquianus is composed of two subspe-
cies, C. b. bosquianus and C. b. longimaxilla new subspecies. C. b. bosquianus is distributed
on the Atlantic seaboard from Maryland to Marineland, Florida, and C. b. longimaxilla occurs
in the northern Gulf of Mexico from Pensacola, Florida, to Veracruz, Mexico.
Chasmodes bosquianus longimaxilla exhibits character displacement in the zone of sym-
patry with C. saburrae in the Gulf. Increased maxillary length in this area may be contributing
to the divergence of the disjunct populations of C. bosquianus.
The zoogeography of the genus Chasmodes is discussed and recent geological information
is used in a consideration of the events leading to its present distribution.
Experimental hybridization experiments were performed with Chasmodes saburrae and C.
b. longimaxilla, but lack of adequate facilities prevented rearing the larvae.
RESUMEN: El g6nero Chasmodes de peces bleniidos del Atlantico Occidental esta represen-
tado por dos species: C. bosquianus y C. saburrae. C. saburrae es la mas tropical de las
species y se encuentra en la costa este de Florida, desde Edgewater hasta el extreme sur de
Florida; y hacia el norte y este hasta las Islas Chandeleur, en Louisiana. C. bosquianus se
compone de dos subespecies: C. b. bosquianus y C. b. longimaxilla.
C. b. bosquianus se distribuye cerca a las costas atlanticas desde Maryland a Marineland,
en Florida; y C. b. longimaxilla se distribuye en la parte norte del Golfo de Mexico, desde
Pensacola en Florida, hasta Veracruz en M6xico.
Chasmodes bosquianus longimaxilla exhibe desplazamiento de caracteres en la zona de
simpatria con C. saburrae en el Golfo de Mexico. El aumento de longitud maxilar en esta area
podria contribuir a la divergencia observada en poblaciones disyuntivas de C. bosquianus.
Se discute la zoogeografia del genero Chasmodes; asi como informaci6n geol6gica
reciente al considerar los events que conducen a su actual distribuci6n.
Se Ilevaron a cabo experiments de hibridaci6n utilizando Chasmodes saburrae y C. b.
longimaxilla, pero debido a la falta de facilidades adecuadas las larvas no pudieron ser criadas
con exito.
TABLE OF CONTENTS
INTRODUCTION .............................. .................... 66
ACKNOWLEDGEMENTS ................ .. ......................... 67
METHODS ................... .................. ....................... 68
GENUS CHASMODES ................... .................... ................ 69
KEY TO THE SPECIES AND SUBSPECIES OF CHASMODES.....................70
The author is with the Division of Fishes, National Museum of Natural History, Smithsonian Institution, Washington, D.C.
20560. The study was begun at the University of South Alabama, Mobile, and completed at the University of Florida,
Gainesville, in partial fulfillment for the degree of Master of Science.
WILLIAMS, JEFFREY T. 1983. Taxonomy and Ecology of the Genus Chasmodes (Pisces:
Blenniidae) with a Discussion of Its Zoogeography. Bull. Florida State Mus., Biol. Sci. 29(2):
65-101.
BULLETIN FLORIDA STATE MUSEUM
CHASMODES BOSQUIANUS ...................... ...........................71
CHASMODES BOSQUIANUS BOSQUIANUS ................................72
CHASMODES BOSQUIANUS LONGIMAXILLA new subspecies ..................85
CHASMODES SABURRAE............. ... ... ............................91
ZOOGEOGRAPHY .............................................................95
EXPERIMENTAL HYBRIDIZATION ...........................................96
SUM MARY ................................................. .................97
LITERATURE CITED ................ .................... ................. 97
INTRODUCTION
Chasmodes Valenciennes 1836 is a ditypic genus commonly encoun-
tered in bays and estuaries from Chesapeake Bay, Maryland, around the
southern tip of Florida, and westward to Veracruz, Mexico. Although the
species are usually found in salinities ranging from 15 to 25 ppt, one has
been recorded from salinities as low as 5.5 ppt in Whitewater Bay, Florida
(Springer 1959), and can survive salinities in excess of 35 ppt. They are
usually found in grassbeds or over hard substrates permeated with small
holes and crevices (such as oyster reefs and rock jetties) and have occasion-
ally been collected over open mud or sand bottom (Springer 1959).
Springer (1959) reviewed the genus, but subsequent workers have
experienced difficulty identifying specimens. The mandibular tooth and
maxillary length characters Springer used permit identification of most
adult specimens, but do not always provide a distinct separation when
applied to young individuals or specimens from the eastern coast of Flor-
ida. My study describes several characters not discussed by Springer,
which permit specific identification of any postlarval specimen of Chas-
modes regardless of locality and presents a key to species and subspecies.
Springer recognized the distinctness of the two species in the area of
sympatry in the northern Gulf of Mexico, but offered no explanation for
this phenomenon. A mechanism is proposed herein to explain how genetic
integrity of the two species is maintained within this area. Springer's
observation on the divergence of the Texas population of C. bosquianus is
supported and, in addition, the northern Gulf of Mexico population is
recognized as a distinct subspecies. Analysis of the Everglades population
supports Springer's contention that this population is not taxonomically
distinct from neighboring populations.
The two species of Chasmodes are of special interest from a zoogeo-
graphic point of view. Their distribution patterns coincide with those
described for other fishes (e.g. Dawson 1964, Relyea 1965, Shipp and
Yerger 1969, Hardy 1980) and help to define previously proposed faunal
transition areas. In order to elucidate the zoogeographical implications of
the distribution patterns, the evolutionary histories of the species are ana-
lyzed using geological and climatological evidence.
Since the original description of the genus Chasmodes in 1836, the only
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
comprehensive taxonomic treatment of the genus is that of Springer (1959),
who recognized two species, C. bosquianus and C. saburrae. Bath (1977)
assigned eastern Pacific species (C. jenkinsi and C. maculipinna) to Hyp-
soblennius. Springer and Gomon (1975) placed C. herklotsi in the synon-
ymy of Omobranchus fasciolatoceps.
Norman (1943) provided a synopsis of the genera in the family Blennii-
dae. Included were three species in the genus Chasmodes: C. bosquianus,
C. novemlineatus, and C. maculipinna. He apparently overlooked C.
saburrae. Springer (1959) placed C. novemlineatus in the synonymy of C.
bosquianus.
In an osteological analysis of the family Blenniidae, Spring (1968)
mentioned Chasmodes several times, but only in regard to intergeneric
relationships.
Bath (1977), in his review of the tribe Blenniini, analyzed phylogenetic
relationships within the group. He suggested that Chasmodes serves as a
link between the generaHypsoblennius andHypleurochilus, but he did not
examine the relationship between the two species of Chasmodes.
In a review of the western Atlantic species of Hypsoblennius, Smith-
Vaniz (1980) presented characters to distinguish Chasmodes from Hypso-
blennius and Hypleurochilus.
Ecological studies dealing with Chasmodes saburrae appear in Reid
(1954) and Springer and Woodburn (1960). Tavolga (1958) presented his
results of studies on underwater sounds produced by specimens of C.
saburrae (erroneously called C. bosquianus). Phillips (1971a, 1971b, 1974,
1977) reported on the social behavior and utilization of space by C.
bosquianus.
Hildebrand and Cable (1938) presented information on the develop-
ment and life history of Chasmodes bosquianus.
Additional references to the genus Chasmodes are in the form of geo-
graphical records and/or ecological notes (Weymouth [1910] for Louisi-
ana; Baughman [1950] and Hoese [1958] for Texas; Tabb and Manning
[1961] for Whitewater Bay, Florida; and Ogren and Brusher [1977] for
northwest Florida).
In this paper, Chasmodes bosquianus (Lacepede 1800) and C. saburrae
Jordan and Gilbert 1883 are considered valid species, and a new subspe-
cies, C. b. longimaxilla, is described from the northern Gulf of Mexico.
ACKNOWLEDGEMENTS
I wish to express my appreciation to Robert L. Shipp, who provided encouragement and
guidance during this research and also arranged for financial assistance. I thank Joseph F.
Fitzpatrick, Jr., and Stephen A. Bloom for assistance with statistical analysis of the data; Alvin
P. Rainosek for consultation on statistical procedures; and James M. Boyles, David M. Dean,
and George F. Crozier, for advice and suggestions regarding my research.
Chris Benedict kindly provided the material that Kerry MacPherson of North Carolina
Power and Light Company donated to the Florida State Museum.
BULLETIN FLORIDA STATE MUSEUM
I thank the curators and assistant curators of the fish collections from which material was
borrowed. Special appreciation is accorded Victor G. Springer for providing access to
specimens in the United States National Museum of Natural History.
I wish to express special thanks to R. Douglas Nester for his assistance in collecting
specimens and for his optimism and encouragement regarding my research. Thanks are also
accorded Douglas Clarke for providing the holotype of the new subspecies. Carter R. Gilbert,
George H. Burgess, and Victor G. Springer critically reviewed and improved earlier versions
of the manuscript. Grace Russell typed and helped edit the manuscript.
I thank my parents for their support and encouragement. Special consideration is due my
wife Karen for her understanding and support throughout this study.
METHODS
Measurements were made with dial calipers and recorded to the nearest 0.1 mm. All counts
and most measurements were made using a Wild M5 stereoscope. Meristic counts and
morphometric measurements follow Hubbs and Lagler (1958) and Smith-Vaniz (1976).
Maxillary length was measured from the tip of the snout to the posterior edge of the left
maxilla. The edge was occluded from view by an opaque covering of skin. The posterior edge
was located with a sharp probe or jewelers forceps. Counts of pores in the mandibular series
(Fig. 1A, B) include all pores anterior to the narrow space between the preopercular and
articular bones on each side of the head (Fig. 1). This space can usually be seen as a narrow
translucent area immediately ventrad to the posterior edge of the maxilla. Caudal ray counts
include all segmented rays. The branchiostegal membrane is broadly fused to the body and
was cut on the right side to provide access to the branchial cavity. Counts of gill rakers and
pseudobranchial filaments refer to those on the right side of the body. Gill raker counts
include those rakers on the upper and lower limbs of the first arch (no rudiments were
observed). Upper pharyngeal teeth are located by separating the second and third gill arches
with jewelers forceps and, using illumination from beneath the specimen, are found in two
small semicircular clusters on the posterodorsal roof of the branchial cavity. As the lower
pharyngeal teeth must be dissected out of the specimen to be counted, my counts are based on
examination of four specimens of each subspecies of Chasmodes bosquianus and eight
specimens of C. saburrae. Vertebral counts were taken from radiographs and cleared and
stained specimens. Precaudal vertebrae included vertebrae lacking a well-developed hemal
spine. Caudal vertebrae include vertebrae with a well-developed hemal spine; posteriormost
caudal vertebra bears the dorsal and ventral hypural plates.
Information regarding embryonic development and posthatching characters was obtained
from eggs and larvae taken from experimental laboratory aquaria.
Numbers in parentheses in material examined sections refer to the number of specimens in
that particular lot.
Linear regression equations and their correlation coefficients, and covariance comparisons
(Sokal and Rohlf 1969) were computed using a covariance program written by Dr. S. A.
Bloom (Zoology Department, University of Florida). Computations were made utilizing the
facilities of the Northeast Regional Data Center of the State University System of Florida,
located on the campus of the University of Florida. F values are not included for covariance
comparisons presented in matrix form, but are available upon request.
