|
BULLETIN
OF THE
FLORIDA STATE MUSEUM
BIOLOGICAL SCIENCES
Volume 5
Number 4
MIDDLE-AMERICAN POECILIID FISHES
OF THE GENUS XIPHOPHORUS
Donn Eric Rosen
UNIVERSITY OF FLORIDA
Gainesville
1960
The 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, Jn., Editor
VWILLIAM J. R'EMER, Managing Editor
All communications concerning purchase or exchange of the publication should
be addressed to the Curator of Biological Sciences, Florida State Museum, Seagle
Building, Gainesville, Florida. Manuscripts should be sent to the Editor of the
BULLETIN, Flint Hall, University of Florida, Gainesville, Florida.
Price for this issue $2.80
Published 14 June 1960
MIDDLE-AMERICAN POECILIID FISHES
OF THE GENUS XIPHOPHORUS
DONN ERIC ROSEN 1
SYNOPSIS. Drawing upon information from the present studies of the com-
parative and functional morphology, distribution, and ecology of the forms of
Xiphophorus (Cyprinodontiformes: Poeciliidae) and those made during the last
quarter of a century on their genetics, cytology, embryology, endocrinology, and
ethology, the species are classified and arranged to indicate their probable phylo-
genetic relationships. Their evolution and zoogeography are considered in rela-
tion to a proposed center of adaptive radiation on Mexico's Atlantic coastal plain.
Five new forms are described: X. variatus ecelynae, new subspecies; X. miller,
new species; X. montezumae cortezi, new subspecies; X. pygmaeus nigrensis, new
subspecies; X. helleri alvarezi, new subspecies.
To the memory of
MYRON GORDON, 1899-1959
for his quarter century of contributions to
the biology of this and other groups of fishes.
1 This paper is a revision of a thesis presented to the Biology Department of
the Graduate School of Arts and Sciences of New York University in partial fulfill-
ment of the Ph.D. degree. The work was carried out while the author was associ-
ated with New York University and the Genetics Laboratory of the New York
Aquarium. It was supported by a National Science Foundation grant to the New
York Zoological Society for Dr. Myron Gordon (project: A Biological Synthesis
of the Poeciliid Fishes). Now at the University of Florida, the author serves the
Department of Biology as Assistant Professor and the Florida State Museum as
Assistant Curator of Natural Sciences. Manuscript submitted 24 July 1959.
BULLETIN FLORIDA STATE MUSEUM
TABLE OF CONTENTS
PAGE
Introduction -------------------------- ---- ------ 59
Acknowledgments ......---------------------- ---------- 61
Materials and methods ---.----------------------- --- 62
The genus Xiphophorus --------------------------- 64
Accounts of the species and subspecies ------------------- 66
Xiphophorus couchianus (Girard) -------------------- --- 66
Xiphophorus maculatus (Guenther) ------------------ --- 69
Xiphophorus variatus (Meek) ----------------------- -- 78
Xiphophorus variatus xiphidium (Gordon) ---------------- 81
Xiphophorus variatus variatus (Meek) -------------------- 84
Xiphophorus variatus evelynae, new subspecies ..------------ ---- 87
Xiphophorus miller, new species .......----------------- ---- 89
Xiphophorus montezumae Jordan and Snyder ..------------ -- 92
Xiphophorus montezumae montezumae Jordan and Snyder 93
Xiphophorus montezumae cortezi, new subspecies .. ---------- 96
Xiphophorus pygmaeus Hubbs and Gordon ---------------- 98
Xiphophorus pygmaeus nigrensis, new subspecies 100
Xiphophorus pygmaeus pygmaeus Hubbs and Gordon 102
Xiphophorus clemenciae Alvarez ........... --------------------- 104
Xiphophorus helleri Heckel ------------------------- 106
Xiphophorus helleri helleri Heckel 116
Xiphophorus helleri strigatus Regan ------------------- 118
Xiphophorus helleri guentheri Jordan and Evermann 122
Xiphophorus helleri alvarezi, new subspecies 126
Distribution and ecology of Xiphophorus 156
Distribution patterns of Xiphophorus in relation to habitat preferences -- 156
Ecological restrictions -------------------- 156
Types of topographical and ecological isolation 162
Adaptive modifications to the environment and patterns of distribution-- 169
Routes of dispersal ..------------------- ---------- 172
Relationships within Xiphophorus 175
Assessment of characters 175
Gonopodium -------------------------------- 175
Pigmentation ------------------- ------------- 180
Metamerism and body form in relation to habitat 182
Comparative behavior ...----...------------------------ 188
Skeletal elements --------------.--------------.-- 189
Viability, development, and morphology of interspecific hybrids as
systematic criteria in Xiphophorus 201
Embryonic development, fertility, and sexuality of interspecific
hybrids in relation to their genetic background 201
Pigment cell abnormalities in hybrids -------------------- 205
Abnormalities of the external genitalia of hybrids 208
Species groups in Xiphophorus ------------------------ 212
The maculatus species group (couchianus, maculatus, variatus) ---- 214
The montezumae species groups millerr, montezumae, pygmaeus)---- 216
The helleri species group (clemenciae, helleri) -------------- 218
Summary and conclusions ----------------------------- 219
Appendix 1: Notes on meristic characters 223
Appendix 2: Key to the species Xiphophorus 226
Literature cited ..... ----------------------------- ----- 229
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
INTRODUCTION
The genus Xiphophorus (Cyprinodontiformes: Poeciliidae), which
includes the platyfishes and swordtails, presents an exceptional op-
portunity for taxonomic research. It is outstanding among fish groups
in that its natural range is comparatively well known and its members
adapt readily to laboratory conditions. During the last quarter cen-
tury it has been the subject of an unusual number of diverse experi-
mental and descriptive studies. Investigations of its genetics, cytol-
ogy, embryology, endocrinology, and ethology have produced a back-
ground of general biological information that gives an added per-
spective to the basic morphologic and zoogeographic criteria of the tax-
onomist. The objective of this revision is a systematic reappraisal of
Xiphophorus that combines data from all these diverse yet comple-
mentary fields. The following species and subspecies are recognized:
Xiphophorus couchianus (Girard) (Northern platyfish)
Xiphophorus maculatus (Guenther) (Southern platyfish)
Xiphophorus variatus (Meek) (Variable platyfish)
Xiphophorus variatus variatus (Meek)
Xiphophorus variatus evelynae, new subspecies
Xiphophorus variatus xiphidium (Gordon)
Xiphophorus miller, new species (Catemaco livebearer)
Xiphophorus montezumae Jordan and Snyder (Montezuma swordtail)
Xiphophorus montezumae montezumae Jordon and Snyder
Xiphophorus montezumae cortezi, new subspecies
Xiphophorus pygmaeus Hubbs and Gordon (Pygmy swordtail)
Xiphophorus pygmaeus pygmaeus Hubbs and Gordon
Xiphophorus pygmaeus nigrensis, new subspecies
Xiphophorus clemenciae Alvarez (Yellow swordtail)
Xiphophorus helleri Heckel (Green swordtail)
Xiphophorus helleri helleri Heckel
Xiphophorus helleri strigatus Regan
Xiphophorus helleri guentheri Jordan and Evermann
Xiphophorus helleri alvarezi, new subspecies
The formal generic separation of the swordtails (Xiphophorus
Heckel, 1848) and the platyfishes (Platypoecilus Guenther, 1866) grew
unchallenged out of early imperfect diagnoses of the species for which
the two genera were first erected. Xiphophorus was applied by Heckel
to three quite unrelated forms, X. helleri, Pseudoxiphophorus bimacu-
latus, and Gambusia gracilis, a species of uncertain status. Weyen-
burgh (1874) described three new forms as Xiphophorus Heckelii, X.
obscurus and X. minor, species since assigned to the Jenynsiidae.
For many years, Platypoecilus was misapplied to various cyprinodonti-
forms, as follows:
BULLETIN FLORIDA STATE MUSEUM
Platypoecilus mentalis Gill (1876: 335) = Mollienesia sphenops Cuvier and
Valenciennes, a poeciliid.
Platypoecilus quitzeoensis Bean (1898: 540, 1 fig.) = Zoogoneticus quitzeoen-
sis (Bean), a goodeid.
Platypoecilus nelsoni Meek (1904: 147, Fig. 46) = Mollienesia sphenops
Cuvier and Valenciennes, a poeciliid.
I[.Ldi..... .,i perugiae Evermann and Clark (1906: 851-52, Fig. 1) = Limia
perugiae (Evermann and Clark), a poeciliid.
Platypoecilus dominicensis Evermann and Clark (1906: 852-53, Fig. 2) =
Mollienesia (Psychropoecilia) dominicensis (Evermann and Clark), a poeciliid.
Platypoecilus tropics Meek (1907: 146-47) = Mollienesia sphenops Cuvier
and Valenciennes, a poeciliid.
The above corrections in identification were indicated by Meek
(1902, 1904), Regan (1907, 1913), and Myers (1935). Not until the
gonopodial characters of the Poeciliidae were thoroughly studied
(Regan, 1913) were the relationships of Platypoecilus demonstrated.
Early knowledge of Xiphophorus came largely from its comparison
with Poecilia and Mollienesia to which some thought it bore a close
relationship. The delimiting criteria for Poecilia eventually served
also, in part, for Xiphophorus. When Platypoecilus was named dur-
ing this early period and separated nomenclaturally from Poecilia and
its allies, it was natural to regard Platypoecilus and Xiphophorus as
distinct.
Summarizing the historical events up to 1911, three distinct groups
of Xiphophorus (as presently constituted) were formally recognized,
largely on the basis of their fin structure and dentition. 1) Both
helleri and montezumnae possess the the caudal appendage or sword
and biserial dentition (Xiphophorus); 2) maculatus and variatus lack a
sword and have but a single series of small teeth (Platypoecilus); 3)
couchianus also lacks the sword but possesses biserial dentition (Po-
ecilia).
Regan (1913) decided that the species of Platypoecilus and Xiph-
ophorus, which previously had been widely separated, were in reality
closely related. He also showed that the alleged dentitional differ-
ences in the two genera are not real, and that all the members of both
genera do, in fact, have biserial dentition, a fact which was presaged
when he (1907) distinguished between an outer row and inner band
of teeth in P. maculatus. He showed that Xiphophorus helleri Heckel
(1848) and Xiphophorus montezumae Jordan and Snyder (1899) were
correctly associated as were Platypoecilus maculatus Guenther (1866)
and Platypoecilus couchianus (Girard, 1859).
Langer (1913), after analyzing the anatomy of the gonopodia and
their suspensoria in Xiphophorus helleri and Platypoecilus maculatus,
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
suggested joining the genera, but did not do so. The similarity of
these genera was appreciated also by Hubbs (1924) who erected the
tribe Xiphophorini to contain them.
When Hubbs and Gordon (1943) described Xiphophorus pygmaeus,
they pointed out that the distinctions between Platypoecilus and
Xiphophorus are uncertain, and assigned pygmaeus to the older genus,
Xiphophorus, in anticipation of possible taxonomic joining of the
platyfishes and swordtails.
Gordon and Rosen (1931) studied the genetics of species differences
in the morphology of the gonopodia of two "structurally divergent"
species, X. helleri and P. maculatus. They pointed out that even the
measurable differences in genital structure between these two species
were not sufficient to prevent hybridization under laboratory condi-
tions. They recommended that, in view of the genetic continuity
within the xiphophorin species, a single generic term, Xiphophorus,
be applied to them all. They suggested, however, that the name Platy-
poecilus be retained as a subgeneric grouping to reflect what they
believed to be a real but subtle difference between the platyfishes and
swordtails, though statistical and taxonomic evaluations showed no
line of demarcation between them.
It is now believed that even subgeneric separation is unwarranted.
ACKNOWLEDGMENTS
It is with great pleasure that I extend my warmest thanks to many
persons whose generosity and interest facilitated this work. I am par-
ticularly indebted to the late Myron Gordon under whose guidance
this work was begun. In addition to making available an unusual
variety of equipment and technical facilities, and an always congenial
and stimulating atmosphere in which to work, Dr. Gordon provided
for my use much unpublished information on the biology of this
group of fishes. Among other things, this included a folder of type-
written material relating to the ecology, distribution, and polymorph-
ism of six of the eight recognized species of Xiphophorus. These
notes have been incorporated wherever possible.
To Reeve M. Bailey I offer special thanks for numerous stimulating
discussions of the present study and for reviewing in detail the manu-
script as a whole during the final stages of the work. For reading and
criticizing various parts of the manuscript I thank James W. Atz, H.
Clark Dalton, Hugh M. Gordon, Robert R. Miller, Bobb Schaeffer, and
Pamela Alexander.
1960
BULLETIN FLORIDA STATE MUSEUM
Robert R. Miller brought to my attention material from two of his
recent expeditions to Middle America that included a new species
of Xiphophorus. I also thank Carl L. Hubbs for making available
detailed locality maps of his and Myron Gordon's early collections
of Xiphophorus in Mexico.
For the loans or gifts of specimens used in this study, I thank Jose
Alvarez del Villar, Reeve M. Bailey, James Boehlke, Richard T. Gregg,
Ernest A. Lachner, Robert R. Miller, George S. Myers, Leonard P.
Schultz, William I. Follett, Loren P. Woods, Myvanwy Dick, and
Margaret Story, and the institutions that they represent: The Academy
of Natural Sciences of Philadelphia, California Academy of Sciences,
Chicago Natural History Museum, Louisiana State University Museum
of Zoology, Museum of Comparative Zoology, Stanford University
Natural History Museum, United States National Museum, and the
University of Michigan Museum of Zoology.
To my wife, Mel, I extend deepest thanks for her editorial aid.
Elwood Logan supplied numerous and excellent photographic
services.
MATERIALS AND METHODS
MATERIAL. Preserved material of Xiphophorus used in this study
is from the University of Michigan Museum of Zoology (UMMZ),
Louisiana State University Museum of Zoology (LSU), United States
National Museum (USNM), Stanford University of Natural History
Museum (SU), Chicago Natural History Museum (CNHM), and the
Academy of Natural Sciences of Philadelphia (ANSP). Many speci-
mens formerly maintained in the collection of the Genetics Laboratory
of the New York Zoological Society have been deposited in the Uni-
versity of Michigan Museum of Zoology; these are herein designated
by the symbol UMMZ-NYZS-GL.
Laboratory born and reared inter- and intra-specific hybrids of
Xiphophorus also were studied (table 34); this is the same material
examined in greater detail by Atz (Ms.) in his analyses of the effects
of hybridization in this genus.
Laboratory reared descendants of wild caught fish also were
analyzed. Living Xiphophorus maintained in the Genetics Laboratory
of the New York Aquarium were derived from nine expeditions to
Mexico and northern Central America, as follows:
Clarence L. Turner and party, 1932; Mexico.
Myron Gordon, John Ross, and Joseph Whetzel, 1932; Mexico.
Myron and Evelyn Gordon, and James W. Atz, 1939; Mexico.
Myron Gordon, James W. Atz, and Forest G. Wood, 1948; Mexico.
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
Myron Gordon and Gerald Fairweather, 1949; British Honduras.
Myron Gordon and Alphonse Chable, 1951; Honduras.
Myron Gordon and Jesus Garcia, 1952; Mexico.
Myron Gordon and party, 1954; British Honduras and Guatemala.
Myron Gordon, Malcolm Gordon, and Donn E. Rosen, 1957; Mexico.
COUNTS AND MEASUREMENTS. The methods of counting and meas-
uring are those described by Miller (1948: 8-14). Dorsal fin ray
counts include small anterior rays; the last ray as counted actually con-
sists of two lepidotrichia that are separate to their bases. Scales in a
lateral series are counted from the upper angle of the gill aperture to
the caudal base at midside; each scale in this row almost invariably
contains a central pit. The vertebral count includes all cervical ele-
ments as well as the urostylar vertebra. Standard length is measured
from snout tip to caudal base. Head length is taken to the opercular
margin. Length of caudal peduncle is measured from the anal origin
to the caudal base at midside. Greatest depth is taken from the origin
of the dorsal to the origin of the anal fin. Least depth is taken
across the caudal peduncle just forward of the caudal base. Caudal
fin length is measured from the caudal base to the distal tip of the
longest bifurcate ray at midfin. Length of caudal appendage of adult
males is taken from the ventral edge of the caudal base to the tip of
the longest modified fin ray. All measurements are recorded to tenths
of millimeters. Indices of body proportion are obtained by arithme-
tically dividing each measurement into standard length.
SKELETAL MATERIAL. Preparation of skeletal material follows the
method outlined by Rosen and Bailey (1959). Terminology of ele-
ments in the gonopodial suspensorium is that employed by Gordon
and Benzer (1945) as modified by Rosen and Gordon (1953). Termin-
ology of cranial elements is that used by Ramaswami (1946).
PREPARATION AND STUDY OF GONOPODIA OF ADULT MALES. Gono-
podia to be used in statistical comparisons of species or populations
were always removed from the fish and mounted on standard micro-
scope slides in the following manner:
1. The fin was excised as close as possible to the body without
damaging scales or adjoining bones of the gonopodial suspensorium.
2. Each fin was immersed successively for 15 minutes in each of
a graded series of alcohols (70%, 95%, 100%, 100%) and then trans-
ferred for 20 to 30 minutes to xylol.
3. When completely cleared in xylol, the fin was transferred to a
drop of Canada balsam on a microscope slide and then covered with
a number 2 cover slip.
1960
BULLETIN FLORIDA STATE MUSEUM
4. The cover slip was pressed down with the blunt end of a pencil,
or other wooden rod, to flatten the gonopodium and spread the balsam.
5. Slides prepared in this way were transferred quickly to the
warming table where the mounting medium hardened in about 2 days.
The terminology used for the bony elements at the distal tip of
the gonopodium of Xiphophorus is that given by Rosen and Gordon
(1953: 8, text-figure 12A).
THE GENUS XIPHOPHORUS
Xiphophorus Heckel, 1848: 291 (original description of X. helleri, designated
type species).
Platypoecilus Guenther, 1866: 350-351 (original description of P. maculatus.)
Figure 1. Radiograph of an adult male Xiphophorus variatus evelynae, n.
ssp. (UMMZ: 124317), 36.2 mm. in standard length, illustrating skeletal plan
in Xiphophorus.
Body moderately deep and compressed, without severely angulated
dorsal and ventral margins, covered with large cycloid scales. Caudal
fin usually with variously developed extensions of the lower rays as a
"sword." Dorsal fin usually rounded, the first two rays simple, other
rays bifurcate one or more times (in adults). In adult males pelvic
fin with fleshy appendage developed along distal third of first short
unbranched ray, the second and third rays somewhat prolonged.
Anal rays 9 or 10, rarely 11. Gonopodium bilaterally symmetrical,
I-beam-like in cross section; ray 5a forming a broad lanceolate plat-
form that is flanked by erect, ridgelike elements of 5p, the whole frame-
work forming a long, shallow, trough-like area; ultimate segment of
5a usually modified as claw or hook; ray 4p terminated by erect distal
and retrorse subdistal serrae, the two series being separated by sev-
eral simple oblong or cuboidal segments; proximal portion of 4p de-
veloped as a compressed ridge of raised, slender segments directly
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
beneath the lanceolate platform of ray 5a; ray 4a simple, except dis-
tally where it is modified as a ramus that arches or hooks ventrally
anteriorlyy); well-developed hook at terminus of ray 3, followed by
several irregularly oval or cuboidal subterminal segments, and a
series of large more or less erect spines with angulated shafts; a dense
blade of tough membranous tissue separating tips of rays 3 and 4a,
its anterior edge merging with hook (ray 3), its posterior edge follow-
ing contour of terminal ramus (ray 4a); posterior margin of ray 3 prox-
imal to spines raised to meet anterior margin of ray 4a, the two rays
together forming a ball and socket joint in cross section; tips of rays
6 and 7 swollen and club-shaped. Gonopodial suspensorium (fig. 1)
with three erect, more or less linear gonapophyses; parapophyses pres-
ent and well-developed on first and second gonapophyses, variously
developed or absent on third; uncini usually absent, minute and ir-
regular when present; ligastyle long and rodlike, almost as long as
first gonapophysis; primary gonactinostal complex moderately dilated
anteroposteriorly, inclined slightly forward, its upper edge deeply
notched. Vertebrae 26 to 31. Pleural ribs approximately 12 to 14
in number; distal tips of ribs 7 to 10, counting anteroposteriorly,
arched slightly downward and forward just above or posterior to
pelvic girdle (fig. 1). Pectoral girdle somewhat triangular in outline,
its longest dimension vertical; four actinosts recessed within posterior
margins of scapula and coracoid, usually discrete but occasionally
showing varying degrees of consolidation; upper part of cleithrum
produced backward above scapula as large spatulate process; posterior
edge of coracoid usually produced backward as flat process similar
in outline to cleithral process but smaller. Skull deep and wedge-
shaped (fig. 1), with well-developed supraoccipital processes and
variously developed epiotic processes; jaws weak, consisting of slender
elements with delicate articulations; premaxillae and dentaries flat-
tened in front, the paired elements not joined at midline and separated
by a distinct tissue space, each with an outer series of movable, com-
pressed or narrow incisorlike teeth in a single, largely transverse row
that is weakly indented near midline, and an inner series of minute
pointed teeth in a band that is broad laterally, becoming narrow
medially.
1960
BULLETIN FLORIDA STATE MUSEUM
ACCOUNTS OF THE SPECIES AND SUBSPECIES
Xiphophorus couchianus (Girard)
Northern Platyfish, figures 2-4, tables 1-6
Figure 2. Diagram of the pigment patterns in a male Xiphophorus couchianus
(Girard). Note three broken rows of black dashes posteriorly. These markings
are deep-lying dermall and subdermal) and are composed entirely of micromelan-
ophores.
Limia couchiana Girard, 1859: 116 (original description; Monterrey).
Gambusia couchiana, Bleeker, 1860: 485 (characters).
Mollienesia couchiana, Jordan and Copeland, 1876: 143 (Nuevo Leon); Jordan,
1878: 434 (Nuevo Leon).
Poecilia couchiana, Jordan and Gilbert, 1883: 348 (description; Nuevo Leon);
Jordan, 1885: 838 (name only); Evermann and Kendall, 1894: 76, 84, 87,
89, 92, 108 (Rio San Juan at Cadareita and Monterrey); Jordan and Ever-
mann, 1896: 695 (description; Cadareita, Monterrey); Meek, 1904: 152 (de-
scription; Monterrey); Regan, 1906-1908: 104-105 (description).
Platypoecilus couchianus, Regan, 1913: 1003-1004 (description; gonopodium;
close relative of P. maculatus); Gordon, 1932a: 287 (characters; distribution);
1932c: 89 (distribution); 1932b: 5 (ecology; hybrids); Breder, 1932: 7-9
(figure); Ralston, 1933: 124-125 (misspelled couchiana; chromosomes); Stoye,
1933: 307 (aquarium introduction); 1934: 86 (characters); Friedman and Gor-
don, 1934: 446-455 (chromosomes; distribution); Gordon, 1935a: 8-9 (char-
acters; distribution); 1935b: 186 (ecology); Stoye, 1935: 53 (description);
"Anonymous," 1937: 46 (figure; record mapped); Gordon and Smith, 1938a:
543-565 (distribution; hybridization; melanosarcoma); De Buen, 1940: 43
(name only); Dobzhansky, 1941: 280 (genetics; species or race); Gordon,
1941a: 39 (color factor E); 1941b: 112-113 (Rio Grande; hybridization with
Platypoecilus maculatus); Mayr, 1942: 171 (member of superspecies or sub-
species); Gordon, 1943c: 65, 68 (ecology, variation); Gordon, Cohen and
Nigrelli, 1943: 569 (caudal fin; Rio San Juan system); Gordon and Lansing,
1943: 231-235 (tumors in hybrids with Platypoecilus maculatus); Gordon and
Benzer, 1945: 57-70 (vertebrae; ribs; gonopodial suspensoria); Gordon,
1946d: 81, 83 (wagtail pattern in hybrids; introgressive hybridization); 1946b:
317-320 (sex-determining mechanism); 1947a: 117, 129 (ecological communi-
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
ties); 1950b: 19 (hybridization with Platypoecilus maculatus; melanomas);
Gordon and Rosen, 1951: 417, 423-424, 427-429, 431-433, 457-458 (genetics
and morphology of gonopodia; Platypoecilus as a subgenus of Xiphophorus);
Gordon, 1951c: 194-197, 216 (genetics of atypical pigment cell growth);
1953c: 789-790 (distribution; taxonomic history).
Xiphophorus couchianus, Rosen and Gordon, 1953: 23 (gonopodial specializa-
tions).
Poecilia couchii, Guenther, 1866: 347 emendationn of couchiana).
Platypoecilus variatus, Rachow, 1936: sheets 145-146 (in part; erroneous synonymic
reference).
MATERIAL EXAMINED
Mexico, Nuevo Leon
Rio Grande system
1. Springs of Huasteca Canyon, 2 mi. W Santa Catarina, Rio San Juan. UMMZ
124163. 8 half-grown to subadult males; these constitute basis of laboratory
stock.
2. Rio Santa Catarina, 1% mi. S Santa Catarina, Rio San Juan. UMMZ 97572.
25 half-grown to adults.
3. Branch of Rio San Juan in Huasteca Canyon, 4 mi. W Monterrey. UMMZ
108609. 20 half-grown to adults.
4. Headwaters of Rio Santa Catarina, mouth of Santa Catarina, Rio San Juan.
UMMZ 97571. 432 half-grown to adults.
5. Rio Santa Catarina in Huasteca Canyon. UMMZ NYZS-GL. 30 young to
adults, 15 taken alive.
DIAGNOSIS. A small, moderately deep-bodied species of Xipho-
phorus (tables 4 and 5). Lower caudal rays not developed as a
"sword" in adult males. With a weakly developed zigzag midlateral
stripe separating the darker heavily reticulated dorsum from the
paler ventral surface; usually with a variable number of deep-lying
black spots in one to three rows on the caudal peduncle along the line
where the myomeres are abruptly angulated. Without a deep-lying
ridge of black pigment midventrally on caudal peduncle. Dorsal fin
distinctly rounded, occasionally with a weakly developed row of spots
or more often a diffuse band of dusky pigment near base. Claw want-
ing at tip of ray 5a of gonopodium; distal serrae of ray 4p well devel-
oped, 4 to 6 in number, their tips converging, the spread of the two
outermost ones from base to base four times the distance from tip to
tip; ramus of ray 4a curved downward over blade, not hooked; hook
of ray 3 long and slender. Vertebrae 28, rarely 27. Scales in a lateral
series 26 or 27, usually 26. Dorsal fin rays 8 to 11, usually 9 or 10.
GENERAL ACCOUNT. Xiphophorus couchianus agrees with X. vari-
atus in having a slender caudal peduncle. From variatus it may
usually be told by the lack of distinctive horizontal zigzag dark
stripes, and in being browner in life and more sharply bicolored-
1960
BULLETIN FLORIDA STATE MUSEUM
F
Figure 3. Distal tips of the gonopodia of males of the species and subspecies
of platyfishes. A. X. couchianus (Girard) (UMMZ: 97571). B-C. X. maculatus
(Gunther) (UMMZ: 124255). D-E. X. variatus xiphidium (Gordon) (UMMZ:
124416). F-H. X. variatus variatus (Meek) (UMMZ:NYZS-GL, Rio Nautla).
I-J. X. variatus evelynae, n. ssp. (UMMZ: 124318).
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
darker above and paler below. Rarely, individuals of variatus may
be as dark and bicolored as couchianus. This northern species differs
from the allied southern species cariatus and maculatus in lacking
the variable black markings, other than vertical bars, that appear in
varying frequency in those species. X. couchianus usually shows a
variable number of deep-lying black spots on the caudal peduncle,
in one to three rows along the line where the myomeres are abruptly
angulated. Specimens collected at Monterrey by Cope show in ad-
dition a similar but more conspicuous row of spots or streaks along
the line where the superior muscles are sharply bent. All three rows
of spots are present in some live specimens of the stock from Huasteca
Canyon. These markings, lying deep in the muscles, probably have
a histological and a genetic basis different from that of the more super-
ficial black spotting (Sp) of the other species.
This species is confined to the Rio Grande basin in Nuevo Leon,
Mexico.
Xiphophorus maculatus (Guenther)
Southern Platyfish, figures 3-5, tables 1-5, 7-8
Platii .... .i/1 maculatus Guenther, 1866: 350-351 (original description; Mexico);
Eigenmann, 1893: 57 (Mexico); Garman, 1895: 48-49 (after Guenther; except
reference to P. mentalis in synonymy); Jordan and Evermann, 1896: 686
(after Guenther); Meek, 1904: 145-146 (records; description; variation); Ger-
lach, 1909: 25-27 (aquarium introduction); Regan, 1913: 1004 (gonopodium;
synonymy; description; range in part); Brind, 1914: 22 (as aquarium fish);
Gerschler, 1914: 73-96 (hybridization); Bellamy, 1922: 419-420 (genetics);
1924: 513-529 (biometry); Gordon, 1927: 253-283 (history; taxonomy; gen-
etics); Bellamy, 1928: 226-232 (genetics); Goodrich, 1929: 83-99 (review of
genetics); Gordon, 1931a: 732-787 (morphology of color patterns); 1931b:
1495-1523 (history of intergeneric hybrids); Gordon and Fraser, 1931: 169-
185 (genetic analysis of natural color variations); Myers, 1932: 403 (reference
to P. nigra Brind); Shaw, 1932: 263-278 (conditioned water and growth rate);
Gordon, 1934b: 79-92 (history of introduction of color varieties; habitat);
Stoye, 1934: 84-86; 1935: 50-52 (color varieties); Kosswig, 1935: 40-48 (gen-
etics); Hubbs, 1935: 10 (characters; aguada at Uaxactun, Guatemala); "Anony-
mous," 1937: 46 (figure; records mapped); Goldschmidt, 1937: 429-39 (sex
determination); Gordon, 1937: 376-392 (genetics; references); Breider, 1938:
784-828 (genetic and cytological basis of tumors); Brind, 1938: 40 (color
variation); Gordon and Smith, 1938b: 255-272 interspecificc hybridization and
tumors); 1938a: 543-565 (hybridization and tumors); Crozier and Wolf, 1939a:
463-485; 1939c: 143-163; 1939d: 176-179 (flicker response); Goodrich, 1939:
198-207 (chromatophores); Grobstein and Bellamy, 1939: 363-365 (thyroid
feeding); De Buen, 1940: 43 (literature records, in part); Gordon, 1940b: 3-5;
1940c: 22-24; 1940d: 63-64 (genetics of pigment cells; habitats); 1940e: 171-
172 (ecological isolation; no hybrids in nature; Rios Papaloapan, Coatzacoal-
cos, Usumacinta); Grobstein, 1940a: 1-22 (structure and development of
BULLETIN FLORIDA STATE MUSEUM
gonopodium); 1940b: 484-486 (testosterone, effects on gonopodium devel-
opment); Turner, 1940: 64-67 (fetal membrane); Dobzhansky, 1941: 279-
280 (genetics; hybridization); Baker and Furgeson, 1942: 116-119 (growth
in bacteria-free cultures); Goodrich, Hill and Arrick, 1941: 573-586 (chem-
istry of pigments); Gordon, 1941a: 38-39 (abstract, hybridization); 1942c: 76
(gene frequencies; genetics; Rios Jamapa, Papaloapan, Coatzacoalcos, Usuma-
cinta); Grobstein, 1942a: 477-478; 1942b: 305-328 (effect, steroids; gonopodia
in females); Needham, 1942: 387, 442 (tumors; regeneration of gonopodium);
Mayr, 1942: 75-76, 79, 171 (genetic phases; member of subspecies or super-
species); Gordon, 1943a: 71-72 (mutant strain); 1943b: 28 (genetics in relation
to melanomas); 1943c: 64-71 (variation; dorsal rays; hybridization; origin of
races; genetics; Rios Jamapa, Papaloapan, Coatzacoalcos, Grijalva, Usuma-
cinta); Gordon and Flathman, 1943: 9-12 (genetics of melanomas in hybrids);
Goldsmith, Nigrelli, Gordon, Charipper and Gordon, 1944: 132 (effects,
thiourea); Nigrelli and Gordon, 1944: 45 (tumors); Gordon and Benzer, 1945:
57-61, 63-70 (vertebrae; ribs; gonopodial suspensoria); Levine and Gordon,
1946: 197-198, 200-204 (ocular tumors); Gordon, 1946c: 87, 92 (ocular
tumors); 1946d: 77-78, 81, 83-86 (introgressive hybridization); 1946b: 307-
313, 315-320 (sex determination); 1947a: 95-101, 103, 105-112, 114-131
speciationn); 1947b: 8-9, 12, 14-17 (sex determination); 1948a: 216, 219-221,
223, 225, 227, 245 (color genes and melanomas); 1948b: 254, 256, 258 (genetics
of tumor development); 1950a: 405, 410-413 (genetics, in part); 1950b: 19
(pigment cell growth); 1950c: 26-34 (heredity of tumors); Gordon and Gor-
don, 1950: 61, 64, 71-72 (gene frequencies); Gordon and Rosen, 1951: 414-
417, 423, 427-430, 432-436, 438-451, 454, 456-459 (genetics of gonopodia;
Platypoecilus as subgenus of Xiphophorus); Nigrelli and Gordon, 1951: 122
(tumors); Gordon, 1951a: 127, 129-130, 133 (sex determination); 1951b: 155,
157-158, 160, 163, 194-197, 199-206, 211, 216 (pigment cell growth; geo-
graphical distribution; hybridization); 1951b: 676-677, 679 (genetic ex-
pressivity); Chavin and Gordon, 1951: 135-145 (genetics; sex determination);
Nigrelli, Jakowska and Gordon, 1951: 54 (peoplastic growths); Gordon, 1953c:
787-792 (pigment cell growth).
Platypoecilia maculata, Taubles, 1916: 16 (aquarium notes).
Poecilia maculata, Regan, 1906-1908: 105 (synonymy; description; records, in
part; El Hule, Obispo, Perez).
Xiphophorus (Platypoecilus) maculatus, Gordon, 1952b: 988 (melanomas);
1952a: 91, 94-98 (sex determination); Berg and Gordon, 1953: 44, 46, 58
(gonads).
Xiphophorus maculatus, Gordon, 1953a: 192 (pigment cell growth); Rosen and
Gordon, 1953: 20, 22-23, 38-40, 42 (morphology and mechanics of gonopo-
dia); Rubin and Gordon, 1953: 646-647 (gonopodial suspensorium); Gordon
and Gordon, 1954: 37-58 (biometry).
Platypoecilia var. maculatus, Beldt, 1917: 17 (red spotted form).
Platypoecilus maculata nigra, Brind, 1919: 167 (lapsus for gender of specific
name; black banded form).
Platypoecilus maculatus var. aurata Stoye, 1935: 51 (aquarium variety).
Platypoecilus maculatus var. cyanellus Meinken, 1934: 261-263 (description;
aquarium introduction).
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
Platypoecilus maculatus immaculatus Myers, 1922a: 1-22 (aquarium fish) 1922b:
89 (aquarium fish); Bayliss, 1929: 76-78 (characters); 1943: 74 (description).
Platypoecilus maculatus niger, Myers, 1922a: 1-22 (characters).
Platypoecilus maculatus nigra, Stove, 1935: 51 (characters).
Platypoecilus maculatus pulcher, Myers, 1922a: 1-22 (characters).
Platypoecilus maculatus pulchre, Troemner, 1943: 1-3 (aquarium notes).
Platypoecilus maculatus ruber, Myers, 1922a: 1-22 (characters).
Platypoecilus maculatus var. sanguine, Stoye, 1935: 51 (description).
Platypoecilus nigra Brind, 1914: 22 (original description; southern Mexico).
Platutpoccilus pulchra Brind, 1914: 22 (as aquarium fish).
Platypoecilus rubra Brind, 1914: 22 (as aquarium fish).
Platypoecilus sp., Schreitmuller, 1910a: 177-178; 1910b: 573-576 (spotted variety
later called pulchra).
MATERIAL EXAMINED
Mexico, Veracruz
Rio Jamapa system
1. Plaza de Agua near El Tejar, 8 mi. S Veracruz. UMMZ 124296. 828 half-
grown to adults; others kept alive.
Rio Papaloapan system
2. Southern tributary of Rio Tonto, 10 km. upstream from Papaloapan. UMMZ
108583. 10 half-grown to adults preserved; 91 taken alive.
3. Same locality. UMMZ 124178. 1 mature male.
4. Rio Papaloapan at San Bartolo. UMMZ 124473. 1 transforming male.
5. Lagoon beside Rio Papaloapan, 3 km. N San Bartolo. UMMZ 124255. 567
young and adults.
6. Pool, 5 km. S Papaloapan at 148Y2 km. mark on railroad. UMMZ 124284.
5,148 young and adults preserved; others taken alive.
7. Lagoon and pool, 4 km. S Papaloapan at 149 km. mark on railroad. UMMZ
124227. 1,473 young and adults preserved; others taken alive.