Because of editorial problems, all the populations of each form are not differentiated in
Figs. 3 and 4. Maxillary length as percent of SL is presented in tabular form for selected
populations (Table 8). Graphs of the number of mandibular teeth vs. SL were prepared for
each population to facilitate comparisons but are not included owing to space limitations.
Repositories, with abbreviations in parentheses, of specimens examined are: Academy of
Natural Sciences of Philadelphia (ANSP), British Museum (Natural History) (BMNH), Cali-
fornia Academy of Sciences (CAS-SU), Field Museum of Natural History (FMNH), Florida
Department of Natural Resources, Marine Research Laboratory, St. Petersburg (FSBC),
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
Florida State Museum, University of Florida (UF), Florida State University (FSU), Grice
Marine Biological Laboratory, College of Charleston, South Carolina (GMBL), Gulf Coast
Research Laboratory, Ocean Springs, Mississippi (CCRL), Indian River Coastal Zone Study
Reference Museum, Harbor Branch Foundation, Ft. Pierce, Florida (IRCZ), Museum
National d'Histoire Naturelle, Paris (MNHN), Northeast Louisiana University, Baton Rouge,
Louisiana (NLU), Texas Cooperative Wildlife Collection of Texas A&M University (TAMU),
Tulane University (TU), University of Massachusetts (UMS), University of Miami Rosenstiel
School of Marine and Atmosphere Sciences (UMML), University of Michigan Museum of
Zoology (UMMZ), United States National Museum of Natural History (USNM), University
of North Carolina, Morehead City (UNC), University of South Alabama Ichthyological
Collection, Mobile (USAIC), University of Texas Institute of Marine Science (UTMSI),
Virginia Institute of Marine Science (VIMS and CBL for specimens donated to VIMS by
Chesapeake Biological Laboratory). The Florida fishes included in the GCRL material were
from the Raney-Wang Collection donated to GCRL by the Mote Marine Laboratory.
GENUS CHASMODES VALENCIENNES 1836
Chasmodes Valenciennes (in Cuvier and Valenciennes) 1836: 295. (Type
species: Blennius bosquianus Lacepede 1800 by original designation).
Blennitrachus Swainson 1839:78, 274. (Type species: Pholis quadrifasciatus
Wood 1825, by original designation).
A genus of blenniid fish (sensu Springer 1968) characterized by dorsal
elements X to XII, 16-20 (XI spines in 99% of specimens examined, usually
18 or 19 rays; Table 1), occasionally 2 or 3 rays bifurcate; anal elements II,
16-20 (usually 18 or 19 rays; Table 1); pectoral-fin rays 11 to 13 (99% of
specimens examined have 12); segmented caudal-fin rays 10 to 13 (99% of
those examined have 11, one specimen has 9 due to an anomalous condi-
tion); branched caudal-fin rays 0-10; dorsal procurrent caudal-fin rays 4-5;
ventral procurrent caudal-fin rays 3-5; precaudal vertebrae 10; caudal
vertebrae 24-26; epipleural ribs 12; branchiostegal rays 6; teeth in jaws
typically uniserial and uniformly spaced; one specimen examined has two
teeth immediately behind uniserial row of teeth in anterior portion of lower
jaw. This is probably anomalous, as various anomalies were observed in
other specimens. Mandibular teeth distributed over approximately ante-
rior half of dentary; premaxillae toothed over anterior three-quarters;
vomer and palatines toothless; upper pharyngeals each have 5-7 enlarged
canines and 0 to 5 small canines; each lower pharyngeal has 4-6 enlarged
canines and 4-6 small ones, number of enlarged canines on upper pharyn-
geals relatively constant, but count of small pharyngeal teeth highly varia-
ble (Table 1); gill openings restricted by broad fusion of branchiostegal
membranes with body, opening never extends ventrad below ventral level
of pectoral-fin base.
Chasmodes is, perhaps, unique in the blenniid tribe Blenniini, in having
the ventral hypural plate fused to the complex urostylar centrum (Peters
1981; Springer, pers. comm.). Springer (1968) erroneously reported that
the ventral plate was autogenous in Chasmodes.
BULLETIN FLORIDA STATE MUSEUM
Lateral line development is described by Bath (1977). He refers to the
development of the lateral line as "type C," which is a reduced form of the
lateral line consisting anteriorly of alternating narrow branches from the
main canal. This branching pattern terminates immediately behind the
pectoral-fin base, and the lateral line continues posteriorly in the form of
short straight tubes to its terminus at about the middle of the body.
Primary sexual dimorphism is in the form of a greatly reduced first anal
spine in females, and in nuptial males having enlarged, fleshy structures
with many longitudinal folds present on both anal spines (see Smith-Vaniz
1980: Fig. lb).
Secondary sexual dimorphism was found in color pattern (see discus-
sion of coloration), maximum length, and maxillary length (in some popu-
lations of Chasmodes saburrae). The SL of the largest male in any given
population of Chasmodes exceeds that of the largest female by 2 to as much
as 25 mm. This difference in length may be related to territory maintenance
by males.
Statistical analysis of the maxillary length character (Tables 2, 3)
revealed a significant sexual difference in 4 of 7 populations of Chasmodes
saburrae, but no significant sexual dimorphism was found in populations of
C. bosquianus. A similar analysis of the number of mandibular teeth
(Tables 2, 3) showed no sexual dimorphism for any population of Chas-
modes. I am unable to explain the appearance of sexually dimorphic
maxillary lengths in only 4 of the 7 populations of C. saburrae tested. The
biological significance of this sexual dimorphism is unknown.
Coloration shows no variation among the species and subspecies. Colo-
ration of males varies from light lines running longitudinally on a dark
background to a mottled pattern. Females of all sizes are variously mottled.
In life, territorial adult males have light longitudinal lines on an olivaceous
background with an iridescent blue spot on the membrane between the
first and second dorsal spines and an orange streak extending posteriorly
from this spot to about the tenth dorsal spine. The chest and branchiostegal
membranes of these males are orange. Females, immature males, and
non-territorial males have dark brown and white mottled pigmentation.
(Information on coloration is modified from Springer [1959].)
KEY TO THE SPECIES AND SUBSPECIES OF CHASMODES
A. Pores in both mandibular series combined usually 6 or more (Figure 1B and Table 1), if
less than 6, then lower jaw with prominent lip flaps (Figure IB); mandibular teeth with
broadly rounded tips (Figure 2A) in specimens over 35 mm SL ....................
.................. ............................ saburrae (Edgewater, Florida, to
Chandeleur Islands, Louisiana).
AA. Pores in both mandibular series combined usually 4 (occasionally 5 or 6; Figure 1A and
Table 1), if more than 4, then lower jaw without prominent lip flaps (Figure 1A);
mandibular teeth with sharply pointed tips (Figure 2B) in specimens of any length ...
............................................................ ....... . B
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
B. Maxillary length usually 13 to 16.5% SL (range 11-18.5%, X = 14.9%; Table 4); gill rakers
modally 12 (Table 1); total number of small upper pharyngeal teeth 1-6 (Table 1) ....
.................. .............. .......... bosquianus bosquianus (New York
to Marineland, Florida).
BB. Maxillary length usually 15.5 to 20% SL (range 14-22.5%, X = 17.3%; Table 4); gill rakers
modally 11 (Table 1); total number of small upper pharyngeal teeth usually 0-1 in
specimens over 40 mm SL, 2-5 in those under 40 mm SL (Table 1).................
......................................... bosquianus longimaxilla new subspecies
(Pensacola, Florida, to Veracruz, Mexico).
--- uP
AA B
FIGURE 1.-Ventral aspect of lower jaw of Chasmodes bosquianus (A) and C. saburrae (B).
Abbreviations are as follows: UP = upper lip, LF = lip flap, MD = mandibular pore
(anterior and posterior pores are indicated on one side of the lower jaw), AP = space between
the articular and preopercular bones, POP = preopercular pore.
CHASMODES BOSQUIANUS (LACEPEDE 1800)
(See individual subspecies for synonymy)
DIAGNOSI-Chasmodes bosquianus is readily distinguished from C.
saburrae in usually having 4 (vs. 6) mandibular pores (83% of specimens
examined of each species have these counts; Fig. LA, B and Table 1). C.
bosquianus typically lacks lip flaps, whereas C. saburrae has well-
developed lip flaps (Fig. 1A, B). C. bosquianus can also be separated from
C. saburrae by having 9-13 (vs. 12-14) gill rakers on the first arch (visible
only upon dissection; Table 1). Dissected specimens of C. bosquianus have
4 enlarged canines on the lower pharyngeals, whereas C. saburrae has 6.
STATISTICAL COMPARISONS.-Pairwise covariance comparisons were
made between regression equations (Table 2) for males of each population
of Chasmodes saburrae and males of each population of C. bosquianus,
and the same was done for females of each population. The resulting
maxillary length comparisons showed statistically significant (p<0.001)
BULLETIN FLORIDA STATE MUSEUM
0.5 m
FIGURE 2.-Frontal view of an anterior mandibular tooth from Chasmodes saburrae (A) and
C. bosquianus (B). Stippled area indicates curved portion of teeth.
differences in every interspecific comparison. The lack of statistically
significant sexual dimorphism in the number of mandibular teeth (Table 3)
enabled covariance comparisons of regression equations (Table 5) for
entire populations (males and females). All populations of C. saburrae had
significantly fewer mandibular teeth than populations of C. bosquianus
(Table 6).
Chasmodes bosquianus bosquianus (LACIPIDE)
Blennius bosquianus Lac6epde 1800: 495 (Type locality: South Carolina;
holotype lost?).
Pholis novemlineatus Wood 1825:280 (Type locality: Charleston Harbor,
South Carolina; holotype ANSP 10410).
Pholis quadrifasciatus Wood 1825:282 (Type locality: unknown; holotype
lost?).
VOL. 29, NO. 2
Table 1-Frequency distributions for certain characters in species and subspecies of Chasmodes. (Holotype of new subspecies is designated by a
for characters not requiring dissection.)
Segmented dorsal-fin Total dorsal-fin Segmented anal-fin Caudal
rays elements rays vertebrae
Species 16 17 18 19 20 27 28 29 30 31 16 17 18 19 20 24 25 26
saburrae 1 1 14 5 1 1 1 13 6 1 1 14 3 2 15 7 2
bosquianus
bosquianus 3 24 12 2 3 24 12 2 1 4 19 13 3 6 2 1
h. longimaxilla 1 1 27* 5 9 27" 5 1 6 23* 9 1 6 5
Enlarged upper Small upper pharyngeal teeth (both sides included) Dorsal + ventral
pharyngeal teeth Specimens procurrent
(both sides included) Specimens > 40 mm SL <40 mm SL caudal-fin rays
Species 10 11 12 13 14 0 1 2 3 4 5 6 7 8 9 10 2 3 4 5 6 7 8 9 10
saburrae 3 4 17 6 1 1 1- 7 7 3 2 1 1 2 3 2 1 6 10 2 1
bosquianus
bosquianus 2 17 1 2 3 1 7 2 2 1 2 4 8 1
h. longimaxilla 1 1 18 1 1 5 2 1 1 1 1 1 6 1 3 6 8 2
Pseudobranchial filaments Gill rakers Number of mandibular pores
Species 8 9 10 11 12 13 14 9 10 11 12 13 14 4 5 6 7 8 X
saburrae 6 15 5 2 1 8 13 10 5 33 560 69 6 6.0
bosquianus
bosquianus 2 7 6 5 1 2 7 12 4 136 23 10 4.3
b. longimaxilla 4 5 6 6 2 1 5 17 2 103" 13 1 4.1
Table 2-Regression equations, correlation coefficients (r), and probability values (P<) for maxillary length and number of mandibular teeth (Y)
versus standard length (X) for males and females within populations of Chasmodes.