8. Arroyo Zacatispan and lagoon, 4.5 km. S Papaloapan. UMMZ 124220. 48
young to adults.
9. Pool, part of lagoon, 4.5 km. S Papaloapan. UMMZ 124261. 413 young
to adults.
10. Arroyo Zacatispan 6 km. E Papaloapan. UMMZ 124195. 5 half-grown to
sub-adults.
Rio Coatzacoalcos system
11. Laguna del Aserradero near pool 200 m. from Rio de Jaltepec in vicinity
Santa Lucrecia (Jesus Carranza). UMMZ 157657. 1 subadult male.
12. Lagunitas near Jesus Carranza. UMMZ 124452. 5 half-grown to adults.
13. Arroyo de "La Llorona" about 350 m. from Estacion de Juile. UMMZ
157661. 1 subadult male.
14. Laguna de "El Tule" about 3 to 4 km. from El Juile. UMMZ 157663.
18 young to adults.
15a. Small unnamed arroyo emptying into Arroyo Aguazate near Estacion de
Almagres. UMMZ 157662. 90 young to adults.
Mexico, Tabasco
Rio Tonala system
15b. Arroyo de La Venta at La Venta. 30 adults taken alive.
1960
BULLETIN FLORIDA STATE MUSEUM
Rio Grijalva system
16. Rio Muerto tributary to Rio Teapa on Finca de Pablo Pratts 4 km. N Teapa.
UMMZ NYZS-GL. 38 young to adult males and females.
17. Tierra Colorado pond 5 km. N of Villahermosa. UMMZ NYZS-GL. 203
young to adult males and females.
18. Same locality UMMZ NYZS-GL. 478 young to adult males and females.
Guatemala
Rio Usumacinta system
19. Upper Rio de la Pasion (Rio Chajmayic) at La Ceiba. UMMZ 143788.
1 young.
20. Near mouth of Rio Chajmayic, Cahabon, an upper tributary of Rio de la
Pasion. UMMZ 143787. 16 subadults and adults.
21. Mouth of arroyo tributary to the upper Rio de la Pasion between La Ceiba
and Arroyo San Simon. UMMZ 143789. 109 young to adults.
22. In and just above mouth of first arroyo tributary to upper Rio de la Pasion
from east below Arroyo San Simon. UMMZ 143790. 14 young to adults.
23. Arroyo Subin at Santa Teresa 13 mi. S of La Libertad. UMMZ 143786.
3 subadults and adult.
24. Arroyo Subin at second rapids above mouth in Rio de la Pasion. UMMZ
143791. 3 subadults.
25. Laguna de Yalac, in course of Rio San Pedro de Martir above El Paso de los
Caballos. UMMZ 143782, 143783. 2 adults taken by seine and by poison.
26. Rio San Pedro de Martir about 1 mi. below Laguna de Yalac. UMMZ
143784. 3 half-grown to adult.
27. Rio San Pedro at Desempeno just below El Paso de los Caballos. UMMZ
143781. 17 young to adults.
Interior lakes
28. Laguna de Zotz W Laguna de Peten. UMMZ 143780. 558 young to adults.
29. Laguna de Eckixil E southern arm of Laguna de Peten. UMMZ 143778.
1 adult.
30. Outlet of Laguna de Petenxil, tributary of Laguna de Peten. UMMZ 143779.
20 subadults and adults.
31. Tributary creek to short stream which forms outlet of Laguna de Petenxil.
UMMZ 143785. 31 young to adults.
32. Flooded "rio" at E end southern arm Laguna de Peten. UMMZ 143777.
2 young.
33. Arroyo de Ponteil at head of standing water about 1 mi. from mouth. UMMZ
143774. 1 adult.
34. Arroyo de Ponteil at forks about 3 mi. from mouth, a southern tributary of
southern arm Laguna de Peten. UMMZ 143775. 6 young to adults.
35. Laguna de Peten along airfield near middle southern shore of southern arm
of lake. UMMZ 143771, 143776. 22 young to adults, 1 adult.
36. Arroyo de Puebla Nueva near head of standing water (flood waters of Laguna
de Peten). UMMZ 143773. 9 young to adults.
37. Flores, Lake Peten, Arroyo San Helena, opposite Flores. UMMZ NYZS-GL.
1 male and 2 females.
38. Flores, Lake Peten, Arroyo Ponteil, Peten. UMMZ NYZS-GL. 1 male and
1 female.
Guatemala, British Honduras
Rio Hondo system
39. Aguada at Uaxactun, Guat. UMMZ 97873. 3 adults.
40. Cenote Creek, 8 mi. SE Gallon Jug, Orange Walk, B. Hond. UMMZ NYZS-
GL. 2 adult females.
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
41. Lagoon and creek on E bank of E branch of Rio Hondo opposite San Antonio,
B. Hond. UMMZ NYZS-GL. 9 males, 17 females and 16 half-grown.
42. Lagoon opposite San Antonio, connected with creek to east branch of Rio
Hondo, Orange Walk, B. Hond. UMMZ NYZS-GL. 144 males, 98 females,
150 young to half-grown.
4:3. Chorro 4 mi. N Corozal on highway to Santa Helena (Chetumal), Northern
District, B. Hond. UMMZ NYZS-GL. 1 adult male.
44. Corozal, Back Street (extreme north B. Hond.) UMMZ 158391. 191 young
to adult.
British Honduras
Independent Tributaries to Baluo de Chetumal
45. Chorro at Back and Queen Streets, Corozal, Northern District. UMMZ
NYZS-GL. 40 young to adult males and females.
46. Chorro, Northern District. UMMZ NYZS-GL. 9 young to adult males and
females.
47. Hill Bank opposite campsite, inlet to lagoon of New River. UMMZ NYZS-
GL. 1 male, 1 female and 1 half-grown.
48. San Estevan 1 mile on road to Progreso, Orange Walk. UMMZ NYZS-GL.
8 males, 2 females and 4 half-grown.
Belize River system
49. Lower extension of Gabourel Creek Y2 mi. E Belize airport, Belize. UMMZ
NYZS-GL. 1 half-grown.
50. Gabourel Creek near Belize Airport, Belize. UMMZ NYZS-GL. 500 young
to adult males and females.
51. Johnny Pond Creek 1 mi. from Boom, Belize. UMMZ NYZS-GL. 8 young
to adult males and females.
52. Arroyo W of western branch Belize River, near Benque Viejo, Cayo. UMMZ
NYZS-GL. 26 young to adult males and females.
DIAGNOSIS. A small, deep-bodied species of Xiphophorus (tables
4 and 5). Lower caudal rays not developed as a "sword" in adult
males. Midlateral stripe wanting. Without a deep-lying ridge of
black pigment midventrally on caudal peduncle. Dorsal and anal fins
distinctly rounded. Dorsal fin not usually dimorphic, rounded, usually
with a faint or moderate diffuse dusky band near base. Gonopodium
with claw usually wanting at tip of ray 5a, rarely with a minute scythe-
shaped or crescentic element; distal serrae on ray 4p well developed
and erect, seldom retrorse, 4 to 7 in number; ramus of ray 4a curved
downward over blade, not hooked, and extending to tip of ray 4p;
hook of ray 3 long and slender. Vertebrae usually 27, rarely 26.
Scales in a lateral series 22 to 25, usually 23 or 24. Dorsal fin rays
7 to 11, usually 9 or 10.
GENERAL ACCOUNT. The two females on which Guenther based his
description of this species were examined in the British Museum by
Myron Gordon, who supplied the following information: One, 32 mm
in standard length, has a spotted dorsal fin pattern (Sd), and a one
spot (0) on the tail base. The other, 30 mm in standard length, has
two tail patterns, a comet (Co) and one spot (0), but no macromelano-
1960
BULLETIN FLORIDA STATE MUSEUM
Vol. 5
1-5
-.15A- 15B
15C -17
-18-38
39
40
-41-43
44-46
47
52-5
45-46
39- 44 -4
49-52
XIPHOPHORUS COUCHIANUS -
XIPHOPHORUS V. VARIAreS 0
XIPHOPHORUS V X/PHIO/UM 0
X/PHOPHORUS V EVELYNAE @
XIPHOPHORUS MACULATUS A '
Figure 4. Distribution by record stations for the species and subspecies of
platyfishes. Additional records in the Rio Tonala and Rio Grijalva from Gordon
and Gordon (1954) are included.
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
phore patterns (see below and fig. 5). Both have a dorsal ray count
of 9, treating the last two rays as one.
Also seen was a small series of specimens in rather poor condition
collected at Orizaba in 1867 by Professor Sumichrast (ANSP 14684-85,
14686, 14671-83, 15927-28, 15929, 39893), as well as Meek's specimens
from Obispo (SU 21462).
The color phases (fig. 5) which make this species one of the most
variable of all wild vertebrates occur together in nature, and these
polymorphs differ in no other respects. Some of these polymorphs
have been given scientific names (see synonymy) which have no proper
place in systematic nomenclature. These color phases, however, are
significant in the study of the comparative genetics of species and the
composition of natural populations (Gordon and Gordon, 1950, 1954).
In X. maculatus seven basic tail patterns are known. The morpho-
logical and histological features of four of these patterns were first
described by Gordon (1931a). When, later, he described the remain-
ing three patterns, Gordon (1947a) showed that all the tail patterns
in this species are controlled by autosomal genes that constitute a sin-
gle series of dominant epistatic multiple alleles. All seven patterns
are formed by characteristic groupings of many small melanophores.
A rare and exceptional tail pattern called the Guatemala crescent
pattern was first discovered by Gordon (1947a) in a population of platy-
fish from the Laguna Petenxil and Arroyo Ponteil in Guatemala. This
pattern resembles the simple crescent pattern but is broader through-
out and particularly where the ends of the crescent bulge cephalad
to form two distinct nodes. Its presence in X. maculatus is always
associated with two small black dots in the integument at the tip of
the lower jaw known as lip markings. This is the only known example
of a micromelanophore tail pattern in Xiphophorus associated with a
far-removed accessory pattern.
Other seldom encountered tail patterns are: The anterior or trap-
ezoidal element of the complete crescent occurring alone; the upper
and lower elements of the cut crescent pattern occurring alone; and
the upper and lower elements of the comet pattern occurring alone.
Possibly the occurrence of only one element of either the cut crescent
or comet patterns may be due to localized interference with the ex-
pression of the complete pattern during ontogeny. As yet no evidence
suggests that these patterns reflect distinct genotypes.
Genetic analyses reveal that all the fundamental tail patterns of
X. maculatus are controlled entirely by the action of autosomal genes
(Gordon, 1947a), and that none so far as known is sex limited.
BULLETIN FLORIDA STATE MUSEUM
Gordon (1947a) indicated that in X. maculatus any two of the
seven basic tail patterns theoretically may occur together to produce
a net phenotypical effect of both patterns combined. Such a dual
pattern often is difficult to recognize because most of the seven
basic patterns occupy similar areas at the base of the caudal peduncle
and a certain amount of superimposition is inevitable. This point is
well illustrated by a dual pattern consisting of the complete crescent
and the simple moon patterns in which the single element of the
simple moon pattern occupies approximately the same position on the
caudal peduncle as the anterior or trapezoidal element of the complete
crescent. As the moon pattern is somewhat larger and more rounded
than the crescent pattern, careful observation reveals its presence with
the single complete crescent. But the moon pattern is also known to
occur with the simple crescent in which the anterior trapezoidal ele-
ment is lacking. In such cases only careful analysis of the intensity
and limits of the pigmented mass will show whether or not a super-
imposed pattern is present. Many dual patterns probably are over-
looked or misinterpreted (Gordon, 1931a; Gordon and Gordon, 1950).
In addition to the seven basic tail patterns in X. maculatus, 17
double, but no triple or more complex combinations have been re-
corded. The presence of double, but no more complex patterns,
suggests the operation of seven dominant multiple alleles, as indicated
earlier. Evidence from genetic experiments supports this suggestion.
The following tabulation, from Gordon (1947a), lists all the types
possible in X. maculatus (fig. 5).
0 M Me C Cc Co T
0 0
M OM M
Me OMc (MMc) Mc
C OC MC (McC) C
Cc OCc MCc McCc (CCc) Cc
Co OCo MCo McCo CCo CcCo Co
T OT MT (McT) CT CcT CoT T
0 = one spot; M = simple moon; Mc = complete moon; C = simple
crescent; Cc = complete crescent; Co = comet; T = twin spot. Dual patterns
shown in parentheses are theoretically possible but have never been detected.
The macromelanophore patterns are composed of cells that are
both morphologically and genetically distinct from those responsible
for the tail patterns. The macromelanophore patterns (fig. 5) are sex
linked, or, at least, incompletely sex linked, and they are not all
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
members of a single allelic series.
together in a single individual.
A
B
C
D
Two or more patterns may occur
N-
QJ
R -
Figure 5. Diagrams of the pigment patterns in Xiphophorus maculatus
(Guenther).
A. Unmarked except for basic reticular patterns.
B-M. Micromelanophore tail patterns. B. One spot. C. Twin spot. D.
Complete crescent. E. Simple crescent. F. Cut crescent. G. Ax-head. H.
Guatemala crescent (note lip markings). I. Comet. J. Upper comet. K. Lower
comet. L. Complete moon. M. Simple moon. Only patterns B-F, and L are
common in natural populations and have been studied genetically. The other
tail patterns, except for the Guatemala crescent, are rare and may be only partial
expressions of other more common patterns.
N-R. Macromelanophore patterns. N. Spotted. 0. Spotted dorsal. P.
Striped. Q. Black banded. R. Black belly.
The macromelanophores are chiefly in the dermis, both above and
below the scales. Macromelanophores are sometimes seen in the epi-
dermis, and occasionally, when the scales are removed from their
dermal pockets, they may be seen adhering to the scale substance. In
such cases the pigment granules within the scale macromelanophores
are more diffuse and the cell's dendritic processes are less delicate
than those within the dermal melanophores.
Though micro- and macromelanophores are morphologically and
genetically distinct, the two types of patterns interact when they occur
G -
H
A<---.
/" ,.
BULLETIN FLORIDA STATE MUSEUM
together in the same individual according to observations by Gordon
(1928). He found that the occurrence of a macromelanophore spot
on the side in X. maculatus is correlated with the loss of micromelano-
phore reticulation in the same region. Conversely, he found that the
micromelanophore pattern enhances the expression of the macro-
melanophore spotting in this species. Gordon's experiments show
that these reciprocal relations are referable to the interaction of an
autosomal gene, St, for reticulations with a sex linked gene, Sp, for
spotting.
The southward dispersal of X. maculatus is limited by the swift
waters that arise in the Maya mountains and intervene between British
Honduras and the Honduras lowlands. Recent collections in British
Honduras suggest that the species of Phallichthys compete with macu-
latus and entirely replace it ecologically to the south, thereby re-
inforcing the topographical barriers that block further extension of
its range.
Where this species occurs together with Phallichthys in the New
River and Rio Hondo systems (see Rosen and Bailey, 1959, for distri-
bution of Phallichthys), the two species become stratified and occupy
alternative microhabitats. The platyfish are caught deep in the cooler
mud, and the Phallichthys on top. When separated in different drain-
ages or in different sections of the same drainage, they appear to
occupy almost identical ecological niches.
This species ranges from Veracruz, Mexico to northern British
Honduras, usually at lower elevations.
Xiphophorus variatus (Meek)
Variable Platyfish, figures 1, 3-4, 6-7, tables 1-5, 9-13
Platypoecilus variatus Meek, 1904: 146-147 (original description; records, includ-
ing those of the Rio Soto la Marina system).
DIAGNOSIS. A small to moderate, deep-bodied species of Xipho-
phorus (tables 4 and 5), rarely with a minute unpigmented caudal
appendage in adult males (Rio Soto la Marina system only); usually
with two or more variably complete zigzag black midlateral stripes
running from base of caudal peduncle to opercular membrane; with
two or more oblique black lines extending downward from the mid-
lateral stripes just behind pectoral base. Without a deep-lying ridge
of black pigment midventrally on caudal peduncle. Dorsal fin usually
rounded, higher and more smoothly rounded in adult males than
females. Claw wanting at tip of ray 5a of gonopodium; distal serrae
on ray 4p well developed and retrorse, 5 to 8 in number; ramus of ray
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
4a curved downward over blade, not hooked, and extending to tip of
ray 4p; hook of ray 3 long and slender. Vertebrae 28 or 29, rarely 27
or 30. Scales in a lateral series 24 to 28, usually 26. Dorsal fin rays
9 to 14, rarely 14.
GENERAL ACCOUNT. Most of the present material corresponds well
with Meek's (1904) account of Platypoecilus variatus. Three distinct
subspecies may be separated however, X. v. xiphidium, long regarded
by authors as a full species, X. v. evelynae from the headwaters of the
Rio Tecolutla, and the nominate form. Even after excluding xiphidium
and evelynae, variatus remains far from a simple unit. It shows
marked geographical variation, as well as the spectacular individual
variation in micromelanophore tail patterns and macromelanophore
spotting exhibited by maculatus (see fig. 5). Aquarists have tended
to select the brightest of the distinctive color phases of variatus.
Frequently, large males have lemon yellow or orange caudal and
dorsal fins, and a coppery red spot on the base of the caudal peduncle
never seen in maculatus.
The position of variatus in the genus Xiphophorus is perhaps
unique, for it combines the pigmentary polymorphism of maculatus
with the structural and metameric variability of helleri. Unlike mem-
bers of the helleri complex, however, the morphometric differences
distinguishing the subspecies of variatus show no evidence of any
common trends or uniting features. Only within X. v. variatus does
there appear a marked north-south trend in values of least relative
depth of caudal peduncle (table 5) and in the frequencies of the micro-
melanophore tail patterns. The geographical distribution of other
traits is distinctively mosaic (see tables).
Gordon and others early appreciated the special features of the
numerous forms of X. v. variatus. These were found to differ chiefly
in average numbers of dorsal fin rays and certain pigmentary traits,
but a number of other features have nevertheless been examined.
This information may be summarized as follows:
1. Greatest depth of body. X. c. variatus in the Rio Axtla (Rio
Panuco system), Estero Tancochin system, Rio Cazones system, and
Rio Nautla system are most slender bodied.
2. Least depth of caudal peduncle. In X. v. variatus the least
depth of the caudal peduncle shows a gradual increase from north
(Rio Tamesi) to south (Rio Nautla).
3. Length of head. Members of the Estero Tancochin and Rio
Cazones populations have relatively long heads.
1960
BULLETIN FLORIDA STATE MUSEUM
4. Length of caudal peduncle. Platyfish in the Arroyo Sin Nombre
(lower reaches of the Rio Tecolutla system) have the shortest average
length of caudal peduncle.
5. Length of caudal fin. Platyfish in the Rio Cazones system pos-
sess the longest caudal fins.
6. Dorsal fin ray counts. X. v. variatus in the Rio Cazones sys-
tem average one more fin ray than the other nine populations (table 9).
7. Scales in a lateral series. Scale counts in all populations are
relatively uniform (table 10).
Of the three subspecies, only variatus and xiphidium possess micro-
melanophore tail patterns. Four have been identified so far, all evi-
dently members of a single allelic series. A tabulation of their tail
pattern combinations follows:
0 Cu Ct C
0 0
Cu CuO Cu
Ct CtO CtCu Ct
C CO (CCu) CCt C
O = one spot; Cu = upper portion only of cut-crescent; Ct = cut crescent;
C = simple crescent. Dual patterns shown in parentheses are theoretically pos-
sible but have never been detected.
In variatus from the Rio Nautla, a pattern occurs that strongly
resembles the complete crescent of maculatus, having a peduncular
segment and crescentlike extension on the base of the caudal fin.
The crescentic portion is believed to result from a "spilling over" of
pigment from an unusually well developed one spot onto the fin base.
In addition, some one-spot patterns of variatus may become sufficiently
large and well defined that they resemble the simple moon pattern
of maculatus, or by conforming to the limits of the sharply bent poste-
rior myomeres they resemble the axlike portion of the complete
crescent of maculatus.
As indicated, only X. v. evelynae lacks tail patterns. X. v. variatus
from the Rio Moctezuma and Rio Calaboza (Rio Panuco system) have
but a single pattern, though a different one in each case. Three pop-
ulations, those of the Rio Tuxpam,, the lower reaches of the Rio
Tecolutla, and the headwaters of thei Rio Nautla, have only two pat-
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
terns, and the remaining populations of X. v. variatus have three,
four, and five in various combinations.
Four macromelanophore patterns occur in variatus, a caudal spot,
a black banded pattern, and two types of spotting (blotches and
speckles) along the sides. The banded and speckled patterns and cau-
dal spot are found only in the Rio Cazones population, and in less
than 10 percent of the adult individuals. The blotched pattern occurs
throughout all populations of variatus in numbers ranging from 15
to 96 percent of each adult sample.
Atz (Ms.) has found genetic evidence that some of the tail pat-
terns, as well as certain macromelanophore markings, are controlled
by homologous genes in variatus variatus, v., xiphidium, and maculatus.
He also provides evidence that the genetic sex-determining mechan-
isms (female homogametic, male heterogametic) in variatus and in
maculatus north of the Rio Usumacinta system are similar, while
maculatus from the rivers of British Honduras are female heteroga-
metic, and those from the Peten district of Guatemala do not behave
consistently according to either system. This genetic information,
considered in the light of the morphological and ecological similarities
between maculatus and variatus, is taken as direct evidence of their
close relationship. Nevertheless it will be shown below that, despite
their many similarities and their allopatric distributions, good evi-
dence favors recognizing maculatus and variatus as distinct species.
Xiphophorus variatus ranges throughout most of the Atlantic
coastal rivers and streams of southern Tamaulipas, eastern San Luis
Potosi, and northern Veracruz over a straight line distance of approx-
imately 350 miles. Within this range it occurs at altitudes ranging
from 100 to 4000 feet (table 33).
Xiphophorus variatus xiphidium (Gordon)
Figures 3-4, 6, tables 3-5, 9-11
Platypoecilus xiphidium Gordon, 1932a: 287 (original description; small sword
mentioned); 1932c: 89 (Rio Soto la Marina); 1933a: 143-146 (habitat; char-
acters; not mentioned by name); 1933c: 77-78 (polymorphism); Stoye, 1934:
86 (characters); Friedman and Gordon, 1934: 446-455 (chromosomes and dis-
tribution); Stoye, 1935: 53 (description); Kosswig, 1935: 46-48; 1937: 491-
528 (hybrids); Gordon, 1935a: 8-9; 1937: 388-389 (characters and sex de-
termination); Anonymous, 1937: 46 (figure; records mapped); Hubbs, 1937:
296 (tributary of Arroyo San Carlos); Gordon, 1938: 21 (coloration); Gordon
and Smith, 1938a: 543-565 (hybridization); Gordon, 1940a: 96-106 (habitat);
De Buen, 1940: 44 (cuenca del Rio Soto la Marina); Dobzhansky, 1941: 280
(genetics; species or race); Gordon, 1941b: 112-113 (Rio Soto la Marina;
hybridization); 1943c: 65-66 (Rio Soto la Marina system; patterns; variation;
BULLETIN FLORIDA STATE MUSEUM
dorsal fin rays); Gordon, Cohen and Nigrelli, 1943: 569-572 (xiphidium-like
sword in maculatus); Hubbs and Gordon, 1943: 31 (caudal appendage); Gor-
don and Benzer, 1945: 58-70 (vertebrae; gonopodial suspensorium); Gordon,
1946b: 316-317, 320 (sex determination); 1947a: 117, 129 speciationn); Gor-
don and Rosen, 1951: 417, 423-433, 456-458 (genetics of gonopodia); Nigrelli
and Gordon, 1951: 121-125 (spontaneous neoplasms; hybrids with variatus);
Gordon and Aronowitz, 1951: 151 (sex determination); Gordon, 1951c: 194-
197 (species hybridization; distribution).
Xiphophorus xiphidium, Rosen and Gordon, 1953: 23, 47 (evolution of geni-
talia); Gordon, 1953c: 789 (mating systems; distribution).
F'i.iri...,..,i, variatus (in part), Meek, 1904: 146 (records for Garza Valdez,
Santa Engracia, Victoria).
Poecilia maculata misidentificationn; not Platypoecilus maculatus Guenther), Re-
gan, 1906-1908; 105 (in part; records for Garza Valdez, Santa Engracia, and
Victoria).
Platypoecilus maculatus, Regan, 1913: 1004 (in part); De Buen, 1940: 44 (litera-
ture records; in part).
Platypoecilus xiphodium (misspelling), Needham, 1942: 387 (tumors).
Platypoecilus xiphidinus (lapsus for xiphidium), Mayr, 1942: 171 (subspecies or
member of superspecies).
Platypoecilus n. sp. Stoye, 1933: 307 (discovery; coloration).
MATERIAL EXAMINED
Mexico, Tamaulipas
Rio Soto la Marina system
1. Arroyo Marmolejo, tributary of Arroyo San Carlos. UMMZ 92310. 4 half-
grown to adult females.
2. Rio Santa Lucia at Villagran. UMMZ 108664. 8 half-grown to adult males
and females.
3. Rio Purificacion above Cruz. UMMZ 108662. 121 half-grown to adult
males and females.
4. Spring tributary of Rio Purificacion 2 mi. W Cruz. UMMZ 124168. 2 half-
grown.
5. Rio Purificacion, above Cruz. UMMZ 108661. 12 half-grown to adult males
and females.
6. Chorro, V2 mi. above junction Rios Pilon and Purificacion, Rancho Santana,
8 mi. SE Padillo. UMMZ 169671. 17 half-grown.
7. Manantial de Rio Santa Engracia near Hacienda Santa Engracia. UMMZ
124416. 282 young to adult males and females.
8. Rio Corona (Rio Santa Engracia) at La Corona. UMMZ 108663, 97573.
164 half-grown to adult males and females plus the lectotype.
9. Arroyo de la Presa 16 km. N Ciudad Victoria. UMMZ 124404. 54 large
young to adult males and females.
10. Rio Corona 3 mi. NE Guemes. UMMZ 124398. 36 large young to adult
males and females.
11. Rio San Marcos S Ciudad Victoria. UMMZ 124389. 17 half-grown to
adult males and females.
12. Rio Santa Engracia, 17 mi. N Victoria at bridge, highway 1 to Monterrey.
UMMZ 169600. 13 half-grown to adults.
13. Rio Corona where road from Guemes to Padilla crosses Rio Corona. UMMZ
169610. 8 adults.
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
14. Rio Caballero, 729 km. N Mexico City. UMMZ 162151. 234 half-grown
to adult males and females.
DIAGNOSIS. A consistently deep-bodied, robust subspecies of vari-
atus; with a single weak zigzag dark mid-lateral stripe, two or more
stripes, or with a diffuse, dusky midlateral black band; frequently
with irregular black speckling along middle sides and on caudal ped-
uncle. Adult males with a short, broad, pointed appendage developed
from the lower angle of the caudal fin as a sword. Vertebrae 27 or
28, usually 28. Dorsal fin rays 9 to 11, usually 10.
REMARKS. Though in all sections of the variatus complex the
adult male shows distinctive "parr" markings along the sides and
flanks, these are intensified and very broad in xiphidiumi. They vary
with age and in very old males the vertical bars become broad bands
of deep purple or black. This subspecies differs from variatus proper
in the above mentioned color character, in averaging one less dorsal
fin ray and one less vertebra, in the invariable presence of the minute
caudal sword in adult males, and in proportional measurements (see
table 29).
Though this well known form has been described many times
(see synonymy above), no holotype was ever designated and the name
xiphidium thus rests upon a syntypic series from the Rio Corona at
La Corona, Tamaulipas, Mexico. From this syntypic material a lecto-
type is herewith designated:
Figure 6. Lectotype of Xiphophorus variatus xiphidium (Gordon),
(UMMZ: 9757.3), adult male 29.0 mm in standard length.
Lectotype, an adult male (UMMZ 97573) 29.0 mm in standard
length, collected at Rio Corona (Rio Santa Engracia) at La Corona,
15 miles north of Ciudad, Victoria, Rio Soto la Marina system, Ta-
mualipas, Mexico, by Gordon, Creaser, and Ostos on 25 April 1930,
BULLETIN FLORIDA STATE MUSEUM
together with 164 half-grown to adult males and females (UMMZ
108663).
Reasons for regrouping xiphidium and variatus as a single species
are discussed below. The specimens reported by Meek as Platypoeci-
lus variatus from Garza Valdez, Santa Engracia, and Victoria, repre-
sent this subspecies.
Xiphophorus variatus variatus (Meek)
Figures 3-4, tables 3-5, 9-13
Platypoecilus variatus Meek, 1904: 146-147 (in part; original description; Valles;
records, excluding those of Rio Soto la Marina system); Gordon, 1932a: 287
(Valles); 1932c: 89 (distribution); 1933b: 518-522 (habitat; colorvariations);
Meinken, 1933: 89-92 (nomenclature; as aquarium fish; Tampico); Peters,
1933: 9-10 (variations lowermostt figure X. montezumae]); Friedman and
Gordon, 1934: 446-455 (chromosomes; distribution; hybrids); Kosswig, 1935:
43-46 (distribution; synonymy in part; characters); Gordon, 1935a: 7-10 (dis-
tribution; synonymy in part; characters); Bellamy, 1936: 531-535 (hybrids);
Anonymous, 1937: 46 (figure; records mapped); Goldschmidt, 1937: 429-439
(sex determination); Gordon, 1937: 388-389 (sex); 1938: 21 (coloration);
Breider, 1938: 784-828 (in part; genetic and cytological basis of tumors); Grob-
stein and Bellamy, 1939: 363-365 (effect of thyroid feeding); Rust, 1939: 172-
176 (heterogamety); Gordon and Smith, 1938a: 543-565 (hybridization); Grob-
stein, 1940a: 1-22 (structure and development of gonopodium); 1940b: 484-
486 (hormones; gonopodium; gonopodium in females); Turner, 1940: 64-67
(fetal membranes); Dobzhansky, 1941: 280 (genetics; species or race); Gordon,
1941a: 38-39 (sex); 1941b: 112-113 (Rio Panuco, Rio Cazones; hybridization);
Mayr, 1942: 171 (as subspecies or member of superspecies); Potts, 1942: 85-
88, 90-93, 95 hypophysiss); Needham, 1942: 387 (tumors); Hubbs and Gordon,
1943: 31 (comparisons); Gordon, Cohen and Nigrelli, 1943: 569 (hormones;
caudal fin appendage); Gordon and Benzer, 1945: 58-70 (gonopodial suspensor-
ium; vertebrae); Gordon 1946b: 316-318, 320 (sex determination); 1947a: 117,
129 speciationn; three subspecies); 1950a: 408 (25 pairs of chromosomes);
1950b: 19 (hybrids; Rio Panuco); Gordon and Rosen, 1951: 415, 417, 424-425,
427-433, 454, 457 (genetics of gonopodia); Nigrelli and Gordon, 1951: 121-124
(spontaneous neoplasm in xiphidium-variatus hybrid); Gordon and Aronowitz,
1951: 151 (sex determination); Gordon, 1951c: 194-197 (hybridization; distri-
bution; spotting); 1953c: 789-790 (spotting; distribution; hybrids).
rlg,i._.... .i',, variatus variatus Gordon, 1940e: 172 (habitat; ecological segrega-
tion; Rio Axtla); 1943c: 65-67 (dorsal rays; habitat; Guayalejo, Panuco, Valles,
Axtla, Cucharas, Tancochin, Tuxpam).
Platypoecilus variatus dorsalis Stoye, 1933: 306-307 (misuse of name in 1932
for P. v. variatus); Gordon, 1933b: 6 (explanation of name).
Platypoecilus variatus ssp. Gordon, 1943c: 65-67 (Rio Cazones).
Xiphophorus (Platypoecilus) variatus, Aronowitz, Nigrelli and Gordon, 1951:
239-240 (spontaeous epithelioma); Gordon, 1953b: 148-150 (habitat; associ-
ation with X. montezumae, X. pygmaeus).
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
Platypoecilus maculatus, Regan, 1913: 1004 (in part; diagnosis, Meek's specimens
only).
Poecilia maculata (not of Guenther) Regan, 1906-1908: 105 (description; records).
Mollienisia formosa misidentificationn, not Limia formosa Girard), Rachow, 1932:
705; Roth, 1932: 249 (as aquarium fish).
Mollienesia formosa, Wheeler and Ingle, 1953: 266 misidentificationn).
MATERIAL EXAMINED
Mexico, Tamaulipas
Independent Atlantic tributaries
15a. Tributary of Rio San Rafael, 3 mi. S San Rafael. UMMZ M59 195 young to
adult.
15b. Rio Tigre at road crossing just north Aldama. UMMZ 59 22 young to adult.
Rio Tamesi system
15c. Rio Guayalejo near Llara, 45 mi. SE Ciudad Victoria. UMMZ 108669. 19
half-grown to adults.
16. Tributary of Rio Guayalejo 16 mi. N Xicotencatl. UMMZ 108668. 75 half-
grown to adults.
17. Canal 5 mi. S Limon. UMMZ 97574. 1 young.
Mexico, San Luis Potosi and Veracruz
Rio Panuco system
18. Rio Valles at Valles. UMMZ 108667. 115 half-grown to adults.
19. Same locality. UMMZ 108616. 2 adult females.
20. De los Puercos, Valles. UMMZ NYZS-GL. 21 young to adults.
21. Arroyo near Valles. UMMZ 124437. 3 subadult to adult.
22. El Nilo, Pujal. UMMZ NYZS-GL. 25 half-grown to adult.
23. Pool reputed to have connection with Cueva Chica, Pujal. UMMZ NYZS-
GL. ? half-grown to adults.
24. Open pond at Pujal near Cueva Chica. UMMZ NYZS-GL. 2 half-grown.
25. Arroyo Santa Isabel, tributary of Rio del Coy. UMMZ 124383. 10 half-
grown to adults.
26. Small arroyo between Rio Axtla and Rio Moctezuma. UMMZ NYZS-GL.
31 half-grown to adults.
27. Rio Matlapa at Matlapa, 8 mi. N Tamazunchale. UMMZ 124338. 11 half-
grown to adults.
28. Rio Matlapa. UMMZ NYZS-GL. 9 half-grown to adults.
29. Arroyo Plan de Jalpilla, 18 mi. N Tamazunchale. UMMZ 124351. 18 half-
grown to adults.
30. Danubio, Rio Axtla 4 mi. upstream from Axtla. UMMZ NYZS-GL. 22 young
to adult males and females.
31. Rio Axtla 2 mi. W Axtla. UMMZ 108604. 123 half-grown to adults.
32. Arroyo Matlapa at Comeca 2 mi. NW Axtla. UMMZ 124375. 3 adults.
33. Rio Axtla at Axtla. UMMZ 124363. 1 mature male.
34. Rio Calaboza 9 mi. SW Tantoyuca. UMMZ 108666. 32 half-grown to
adults.
35. Rio de los Hules 11 mi. SW Tantoyuca. UMMZ 108670. 3 adults.
36. Junction of Rio de los Hules and Rio Calaboza. UMMZ 108665. 7 half-
grown to adults.
37. Tributary of Rio Tempoal 2.5 mi. N Tempoal. UMMZ 108672. 184 half-
grown to adults.
38. Tributary of Rio Panuco 3 mi. N El Higo. UMMZ 108671. 5 half-grown
to adults.
1960
BULLETIN FLORIDA STATE MUSEUM
Mexico, Veracruz
Estero Cucharas system
39. Estero Cucharas near Cucharas. UMMZ 97575. 203 half-grown to adults.
Estero Tancochin system.
40. Tributary of Estero Tancochin E Naranjos. UMMZ 108675. 92 half-grown
to adults.
Rio Tuxpan system
41. Tributary of Rio Tuxpan 14 mi. S Alamo. UMMZ 108676. 27 half-grown
to small adults.
42. Rio Frio, 10 mi. N Juarez (or Mante). UMMZ 162105. 5 specimens not
seen but probably subadult since their stand. lengths are given as 24.0-
25.0 mm.
43. Drainage canal, 1 mi. N Juarez (or Mante). UMMZ 162119. 4 adults.
Rio Cazones system
44. Tributary of Rio Cazones 12 mi. S Miahuapam. UMMZ 108673. 28 half-
grown to adults.
45. Tributary of Rio Cazones near Tihuatlan. UMMZ 108674. 47 half-grown
to adults.
46. Arroyo Mariandrea on road to Apapantilla from Papantla at Mariandrea about
10 mi. W Poza Rica (Puebla, Mexico). UMMZ NYZS-GL. 63 half-grown
to adult males and females.
Rio Tecolutla system
47. Arroyo Sin Nombre between Guitierrez Zamora and Papantla. UMMZ
NYZS-GL. 87 half-grown to adult males and females.
Rio Nautla system
48. Rio Nautla 6 mi. from Tlapacoyan on road to Nautla. UMMZ NYZS-GL.
146 adult males and females.
49. Rio Nautla 4 mi. from Martinez de la Torre on road to Nautla. UMMZ
NYZS-GL. 235 young to adult males and females.