Regression equation Regression equation
Species, population, sex N maxillary length r P< mandibular teeth r P<
bosquianus bosquianus
Marineland, Florida
males 16 Y = -0.9520 + 0.1669X
females 21 Y = 0.0840 + 0.1399X
N. E. Florida and Georgia
males 8
females 7
Y = -2.6378 + 0.2059X
Y = -0.1528 + 0.1416X
South Carolina
males 10 Y = -1.4728 + 0.1842X
females 10 Y = -0.1885 + 0.1528X
North Carolina
males 22 Y = -1.3193 + 0.1803X
females 11 Y = -0.7484 + 0.1608X
Chesapeake Bay to
New York
males 51 Y = -1.9534 + 0.1910X
females 48 Y = -0.9159 + 0.1650X
bosquianus longimaxilla
Pensacola, Florida to
Louisiana
males 20 Y = -1.6030 + 0.2164X
females 11 Y = -4.4562 + 0.2631X
Matagorda Bay area,
Texas
males 25 Y = -1.2860 + 0.2072X
females 8 Y = -0.7269 + 0.1830X
0.959 0.01 Y = 20.3108 + 0.4950X
0.862 0.01 Y = 22.5687 + 0.4458X
0.970 0.01 Y = 21.1373 + 0.4911X
0.997 0.01 Y = 24.9190 + 0.4209X
0.962 0.01 Y = 18.8009 + 0.5621X
0.918 0.01 Y = 21.2808 + 0.5150X
0.972 0.01
0.986 0.01
0.964 0.01
0.971 0.01
0.943 0.01
0.962 0.01
0.928 0.01
0.984 0.01
Y = 21.6752 + 0.4948X
Y = 14.7354 + 0.6340X
Y = 18.4520 + 0.4753X
Y = 18.8321 + 0.4779X
Y = 25.5717 + 0.4622X
Y = 27.0016 + 0.4493X
0.920 0.01
0.718 0.01
0.773 0.05
0.789 0.05
0.854 0.01
0.838 0.01
0.808 0.01
0.913 0.01
0.867 0.01
0.911 0.01
0.921 0.01
0.770 0.02
Y = 31.5014 + 0.3560X 0.711 0.01
Y = 28.5067 + 0.4183X 0.963 0.01
Aransas Bay area, Texas
males 12 Y = -1.3109 + 0.1928X
females 16 Y = -1.5726 + 0.2063X
Southern Laguna Madre,
Texas
males 25
females 21
Y = -1.2872 + 0.1950X
Y = -1.1026 + 0.1857X
0.948 0.01
0.983 0.01
0.970 0.01
0.962 0.01
Y = 31.5834 + 0.3215X
Y = 26.2605 + 0.4383X
Y = 15.7722 + 0.6451X
Y = 17.7885 + 0.6597X
saburrae
Louisiana to Pensacola,
Florida
males 43 Y = -0.6861 + 0.1340X
females 60 Y = -0.0690 + 0.1116X
Ft. Walton area to
Cedar Key, Florida
males 116 Y = -0.7615 + 0.1339X
females 82 Y = -0.1677 + 0.1139X
Tampa Bay area
males 55 Y = -0.8562 + 0.1395X
females 45 Y = -0.1698 + 0.1170X
Charlotte Harbor
males 18 Y = -0.6510 + 0.1357X
females 17 Y = 0.4354 + 0.1257X
Everglades
males 28 Y = -0.4289 + 0.1269X
females 18 Y = 0.1629 + 0.1086X
Biscayne Bay area
males 6 Y = -0.9597 + 0.1428X
females 9 Y = 0.1289 + 0.1026X
Indian River area
males 29 Y = -0.5455 + 0.1373X
females 26 Y = 0.1782 + 0.1082X
0.982 0.01
0.966 0.01
0.984 0.01
0.981 0.01
0.986 0.01
0.982 0.01
0.996 0.01
0.990 0.01
0.951 0.01
0.966 0.01
0.982 0.01
0.949 0.01
0.992 0.01
0.991 0.01
Y =16.7828 + 0.3289X
Y = 13.1644 + 0.4274X
Y = 14.0077 + 0.3906X
Y = 14.4744 + 0.3773X
Y = 13.3902 + 0.4070X
Y = 13.4447 + 0.4362X
Y = 16.8821 + 0.3266X
Y = 14.6048 + 0.4007X
Y = 15.0054 + 0.4161X
Y =12.1411 + 0.5037X
Y = 14.4144 + 0.3866X
Y = 15.1999 + 0.3449X
0.894 0.01
0.880 0.01
0.903 0.01
0.926 0.01
0.928 0.01
0.924 0.01
0.939 0.01
0.934 0.01
0.844 0.01
0.914 0.01
0.977 0.02
0.907 0.01
Y = 13.6418 + 0.4745X 0.966 0.01
Y = 13.8857 + 0.4654X 0.941 0.01
0.714 0.02
0.915 0.01
0.934 0.01
0.828 0.01
Table 3-F values for covariance comparisons of regression equations (Table 2) of maxillary length and number of mandibular teeth between males
and females within populations of Chasmodes (** = P<.001; NS = not significant). F values for testing parallelism of the regression lines
are given only when the lines are not coincident.
F values F values
Species, population Abbreviation df coincidence df parallel
bosquianus bosquianus
Marineland, Florida
maxillary length
mandibular teeth
N. E. Florida and Georgia
maxillary length
mandibular teeth
South Carolina
maxillary length
mandibular teeth
North Carolina
maxillary length
mandibular teeth
Chesapeake Bay to New York
maxillary length
mandibular teeth
bosquianus longimaxilla
Pensacola, Florida to Louisiana
maxillary length
mandibular teeth
Matagorda Bay area, Texas
maxillary length
mandibular teeth
Aransas Bay area, Texas
maxillary length
mandibular teeth
2/35 1.16 NS
2/35 0.10 NS
2/13 4.47 NS
2/13 0.07 NS
2/18 0.82 NS
2/18 0.04 NS
2/31 3.18 NS
2/31 0.66 NS
2/97 6.16 NS
2/97 0.32 NS
2/29 0.79 NS
2/29 0.12 NS
2/31 3.64 NS
2/31 0.27 NS
2/26 1.38 NS
2/26 0.62 NS
Southern Laguna Madre Bay, Texas
maxillary length
mandibular teeth
saburrae
Louisiana to Pensacola, Florida
maxillary length
mandibular teeth
Ft. Walton area to Cedar Key, Florida
maxillary.length
mandibular teeth
Tampa Bay area
maxillary length
mandibular teeth
Charlotte Harbor
maxillary length
mandibular teeth
Everglades
maxillary length
mandibular teeth
Biscayne Bay area
maxillary length
mandibular teeth
Indian River area
maxillary length
mandibular teeth
2/44 0.99 NS
2/44 3.43 NS
2/101 16.71 **
2/101 3.33 NS
2/196 33.72 **
2/196 0.19 NS
2/98 10.88 **
2/98 4.58 NS
2/33 3.35 NS
2/33 1.11 NS
2/44 4.18 NS
2/44 2.03 NS
2/13 4.37 NS
2/13 0.51 NS
2/53 25.39 *
2/53 0.03 NS
1/101
1/196
15.85 **
30.68 **
17.80 **
34.99 **
BULLETIN FLORIDA STATE MUSEUM
Table 4-Frequency distribution of maxillary length expressed as percent of standard length
in Chasmodes bosquianus.
C. b. longimaxilla
Florida,
Maxillary Length Mississippi,
as % SL C. b. bosquianus Louisiana Texas
12.5 6
13.0 9
13.5 13
14.0 28 1 1
14.5 32 4
15.0 34 3
15.5 29 1 13
16.0 15 11
16.5 11 1 9
17.0 5 1 12
17.5 3 3 11
18.0 2 2 13
18.5 2 2 4
19.0 3 9
19.5 3 1
20.0 2 6
20.5 3 1
21.0 1
21.5 1
22.0 1
22.5 2*
*indicates the holotype
DIAGNOSIs.-Chasmodes bosquianus bosquianus differs from C. b. lon-
gimaxilla in having a shorter maxillary length (X = 14.9 vs. 17.3% SL; Fig. 3,
Table 4), and in having a total of 1-6 small canines (Table 1) in addition to
enlarged canines on upper pharyngeals, whereas C. b. longimaxilla has 0-5
(most specimens examined over 40 mm SL with 0-1). Gill rakers 11-13
modallyy 12) for C. b. bosquianus vs. 9-12 modallyy 11) for C. b. longimax-
illa (Table 1).
DEscRIPTIoN.-(See also Table 1). Maxillary length expressed as per-
cent of SL 12.5-18.5 (X = 14.9%, number [n] = 189; Table 4); number of
mandibular teeth varies ontogenetically, number increasing with increased
size (Fig. 4); mandibular pores 4 to 6 (X = 4.3, n = 169; Table 1); enlarged
upper pharyngeal teeth 6 or 7 on each side; branched caudal-fin rays 0-10;
small upper pharyngeal teeth 0-4 on each side (Table 1); enlarged lower
pharyngeal teeth 4 (n = 4); maximum size approximately 80 mm SL.
Twelve larvae (5.4 to 7.3 mm SL) have 5 to 7 spines radiating posteriorly
from preopercle, with spine at angle of preopercle enlarged. Small man-
dibular teeth range in number from approximately 8 to 12. Maxillary length
VOL. 29, NO. 2
Table 5- Regression equations, correlation coefficients (r), and probability values (P<) for maxillary length and number of mandibular teeth (Y)
versus standard length (X) for those populations (males and females combined) of Chasmodes in Table 3 showing no significant sexual
dimorphism in these characters. (Species and populations are designated using the abbreviation in Table 3.)
Species, Regression equation Regression equation
population N maxillary length r P< mandibular teeth r P<
BM 37 Y = -0.9274 + 0.1648X 0.950 0.01 Y = 21.0121 + 0.4817X 0.887 0.01
BF 15 Y = -1.9437 + 0.1924X 0.985 0.01 Y = 23.8910 + 0.4453X 0.882 0.01
BS 20 Y = -1.0009 + 0.1735X 0.950 0.01 Y = 20.2408 + 0.5368X 0.865 0.01
BN 33 Y = -1.3681 + 0.1796X 0.980 0.01 Y = 18.0279 + 0.5608X 0.886 0.01
BC 99 Y = -1.6662 + 0.1838X 0.969 0.01 Y = 19.0285 + 0.4688X 0.896 0.01
LS 31 Y = -2.1628 + 0.2253X 0.945 0.01 Y = 26.0300 + 0.4585X 0.886 0.01
LM 33 Y = -1.7690 + 0.2126X 0.963 0.01 Y = 30.3805 + 0.3748X 0.859 0.01
LA 28 Y = -1.3976 + 0.1987X 0.970 0.01 Y = 27.8290 + 0.4008X 0.859 0.01
LL 46 Y = -1.5185 + 0.1980X 0.976 0.01 Y = 19.6115 + 0.5889X 0.901 0.01
SS 103 Y = 15.5021 + 0.3599X 0.894 0.01
SN 198 Y = 14.1252 + 0.3874X 0.920 0.01
ST 100 Y = 14.0308 + 0.4022X 0.921 0.01
SC 35 Y = 16.3956 + 0.3416X 0.938 0.01
SE 46 Y = 14.9886 + 0.4242X 0.871 0.01
SM 15 Y = 13.9553 + 0.3857X 0.963 0.01
SF 55 Y = 13.7037 + 0.4722X 0.961 0.01
BULLETIN FLORIDA STATE MUSEUM
varies from 8.5 to 10% SL (X = 9.1%, n = 12). Twelve pectoral-fin rays
present in all larvae.