50. Stream 2.3 mi. W Ixtacocoa, 3.9 mi. W Martinez de la Torre, on Teziutlan-
Casita Highway. UMMZ 167485. 11 half-grown to adult.
51. Ditch 1.8 mi. SW Nautla turnoff towards Martinez de la Torre. UMMZ
167496. 4 adults.
DIAGNOSIS. A variably deep-bodied, moderately robust subspecies
of variatus; with two or more variably developed zigzag black mid-
lateral stripes; without black speckling along middle sides or on
caudal peduncle. Adult males lacking short, broad, pointed caudal
appendage or sword. Vertebrae 28 or 29, rarely 30. Dorsal fin rays
9 to 14, usually 11.
REMARKS. Also examined were some of Meek's original collections
of this form from Tamaulipas and San Luis Potosi (CNHM 4449,
4461, 4471, 4486, 4502, 14360, 14361, 14435, 14436). An adult speci-
men from "30 kilometers back of Tampico" (Stoye, 1935: 52) was from
the German aquarium stock introduced first as Mollienisia formosa,
according to an identification by Rachow (1932). The first specimens
introduced into German aquaria in 1931 were all young. They were
collected by Conrad from a cattle hole near Tampico (Rachow, 1936:
145-146).
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
Wheeler and Ingle (1953) misidentified variatus as Mollienesia
formosa in a published photograph, showing a female (above) and a
male (below), that bears the caption "Male and female M. formosa."
The name dorsaliss" is an erroneous application of a manuscript
name, but apparently sufficient to fix the name on X. v. variatus, which
was the only other form mentioned in connection with this first use
of the name dorsaliss". The originally intended (manuscript) use of
the name dorsaliss" was later explained by Gordon (1933b: 6). The
fish to which this name refers are confined entirely to the Rio Cazones
basin and are not at present regarded as being subspecifically sep-
arable from other geographical populations of X. v. variatus.
Xiphophorus variatus evelynae, new subspecies
Figures 1, 4, 7, tables 3-5, 9-13
Platypoecilus variatus, Gordon, 1941b: 112-113 (Rio Tecolutla record only).
Platypoecilus variatus ssp., Gordon, 1943c: 65-67 (Rio Necaxa).
Figure 7. Holotype of Xiphophorus variatus eveliynae, n. ssp.
(UMMZ: 177306), adult male 33.0 mm in standard length.
TYPES. Holotype, an adult male 33.0 mm. in standard length
(UMMZ 177306) from station number 58 below. Allotype, an adult
female 36.3 mm. in standard length (UMMZ 177307) from same
locality.
MATERIAL EXAMINED
Mexico, Puebla
Rio Tecolutla system
52. Rio Tezcapa (= Rio Necaxa) and mouths of two small tributaries 4 mi. W
La Mesa ( Necaxa). UMMZ 124313. 20 half-grown to adults.
53. Rio Acatlan entering Vaso de Tenango near La Mesa. UMMZ 124312. 25
subadults to adults.
BULLETIN FLORIDA STATE MUSEUM
54. Vaso de Necaxa at La Mesa. UMMZ 124309. 1 mature male.
55. Old Rio Necaxa at foot of dam of Vaso de Necaxa at La Mesa. UMMZ
124303. 12 adults.
56. Tributary of Rio Necaxa draining Vaso de Tenango, ca. 2 km. E Necaxa. 18
adult males and females shipped alive to Genetics Laboratory, New York
Zoological Society; this is the basis for present living stock.
57. Rio Necaxa at Tepexic 2 mi. downstream from La Mesa. UMMZ 124317.
40 half-grown to adult.
58. Side branch of the Rio Xaltepuztla where it meets the Rio Necaxa at Tepexic.
M. and E. Gordon. 6 April 1939. UMMZ 124318. 9 subadults to adults,
excluding holotype and allotype.
59. Manantial de Tepexic entering Rio Necaxa below mouth of Rio Xaltepuztla.
UMMZ 124321. 17 half-grown to adults.
DIAGNOSIS. A moderately deep bodied and long headed subspecies
of variatus, with a relatively short caudal peduncle and long caudal
fin (tables 12 and 13). Adult males consistently with 8 to 12 slender,
close-set dusky to black vertical bars along midsides that are best
developed anteriorly. Vertebrae 28, rarely 27. Dorsal fin rays 12
or 13, rarely 11 or 14.
REMARKS. In addition to the distinguishing criteria listed in the
diagnosis, this subspecies is exceptional in lacking micromelanophore
tail patterns. Some specimens of each sex show along the sides a
speckling of small spots composed of macromelanophores. In adult
males, orange-yellow coloring is confined to the proximal half of the
caudal fin as a diffuse band whereas in X. v. variatus the entire fin
becomes pigmented.
Most of the known specimens were collected above a series of
cataracts at or around the village of Necaxa (= La Mesa), principally
in the Mexican Light and Power Company's reservoirs (vasos) and
their tributaries at elevations estimated higher than 4000 feet (Stations
52-56). The waters that at one time cascaded sharply from the pla-
teaulike reservoir area down into the gorge of the Rio Necaxa are
now deflected into large conduits that lead to the hydroelectric gen-
erators. At a small village, Tepexic, on the banks of the Rio Necaxa,
almost at the foot of the precipitous 1900-foot drop below the reservoir
area (stations 57-59), approximately 70 fish were collected in 1939,
some in manantiales near the base of the mountain, others directly in
the torrential boulder-strewn riverbed. The latter specimens are the
largest and most strikingly marked X. v. evelynae in the total sample;
the dorsal fins of adult males are greatly developed, extending posteri-
orly in two examples almost to the caudal base. In 1957 no platyfish
were taken below the falls at Tepexic in 2 days of seining, although
Mollienesia sphenops and Pseudoxiphophorus bimaculatus were
abundant in both the river and the manantiales. Many X. v. evelynae
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
were seined, however, above the falls in the vicinity of Necaxa; this
form is undoubtedly plentiful in the reservoirs. According to the
engineers of the Mexican Light and Power Company, the Rio Necaxa
valley was scoured out and locally modified by flood waters that
followed a severe hurricane during the 1940's.
Apparently variatus has not become reestablished in its former
habitat at Tepexic, although two other poeciliid species have done so.
This raises the question of the source of the original Necaxa (reservoir
area) and Tepexic populations. Did X. v. evelynae derive from a low-
land Rio Tecolutla cariatis population, or from one in a neighboring
drainage by headwater transfer? It is difficult to see how lowland
forms reaching Tepexic in the past could possibly have overcome the
1900-foot cataracts to establish a population above the falls at Necaxa;
also why some variatus have not repenetrated to Tepexic from the
lowlands.
On the other hand, derivation of X. c. evelynae from the head-
waters of an adjacent drainage at some former time could account
for the large populations in the reservoir area as well as for smaller
numbers at Tepexic that floods may have swept downstream. Though
they do not now merge, the headwaters of the Rio Necaxa and Rio
Cazones approach one another to the north in a network of small
streams and tributaries. The fish in these two systems have high
dorsal fin ray counts, low vertebral counts, and rather long caudal
fins, but the two populations differ in many relative body measure-
ments and in several micromelanophore tail patterns and macromelan-
ophore spotted patterns. It may also be asked why some of the Necaxa
fish have not again been washed downstream into Tepexic since their
habitat was disrupted. Perhaps the deflection of the larger streams
into the conduits and generators of the hydroelectric plants has effec-
tively reduced or eliminated this possibility. In this regard it would
be important to know definitely if the Mollienesia and Pseudoxipho-
phorus at present around Tepexic are recent immigrants (within the
last 10 years) and, if so, from where they came.
This subspecies is named for Mrs. Evclyn Gordon, who in 1939,
was instrumental in collecting the first known examples.
Xiphophorus milleri, new species
Catemaco Livebearer. Figures 8-10, tables 1-5, 14
TYPES. Holotype, an adult male kUMMZ 177308) 23.2 mm in
standard length (station 1 below); allotype (UMMZ 177309), an adult
female 28.5 mm in standard length (same data).
BULLETIN FLORIDA STATE MUSEUM
MATERIAL EXAMINED
Mexico, Veracruz
Rio Papaloapan system
1. A small tributary of Laguna Catemaco about 2 mi. SE Catemaco. R. R. and
M. Miller; 29 March 1957. UMMZ 177310. 241 young to adult males and
females, excluding holotype and allotype.
2. Same locality. UMMZ M59-25. 93 young to adult males and females.
4r~
Figure 8. Holotype and allotype of Xiphophorus miller, n. sp. Holotype
above, adult male (UMMZ: 177308), 23.2 mm in standard length. Allotype be-
low, adult female (UMMZ: 177309), 28.5 mm in standard length.
DIAGNOSIS. A small to medium, slender-bodied species of Xipho-
phorus (tables 4 and 5), with caudal peduncle extremely slender and
much compressed laterally; with 6 to 8 variably developed dusky
bands along sides that follow the scale rows and a deep lying ridge
of darker black pigment midventrally on caudal peduncle. Adult
males without pointed appendage on ventral rays of the caudal fin.
Dorsal fin with a broad marginal dusky band and a median row of
distinct black daFbes that are oriented parallel with the fin rays and
lie in the interradial membrane. Claw at tip of ray 5a of gonopodium
thin, one half as thick as hook on ray 3, with weakly developed or
obsolescent terminal cusp, present only in about half of the specimens;
distal serrae on ray 4p well developed, 8 to 11 in number; ramus of
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
ray 4a curved downward over blade, extending to tip of ray 4p; hook
on ray 3 long and slender. Vertebrae 28, rarely 29. Scales in a
lateral series 25 to 27, usually 26. Dorsal fin rays 9 to 11, usually 10.
Figure 9. Distal tips of the gonopodia of males of
Xiphophorus miller, n. sp. (UMMIZ: 177310).
GENERAL ACCOUNT. Among the many distinctive features of this
rather small species is the presence in slightly less than half the speci-
mens of both sexes of several different micro- and macromelanophore
markings. There are three tail patterns (composed of micromelano-
phores), a one spot similar to that of maciulatus and variatus, a large
one spot similar to that of some variatus, and a barlike pattern re-
sembling one found in montezumae and pygmaeus. Two macromel-
anophore patterns may be distinguished phenotypically, one of ir-
regular spots on the body, and another of more or less definite rows of
spots arranged along the dusky bands on the side. Similar macromela-
nophore patterns are found in montezumae, maculatus, and variatus.
According to the field notes of R. R. Miller, who collected this species,
the side and belly of adult males are deep yellow-orange (in late
March).
BULLETIN FLORIDA STATE MUSEUM
This species is of particular importance for establishing the tax-
onomic integrity of the platyfishes and swordtails. It combines traits
that in each group were hitherto used for generic separation. Body
form, pigmentation, and ecology ally it to the swordtails. It resembles
the platyfishes in certain secondary sexual characters, in the absence
of a well-developed caudal appendage, and in showing pigmentary
polymorphism. Its probable relationships are discussed below.
Xiphophorus miller, known only from the single locality listed,
is named for Dr. Robert Rush Miller who collected this and many
other forms of Xiphophorus.
Xiphophorus montezunae Jordan and Snyder
Montezuma Swordtail, figures 10-13, 30, tables 1-5, 15-19
Xiphophorus montezumae Jordan and Snyder, 1900: 131-133 (original description;
color variation; Rio Verde near Rascon, San Luis Potosi).
DIAGNOSIS. A moderate to large, rather deep-bodied species of
Xiphophorus (tables 4 and 5) with a moderately to well developed
caudal appendage that is conspicuously edged with black on its upper
and lower margins, and having a continuous solid or zigzag black mid-
lateral stripe running from base of caudal peduncle through eye to
lips. With a deep-lying ridge of black pigment midventrally on caudal
peduncle. Dorsal fin rounded, with broad marginal dusky band. Cau-
dal fin above swordlike appendage without marginal pigment. Claw
at tip of ray 5a of gonopodium thin, one half as thick as hook on ray
3, with weakly developed terminal cusp; distal serrae on ray 4p well
developed, 5 to 12 in number; ramus of ray 4a curved downward over
blade, extending to tip of ray 4p; hook on ray 3 long and slender. Ver-
tebrae 28 to 30, usually 29. Scales in a lateral series 25 to 28, usually
27 or 28. Dorsal fin rays 10 to 14, usually 11 or 12.
GENERAL ACCOUNT. This species apparently is confined to the Rio
Panuco drainage system. The typical form occurs in the headwaters
of the Rio Tamesi, and in the Panuco proper, in the headwaters of a
tributary to the Rio Tampaon, the type locality. A second form of
montezumae occurs to the south in the Rio Moctezuma. Each of these
populations is distinct in pigmentation, in shape and relative size of
the caudal appendage, and to a lesser extent in morphometric traits
(see tables 15-19 and discussion above). The form from the Rio
Moctezuma is treated here as a distinct subspecies.
This species has become well known from genetic and correlated
studies by Gordon and others on m. cortezi. This is the most southern
representative (Rio Moctezuma) and until recently the only subspecies
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
maintained alive for laboratory study. Living examples of m. monte-
zumae recently secured in Mexico have been successfully crossed
with m. cortezi. So far 200 fertile hybrids have been obtained.
Four macromelanophore patterns occur in this species, a striped
pattern, two kinds of spotting on the side, and a caudal marking.
Morphologically the cells composing these patterns resemble those
of the platyfish (see above). At least one of the montezumae patterns,
spotted caudal (Sc), produces melanotic and melanomatous growth in
montezumae x helleri hybrids.
A single micromelanophore tail pattern occurs in this species,
exclusive of pigmentation associated with the "sword." This pattern
is indistinguishable from one occurring in pygmaeus nigrensis. It is
similar to the moon pattern of maculatus, though larger and more
diffuse, and usually dusky rather than black.
Xiphophorus montezumae montezumae Jordan and Snyder
Figures 11-12, 30, tables 3-5, 15-19
Xiphophorus montezumae Jordan and Snyder, 1900: 131-1.33 (original description;
Rio Verde near Rascon); Jordan and Evermann, 1900: 3154-3156 (after Jor-
dan and Snyder); Meek, 1904: 158-159 (description; Rascon); Regan, 1906-
1908: 107 (diagnosis; Rascon); 1913: 1004 (diagnosis; relationships; Rio
Panuco).
MATERIAL EXAMINED
Mexico, Tamaulipas
Rio Panuco system
Rio Tamesi
1. Nacimiento of Rio Sabinas, about 5 mi. W N\V of El Porvenir (near km. mark
613 on Pan American Highway). 30 young to adult males and females
taken alive.
2. Nacimiento of Rio Sabinas, tributary to Rio Guayalejo, 11 mi. N Gomez
Farias. UMMZ NYZS-GL. 45 half-grown to adult males and females.
3. Pano Ayuctle, Rio Sabinas about 2 mi. W of km. mark 610 on Pan American
Highway. 1 adult male taken alive.
Mexico, San Luis Potosi
Rio Panuco system
Rio Salto
4. Pools adjacent to Rio Salto de Agua 12 km. NW of road from Antigua
Morelos to San Luis Potosi. 26 young to adult males and females taken
alive.
5. Same locality and pools as above. LSU 6262. 55 half-grown to adult males
and females.
6. Rio Verde near Rascon. SU 6145, 6146, 6194. 132 half-grown to adult
males and females, including holotype and paratypes.
7. Rascon. SU 23876. 13 subadults.
DIAGNOSIS. A short-headed subspecies of montezumae, adults
ranging from 27 to 54 mm. in standard length. Caudal appendage
BULLETIN FLORIDA STATE MUSEUM
XIPHOPORUS MO~EZ'MA
XIPOPORU M COr-I
/IHPOU PPGAU
XIPHPHORS PNIGRNS2
XIPHOPHORUS M. MONEZUAE*
Figure 10. Distribution by record stations for X. miller,
and the forms of X. montezumae and X. pygmaeus.
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
B
Figure 11. Distal tips of the gonopodia of males of the forms of X. monte-
zuimae and X. pygmaeus. A-C. X. montezumae cortezi, n. ssp. (LSU: 6261).
D. X. montezumae montezumae Jordan and Snyder (SU: 6146). E. X. pygmaeus
nigrensis, n. ssp. (UMMZ:NYZS-GL, Rio Choy). F. X. pygmaeus pygmaeus
Hubbs and Gordon (laboratory stock).
Figure 12. Xiphophorus montezumae montezumae Jordan and Snyder (LSU:
6262), from the Rio Salto de Agua. Male above, 55.0 mm, and female below,
54.0 mm in standard length. Retouched photo.
1960
BULLETIN FLORIDA STATE MUSEUM
always long, straight, and rather wide, particularly at the base. Black
markings along sides and flanks arranged in more or less definable
rows that are joined together to form a reticulum, the midlateral row
always the most prominent and developed as a zigzag line, never as a
solid band. In adult males, dorsal fin with a distinct median row of
black spots on interradial membrane. Vertebrae 28 to 30, usually 29.
REMARKS. I found this form to be most common in small spring
pools that are periodically connected with the main river at times of
high water. They occur in small numbers together with Astyanax,
Dionda, and Gambusia in the main channel.
Two macromelanophore patterns have been recorded, a striped
pattern in which the spots are confined to regions delimited by the
horizontal sections of the reticulum, especially the midlateral stripe,
and a spotted pattern in which larger irregular blotches dot the
middle side below the dorsal fin.
Xiphophorus montezumae cortezi, new subspecies
Figures 10-11, 13, 30, tables 3-5, 15-19
Figure 13. Holotype and allotype of Xiphophorus montezumae cortezi, n.
ssp. Holotype above, adult male (UMMZ: 177302), 38.5 mm in standard length.
Allotype below, adult female (UMMZ: 177303), 34.9 mm in standard length.
Xiphophorus montezumae, Gordon, 1933b: 518-522; 1934a: 107-110 (habitat;
coloration; "X. montezumae" of aquarists cleared from synonymy); Friedman
and Gordon, 1934: 446-455 (chromosomes; distribution); Gordon, 1938: 19-29
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
(history of name "montezumae" as applied to a variety of X. helleri); 1940e:
172 (habitat and ecological separation; Rio Axtla); 1941b: 112-113 (northern
and upland parts of Rio Panuco system; differential ecology; hybrids with
X. maculatus); 1942b: 196-197 (as aquarium fish; confusion of name; Rio
Axtla); 1943c: 68-69 (habitat; gonopodium; variant patterns; Rio Axtla; no
hybrids in nature); Gordon, Cohen and Nigrelli, 1943: 571 (caudal fin append-
age in light of hormone experiments); Hubbs and Gordon, 1943: 31-32 (gono-
podium; dorsal rays); Gordon and Benzer, 1945; 58, 60-61, 63-67, 69-70
(vertebrae; ribs; suspensorium; comparisons); Gordon, 1946d: 84-85 (intro-
gressive hybridization; color patterns); 1948a: 258 (melanomas and hybrids);
1950a: 406 (spotting in hybrids with X. helleri); Gordon and Rosen, 1951: 414,
417, 424-429, 432-433, 454, 456-457 (genetics and morphology of gonopodia);
Gordon and Aronowitz, 1951: 151 (sex determination unknown); Aronowitz,
Nigrelli and Gordon, 1951: 240 (tumors); Gordon, 1951c: 172, 196-198
(tumors; distribution; hybrids); Rosen and Gordon, 1953: 22-23 (gonopodial
adaptations); Gordon, 1953b: 148-150 (habitat); 1953c: 789-790 (distribution;
spotting).