Larvae exhibit the following pigment pattern in alcohol: membrane
between lower 6 pectoral rays pigmented to tips, membranes between
sixth-seventh and seventh-eighth rays pigmented only on inner half and
fourth respectively, with outer portion of membrane transparent; mem-
brane covering brain with several large pigment spots; two pigment spots
on either side of tip of upper jaw, immediately anterior to oribital margin; a
pigment spot at dorsal edge of each opercle at upper angle of gill opening.
Table 6-Matrix of covariance comparisons of regression equations (Table5) for number of
mandibular teeth between different populations of Chasmodes (P<0.001; see
Table 3 for explanation of abbreviations). 0 = coincident regression lines;
= noncoincident but parallel lines; + = noncoincident and nonparallel lines.
BM BF BS BN BC LS LM LA LL SS SN ST SC SE SM SF
BM 0 0 0
BF 0 0 0 0 0 0
BS 0 0 0 0 0 +
BN 0 0 0 + + + +
BC + +
LS 0 0 0
LM 0 0
LA 0
LL + + + +
SS 0 0 0 0 +
SN 0 0 0 +
ST 0 0
SC 0 +
SE 0
SM
SF
STATISTICAL COMPARISONS.-The results of statistical comparisons of the
populations of Chasmodes bosquianus are summarized in Tables 6 and 7.
Populations of C. b. bosquianus have significantly shorter maxillary lengths
than populations of C. b. longimaxilla (Table 7). No consistent statistical
differences were obtained for the number of mandibular teeth between C.
b. bosquianus and C. b. longimaxilla populations (Table 6). Two popula-
tions of C. b. bosquianus (Chesapeake Bay area and Marineland) have
significantly fewer mandibular teeth than all populations of C. b. longimax-
illa. The North Carolina population of C. b. bosquianus significantly
differs in number of mandibular teeth from only one population (Laguna
Madre) of C. b. longimaxilla. Comparisons of the number of mandibular
teeth between other populations of the two subspecies did not show
significantly different regression equations for number of mandibular
VOL. 29, NO. 2
o o
a
o o
0 0 0
0 0
0 0o0
0 0 00 AAA
0 0a 'n' 0 NAA A 0
0 0 0A &AA A AA60
00 0AAAOA 0 0 0
o 0AA0 o
o A o 00 00
^n AZ A o o oo
0 0
00
Bo00 o
0 1 1 1 I I I 1 i
0 10 20 30 40 50 60 70 80
STANDARD LENGTH (mm)
FIGURE 3.-Graph of maxillary length vs. standard length for Chasmodes: C. saburrae, circles, closed circles represent Everglades specimens; C.
bosquianus bosquianus, triangles; C. b. longimaxilla, squares.
17.5 -
15.0
S12.5
E
I
10.0
w
IC
_J
S7.5
.J
5.0
5.0
2.5 -
70- 0
A
60- o oO o
A 0 o
-- o 0o A A 0
S0 08
o ooo oooo O O 8 a& o
z 0 08 O 8 A A 0oO
So808 8 a8o ao
o-oo o o oo0o A
S40 A oA A 0 0 0
A 0 oA8 8A 0 00 00 0
a o 8 AA A 8%oAOo-ooOo0 8 880 0000 0
S000000 00 o
o A 80 O 20Ot 0000 0 0
00 0 000
0 88 0 Ac 00 0
SQ O 000 000000 000 0
0 2 0 0 0 40 50 60 70 80
A O 0 000 00 0 0
08 oooooooo o o oo
S20- o S0 0 0
E3 0 000 0 0 0
3o 0o oD 0
0
10-
0*
0 10 20 30 40 50 60 70 80
STANDARD LENGTH (mm P
z
FIGURE 4.-Graph of number of mandibular teeth vs. standard length for Chasmodes: (symbols as in Fig. 3). to
WILLIAMS: CHASMODES TAXONOMY
Table 7-Matrix of covariance comparisons of maxillary length regression equations (Table
5) for populations of Chasmodes bosquianus bosquianus and C. b. longimaxilla
(P<0.001). 0 = coincident lines; = noncoincident but parallel lines; + = non-
coincident and nonparallel lines. (Population abbreviations defined in Table 3.)
BM BF BS BN BC LS LM LA LL
BM 0 0 0 0
BF 0 0 0
BS 0 0
BN 0
BC + -
LS 0 0
LM 0 0
LA 0
LL
SCALE
0 250 500KM
94" 90* 86*
4~
4i
/
a
;iy
82 78' 74"
FIGURE 5.-Distributional records for species and subspecies of Chasmodes. Grand Terre,
Louisiana, record based on University of Southwestern Louisiana specimen (cat. no. 3496) not
included in material examined.
teeth.
Comparisons of regression equations for maxillary length (Table 5)
show no statistically significant differences between populations of C. b.
bosquianus (Table 7). Significant differences were found for comparisons
SC.B. BOSQUIANUS
* C.B. LONGIMAXILLA
* C. SABURRAE
* AREAS WHERE BOTH C. SABURRAE AND
C. B. LONGIMAXILLA HAVE BEEN COLLECTED
o SPECIMENS OF C.B. LONGIMAXILLA
REPORTED BY BATH (1977)
20"
' ' '
L
ir
9'
I
'a
^-^
BULLETIN FLORIDA STATE MUSEUM
of regression equations for number of mandibular teeth between popula-
tions of C. b. bosquianus (Table 6). The Chesapeake Bay area population
has significantly fewer mandibular teeth (Tables 5, 6) than other popula-
tions of this subspecies. I believe that this difference is due to a collecting
bias. I was unable to locate specimens of Chasmodes from the Cape
Hatteras area (Fig. 5), but the fact that C. b. bosquianus usually is found in
an estuarine habitat, makes it probable that the extensive estuaries of the
Cape Hatteras area support this form. An analysis of specimens from this
area is needed to determine the nature of the relationship of the Chesa-
peake Bay area population to the other populations of C. b. bosquianus.
DISCUSSION OF SYNONYMY.-Examination of the holotype of Pholis
novemlineatus Wood 1825 substantiates Springer's (1959) conclusion that
this species is a synonym of Chasmodes bosquianus. Based on the maxillary
length, the mandibular tooth count, and the type locality, I identify this
form as C. b. bosquianus.
The holotype of Pholis quadrifasciatus Wood 1825 is missing (lost?).
Only two meristic characters (caudal and anal ray counts) given in the
original description vary from those of C. bosquianus. The count of 9
caudal-fin rays and 15 anal-fin rays for P. quadrifasciatus are below the
minimum for Chasmodes, but it is not uncommon for specimens of Chas-
modes to have anomalous fin ray counts. If Wood included only branched
rays in his caudal-fin ray count, the count of 9 lies within the observed range
(0-10) for the genus Chasmodes. Inasmuch as all other characters, including
color pattern, agree with those of C. b. bosquianus, I believe that the
original anal-fin count for P. quadrifasciatus was either in error or anomal-
ous. The original description and illustration of P. quadrifasciatus leave no
doubt that this is a specimen of C. bosquianus. I tentatively place P.
quadrifasciatus in the synonymy of C. b. bosquianus, because it is probable
that the holotype was collected on the Atlantic coast (Wood having
obtained it from a museum in Baltimore).
ECOLOGY AND DISTRIBUTION.-Chasmodes bosquianus bosquianus is
found primarily in bays and estuaries of the Atlantic seaboard. As is true of
the genus as a whole, it usually inhabits waters with salinities between 15
and 25 ppt. The only reliable record of its capture in lower salinities is that
of Dr. William D. Anderson, Jr. Two specimens (27 and 35 mm SL) were
collected in Charleston Harbor 4 May 1971 in water with a salinity of 9.7
ppt.
Hildebrand and Cable (1938) stated that the spawning season extends
from May to August. Acquisition of several 5-7 mm larval Chasmodes
bosquianus bosquianus collected in Boque Sound, North Carolina, on 23
April 1957 indicates that the spawning season begins much earlier, proba-
bly in mid-March.
I have examined only five specimens of Chasmodes bosquianus bos-
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
quianus from waters north of Chesapeake Bay, Maryland: three from
Delaware, one from New Jersey, and one from New York. The last is the
northernmost record. All likely are stragglers to these areas, as they were
collected either in late summer or early fall, which would have allowed
sufficient time at sufficiently high temperature for migration or passive
larval transport to these northern waters. C. b. bosquianus is primarily
distributed from Chesapeake Bay, Maryland, south to Marineland, Flor-
ida. I have identified, as C. b. bosquianus, one specimen from a series of C.
saburrae (USNM 48783) from Indian River, Florida. If this specimen was
not erroneously mixed with this lot (collection date: 14 January 1896), it is
the only record of sympatric occurrence of C. b. bosquianus with C.
saburrae, and is the only record of C. b. bosquianus south of Marineland,
Florida.
MATERIAL EXAMINED.-TYPES: ANSP 10410 (1), Charleston Harbor,
South Carolina, holotype of Pholis novemlineatus Wood. NEW YORK:
MNHN A-2206 (1). NEW JERSEY: UMMZ 109932 (1). DELAWARE:
USNM 187154 (1), UMS 24-598-5-1 (1), 24-589-5-3-(1). MARYLAND: FSU
17747 (4), USNM uncatalogued (2), 68180 (1), 74857 (1), 91140 (1), 91149
(5), 91150 (17), 100728 (1), 116659 (1), 201352 (8), 205137 (1), CBL 1952(2),
1971 (2), 1990 (4), 2320 (2), 2387 (12), 2390 (4), 2501 (26), 3601 (1), 4043 (3).
VIRGINIA: USNM 91141 (2), 91146 (4), 91151 (1), 116660 (1), 156651 (6),
VIMS 559 (1), 879 (1), 881 (3), 2694 (2). NORTH CAROLINA: USNM
198245 (2), UF 28282 (1), 28283 (1), 28284 (1), 28285 (1), 28286 (4), 28287 (3)
28288 (1), 28296 (1), UNC 36 (1), 1114 (3), 1472 (1), 1817 (2), 2065 (4), 2154
(2), 3846 (3), 6262 (3), 6279 (1), 6348 (2), 8510 (1), 8562 (2), 8558 (1), 9630
(1), 10430 (1). SOUTH CAROLINA: USNM 195819 (1), GMBL 70-23 (1),
70-157 (1), 71-67 (2), 71-85 (1), 71-110 (5), 74-173 (9), 152-FB-69 (1).
GEORGIA: UF 23684 (3), USNM 91976 (1), 127459 (1), UMML uncata-
logued (3), uncatalogued (1), uncatalogued (4). FLORIDA: ANSP 106142
(12), FSBC 15 (1), 17 (19), 21 (2), 22 (15), UF 11795 (2), 11797 (1).
Chasmodes bosquianus longimaxilla NEW SUBSPECIES
FIGURE 6
DIAGNOSIs.-Chasmodes bosquianus longimaxilla differs from C. b.
bosquianus and C. saburrae in having 0-1 (range 0-5) total small upper
pharyngeal teeth in most specimens over 40 mm SL (2-5 in specimens under
40 mm SL) vs. 1-10 in its congeners (Table 1), and in having a significantly
longer maxillary length (17.3% SL versus 14.9 and 11.3% SL, respectively;
Fig. 3, Tables 4,8). Gill rakers 9-12 modallyy 11) vs. 11-13 modallyy 12) for
C. b. longimaxilla and C. b. bosquianus respectively.
DESCRIPTION (See also Table 1).-Characters appearing in the generic
description and subspecies diagnoses are not repeated here. In addition,
the maxillary length varies from 14 to 22.5% SL (X = 17.3% SL, n = 125;
Table 8-Frequency distribution with respect to maxillary length expressed as percent of standard length (specimens >30 mm SL) for selected
populations of Chasmodes. C. b. 1. = Chasmodes bosquianus longimaxilla; C. b. b. = C. b. bosquianus; C. s. = C. saburrae.
Allopatric
Southern Central Area of Sympatry East Coast Populations
Maxillary Texas Texas Miss. Sd. & N. W. Indian River Marineland,
Length coast coast Mobile Bay Pensacola, Fl. Fl. area, Fl. Fl.
as % SL C. b. 1. C. b. 1. C. b. C. s. C. b. 1. C. s. C. s. C. s. C. b. b.
9.0 1
1
1
4 13
4 6
3 3
2
1
1 -
3 1
1 2
3 4 2 4
2 5 2 2
2 4 2 1 1
5 2 2 3 1
3 5 3 3
4 1 6 2 2
1 3 2
1 6 2 1 2
1 3 -
5 1 1 1
1 2 1
1
1
1
2'
2
1 2 7
1 8 10 30
2 5 33 32 3
5 3 49 14 2
4 2 29 7 3
4* 19 2 4
1 12 5
2 1
1
O
3 3
1 2
1 2 >
2 2
3 6
2 3 C
3 2
2 -
2 4
1
*indicates a holotype value.
C
WILLIAMS: CHASMODES TAXONOMY
FIGURE 6.-Holotype of Chasmodes bosquianus longimaxilla new subspecies. Adult male, 86
mm SL, from Dauphin Island, Alabama (USNM 219830).
Table4); like C. b. bosquianus, number of mandibular teeth increases with
increased size (Fig. 4); pores in both mandibular series combined 4 to 5
(rarely 6; X = 4.1, n = 115; Table 1). Of the three taxa referable to Chas-
modes, C. b. longimaxilla alone shows a difference in number of small
upper pharyngeal teeth between size classes (Table 1). Enlarged upper
pharyngeal teeth 5-7 on each side (Table 1); each lower pharyngeal with 4-5
small teeth anterior to 4 large, recurved canines; maximum size approxi-
mately 85 mm SL.
Pigmentation and other characters of larvae immediately after hatching
are similar to those described by Hildebrand and Cable (1938) for C.
bosquianus (= C. b. bosquianus) from Chesapeake Bay.
The only small (<10 mm) posthatching, wild-caught specimen of C. b.
longimaxilla examined measures 8.9 mm SL, and has the following charac-
ters: maxillary length 11% SL, 11 mandibular teeth, 4 pores in mandibular
series, and no developed lip flaps on lower jaw. The 5 to 7 preopercular
spines present in larval specimens of C. b. bosquianus are represented by 5
serrations in this specimen.
Pigmentation of this small individual is consistent with that described
previously for adult female Chasmodes. All young of this subspecies, as
well as other members of the genus, have the mottled pigmentation pattern
until they attain a length of 35 to 40 mm SL, at which time the males begin to
show the characteristic adult color pattern (described previously).
STATISTICAL COMPARISONS.-Statistical comparisons of populations of
Chasmodes bosquianus longimaxilla with populations of C. saburrae and
C. b. bosquianus are discussed in the generic and the other subspecies
accounts, respectively. A statistically significant difference in comparison
of regression equations (Table 5) for maxillary length and number of
mandibular teeth between populations of C. b. longimaxilla was found in
only one comparison (Tables 6, 7). A significant difference in maxillary
BULLETIN FLORIDA STATE MUSEUM
Table 9-Matrix of covariance comparisons of maxillary length regression equations (Table
2) for populations of Chasmodes saburrae (P<0.001). Comparisons above the
diagonal are between males in the different populations, and those below the
diagonal between females (see Table 3 for explanation of abbreviations).
0 = coincident regression lines; = noncoincident but parallel lines; + = noncoin-
cident and nonparallel lines.
SS SN ST SC SE SM SF
SS 0 0 0 0 0 0
SN 0 0 0 0 0 -
ST 0 0 0 0 0 0
SC 0 0 0 0 0 0
SE 0 0 0 0 O 0
SM 0 0 0 0 0 0
SF 0 0 0 0 0 0
length comparisons is shown (Table 7) between the population in the
southern Laguna Madre Bay (LL) and the population in the area of sympa-
try (LS). The lack of significant difference between any of the adjacent
populations, coupled with the difference between LL and LS, suggests the
existence of an east-west dine within this subspecies (see Table 8). This
type of variation in maxillary length is not found in C. b. bosquianus or C.
saburrae (Tables 7, 9) and may be related to character displacement in the
area of sympatry (discussed later).
ECOLOGY AND DISTRIBUTION.-Like Chasmodes bosquianus bosquianus,
C. b. longimaxilla typically inhabits the estuarine areas that have either a
rock-shell bottom or a grassbed habitat. Unlike C. b. bosquianus, this
subspecies has not been collected in salinities less than 10 ppt. It is usually
captured in salinities ranging from 18 to 28 ppt.
Based on my observations, together with information from preserved
specimens, the spawning season of Chasmodes bosquianus longimaxilla
begins in March and extends over the summer months into October or early
November. The longer spawning season may be attributable to water
temperature. Water temperature in the northern Gulf of Mexico often
remains warm until October or November (Bault 1972).
Male Chasmodes bosquianus longimaxilla over 35 mm SL are highly
aggressive and maintain territories year-round. Females seek shelter
among rocks and shells, but normally do not actively defend a territory.
Females are constantly on the move during the spawning season. At their
reproductive peak in early summer, I have noted one female spawning
with more than one male. On several occasions, I have seen different
females spawning with the same male. Each female laid a number of eggs
on the inner surfaces of the shell being defended by the male, where he
would then fertilize them. Thus, neither male nor female shows a prefer-
ence for one partner during the spawning season.
The fish's pigmentation pattern renders it practically invisible against a
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
background of oyster shell or aquatic grasses. The greenish color of adult
males blends with the algae covering the oyster shells and aquatic grasses.
This cryptic coloration probably helps to protect the male from predation,
and thus indirectly protects the eggs being guarded during the spawning
season. The dark and light mottled color pattern of the females and
younger fishes is also cryptic and aids in concealment.
Chasmodes bosquianus longimaxilla is known from Pensacola, Florida,
westward to the Texas coast and southward to Veracruz, Mexico. Very few
specimens of C. b. longimaxilla are known from Pensacola, and I believe
these are stragglers to the area. Large aggregations of C. b. longimaxilla are
encountered around oyster shell laden regions in Mobile Bay, Alabama,
where very few C. saburrae occur. The zone of sympathy of C. b. longi-
maxilla and C. saburrae extends from Pensacola, Florida, where C. sabur-
rae predominates, to the Chandeleur Islands, Louisiana. Individuals found
in this zone exhibit no overlap in diagnostic characters and show almost
exclusive segregation by habitat. Based on my collections in Mobile Bay
and ecological data on preserved specimens, C. saburrae appears to be
restricted to the grassbeds, whereas C. b. longimaxilla is associated with the
oyster reef community. This habitat specificity does not seem to hold
outside the zone of sympatry, allopatric populations of both species being
found in either of the two communities.
The habitat specificity in the zone of sympathy is coupled with charac-
ter displacement in Chasmodes bosquianus longimaxilla. The populations
of C. b. longimaxilla in the zone of sympatry, although exhibiting a wide
range of maxillary lengths (14 to 22.5% SL), tend to have longer maxillary
lengths than other populations of this subspecies (Table 8). This difference
may be due to habitat segregation in the zone of sympatry. There is a
tendency for the maxillary length to decrease with increasing geographic
distance from the zone of sympathy (Table 8). This decrease is not statisti-
cally significant until the populations at either end of the geographic range
(excluding Veracruz) are compared (Table 7). This type of variation
provides support for the existence of an east-west dine in the northern Gulf
of Mexico. One possible explanation for this dine is that it is the result of the
effect of character displacement on the gene pool in the zone of sympatry,
together with a decreasing impact of this gene pool on increasingly distant
populations. Biochemical studies of these populations are needed to either
confirm or refute this hypothesis.
Bath's (1977) report of two specimens of Chasmodes bosquianus (= C.
b. longimaxilla) from Veracruz, Mexico, constitutes the only record from
waters south of Laguna Madre, Texas. It is likely that additional specimens
will be collected from the northeastern coast of Mexico when these areas
are more intensively sampled. Although not reported in the few ichthy-
ofaunal studies done in this region (e.g. Resendez 1970, 1973), it should be
1983
BULLETIN FLORIDA STATE MUSEUM
noted that collecting was done mostly with trawls. Since C. b. longimaxilla
is rarely captured in trawls, its absence is not surprising.
A specimen (FMNH 80505) supposedly collected at Key Largo, Flor-
ida, on 10 April 1968, and identified as Chasmodes saburrae, proves upon
re-examination to be C. bosquianus (probably C. b. longimaxilla). As this
record, if valid, would be the only instance of occurrence of C. bosquianus
within the exclusive range of C. saburrae, I assume this to have resulted
from a transposition of locality data.
REMARKS.-Although Chasmodes bosquianus longimaxilla may in fact
be specifically distinct from C. b. bosquianus, I do not feel that specific
recognition can be justified based on the information presented herein.
Based on its allopatric distribution, statistical differences in maxillary
length comparisons, modally different gill raker counts, and differences in
pharyngeal tooth counts, I believe that the Gulf of Mexico population of C.
bosquianus warrants subspecific recognition.
ETYMOLOGY.-The Latin name, a combination of longus (long) and
maxilla (jawbone), refers to the long maxillary bones. The name is to be
treated as a noun in apposition.
MATERIAL EXAMINED.-HOLOTYPE: USNM 219830, male, 86 mm,
30015'30"N, 88005'W, rock jetties on northeast side of Dauphin Island,
Mobile County, Alabama; 20 October 1977; collected by Douglas Clarke.