Platypoecilus variatus, Peters, 1933b: 9 (in part; misidentification; lowermost figure
only).
TYPES. Holotype, an adult male (UMMZ 177302) 38.5 mm. in
standard length and allotype, an adult female (UIMMIZ 177303) 34.9
mm. in standard length, both from station 16.
MATERIAL EXAMINED
Mexico, San Luis Potosi
Rio Panuco system
Rio Moctezuma
8. Arroyo near Valles. UMMZ 124436. 3 subadults.
9. Arroyo Palitla at Palitla, 8 mi. N Tamazunchale. UMMZ 124331. 312
young to adult males and females.
10. Small arroyo between Rio Axtla and Rio Moctezuma. UMMZ NYZS-GL.
67 half-grown to adult males and females.
11. Rio Matlapa at Matlapa, 13 mi. N Tamazunchale. UMMZ 124341. 46 half-
grown to adult males and females.
12. Arroyo Matlapa. UMMZ NYZS-GL. 37 half-grown to adult males and
females.
13. Arroyo Plan de Jalpilla, 18 mi. N Tamazunchale. UMMZ 124353. 55 large
young to subadults.
14. Danubio, Rio Axtla 4 mi. upstream from Axtla. UMMZ NYZS-GL. 53
young to adult males and females.
15. Rio Axtla 2 mi. W Axtla. UMMZ 108602. 21 half-grown to adult males
and females.
16. Arroyo Matlapa at Comoca, 2 mi. N of Axtla. M. Gordon, S. Coronad, and
H. F. Gandy; 14-15 April 1939. UMMZ 124374. 128 young to adult males
and females excluding holotype and allotype.
17. Rio Axtla at Axtla. UMMZ 124364. 9 young to adult males and females.
18. Pool in river bed of Rio Axtla at Axtla. UMMZ 124174. 4 large young
to subadults.
19. Tributary of Rio Calaboza 20 mi. S Tantoyuca. UMMZ 108679. 3 half-
grown.
20. Rio de los Hules 11 mi. SW Tantoyuca. UMMZ 108678. 1 half-grown.
BULLETIN FLORIDA STATE MUSEUM
DIAGNOSIS. A long-headed subspecies of montezumae, adults vari-
able in size, averaging about 30 mm. in standard length. Caudal ap-
pendage usually short and curved upward at tip; and rather slender.
Black pigment along sides and flanks evenly arranged in a uniform
reticulum, except along midlateral line where it usually forms a prom-
inent zigzag line or a solid black band. In adult males, dorsal fin
lacking a distinct median row of black spots on interradial membrane.
Vertebrae 29, rarely 28.
REMARKS. Though never before recognized as a distinct subspe-
cies, this form is well known in the literature and has been described
many times. The above diagnosis is sufficient to set it off clearly from
the lesser known typical form.
The midlateral stripe in adult male cortezi may be rather faint,
incomplete, or lacking. Its expression is affected by the physiological
state of the animal at time of fixation, i.e., excitation versus quiescence,
and by the background shade of the environment. Some living speci-
mens maintained in the laboratory for years show heavy deposition of
guanine along the side and flank giving the fish a silvery hue. Per-
haps guanine competes with the black pigment, for in some instances
where the silvery coloring pervades the entire fish even the character-
istic reticulum may be absent.
Two macromelanophore patterns and one micromelanophore tail
pattern occur in this subspecies. One or more variable black spots
may occur on the interradial membrane of the caudal fin. In both
size and position on the fin this pattern is inconstant, though it usually
follows the fin's long axis. A spotted pattern on the upper side near
the dorsal fin may spread to the base of this fin, rarely to its distal
margin.
The name cortezi is a historical reference to the Spanish conquista-
dor, Hernando Cortez (1485-1547).
Xiphophorus pygmaeus Hubbs and Gordon
Pygmy Swordtail, figures 10-11, 14-16, 31, tables 1-5, 20-23
Xiphophorus pygmaeus Hubbs and Gordon, 1943: 31-33 (Rio Axtia).
DIAGNOSIS. A small, slender species of Xiphophorus (tables 4 and
5) with a slightly to well-developed caudal appendage, having a con-
tinuous solid black midlateral stripe running from base of caudal
peduncle through eye to lips that is separated from a series of black
to dusky blotches on the dorsum by a light, almost unpigmented area.
Without a deep-lying ridge of black pigment midventrally on caudal
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
A
Figure 14. Living laboratory born and reared breeding pairs of Xiphophorus
pygmaeus pygmaeus Hobbs and Gordon (above) and Xiphophorus pygmaeus
nigrensis, n. ssp. (below). Males are on top; note thin line of dark pigment along
the edges of their dorsal and caudal fins and their partly faded midlateral black
stripes.
BULLETIN FLORIDA STATE MUSEUM
peduncle. Dorsal fin rounded, with fine dusky margin and basal or
median row of black or dusky spots on interradial membrane in adult
males. Caudal fin above "sword" truncate, usually with a fine dusky
margin on posterior edge in adult males. Claw at tip of ray 5a of gon-
opodium thick, two times as thick as hook on ray 3, with moderately
developed terminal cusp; distal serrae on ray 4p only moderately de-
veloped, approximately 4 in number; ramus of ray 4a slightly curved
downward over blade, extending to tip of 4p but sometimes falling
slightly short; hook on ray 3 long and slender. Vertebrae 28 or 29.
Scales in a lateral series 25 to 28. Dorsal fin rays 9 to 14.
GENERAL ACCOUNT. Until recently this species was thought to
be confined to the Rio Axtla (Rio Panuco system) above the town of
Axtla, the type locality of this small swordtail (Hubbs and Gordon,
1943). In 1952 R. T. Gregg collected another distinctive form of
the pygmy swordtail in the north arm of the Rio Panuco below Valles,
a little over 100 miles by water from Axtla.
Both forms of the pygmy swordtail are restricted, for the most
part, to deep pools and the waters along undercut slopes of river and
stream beds. X. p. pygmaeus is infrequently encountered in dense
stands of submerged aquatics and along exposed shoreline in shallow
water on the slip-off slope of the Rio Axtla at Axtla. X. p. nigrensis
is known only from deep, swiftly flowing water at depths as great as
10 feet.
Xiphophorus pygmaeus nigrensis, new subspecies
Figures 10-11, 14-15, 31, tables 3-5, 20-23
Figure 15. Holotype of Xiphophorus pygmaeus nigrensis, n. ssp. (UMMZ:
177301), adult male 29.0 mm in standard length. Retouched photo.
HOLOTYPE. An adult male (UMMZ 177301) 29.0 mm standard
length from Station 1.
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
MATERIAL EXAMINED
Mexico, San Luis Potosi
Rio Panuco system
Rio Choy
1. Nacimiento del Rio Choy, 4 km. N Hotel Taninul, -3 kin. N of road from Route
110 (Valles-Tampico Highway). D. E. Rosen, M. S. Gordon, and M. Gordon;
19 January 1957. UMMZ NYZS-GL. '30 adults to half-grown including
holotype; 8 adult males and 12 adult females and young taken alive.
2. Nacimiento del Rio Choy. LSU 6265, 6388. 6 adults to half-grown.
:3. Rio Choy at Rancho Colandria 2 mi. NV Tanmuin. LSU 6264. 1 adult male.
DIAGNOSIS. A large, deep-bodied subspecies of pygmaeus, with
a moderately to well-developed caudal appendage that is set off by a
distinct black line on its ventral margin. In pregnant adult female,
a small indistinct dusky patch or weakly developed triangular black
mark on side directly above vent. Vertebrae 29. Scales in a lateral
series 25 to 28, usually 27. Dorsal fin rays 11 to 14, usually 12.
REMABKS. A large oval or rectangular black spot may occur at
the base of the caudal peduncle. In the two available samples it
appears to be normal for a single population of nigrensis to contain
adult males of quite diverse sizes (fig. 28). The smallest ones (about
20 mm) are slender, have short swords and an intensely black lateral
stripe. With increase in size (standard length) the body becomes pro-
portionally deeper, the sword longer, and a brilliant iridescent blue
develops along the side and flank, partly obscuring the midlateral
black stripe. In the largest males (30 mm plus) the lateral stripe may
not be in evidence during times of inactivity, but during courtship or
feeding it may be pronounced. The polymorphism exhibited by adult
males of p. nigrensis appears to be most closely related to factors
affecting absolute size, e.g. body growth in the vertical dimension
and sword growth are positively allometric with axial lengthening.
(See values of maximum and minimum relative depths, tables 4 and
5, and discussion below).
This distinctive form is closely allied to the pygmy swordtail on
the basis of overall pigmentation, the structure of the gonopodium of
the adult male, and ecology. In coloration it differs primarily in
having a pigmented caudal appendage and in the character of the
anal blotch in adult females, as noted in the diagnoses. Its gono-
podium cannot be clearly distinguished from that of the nominate
form. Like p. pygmaeus, it prefers deep and clear fast waters along
undercut slopes. Other characters are listed in the diagnosis and in
tables 7, 8, 21, 22, 23, and 24.
The young of p. nigrensis apparently are born larger than those
of the nominate form, and they may be slower growing. Precise data
BULLETIN FLORIDA STATE MUSEUM
on these two points are not yet available. Despite these and other
differences, the two pygmy swordtails form such a manifestly compact
group that they are treated as conspecific. On the one hand, their
differences are not greater than those that separate the subspecies of
helleri, montezumae or variatus, whereas on the other, heterotypic
matings are as readily effected as homotypic matings of p. nigrensis,
and the comparatively large numbers of young so produced (about
14 per brood as compared with 11, 9, 7, and 4) develop into vigorous
and healthy adults. The two forms are allopatric.
As discussed below, large adult males of p. nigrensis are strikingly
similar to the smaller, silvery males of m. cortezi, and in this way aid
in defining the relationships of these two swordtail species.
The name nigrensis, compounded from the Latin, niger = black,
and ensis =- sword, refers to the diagnostic ventral margin of black
pigment on the caudal appendage of adult males.
Xiphophorus pygmaeus pygmaeus Hubbs and Gordon
Figures 10-11, 14, 16, 31, tables 3-5, 20-23
Xiphophorus pygmaeus Hubbs and Gordon, 1943: 31-33 (original description;
comparisons; variation; Rio Axtla at Axtla; between Rio Axtla and Rio Mocte-
zuma); Gordon, Cohen and Nigrelli, 1943: 571 (caudal fin character in light
of hormone experiments with Platiypocciluis [=Xiphophorus] maculatus; Gor-
don and Rosen, 1951: 424-433 (gonopodium); Gordon, 1953: 148-150 (ecol-
ogy); Rosen and Gordon, 1953: 21-23 (mechanics of gonopodium).
Xiphophorus sp., Gordon, 1940e: 172; 1943c: 68 (ecological isolation; habitat).
Figure 16. Adult male of Xiphophorus pygmaeus pygmaeus Hubbs and Gordon,
22.0 mm in standard length. Wild caught specimen from the Rio Axtla (NYZS-GL).
MATERIAL EXAMINED
Mexico, San Luis Potosi
Rio Panuco system
Rio Axtla
4. Small arroyo between Rio Axtla and Rio Moctezuma. UMMZ NYZS-GL.
1 adult female.
5. Rio Matlapa at Matlapa 8 mi. N Tamazunchale. UMMZ 124:340. 1 adult
female.
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
6. Danubio, Rio Axtla, 4 mi. upstream from Axtla. UMMZ NYZS-GL. 29
young to adult males and females.
7. Rio Axtla, 2 mi. W Axtla. UMMZ 108601. 1 subadult male.
8. Rio Axtla at Axtla. UMMZ 124366, 124365. 174 young to adult males
and females, one of which, a male, is holotype.
9. Rio Axtla at Axtla. UMMZ NYZS-GL. 28 young to adult males and females.
10. Rio Axtla at ferry to Xilitla. LSU 6311. 23 half-grown to adult males and
females.
11. Rio Axtla at ferry to Xilitla. UMMZ NYZS-GL. 50 adult males and females,
some taken alive.
DIAGNOSIS. A small, slender form of pygmaeus, with a weakly
developed caudal appendage that lacks black ventral emargination.
In pregnant adult female, a well-developed subtriangular black patch
on side directly above vent. Vertebrae 28, rarely 29. Scales in a
lateral series 25 or 26, usually 26. Dorsal fin rays 9 to 12, usually
10 or 11.
REMARKS. The nominate form of the pygmy swordtail appears to
have a restricted distribution in the upper part of the Rio Panuco
system of northeastern Mexico. Most of the known specimens have
come from the Rio Axtla, tributary of the Rio Moctezuma, San Luis
Potosi. Here it was found in the same cross section of the stream
with Xiphophorus v. variatus and Xiphophorus montezumae cortezi,
but almost always in a different ecological niche.
Since, with one exception, all of the known specimens came from
the Rio Axtla at Axtla, and from an "arroyo between Rio Axtla and Rio
Moctezuma," or have been bred from stock obtained at Axtla, little
can be said regarding geographical variation. The one fish, an adult
female 40 mm. in standard length, that was taken in the Rio Matlapa
at Matlapa (misspelled Matalpa in Hubbs and Gordon, 1943), is 8 mm.
longer than any of the large series from Axtla and is not fully typical
in appearance. The black band is not set off as strongly as in most of
the specimens, though it can be matched by some of the Axtla fish.
Its low dorsal fin ray count of 10 suggests that it is correctly assigned
to this subspecies.
Like other Xiphophorus, with the probable exception of couchi-
anus, the stock of Xiphophorus pygmaenus from Axtla is polymorphic.
Golden or slaty-blue variants appear and lines of each color have been
established by breeding. The usually broad, black, rather even-edged
lateral band may be narrow and zigzag, much as in X. montezumae.
The dark blotches on the upper parts may be absent, or may connect
with the lateral band. A ventrolateral band of dusky blotches de-
velops in some fish.
BULLETIN FLORIDA STATE MUSEUM
Xiphophorus clemenciae Alvarez
Yellow Swordtail. Figures 17-19, tables 1-5, 24
Xiphophorus clemenciae Alvarez, 1959: 69-71 (original description; Rio Sarabia,
Oaxaca).
Figure 17. Topotypes of Xiphophorus clemenciae Alvarez. Above, adult
male (UMMZ: 177311), 30.0 mm in standard length, and below, an adult male
(UMMZ: M57-55) 33.9 mm. in standard length.
MATERIAL EXAMINED
Mexico, Oaxaca
Rio Coatzacoalcos system
1. Small tributary of Rio Sarabia, about 75 mi. S Acayucan (Veracruz). UMMZ
M57-55. 33 young to subadult.
2. Same locality. UMMZ 177311, 177312. 15 young to adult.
3. Stream running east-west crossed by Trans-Isthmian Highway, 5 mi. N Donaji.
UMMZ M57-56. 13 young to subadults.
DIAGNOSIS. A small to moderate, slender bodied species of Xipho-
phorus (tables 4 and 5), with a well-developed caudal appendage that
is conspicuously edged with black on its upper and lower margins;
two continuous solid red lateral stripes extending from caudal base
to opercle; a small black band passing from in front of eye forward
to lips; a well-developed dusky axillary stripe extending from ventral
margin of pectoral base posteriorly along belly to base of gonopodium
or beyond; a deep-lying ridge of black pigment midventrally on cau-
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
dal peduncle; and a group of from 3 to 6 small red or orange spots
proximally on base of caudal fin. Dorsal fin with last few rays elongate
in adult male and rounded in female and young, with a row of tri-
angular black or dusky markings along its proximal third and a diffuse
band of dusky pigment along its distal third. Caudal fin above
"sword" distinctly convex without definite marginal pigment. Claw
at tip of ray 5a of gonopodium variable, but occasionally moderately
developed and formed as in helleri; distal serrae on ray 4p moderately
developed, 5 or 6 in number; ramus of ray 4a slender, but definitely
hooked backward over bluntly pointed blade, extending slightly be-
yond tip of ray 4p; hook on ray 3 rather short and broad, sickle-shaped.
Figure 18. Distal tips of the gonopodia of males of Xiphophorus clemenciae
Alvarez (UMMZ: 177312).
BULLETIN FLORIDA STATE MUSEUM
Vertebrae 29, rarely 28. Scales in a lateral series 26 or 27, usually 27.
Dorsal fin rays 11 or 12, rarely 10 or 13.
GENERAL ACCOUNT. When first encountered in 1957 in collections
of swordtails from the Rio Coatzacoalcos system, this species was
overlooked, primarily because of its strong superficial resemblance in
size, coloration, and proportion to X. helleri strigatus. It is known
only from three collections in Oaxaca and was taken together with
the green swordtail each time. It lives sympatrically with helleri
and is certainly that species' closest relative. The two are readily
separated by gonopodial characters and dorsal fin markings.
According to field observation of Robert R. Miller, who first col-
lected this new species, its habitat appears to be typical of helleri and
other swordtail species. His records for March 1957 state the water
was clear, vegetation none, the bottom consisted of mud, silt and
brush, and the current was moderate to almost none. The stream
averaged about 4 feet wide and was bordered by a steep bank of
dense, tropical vegetation (rain forest). In February 1959 he ob-
tained many living specimens in the same stream, which was then a
torrent of significant proportions.
Alvarez (1959) reported that some of the males were bright yel-
low, a color not noted in my specimens. A male and female paratype
of clemenciae are identical to my material in all other respects.
Xiphophorus helleri Heckel
Green Swordtail, figures 19-23, 32, tables 1-5, 25-31
Xiphophorus hellerii Heckel, 1848: 291, pl. 8.