PARATYPES: FLORIDA: FMNH 57043 (1), Pensacola. ALABAMA: USNM
219831 (1), female taken with holotype; UF 31478 (1; ex USAIC 3980),
31479 (1; ex USAIC 4267), USNM 5721 (1), USAIC 3160 (2), 3164 (1), 4059
(1), 4268 (1), 4269 (2), 6003 (2), 6004 (1), 6274 (1), 6275 (1), 6276 (1), Mobile
Bay; USAIC 3163 (1), Mississippi Sound. MISSISSIPPI: GCRL V61:514 (1),
V67:2061 (1), V68:2903 (1) Mississippi Sound; GCRL V76:15103 (1), Davis
Bayou; FMNH 62465 (1), Ocean Springs; USNM 72333 (1), Three Mile
Bayou. LOUISIANA: GCRL V68:2711 (1), TU 91945 (1), Lake Pontchar-
train; TU 5244 (1),5564 (1), Oyster Bayou; USNM 162591 (1), off Schriever
and Houma. TEXAS: TU 21596 (1), Galveston Bay; TAMU 690.1 (1), near
Matagorda; USNM 69338 (3), Matagorda Bay; USNM 217325 (1), Port
Lavaca; UMMZ 114470 (5), 114474 (1), 145880 (1), 145892 (3), 145976 (3),
170295 (3), Lavaca Bay; UMMZ 145885 (7), 145934 (3), Tres Palacios Bay;
UMMZ 112925 (1), Kellen's Bay; TAMU 919.7 (1), Mustang Island; UTMSI
289 (1), Ransom Island; UMMZ 111743 (3), TU 21665 (2) near Rockport;
ANSP 129692 (1), near Ingleside; FMNH 62464 (9), TAMU 187.3 (1), 700.1
(1), Aransas Bay; FMNH 40311 (2), USNM 156556 (5), 156557 (1), 156611
(1), 188252 (1), near Corpus Christi; ANSP 71181-2 (2), NLU 34793 (1),
Laguna Madre; TAMU 468.3 (1), 784.1 (5), 898.2 (19), 929.6 (16), UF 4233
(1), UTMSI 1358 (1), Redfish Bay; NLU uncatalogued (5), Texas (exact
locality unknown).
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
Chasmodes saburrae JORDAN AND GILBERT
Chasmodes saburrae Jordan and Gilbert 1883: 298 (Type locality: Pensa-
cola, Florida; lectotype USNM 30824).
Blennius fabbri Nichols 1910:161 (Type locality: Miami, Florida; holotype
AMNH 2537).
DIAGNOSIs.-Chasmodes saburrae is distinguished from C. bosquianus
in the diagnosis of the latter.
DESCRIPTION. (See also Table 1).-Characters appearing in generic and
Chasmodes bosquianus descriptions are not repeated here. In addition,
enlarged upper pharyngeal teeth 5-7 (each side); small upper pharyngeal
teeth total 1-10 (Table 1); lower pharyngeals each with 5 to 6 small teeth
immediately anterior to 6 enlarged canines (n = 8); number of mandibular
teeth increases with increasing size (Fig. 4); maximum length about 80 mm
SL.
Peters (1981) discussed the developmental osteology and reproductive
biology of Chasmodes saburrae.
STATISTICAL COMPARISONS.-Secondary sexual dimorphism and com-
parisons of Chasmodes saburrae populations with C. bosquianus popula-
tions are discussed in the generic and C. bosquianus accounts, respectively.
Two populations of C. saburrae (Everglades [SE] and Indian River area
[SF] ) were found to show statistically significant differences when com-
pared with other populations of C. saburrae. Results of comparisons of
regression equations for the number of mandibular teeth (Table 6) show
that the SE and SF populations differ from other populations, but not from
themselves. Both the SE and SF populations differ from the geographically
intermediate Biscayne Bay area population (SM), but this may be due to
the relatively small sample size for SM (n = 15). Springer (1959) suggested
that the relatively high number of mandibular teeth found in some speci-
mens from Whitewater Bay, Everglades, may be related to low salinity (5.5
ppt), but this explanation does not hold for the Indian River area popula-
tion (specimens collected in 15-25 ppt).
Analysis of the maxillary length character between populations of C.
saburrae (Table 9) revealed no significant differences between regression
equations for females. Comparisons between the males in these different
populations show a statistical difference in maxillary length (Table 9)
between the Indian River area population (SF) and the populations in the
Everglades (SE) and in the waters between Ft. Walton and Cedar Key
(SN). It is interesting to note that the SF and the SE populations differ in
maxillary length, but not in number of mandibular teeth (Tables 6,9). I can
offer no explanation for these differences.
ECOLOGY AND DISTRBUTION.-Individuals of Chasmodes saburrae are
sometimes found associated with oyster reef communities, but they appear
to prefer grassbed habitats (Springer and Woodburn 1960). Like their
BULLETIN FLORIDA STATE MUSEUM
congeners, they are inclined toward places where small enclosures abound.
A population of Chasmodes saburrae in the Everglades inhabits waters
that become less saline (down to 5.5 ppt) than those inhabited by the other
populations. The combination of this and other environmental factors yet
to be determined has apparently resulted in a slight increase in number of
mandibular teeth in this population (Table 6). This slightly higher number
of mandibular teeth (Fig. 4) than most other populations (Table 6) is the
only character that shows measurable differences (statistical or otherwise).
In addition, I have not examined specimens from the area between the
southern tip of peninsular Florida and Charlotte Harbor (Fig. 5). This
suggests the possibility that the Everglades population is isolated (i.e. no
gene flow) from the Charlotte Harbor population. Although possible, the
existence of estuarine habitat, suitable for Chasmodes, between these areas
argues against this type of isolation. I believe that this disjunction is a
collecting artifact, since most of this area is accessible only by boat and has
apparently not been well sampled.
This difference of the Everglades population from the others is not
unique to the genus Chasmodes. Weinstein and Yerger (1976) commented
on the disparity between the Everglades "subpopulation" of Cynoscion
nebulosus and other subpopulationss" of that species. In an attempt to
explain this difference, they suggested that it results from environmental
factors unique to that area.
Relyea (1967) presented data on an Everglades population of Fundulus
confluentus that clearly show a trend toward an increase in number of
elements in numerous meristic characters, although Relyea made no com-
ment on this disparity.
Life history and reproductive strategies of Chasmodes saburrae are
discussed by Peters (1981), and thus are not treated in this work.
Chasmodes saburrae occurs from Edgewater, Volusia County, Florida,
southward around the southern tip of the state, northward and westward as
far as the Chandeleur Islands, Louisiana. I have examined three young
specimens (8.9 to 9.8 mm SL, collected on 27 May 1935, USNM 188254)
from Matanzas River, Florida. Based on their small size and the fact that no
other C. saburrae have been collected this far north on the east coast of
Florida, I believe that these three specimens were wafted northward
during their planktonic larval stage. Kevin Peters (pers. comm.) has
informed me that C. saburrae larvae settled out of the plankton in three
weeks under ideal conditions. This length of time would be sufficient to
allow the larvae to drift from more southerly waters to Matanzas River,
assuming a drift rate of approximately 3 to 4 miles a day. The absence of
sexually mature C. saburrae in the Matanzas River indicates that this
species does not normally occur this far north, and this region is therefore
not considered to comprise a normal part of the range of the species. In
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
addition to these, there are three specimens of C. saburrae (USNM 4307,
4308,4309) from the "Coast of South Carolina," collected in the late 1800's.
Dr. Victor G. Springer has informed me that it is likely these locality data
are in error, a conclusion with which I agree.
Between Pensacola, Florida, and Chandeleur Islands, Louisiana, Chas-
modes saburrae is sympatric with C. b. longimaxilla. Within this region, the
two species are allotopic (as defined by Rivas 1964), with C. saburrae
almost exclusively inhabiting the grassbeds and C. b. longimaxilla res-
tricted to the oyster shell areas, thus accounting for the genetic integrity of
the two species over the area of sympatry.
The distinctness of the sympatric Gulf of Mexico populations can be
contrasted with the overlapping characters of the allopatric populations on
the Atlantic coast of Florida. This may be seen in Table 8, which shows a
relatively high degree of overlap in maxillary length between Chasmodes
saburrae and C. bosquianus bosquianus, and no overlap between sympat-
ric C. saburrae and C. b. longimaxilla. The phenomenon of character
displacement (see C. b. longimaxilla account) appears to be contributing to
the divergence of the disjunct populations of C. bosquianus. It is interesting
to note that maxillary length is one of the few morphological characters
showing a significant difference between the disjunct populations (see
Tables 4, 7).
DIscussION OF SYNONYMY.-Dr. C. Lavett Smith (American Museum of
Natural History) has been unable to locate the holotype of Blennius fabbri,
but I concur with Springer (1959) in assuming that this specimen, because
of its type locality, must be a young specimen of Chasmodes saburrae and
not C. bosquianus, as suggested by Nichols (1911).
REMARKS.-Bath (1977:181) designated a specimen from the British
Museum (BMNH 1887.5.14.141) as a paratype of Chasmodes saburrae.
This was apparently the first mention of this specimen as a type. Bath
(1977) does not explain his reason for this action. I have examined this
specimen and, although it belongs to C. saburrae, found nothing in the jar
with the specimen to indicate that it was designated as a paratype by
Jordan or Gilbert. Mr. Alwyne Wheeler, at the British Museum, informed
me that he could find nothing in their records to indicate that this specimen
was ever designated as a paratype. It is possible that this specimen was one
of the 14 mentioned in the original description (Jordan and Gilbert 1883:
299), but I do not feel that there is any basis for its being designated as a
paratype.
Another problem exists with regard to the number of paratypes. Sprin-
ger (1959) pointed out that there are 9 type specimens of Chasmodes
saburrae catalogued in the National Museum of Natural History, and 12
paratypes (now at the California Academy of Sciences) listed by Bohkle
(1953). I have examined the latter series and confirm their identification as
BULLETIN FLORIDA STATE MUSEUM
C. saburrae. These specimens are clearly labeled as paratypes. The fact
that 21 type specimens are known to exist, 7 more than the 14 mentioned by
Jordan and Gilbert (1883), suggests to me that more than 14 specimens
were actually used by Jordan and Gilbert in compiling the original descrip-
tion. Those specimens that are clearly labeled as types (i.e. USNM 1779448
[8], and CAS-SU 440 [12] ) are recognized as paratypes in this work.
MATERIAL EXAMINED.-TYPES: USNM 30824 (1), Pensacola, Florida,
lectotype of Chasmodes saburrae (designated by Springer 1959); USNM
1779448 (8), CAS-SU 440 (12), Pensacola, Florida, paratypes of C. sabur-
rae. FLORIDA: IRCZ 107:1739 (1) and UMML 5385 (1), Edgewater;
USNM 1908 (1), Mosquito Lagoon; UMML 457 (1), vicinity of Titusville;
UMML 958 (2), 3245 (5), FSU 19159 (1), 19164 (2), vicinity of Cocoa; IRCZ
107:2409 (1), 107:4338 (5), 107:4809 (2), FSU 4312 (4), 19174 (1), UF 23685
(1), 23686 (5), 23687 (2), 23688 (21), 23689 (1), 24052 (1), UMML 1555 (3),
3875 (5), USNM 48035 (17), 48783 (11), Indian River; UMML 3132 (3),
Jewfish Creek; UMML 40 (1), 44 (2),50 (1), 776 (1), 2728 (1), 4467 (1), 5391
(1), 8183 (1), Biscayne Bay; UMML 5378 (1), Key Largo; UMML24278 (2),
Shark River Delta; UMML 3953 (2), 4171 (17), 4541 (11), 5211 (6), 5478 (1),
6872 (2), 7059 (1), 7636 (3), 10739 (1), 15496 (1), 17360 (2), Everglades;
FMNH 50554 (4), 50555 (2), FSU 4108 (10), 15583 (1), 20644 (6), UMMZ
113268 (1), USNM uncatalogued (1), GCRL V60:144 (1), V71:7126 (1),
V71:7410 (1), V72:8362 (1), V72:8363 (2), V73:10067 (1), V74:13333 (2),
V74:13334 (1), V74:13356 (1), Charlotte Harbor; FSBC 154 (13), 183 (3), 436
(1), 931 (5), 932 (3), 933 (4), 943 (5), 944 (14), 953 (16), 1421 (14), 3177 (2),
3194 (1), 3204 (7), 3260 (37), 3315 (1), 3859 (7), 6186 (13), 7769 (1), USNM
184240 (50), 184242 (12), Tampa Bay; UF 2576 (1), Bayport; FSBC 3843 (1),
7776 (1), 7877 (1), 7915 (1), 9239 (3), 9558 (1), 9657 (2), Crystal River;
FMNH 62461 (3), 62462 (2), FSBC 3 (2), 4 (3), 11 (13), UF 1084 (1), 1889 (2),
2577 (6), 2578 (8), 2579 (11), 2580 (1), 2581 (23), 2582 (20), 3342 (7), 7827 (2),
8634 (1), 8657 (1), 24015 (1), 24016 (1), 24017 (1), 24019 (7), 24020 (2), 24022
(1), 24023 (1), 24024 (3), 24026 (1), 24027 (2), 24028 (6), 24029 (4), 24030 (4),
24031 (8), 24032 (4), 24033 (4), 24034 (7), 24035 (2), 24036 (1), 24037 (3),
24038 (1), 24039 (1), 24040 (5), 24049 (1), 24050 (1), 24051 (2), USNM 39371
(1), Cedar Key; FSU 2467 (1), 2640 (1), 2642 (1), 4179 (1), 14640 (6), 19816
(2), 25175 (3), UTMSI 290 (1), Apalachee Bay; FSU 388 (1), 765 (1), 1672
(6), 1921 (3), 3215 (1), 4693 (1), 7791 (2), 14057 (1), 18271 (26), 18420 (1),
18657 (9), 23509 (3), 25127 (1), Alligator Habor; USNM 101078 (1), Apala-
chicola; UF 4174 (1), St. Joseph's Bay; FSU 15642 (2), 17247 (3), 17258 (1),
17285 (2), 17346 (3), 17351 (1), 17403 (2), 17468 (2), St. Andrews Bay; FSU
1529 (1), 1536 (1), 14019 (2), UNC 4132 (1), Ft. Walton; BMNH 1887.5.14.141
(1), FSU 14167 (16), 14651 (9), USAIC 6135 (1), USNM 31934 (17), 44871
(1), Penscola. ALABAMA: USAIC 3161 (1), Mobile Bay. MISSISSIPPI:
FMNH 46657 (1), GCRL V61:505 (52), USNM 188248 (18), Mississippi
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
Sound. LOUISIANA: TU 12832 (1), 67392 (17), 67648 (2), 68315 (1), 69569
(1), 76615 (1), 76601 (10), 77611 (1), 77736 (2), 77751 (1), 77880 (5), 77918
(1), 77939 (5), 77995 (1), 78008 (1), 78298 (3).
ZOOGEOGRAPHY
Springer (1959) presented a hypothetical scenario of the evolutionary
history of the genus Chasmodes. This hypothesis suggests that a single
Pliocene species of Chasmodes, inhabiting the present range of the genus,
was divided into continental and insular populations by rising sea level in
the Pleistocene. The geographic isolation of the insular population eventu-
ally led to genetic divergence from the continental population. By the end
of the Pleistocene, the insular population presumably had reached the
specific level. The islands were then reunited with the mainland as the sea
level receded to its present height, resulting in the disjunction of the
mainland species, C. bosquianus. After the islands were reunited with the
mainland, C. saburrae has excluded C. bosquianus from the peninsula.
Springer (1959) further suggested that the disjunct populations of C. bos-
quianus have been diverging from each other since the emergence of the
Florida peninsula.
This scenario has been used to describe the evolutionary history of a
broad range of terrestrial and aquatic organisms in the southeastern U.S.
(Ginsburg 1952, Relyea 1965, Zug 1968, Shipp and Yerger 1969, Shipp 1974,
and Burgess and Franz 1978). In view of the number of different forms
exhibiting this distributional pattern this pattern can be considered a gen-
eralized track sensu Croizat et al. (1974).
Only two aspects of this hypothetical scenario have been seriously
criticized, the time frame for the events and the nature of the highest
terrace (i.e. whether or not it is due to high sea level). Deevey (1950)
questioned the marine origin of the high terraces of Cooke (1939) and
suggested that they may be constructional rather than erosional. Zug (1968)
chose to recognize five Pleistocene marine terraces, including the high
terraces. Robertson (1976) confirmed the marine nature of the "90-100
foot" terrace of Alt and Brooks (1965), but suggested a late Pliocene age.
This leads to the controversy over the timing of the events. Cooke (1935,
1939, 1954) and MacNeil (1950) suggest a Pleistocene age while other
geologists prefer a Pliocene or even Miocene age for the terraces higher
than the 10 m terrace (Alt and Brooks 1965; Osmond et al. 1965, 1970;
Bender et al. 1979). Bender et al. (1979, and references therein) provide
data that argue strongly against sea level rising higher than 10 m above the
present level in approximately the last 500,000 years before present (BP),
and probably not in the last one to two million years BP.
Although the timing of the hypothesized events has little effect on the
mechanism, proposed by Springer (1959), for the evolution of Chasmodes
BULLETIN FLORIDA STATE MUSEUM
saburrae, it is very important to the understanding of the relationship
between the disjunct populations of C. bosquianus. The slight morphologi-
cal and meristic divergence of the disjunct populations (discussed pre-
viously) suggests either a relatively short period of isolation, parallel evolu-
tion, or a slow evolutionary rate. Although each of these mechanisms is
possible, I believe the simplest explanation is that the two populations have
only been isolated since the last glaciation, about 14,000 years BP (Brookes
1977).
The two populations of Chasmodes bosquianus could have come in
contact with each other and interbred in the southern part of Florida by
dispersing southward in response to decreased water temperature. This
dispersal in response to decreased water temperature hypothesis was pro-
posed by Walters and Robins (1961) and has been used by Smith-Vaniz
(1980) to explain the lack of divergence between populations of Hypso-
blennius. As water temperature rose following the end of the last glacial
period, the C. bosquinaus populations would disperse northward until
reaching their present geographic positions.
EXPERIMENTAL HYBRIDIZATION
A hybridization experiment was performed using specimens of Chas-
modes saburrae and C. bosquianus longimaxilla collected in Mobile Bay.
Hybridization was observed between a male C. saburrae and a female C.
b. longimaxilla in a 76-liter saltwater aquarium supplied with empty oyster
and gastropod shells. In addition to the two specimens mentioned above,
there was also one adult male C. b. longimaxilla in the aquarium. The three
specimens were introduced into the aquarium simultaneously. Once intro-
duced into the aquarium, the male C. b. longimaxilla established a territory
in one of the empty oyster shells while the C. saburrae male chose a
gastropod shell. This is interesting since the choices correspond to the types
of shells found in their respective habitats. The female spawned indiscrim-
inately with both males, although she also chose the oyster shells for
protection. The eggs were tended by the males until hatching. The newly
hatched conspecific and hybrid larvae were active, but due to a lack of
suitable food or some other factor, they did not survive more than a few
days.
Due to the extreme artificiality of the enclosure in which the interspe-
cific spawning occurred, I do not feel that these results can justifiably be
used to draw taxonomic conclusions regarding the validity of the two
species, but it does at least prove that natural production of hybrids is
possible. Hybridization in marine fish species is a comparatively rare
phenomenon.
In the areas where these species are sympatric, they appear to be
allotopically distributed. This segregation was indicated in the choice of
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
territories by the two males after introduction into the aquarium. In my
experiment I did not have the facilities to simulate their natural habitat
properly. Forcing them together in the confinement of a small aquarium
could have altered any natural isolating mechanisms that might effectuate
genetic isolation in their natural environment.
A more extensive analysis of this problem using large saltwater pools
with both grassbed and oyster reef habitats should provide additional
insight toward an understanding of the natural isolating mechanisms main-
taining the genetic integrity of these species.
SUMMARY
Two valid species are included in the genus Chasmodes and a new
subspecies is described from the northern Gulf of Mexico. C. saburrae is
found from Edgewater, Florida, southward around the tip of Florida,
northward and westward to the Chandeleur Islands, Louisiana. The second
species, C. bosquianus, consists of two allopatric subspecies, C. b. bosqui-
anus inhabits the Atlantic seaboard from Maryland to Marineland, Florida,
and C. b. longimaxilla subsp. nov. is found from Pensacola, Florida, to
Veracruz, Mexico. The new subspecies is differentiated from the Atlantic
form by its decreased number of small upper pharyngeal canines in adults,
a longer maxillary length, and a modally lower number of gill rakers.
Character displacement in the area of sympatry in the Gulf appears to
be contributing to the divergence of the two populations of Chasmodes
bosquianus.
Zoogeography of the genus Chasmodes is discussed and Springer's
(1959) hypothesis is summarized. This hypothesis provides the simplest
explanation for the evolution of C. saburrae, suggesting that it evolved as
an island endemic. Geological views on the timing of the submergence and
emergence of peninsular Florida are discussed in relation to the disjunct
populations of C. bosquianus. The simplest explanation is that the disjunc-
tion originated with the emergence of peninsular Florida, but gene flow
between the disjunct populations has occurred as recently as the last glacial
period by means of southward dispersal due to lowered water tempera-
tures (Walters and Robins 1961).
Results from hybridization experiments were inconclusive with regard
to determination of the taxonomic status of the species currently recog-
nized in the genus Chasmodes. Additional hybridization experiments
combined with an experimental analysis of habitat partitioning should
provide further insight into the relationship between the sympatric
populations.
LITERATURE CITED
Alt, D., and H. K. Brooks. 1965. Age of the Florida marine terraces. J. Geol. 73(2):406-411.
BULLETIN FLORIDA STATE MUSEUM
Bath, Hans. 1977. Revision der Blenniini (Pisces: Blenniidae). Senckenbergiana biol.
57(4/6):167-234.
Baughman, J. L. 1950. Random notes on Texas fishes. Part II. Texas J. Sci. 2:242-263.
Bault, Edward I. 1972. Hydrology of Alabama estuarine areas cooperative Gulf of Mexico
estuarine inventory. Alabama Mar. Res. Bull. 7:1-36.
Bender, M. L., R. G. Fairbanks, F. W. Taylor, R. K. Matthews, J. G. Goddard, and W. S.
Broecker. 1979. Uranium-series dating of Pleistocene reef tracts of Barbados, West
Indies. Geol. Soc. Amer. Bull. 90:577-594.
Bohlke, J. E. 1953. A catalogue of the type specimens of recent fishes in the Natural History
Museum of Stanford University. Stanford Ichth. Bull. 5:1-168.
Brookes, I. A. 1977. Radiocarbon age of Robinson's Head Moraine, west Newfoundland, and
its significance for postglacial sea level changes. Canadian. J. Earth Sci. 14(9):2121-2126.
Burgess, G. H., and R. Franz. 1978. Zoogeography of the St. Johns River system with
comments on adjacent peninsular faunas. Amer. Midl. Nat. 100(1):160-170.
Cooke, C. W. 1935. Tentative ages of Pleistocene shorelines. J. Washington Acad. Sci.
25:331-333.