DIAGNOSIS. A large, slender bodied species of Xiphophorus (tables
4 and 5), with a well-developed caudal appendage that is conspicu-
ously edged with black on its upper and lower margins, and having
one or more continuous solid or zigzag black or red lateral stripes, the
midlateral stripe only running from base of caudal peduncle, where
it may be continuous with pigment of caudal appendage, forward
through eye to lips. With a deep-lying ridge of black pigment mid-
ventrally on caudal peduncle. Dorsal fin usually rounded, rarely with
the last few rays produced; dorsal fin in adult males only, usually with
a median and a subdistal row of subspherical black or red spots. Cau-
dal fin above "sword" without distinct marginal pigment. Claw at
tip of ray 5a of gonopodium large, as large as hook on ray 3, distinctly
scythe-shaped, the terminal cusp being as large as or larger than proxi-
mal section; distal serrae on ray 4p variably developed, usually small
and blunt, rarely obsolescent, 2 to 7 in number when typically formed;
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
ramus of ray 4a heavy, hooked sharply downward over distinctly or
bluntly pointed blade, extending to tip of ray 4a; hook on ray 3 short
and broad. Vertebrae usually 28 to 30, rarely 31. Scales in a lateral
series 25 to 29, usually 26 or 27. Dorsal fin rays 11 to 17, usually
12 to 14.
KIPHOPHORUS H. HELLER/ I
K1PHOPhORUS H. STRIGATUS
XIPHCPHORUS H. GUENTHERI 0
X/PHOPHORUS H. ALVAREZI 0
XIPHOPHORUS CLEMENCIAE 0
Figure 19. Distribution by record stations for X. clemenciae and the forms
of X. helleri. See text for a discussion of the Orizaba swordtail. The Stann
Creek record is from Regan (1907).
-7
BULLETIN FLORIDA STATE MUSEUM
/ A U
Figure 20. Distal tips of the gonopodia of males of the subspecies of X.
helleri. A. X. helleri helleri Heckel (UMMZ: 97579). B. X. helleri strigatus
Regan (laboratory stock, Rio Papaloapan). C. X. helleri guentheri Jordan and
Evermann (UMMZ: 97871). D. X. helleri alvarezi, n. ssp. (UMMZ: 177304).
GENERAL ACCOUNT. The green swordtail has the most extensive
natural range of any species of Xiphophorus, from the Rio Nautla,
Veracruz southward to the independent Atlantic coastal tributaries
of northwestern Honduras, a straight line distance of almost 800 miles.
Within the separate drainage areas of this extensive natural range,
the species breaks up into a great number of morphologically distinct
geographic and microgeographic races. Early recognition of the
green swordtail's great geographic variation led some workers to
describe many of the river populations as distinct species (see synony-
my). Hubbs (1935), Martin del Campo (1938) and Hubbs and Gordon
(1943) are chiefly responsible for indicating the conspecificity of all
these forms. These authors recognized foul subspecies: X. helleri
helleri, X. h. strigatus, X. h. guentheri and X. h. brevis.
A collective view of all the geographic races of X. helleri clearly
shows a north-south oriented morphocline involving a constellation
of traits. In the north, for example, the swordtails are slender with
short heads, intensely black lateral striping and a preponderance of
black in the dorsal fin, relatively short caudal fins (excluding the
appendage or "sword"), few distal serrae on ray 4p of their gonopodia
(the distal serrae may even be obsolescent), relatively high scale count,
and relatively low dorsal fin ray count. In the intermediate popula-
tions the fish are somewhat chunkier with longer heads, have less
black and more red in their dorsal fins and in the striping along the
sides, their caudal fins (excluding "sword') are relatively longer, the
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
distal serrae in the gonopodium are larger and more numerous (they
are never degenerate), the scale counts are slightly lower and dorsal
fin ray counts higher. The southernmost populations are generally
quite robust with a long head, intense red but little black pigment
on the side and dorsal fin, long caudal fin (excluding "sword"), distal
serrae at the tip of the gonopodium still further exaggerated and more
numerous, the scale count lowest and dorsal fin ray count highest of
any of the green swordtail populations.
Yet within each of the northern, intermediate, and southern groups,
the green swordtails of any particular river basin are usually distinct
and characteristic of only that drainage. Descendants of fish from
six major drainage areas, Jamapa, Papaloapan, Coatzacoalcos, Grijalva,
Belize (British Honduras), and Ulua (Honduras), have been main-
tained under laboratory conditions for a considerable period (table
25). The members of each river population not only maintain their
own morphological characteristics (table 26), but can always be cor-
rectly assigned to their appropriate geographical race entirely by
their pigment patterns (fig. 21).
Despite the clinal change from north to south and the great num-
ber of distinct geographical populations within this species, three
groups with relatively abrupt transitions between them can be de-
fined. These three groups correspond exactly with the northern,
intermediate, and southern character complexes indicated above.
The members of each of these major allopatric groups may be desig-
nated as subspecies, as follows:
Xiphophorus helleri helleri Mexico: Rio Nautla, Rio Antigua,
Rio Chachalacas, Rio Jamapa.
Xiphophorus helleri strigatus -- Mexico: Rio Papaloapan, Rio
Coatzacoalcos.
Xiphophorus helleri guentheri -- Mexico: Laguna Sontecoma-
pan, Rio Tonala, Rio Grijalva, Rio Usumacinta (plus tributaries and
interior lakes of Peten, Guatemala). British Honduras: Rio Hondo,
Rio Belize, Rio Grande, Stann Creek. Guatemala: Rio Polochic,
Rio Motagua. Honduras: Rio Ulua, independent Atlantic tributaries.
A fourth group, from the Rio Santo Domingo in northwestern
Chiapas (Usumacinta basin), described below as a new subspecies,
X. helleri alvarezi, constitutes an important exception to the well-
defined step cline already discussed.
1960
/
- ~y.
.,: ..t -
win.
,I
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
MORPHOMETRIC DIFFERENCES
DORSAL FIN RAY COUNTS. Table 27 shows the distribution of
counts of numbers of dorsal fin rays in 37 natural populations of
X. helleri. These data permit separation of the populations into two
major groups: 1) those with 11 to 14 dorsal rays (modal number 12),
distributed continuously from Nautla southward to Minatitlan, and
2) those with 11 to 17 dorsal fin rays (modal number 14) ranging con-
tinuously from Tonala southeastward to Ceiba, Honduras. The rela-
tively sharp transition from low to high fin ray counts occurs in fishes
separated geographically at the Isthmus of Tehuantepec by approxi-
mately 20 miles of coastline (between the Rio Coatzacoalcos and Rio
Tonala). Here two subspecies are separated geographically, X.
helleri strigatus (Rio Papaloapan and Rio Coatzacoalcos) and X. helleri
guentheri (Rio Tonala to Ceiba, Honduras). Table 27 reveals no
obvious quantitative difference between X. helleri helleri, the most
northern subspecies, and X. helleri strigatus, but it does call attention
to the disparate counts of the Rio Santo Domingo (Comitan) sword-
tails within the southern complex. The Comitan swordtail occurs
in the extreme upper reaches of the Rio Usumacinta at an elevation
of about 1500 feet, and is thus isolated within the principal geographi-
cal limits of the southern form. Its fin ray count is exceptionally low
(11 and 12, usually 11) considering its proximity to the southern forms
with high modal counts.
To test the significance of the difference in mean dorsal fin ray
number in natural populations of this swordtail, an estimate of the
standard error of the difference between the means of two samples
was obtained.2 Certain related assumptions were made in setting up
the raw data for analysis. It was assumed that 1) with respect to the
number of dorsal fin rays swordtail samples from separate collecting
2 The calculation of a value of d/-a- is designed to test the hypothesis that
two samples were drawn from one population or from populations with equal
means and variances.
Figure 21. Living laboratory born and reared breeding pairs of three sub-
species of X. helleri. A. X. helleri helleri Heekel from the Rio Jamapa; male
above. B. X. helleri strigatus Regan from the Rio Papaloapan; male below. C.
X. helleri guentheri Jordan and Evermann from the Belize River, British Hon-
duras; male below. Note the progressive loss of black pigment along the sides
from A to C. The midlateral stripe in the male in B is actually dark red or brown
due to the combination of erythrophores and melanophores. In C, the multiple
stripes are orange-red. Note vertical barring in the male and the black caudal
emargination in the female in B.
BULLETIN FLORIDA STATE MUSEUM
stations within a single drainage area are sufficiently alike that they may
be treated as microgeographic races of a single, distinct geographic
race, and that 2) the collective members of all stations within a single
river basin represent distinct biological units that differ genetically
from other such units in adjacent river basins. These assumptions
were based partly upon observations of the striking similarity of the
swordtails from all stations within a single system in overall colora-
tion and body form. All members of local populations of a single
drainage area are now or have recently been in close enough contact
to inhibit significant local variation.
These assumptions are probably valid for the six drainage areas
subsumed by the two northern subspecies, X. h. helleri and X. h.
strigatus. Southward the situation is less clear. Subsamples from
the Rio Usumacinta system (excluding the Rio Grijalva) show a grad-
ual increase in mean fin ray number with eastward penetration of
headwater areas. From the upper Rio Usumacinta in Chiapas to the
streams and lakes of the Peten District of Guatemala, the mean num-
ber of dorsal fin rays rises from 12.67 to 14.00, a significant increase
of at least one fin ray. Similarly in the swordtails from the Atlantic
coastal rivers and streams of Honduras, from the Rio Ulua eastward
to Estero Salada, the mean number of dorsal fin rays rises from 13.34
to 15.33, a significant increase of more than one fin ray. If, however,
a single mean is obtained for the four subsamples from the Rio Usuma-
cinta system and a single mean is obtained for the five subsamples
from Honduras, the overall mean dorsal fin ray counts of the Rio
Usumacinta and Honduras swordtails differ by about one one-
hundredth, 13.76 and 13.77 respectively. The probability that this
small difference between the two mean values represents samples
drawn from different populations is inconsiderable, d/c = 0.11,
where the criterion of significance is taken as 3. The statistical trends
noted within these two drainage areas may indicate the effects of
gradually changing east-west oriented ecogeographical factors. De-
spite this variability in dorsal fin ray number in the helleri popula-
tions from the Rio Usumacinta and from Honduras, their mean values
are uniformly higher than those of populations north of the Isthmus
of Tehuantepec.
Comparisons of the southern members of X. helleri strigatus (Rio
Coatzacoalcos swordtail) with six of the seven natural populations of
X. helleri guentheri yield d/od values ranging from 9.12 to 20.24. A
comparison of the Rio Grijalva population (X. h. guentheri, in part)
with the Rio Santo Domingo population (X. h. alvarezi) gives the
high value of 15.18.
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
LATERAL SCALE COUNTS. Table 28 shows the distribution of counts
of scales in lateral series in 23 natural populations of X. helleri. These
fall into four groups as follows: 1) fish whose scales range from 26 to
29 (modal number 28), from the Rio Nautla, Antigua and Jamapa
(X. h. helleri); 2) those whose scales range from 26 to 28 (modal num-
ber 27), from the Rio Papaloapan and Coatzacoalcos (X. h. strigatus);
3) those whose scales range from 25 to 28 (modal number 26), from
the Rio Tonala to Ceiba, Honduras (X. h. guentheri); 4) fish whose
scales range from 26 to 29 (modal number 27), from the Rio Santo
Domingo (X. h. alvarezi). As in the dorsal fin ray counts, the change
from a lower to a higher mean value of lateral scales is most abrupt
between X. h. strigatus and X. h. guentheri where they are geographic-
ally separated at the Isthmus of Tehuantepec, and between X. h.
guentheri and X. h. alvarezi in the upper reaches of the Rio Usuma-
cinta basin. The separation of X. h. helleri from X. h. strigatus is
poorly defined by both lateral scale and doisal fin ray counts, par-
ticularly as scale counts of fish from Boca del Rio (Rio Jamapa sys-
tem) appear to be intermediate between the typical mean values for
populations north and south of this station.
VERTEBRAL NUMBER. Complementing the dorsal fin ray and scale
counts that separate h. strigatus from the southern forms, vertebral
number and the following study of caudal fin length offer a means of
distinguishing between h. strigatus and the northern h. helleri. The
few X-ray data on vertebral number available show that h. strigatus
from the Rio Coatzacoalcos has one more vertebra than h. helleri
from the Nautla, Antigua, and Jamapa Rivers (table 3). Between
strigatus, guentheri, and alvarezi no difference in vertebral count is
apparent, even in modal number.
RELATIVE LENGTH OF CAUDAL FIN. Relative length of the caudal
fin was determined in 31 populations of X. helleri. As before, these
data are grouped into drainage systems with a single mean for each
(table 29). These relative lengths clearly define the separation be-
tween the two northern subspecies, X. h. helleri and X. h. strigatus.
Fish from the Rio Nautla, Rio Chachalacas and Rio Jamapa have rel-
atively truncate caudal fins (2.9 to 4.5); those from the Rio Papaloapan
and Coatzacoalcos have relatively longer caudal fins in proportion to
body length (2.8 to 3.7). On the basis of the combined indices for
adult males, females, and subadult fish, no clear separation can be
made for the samples of strigatus, guentheri, and alvarezi, though the
increase in low values first appearing in fishes from the Rio Grijalva
suggests a possible separation between strigatus and guentheri at
this point.
1960
BULLETIN FLORIDA STATE MUSEUM
RELATIVE BODY MEASUREMENTS. Relative body measurements may
be used further to define the separation of X. helleri into its four com-
ponent subspecies. Many of the difficulties that beset Gordon and
Gordon (1954) in their use of such measurements in their biometric
studies of X. maculatus are met with here. For reasons discussed be-
low only index values derived from adult males may be safely em-
ployed in intraspecific comparisons. In support of arguments based
upon more precise data for the formal splitting of X. helleri into four
subspecies groups, the data summarized, in part, in tables 4 and 5
are generalized as follows:
1) X. h. helleri may be separated from both strigatus and guentheri
by relative depth of body; strigatus and guentheri are indistinguish-
able on this basis, though the latter is readily separable from alvarezi.
Both helleri and alvarezi are slender fish with greatest body depth
only 25 percent of standard length. Both strigatus and guentheri are
stouter with the greatest body depth about 30 percent of standard
length.
2) X. h. helleri may be separated from both strigatus and guentheri
by relative depth of caudal peduncle; strigatus and guentheri are in-
distinguishable using these measurements, but alvarezi is readily
separable from guentheri. As in relative body depth, caudal peduncle
depth shows both helleri and alvarezi to be slender forms and strigatus
and guentheri to be considerably heavier bodied.
3) The four subspecies do not differ greatly in their relative head
lengths, although the more slender forms, helleri and alvarezi, appear
to have slightly longer heads.
4) An apparent correlation with overall slenderness of form is again
suggested by the relatively greater caudal peduncular lengths of
helleri and alvarezi, as compared with those of strigatus and guentheri.
GENITALIC DIFFERENCES. Differences in the form and frequency
of the bony details at the tip of the adult male's gonopodium may be
used to separate X. helleri into its four component subspecies (table
30). For example, nominate helleri has a low distal serrae count
(ray 4p) (3.10 and 3.50), strigatus a somewhat higher count (3.67), and
guentheri the highest count (3.89, 3.91, 4.47 and 5). The two adult
males of alvarezi at hand both show the low distal serrae count of 3.
The subterminal segment count (ray 3) of nominate helleri is low (5.50
and 5.60), that of strigatus intermediate (5.67 and 6.67), and that of
guentheri high (6.55, 6.60, 8.26 and 10). The two alvarezi males show
the relatively high count of 8.
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
A form analysis of the distal serrae on ray 4p of the gonopodium
permits a relatively sharp separation of the two northern subspecies.
Adult males of nominate helleri show small, poorly developed serrae,
some of which may be degenerate as in the sample from the Rio
Antigua system. Adult males of strigatus show taller, more strongly
developed serrae that rarely are degenerate or anomalous. In gen-
eral, the distal serrae seem the most strongly developed in the adult
males of guentheri; in the two adult males of alcarczi, they are rather
small, as in the Rio Antigua and Jamapa samples of the nominate form.
ECOGEOGRAPHIC DIFFERENCES. A survey of the characters used to
separate the subspecies of X. helleri shows that, with the possible ex-
ception of alvarezi, the subspecies differ from each other in a regular
manner when samples are arranged according to north-south geo-
graphic position. The absence of any important mosaic distributions
of morphological traits within the range of this species and the steady
north-south change in the number and or size or other characters
studied suggests that all the geographic and microgeographic races
of X. helleri are related within a single morphocline. It has been
shown that sudden transitions between character complexes, marking
morphological stepelines, occur at two geographic points in the spe-
cies' range. These zones of transition may be taken as geographic
limits of definable taxonomic units. It has further been shown that
the possibility for gene exchange between members of adjacent drain-
age areas is remote. Information gained by maintaining stocks of six
natural populations of X. helleri under uniform laboratory conditions
and studying various laboratory produced interriver hybrids (tables
25 and 26) shows conclusively that the members of each population
are relatively homogeneous, that the different populations are genetic-
ally distinct from one another, and that the differences are therefore
not directly induced by the environment.
Table 33, which gives estimated altitudes for all stations where
Xiphophorus has been collected, reveals that the samples of helleri
and alcarezi, both slender forms, occur almost entirely at higher alti-
tudes (from 1000 to 5000 feet) and samples of more heavy-bodied
strigatus and guentheri at lower altitudes (mostly from 50 to 800 feet).
There are exceptions: strigatus from Orizaba occurs at 4200 feet; helleri
from the Rio Jamapa system at Boca del Rio at 50 feet; and some
guentheri from the Rio Motagua and Rio Belize at about 2000 and
1000 feet elevations respectively. A few of the Motagua and Belize
fish are noteworthy for their relatively slender bodies, exceptional
pigmentation, and apparently low dorsal ray counts.
BULLETIN FLORIDA STATE MUSEUM
Thus the effects of increasing altitude are similar to those of in-
creasing latitude, as shown by the variations in dorsal fin ray and
lateral scale counts above. This suggests the possibility that tempera-
ture differentials may in part have produced the effects noted (Ap-
pendix I). An analysis of the relationships among four relative body
dimensions suggests that temperature effects may be quite general in
the green swordtail. Not only general body form and metamerism, but
probably pigmentation and, to some extent, configurations of gono-
podial elements may become involved. It now seems likely that dur-
ing the swordtail's course of speciation under geographical isolation,
many of what may originally have been ecophenotypic variations be-
came incorporated into the genotypes of these animals (see Wad-
dington, 1957: 162-187, and discussion below).
Ecological effects other than climatically induced variations may
significantly alter the expressions of certain characters in this species.