1939. Scenery of Florida interpreted by a geologist. State of Florida Dept.
Conserv., Geol. Bull. 17:1-118.
1945. Geology of Florida. Florida Geol. Surv., Geol. Bull. 29:1-339.
Croizat, L., G. Nelson, and D. E. Rosen. 1974. Centers of origin and related concepts. Syst.
Zool. 23(2):265-287.
Cuvier, G., and A. Valenciennes. 1836. Des Chasmode,,. Pp. 295-300 in Histoire Naturelle des
Poissons. Volume II, Paris.
Dawson, C. E. 1964. A revision of the western Atlantic flatfish genus Gymnachirus (the naked
soles). Copeia 1964(4):646-665.
Deevey, E. S., Jr. 1950. Hydroids from Louisiana and Texas, with remarks on the Pleistocene
biogeography of the western Gulf of Mexico. Ecology 31(3):334-367.
Ginsburg, I. 1952. Eight new fishes from the Gulf coast of the United States, with two new
genera and notes on geographic distribution. J. Washington Acad. Sci. 42(3):84-101.
Hardy, J. D., Jr. 1980. Fundulus pulvereus (Evermann), Bayou killifish. P. 525 in D. S. Lee et
al. (eds.) Atlas of North American Freshwater Fishes, North Carolina State Mus. Nat.
Hist., Raleigh, 854 p.
Hildebrand, S. F., and L. E. Cable. 1938. Further notes on the development and life history of
some teleosts at Beaufort, North Carolina Bull. U.S. Bur. Fish. 48:505-642.
Hoese, H. D. 1958. A partially annotated checklist of the marine fishes of Texas. Institute Mar.
Sci., Univ. Texas 5:312-352.
Hubbs, C. L., and K. F. Lagler. 1958. Fishes of the Great Lakes Region (revised edition). Bull.
Cranbrook Inst. Sci. 26:1-186.
Jordan, D. S., and C. H. Gilbert. 1883. Notes on fishes observed about Pensacola, Florida, and
Galveston, Texas, with description of new species. Proc. U.S. Natl. Mus. 5:241-307.
Lac6epde, B. G. E. 1800. Le Blennie Bosquien. Pp. 493-495 inHistoire Naturelle des Poissons.
Volume 2, Paris.
MacNeil, F. S. 1950. Pleistocene shore lines in Florida and Georgia. Geol. Surv. Prof. Pap.
221-F:95-106.
Nichols, J. T. 1910. On two new blennys from Florida. Bull. Amer. Mus. Nat. Hist. 28:161.
1911. Notes on teleostean fishes from the eastern United States. Bull. Amer.
Mus. Nat. Hist. 30:275-278.
Norman, J. R. 1943. Notes on the blennoid fishes. I. A provisional synopsis of the genera of the
family Blenniidae. Ann. & Mag. Nat. Hist., Ser. 11, 10(72):793-812.
Ogren, H. L., and H. A. Brusher. 1977. The distribution and abundance of fishes caught with a
trawl in the St. Andrew Bay System, Florida. Northeast Gulf Science 1(2):83-105.
Osmond, J. K., J. R. Carpenter, and H. L. Windon. 1965. Th230/U234 age of Pleistocene corals
VOL. 29, NO. 2
WILLIAMS: CHASMODES TAXONOMY
and oolites of Florida. J. Geophys. Res. 70:1843-1847.
Osmond, J. K., J. P. May, and W. F. Tanner. 1970. Age of the Cape Kennedy barrier-and-
lagoon complex. J. Geophys. Res. 75:469-479.
Peters, K. M. 1981. Reproductive biology and developmental osteology of the Florida blenny,
Chasmodes saburrae (Perciformes: Blenniidae). Northeast Gulf Science 4(2):79-98.
Phillips, R. R. 1971a. The relationship between social behavior and the use of space in the
benthic fish Chasmodes bosquianus Lacepede (Teleostei, Blenniidae). I. Ethogram. Z.
Tierpsychol. 29:11-27.
1971b. The relationship between social behavior and the use of space in the
benthic fish Chasmodes bosquianus Lacepede (Teleostei, Blenniidae). II. The effect of
prior residency on social and enclosure behavior. Z. Tierpsychol. 29:389-408.
1974. The relationship between social behavior and the use of space in the
benthic fish Chasmodes bosquianus Lacepede (Teleostei, Blenniidae). III. The interac-
tion between attraction/repulsion and prior social experience. Behaviour 49(3-4):205-225.
1977. The relationship between social behavior and the use of space in the
benthic fish Chasmodes bosquianus Lacepede (Teleostei, Blenniidae). IV. Effects of
topography on habitat selection and shelter choice. Behaviour 60(1/2):1-27.
Reid, G. K., Jr. 1954. An ecological study of the Gulf of Mexico fishes, in the vicinity of Cedar
Keys, Florida. Bull. Mar. Sci. Gulf Caribbean 4(1):1-94.
Relyea, K. G. 1965. Taxonomic studies of the cyprinodont fishes, Fundulus confluentus Goode
and Bean and Fundulus pulvereus (Evermann). M. S. Thesis, Florida State Univ., Talla-
hassee. 73 p.
Resendez M., A. 1970. Estudio de la Laguna de Tamiahua, Veracruz, Mexico. An. Inst. Biol.
Univ. Nat. Auton. Mexico 41(1):79-146.
1973. Estudio de los peces de la Laguna de Alvarado, Veracruz, Mexico.
Rev. Soc. Mexicana Hist. Nat. 34:183-281.
Rivas, L. R. 1964. A reinterpretation of the concepts "sympatric" and "allopatric" with
proposal of the additional terms "syntopic" and "allotopic." Syst. Zool. 13(1):42-43.
Robertson, J. S. 1976. Latest Pliocene mammals from Haile XV A, Alachua County, Florida.
Bull. Florida State Mus., Biol. Sci. 20(3):1-186.
Shipp, R. L. 1974. The pufferfishes (Tetraodontidae) of the Atlantic Ocean. Publ. Gulf Coast
Res. Lab. Mus. 4:5-162.
and R. W. Yerger. 1969. A new puffer fish, Sphoeroides parvus, from the
western Gulf of Mexico, with a key to species of Sphoeroides from the Atlantic and Gulf
coasts of the United States. Proc. Biol. Soc. Washington 82:477-488.
Smith-Vaniz, W. F. 1976. The saber-toothed blennies, tribe Nemophini (Pisces: Blenniidae).
Acad. Nat. Sci. Philadelphia, Monog. 19:vii + 1-196.
1980. Revision of western Atlantic species of the blenniid fish genus Hyp-
soblennius. Proc. Acad. Nat. Sci. Philadelphia 132:285-305.
Sokal, R. R., and F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co., San Francisco. 776 p.
Springer, V. G. 1959. Blenniid fishes of the genus Chasmodes. Texas J. Sci. 11(3):321-334.
1968. Osteology and classification of the fishes of the family Blenniidae.
Bull. U.S. Natl. Mus. 284:1-85.
Springer, V. C., and M. F. Gomon. 1975. Revision of the blenniid fish genus Omobranchus
with descriptions of three new species and notes on other species of the tribe Omobran-
chini. Smithsonian Contr. Zool. 177:1-135.
Springer, V. G., and K. D. Woodburn. 1960. An ecological study of the fishes of the Tampa
Bay area. Florida State Bd. Conserv. Mar. Res. Lab., Prof. Paps. Ser. 1:1-104.
Swainson, W. 1839. The natural history of fishes, amphibians, and reptiles, or monocardian
animals. Longman, Orme, Brown, Green and Longmans, London, vol. 2:vi + 448.
Tabb, D. C., and R. B. Manning. 1961. A checklist of the flora and fauna of northern Florida
Bay and adjacent brackish waters of the Florida mainland. Bull. Mar. Sci. Gulf Caribbean
100 BULLETIN FLORIDA STATE MUSEUM VOL. 29, NO. 2
11(4):552-649.
Tavolga, W. N. 1958. Underwater sounds produced by males of the blenniid fish Chasmodes
bosquianus. Ecology 39:759-760.
Walters, V., and C. R. Robins. 1961. A new toadfish (Batrachoididae) considered to be a
glacial relict in the West Indies. Amer. Mus. Nov. 2047:1-24.
Weinstein, M. P., and R. W. Yerger. 1976. Electrophoretic investigation of subpopulations of
the spotted seatrout, Cynoscion nebulosus (Cuvier), in the Gulf of Mexico and Atlantic
coast of Florida. Comp. Biochem. Physiol. 54B:97-102.
Weymouth, F. W. 1910. Notes on a collection of fishes from Cameron, Louisiana. Proc. U.S.
Natl. Mus. 38(1734):135-145.
Wood, W. W. 1825. Description of four new species of the Linnaean genus Blennius and a new
Exocetus. J. Acad. Nat. Sci. Philadelphia 4:278-284, pl. 17.
Zug, G. R. 1968. Geographic variation in Rhineura floridana (Reptilia: Amphisbaenidae). Bull.
Florida State Mus., Biol. Sci. 12(4):185-212.
Contributions to the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL
SCIENCES SERIES, may be in any field of biology. Manuscripts dealing with natural history
or systematic problems involving the southeastern United States or the New World tropics are
solicited especially. Manuscripts should be of medium length circa 35 to 200 pages
(10,500-16,000 words). Examination for suitability is made by an Editorial Board.
The BULLETIN is distributed worldwide through institutional standing orders and exchanges.
It is considered the responsibility of the author to distribute his paper to all interested
individuals. To aid in this the authors) receives) 50 copies free, and may purchase additional
separates at cost, if ordered when galley proof is returned. The author is also responsible for
any charges incurred for alterations made by him on galley or page proofs. The Museum will
send an invoice to the author for this amount upon completion of publication.
PREPARATION OF MANUSCRIPT
Contributors should consult recent numbers of the BULLETIN for preferred style and
format. Highly recommended as a guide is the CBE Style Manual, 3rd Edition, 1972 (Amer.
Inst. Biol., Sci.). More specific instructions are available upon request.
MSS must be submitted in triplicate (NO ONIONSKIN, XEROX COPIES PLEASE) and
satisfy the following minimal requirements: They must be typewritten, double-spaced
throughout (INCLUDING tables, figure captions, and Literature Cited), with triple-spacing
around all headings, on one side of numbered sheets of standard (8-1/2 x 11 in.) bond paper,
with 1-1/2 in. margin on left and 1 in. margins on other three sides. Each table must be typed
on a separate sheet of paper; figure captions should be typed run-on. All illustrations are
referred to as figures. They must comply with the following standards: Photographs should be
sharp, with good contrast, and printed on glossy paper. Drawings must be made with dense
black waterproof ink on quality paper or illustration board and have a cover sheet. All
lettering must be medium weight, sans-serif type (e.g. Futura Medium, News Gothic) in
cutout, dry transfer, or lettering guide letters. Make allowance so that after reduction no
lowercase letter will be less than 1 mm high (2 mm is preferred) nor any capital letter greater
than 5 mm high. The maximum size for illustrations is 8-5/8 x 14 in. (twice typepage size);
illustrations should not be less than typepage width (4-5/16 in.). Designate the top of each
illustration and identify on the back with soft pencil by author's name, MS title, and figure
number.
All manuscripts not submitted in BULLETIN format will be returned to the author for
retyping.
Manuscripts and all editorial matters should be addressed to:
Managing Editor of the BULLETIN
Florida State Museum
University of Florida
Gainesville, FL 32611
|