Green swordtails found in a lily pond at Lancetilla, Honduras, in
which competition for food apparently was minimal, grew to an enor-
mous size, were extremely deep bodied, more intensely blue, and had
a higher dorsal fin ray count than swordtails taken from adjacent
streams (fig. 22). Swordtails taken from the upper Rio Usumacinta
system (station 52), on the other hand, were extremely small and
slender (28 to 32 mm standard length of adults), darkly pigmented,
and had exceptionally short "swords." Although the environmental
conditions in which these fish were living are unknown, possibly their
comparatively small size resulted from a density factor, either crowd-
ing or too keen competition for food. Champy (1924) indicated that
similar proportional changes in body form may be produced experi-
mentally in this species by enforced starvation. These changes may
be regarded as ontogenetic in contrast to the genetically fixed propor-
tional differences discussed above.
Xiphophorus helleri helleri Heckel
Figures 19-21, 32, tables 3-5, 27-30
Xiphophorus hellerii Heckel, 1848: 291, Pl. 8 (original description); Guenther,
1866: 349-350 (description, in part, var. f/ only); Jordan and Evermann, 1896:
70 (after Guenther); Gordon, 1935c: 50-51 (habitat; figure of fish inaccurate);
1941b: 112 (Rio Jamapa); 1943c: 64, 68-71 (type species; status of typical
subspecies; no hybrids in nature).
Xiphophorus helleri, Meek, 1904: 157 (Cordova record only); Regan, 1907: 109
(synonymy; description; Jalapa, Orizaba); 1913: 1004-1005 (in part; specimens
with axillary stripes only); Schauenberg, 1920: 260 (plains near Veracruz
and in high mountain streams; identification by locality); De Buen, 1940:
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
44 (Jalapa record); Turner, 1940: 64-67 (fetal membranes of Cordova speci-
mens).
Xiphophorus hellerii hellerii, Hubbs and Gordon, 1943: 32 (variation, dorsal
rays).
Xiphophorus jalapae Meek, 1902: 107 (original description; Jalapa); 1904: 156-
157, pl. 9 (description; Jalapa).
Xiphophorus strigatus Regan, 1907: 65 (original diagnosis, in part, not types as
hereby restricted to Rio Papaloapan and Rio Coatzacoalcos specimens; Vera-
cruz [Cordova] record only); 1906-1908: 107-108, pl. 14, fig. 7 (diagnosis, in
part; Cordova records and references only).
MATERIAL EXAMINED
Mexico, Veracruz
Rio Nautla system
1. Six mi. from Tlapacoyan on road to Nautla. UMMZ NYZS-GL. 15 young
to adults.
2. Stream 2.3 mi. W Ixtacocoa, 3.9 mi. W Martinez de la Torre, on Teziutlan-
Casita Highway. UMMZ 167487. 18 young to adult.
Rio Chachalacas system
3. L'Encero near Jalapa. UMMZ NYZS-GL. 249 young to adults.
4. Pond at L'Encero 10 mi. SE Jalapa. UMMZ 124288. 93 half-grown to
adults; others taken alive.
5. Tributary of Rio Chachalacas at Plan del Rio. UMMZ 108677. 12 half-
grown to adults.
6. Pond 12 mi. E Jalapa. UMMZ 157650. 210 half-grown to adults.
7. Pond 14 mi. E Jalapa. UMMZ 157647. 140 half-grown to adults.
8. Tributary of Rio Actopan, Jalapa. UMMZ 162464. 10 adult males and
females.
Rio Antigua system
9. Jalapa. UMMZ 66266. 1 adult male.
10. Rio Bejucas 7 mi. S Jalapa. UMMZ 97578. 125 half-grown to adults.
11. Artificial laguna at headwaters of Rio Grande, Jalapa. UMMZ 167701. 131
half-grown to adults.
12. Jalapa. SU 21452, 22338. 18 young and adults. These are topotypes and
possibly paratypes of Meek's X. jalapae.
Rio Jamapa system
13. Rio Chico, on southern edge of Cordoba. UMMZ 108613. 135 half-grown
to adults preserved; others taken alive.
14. Cordoba. USNM 55810. 7 young to adults.
15. Rio Atoyac 4 mi. N Hacienda Portrero Viejo, Paraje Nuevo. UMMZ 162145.
15 young to adults.
16. Rio along highway to Cordoba 7.5 mi. from S shore of Rio Jamapa at Boca
del Rio. UMMZ NYZS-GL. 5 adults.
DIAGNOSIS. A long headed, slender bodied subspecies of helleri.
Midlateral black or dusky stripe strongly developed as a solid band,
rarely as a zigzag line; with a variously developed secondary or
axillary stripe on the lower side behind and below the pectoral fin
base of adult males. Dorsal fin with a median and subdistal row of
red or black spots. Caudal fin excluding "sword" truncate; caudal
BULLETIN FLORIDA STATE MUSEUM
fin of adult females rarely with ventral margin of black pigment. Dis-
tal serrae on ray 4a of gonopodium small, blunt, subtriangular in out-
line, frequently obsolescent, 2 to 4 in number. Vertebrae 30 or 31.
Scales in a lateral series 26 to 29, usually 27 or 28. Dorsal fin rays
11 to 14, rarely 14.
REMARKS. Stations 9, 10, and 11 are provisionally assigned to the
Rio Antigua system despite poor locality records. The adult male
swordtails from these collections show the exaggerated axillary streak
that is not found so well developed to the south.
A series of living specimens of Cordoba origin was maintained by
C. L. Turner for several years at Northwestern University and subse-
quently has been continued at the University of Michigan Museum of
Zoology and at the Genetics Laboratory of the New York Aquarium.
The Rio Bejucos specimens from Station 8 represent the form
called X. jalapae by Meek (1902, 1904).
Heckel wrote that his species was taken "in einem klaren Bache
des Gebirges Orizaba in Mexico." This northern mountain form of
helleri with the axillary stripe was taken by Meek and by Myron
Gordon only in the Rio Nautla, Rio Antigua, Rio Chachalacas, and Rio
Jamapa systems and by Turner at Cordoba. Another subspecies
(strigatus) is also known from near Cordoba, but to the south near
the town of Orizaba, and in waters that drain south into the Rio
Papaloapan rather than northward. Specimens in the U. S. National
Museum collected long ago by Professor F. Sumichrast in Mexico
(USNM 7816) and labelled "Orizaba" appear to represent nominate
helleri, from the north. Of two developed males, one retains a dis-
tinct second or axillary band and the other shows a faint trace of
one (an undeveloped male and 12 females show no second band).
They are all rather slender.
Xiphophorus helleri strigatus Regan
Figures 19-21, tables 3-5, 27-30
Xiphophorus helleri, Meek, 1904: 157 (record from Sanborn, Rio Coatzacoalcos
system); Regan, 1906-1908: 108 (record from Sanborn, Rio Coatzacoalcos sys-
tem); 1913b: 1004-1005 (material and synonymy, in part); De Buen, 1940:
44 (records, except Jalapa).
Xiphophorus hellerii, Gordon, 1941a: 38-39 (material, in part; genetic color
factors); 1941b: 112-113 (Rio Coatzacoalcos record only; abstract on hybridi-
zation); 1942a: 73-74, fig. 1 (genetics); 1942b: 197 (in part, confusion with
X. montezumae); 1942c: 76 (genetics of hybrid mating); 1943b: 28 genetics
in relation to melanomas); 1943c: 68-71 (in part; variations; no hybrids in
nature; origin of domesticated races); Gordon and Flathman, 1943: 9-12,
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
figs. 1-3 (genetics of melanomas in hybrids); Gordon, Cohen and Nigrelli,
1943: 571 (caudal fin characters in light of hormone experiments with X.
maculatus).
Xiphophorus helleri helleri, Martin del Campo, 1938: 226 (Laguna Catemaco,
Veracruz).
Xiphophorus brevis, Arnold, 1914: 125-126, fig. 3 (identification of figure of type
later misidentified as X. montezumae, from near Coatzacoalcos).
Xiphophorus hellerii brevis, Hubbs and Gordon, 1943: 32 (material, in part; vari-
ation, dorsal fin rays).
Xiphophorus strigatus Regan, 1907: 65 hellerii of Meek a synonym; original diag-
nosis; Veracruz and Oaxaca); 1906-1908: 107 (synonymy, in part; description;
Otapa; Motzorongo; Refugio; El Hule [= Papaloapan]; Obispo; Oaxaca).
Xiphophorus helleri strigatus, De Buen, 1940 (after Regan).
Xiphophorus hellerii strigatus Hubbs and Gordon, 1943: 32 (variation, dorsal
rays).
MATERIAL EXAMINED
Mexico, Veracruz, Oaxaca
Rio Papaloapan system
17. Southern tributary of Rio Tonto 10 km. upstream from Papaloapan. UMMZ
108585. 59 young to half-grown and one adult female; others taken alive.
18. Same locality. UMMZ 124177. 5 young to adults.
19. Near Arroyo Zacatispan at its headwaters in the hills of San Bartolo. UMMZ
124271. 33 young to subadult specimens preserved; others taken alive.
20. Same locality. UMMZ 124472. 4 adults.
21. Lagoon beside Rio Papaloapan, 3 km. N San Bartolo. UMMZ 124253. 7
adults.
22. Pool 5 km. S Papaloapan, at km. mark 1481/2 on railroad. UMMZ 124282.
41 half-grown to adults.
23. Lagoon and pool 4 km. S Papaloapan at km. mark 149 on railroad. UMMZ
124236. 60 half-grown to adults.
24. Arroyo Zacatispan and lagoon 4.5 kmn. S Papaloapan. UMMZ 124217. 19
half-grown to adults.
25. Arroyo Zacatispan 6 km. SE Papaloapan. UMMZ 124194. 4 half-grown
to subadults.
26a. Obispo. SU 21463. 23 young to adults.
26b. Laguna Encantada, about 2 mi. by road and trail E SE San Andres Tuxtla.
UMMZ M59-23. 155 young to adult.
26c. Tributary of Laguna Catemaco, on W side about 1.5 mi. S Catemaco. UMMZ
M57-58. 66 young to adult.
26d. Same locality. UMMZ M59-25. 42 young to adult.
26e. Laguna Catemaco at Playa Azul, about 2 mi. E Catemaco. UMMZ M57-57.
21 young to adult.
Rio Coatzacoalcos system
27a. Small tributary of Rio Sarabia about 75 mi. S Acayucan (Veracruz). UMMZ
M57-55. 42 young to adult.
27b. Same locality. UMMZ M59-21. 1 adult male.
27c. Stream running east-west, crossed by Trans-Isthmian highway, 5 mi. N
Donaji. UMMZ M57-56. 6 young to subadult.
27d. Arrovo de "La Llorona" about 350 m. from Estacion de Juile. UMMZ
157660. 6 subadult to adults.
1960
BULLETIN FLORIDA STATE MUSEUM
28. Small unnamed arroyo emptying into Arroyo del Aguagate near Estacion de
Almagres. UMMZ 157658. 6 half-grown to adults.
29. Arroyo de Niscalapa, 1 km. S Estacion de Ojapa. UMMZ 157664. 1 half-
grown and 1 adult male.
30. Arroyo beneath bridge 30a of same railroad about 400 m. N Estacion Hibueras.
UMMZ 157665. 33 half-grown to adults.
31. Acantarilla on railroad 1 km. SW Estacion Hibueras. UMMZ 157659. 8
subadults.
32. Pool along railroad from Jesus Carranza to Veracruz. UMMZ NYZS-GL.
1 subadult male.
33. Arroyo Buena Vista tributary to Rio Uxpanapa 20 km. SE Minatitlan. UMMZ
NYZS-GL. 1 adult female.
34. Under and nearby Culvert "G322-57" on railroad from Jesus Carranza to
Veracruz. UMMZ NYZS-GL. 4 half-grown to subadults.
35. Pool under Culvert "G3T9-76" on railroad from Jesus Carranza to Veracruz.
UMMZ NYZS-GL. 45 young to adults.
36. Arroyo Ojillal tributary to Rio Hondo, Veracruz. UMMZ NYZS-GL. 7 half-
grown to adults.
37. Arroyo Chapala 1%/ km. W Minatitlan. UMMZ NYZS-GL. 7 half-grown to
adults.
38. Rio Viejo about 2 km. E Jesus Carranza, tributary to Rio Jaltepec. UMMZ
NYZS-GL. 3 half-grown to adults.
39. Laguna about % km. NW of Jesus Carranza. UMMZ NYZS-GL. 6 adults.
40. Arroyo running into Arroyo Santa Lucrecia tributary to Rio Jaltepec about
700 m. from river. UMMZ NYZS-GL. 20 young to adults.
41. Arroyo del Aguagate at village of Almagrez. UMMZ NYZS-GL. 23 young
to adults.
42. Arroyo along railorad from Jesus Carranza to Veracruz, tributary to Rio Jalte-
pec. UMMZ NYZS-GL. 22 young to adults.
DIAGNOSIS. A short headed, moderately deep bodied subspecies
of helleri. Midlateral stripe brown or brownish red (dusky in alcohol),
usually developed as a zigzag line, rarely as a solid band; rarely an
axillary stripe in adult males. Dorsal fin with a median and subdistal
row of red or dusky spots, or without definitive spotting. Caudal fin
excluding "sword" long and broad; caudal fin of adult females com-
monly with ventral margin of black pigment. Distal serrae on ray 4a
of gonopodium moderately developed, sharply or bluntly pointed,
higher than wide with a distinct notch separating base from shaft,
never obsolescent, 3 to 5 in number. Vertebrae 28 or 29. Scales in a
lateral series 26 to 28, usually 27 or 28. Dorsal fin rays 11 to 14,
usually 12 or 13.
REMARKS. A spotted form of this swordtail has been found in the
Rio Papaloapan system. Of 14 fish from Motzorongo 4 females are
spotted; of 452 specimens collected at Papaloapan and vicinity 2 fe-
males and 1 male are spotted; of 27 specimens from Obispo 1 female
and 5 males are spotted. This spotting is produced by large black
pigment cells or macromelanophores.
Vol. 5
1960 ROSEN: FISHES OF THE GENUS XIPHOPHORUS 121
In a tributary of the Rio Sarabia (Rio Coatzacoalcos system), Oax-
aca, (stations 27, a, b) R. R. and M. Miller collected a small series of
X. helleri farther upstream than any of the above specimens. In all
ways typical of the swordtails from this drainage, they are especially
interesting because they occur in true sympatricity with the very
similar and closely allied X. clemenciae.
Although many of the swordtails from the Papaloapan and Coatz-
acoalcos rivers show the black or dusky vertical barring that gives
them their subspecific name, and other features listed in the diagnosis,
the living laboratory populations from each river system have a num-
ber of distinctive pigment patterns, many of which, unfortunately, are
lost in preservation.
In a laboratory stock of the Rio Papaloapan swordtails the mid-
lateral stripe may be broad posteriorly near the tail base, but it tapers
to a very thin line of pigment on the precaudal area. It is never in-
tensely black, usually brownish or reddish. It is composed of both
melanophores and erythrophores. Vertical bars, composed of dermal
and subdermal concentrations of micromelanophores, are especially
numerous and well-developed in males; they may number as many
as eight. Dorsal fin spotting consists largely of two or more regular
rows of brilliant red dots sometimes surrounded by a few fin melano-
phores. Adult females show all of these pigmentary characteristics,
but to a lesser extent. In addition, adult females from this and the
Rio Coatzacoalcos system may show a thin line of black pigment on
the lower margin of their caudal fins, a trait found only rarely in the
northern and southern green swordtails. The underlying ground color-
ing of these swordtails is brownish olive with a pale iridescent green
on the caudal peduncle. The interradial membrane of the sword of
adult males is usually bright yellow.
In general coloration the present laboratory stock of the Rio Coatz-
acoalcos green swordtails resembles the above. The adult males
differ principally in having a golden band directly above the mid-
lateral stripe, fewer vertical bars in some instances, and faint dorsal
spotting. The midlateral stripe of the Rio Coatzacoalcos swordtails
tends to be redder posteriorly than in the Rio Papaloapan swordtails,
and some adult females may show a somewhat less developed black
caudal emargination.
The swordtails from a coastal station (43a) between the Papaloa-
pan and Coatzacoalcos rivers are perplexing in that they show definite
affinities to the southern guentheri. They have, for example, a se-
ries of from 3 to 5 distinct red bands that follow the scale rows and
they are all rather deep-bodied (table 31). Their dorsal fin ray and
BULLETIN FLORIDA STATE MUSEUM
lateral scale counts fall within the range of those of strigatus. Speci-
mens from farther south in the Rio del Azufre (Rio Grijalva drainage,
station 44a) are not separable by measurements from guentheri from
that general region (table 31), but are noteworthy because each indi-
vidual possesses both the midlateral black stripe of strigatus and the
multiple red stripes of guentheri. The taxonomic status of these two
populations is uncertain. This suggests that strigatus and guentheri
may have come together secondarily following a period of geographi-
cal isolation. If so, movement of guentheri northward was probably
accomplished coastally as indicated by the apparently restricted dis-
tribution of the deep-bodied, red-striped form between the Papaloapan
and Coatzacoalcos basins in the Laguna de Sontecomapan. These
lowland swordtails are strikingly different from those occurring close
by at higher elevations in the Laguna Catemaco and Laguna Encan-
tada, particularly from the latter which possess not only a black mid-
lateral stripe but a faint trace of dusky axillary pigment. The sword-
tails from stations 43a and 44a are tentatively treated as members of
the X. h. guentheri complex.
Xiphophorus helleri guentheri Jordon and Evermann
Figures 19-22, 32, tables 3-5, 27-31
Xiphophorus helleri, Steindachner, 1863: 184 (coloration; "Gebirgsbachen bei
Tepeaca" = "bei Teapa an der Grenze zwischen Chiapas und Tabasco.";
Schauenburg, 1920: 261 (Rio Pichucalco on boundary between Tabasco and
Chiapas, 3000 feet elevation; natural history).
Xiphophorus hellerii, Guenther, 1866: 349-350 (description in part; var. cr and
7, Rio Chisoy, Central America); 1869: 485, pl. 87, Figs. 2-5; Darwin, 1871:
679, Fig. 30 (sexual dimorphism; figure of black spotted form, from Guenther);
Gordon, 1941b: 112-113 (Rio Usumacinta material only).
Xiphophorus guentheri Jordan and Evermann, 1896: 702 (original diagnosis, based
on Guenther's black spotted variety (7) from Rio Chisoy, Guatemala); Meek,
1904: 158 (nominal species based upon spotting); Regan, 1906-1908: 108
(synonymy; diagnosis, based on both spotted and unspotted forms from Rio
Chisoy).
Xiphophorus hellerii guntheri, Hubbs, 1935: 10-11 (characters; dwarfed and
large races; tributary of Belize River, Mountain Pine Ridge, British Honduras;
aguada at Uaxactun, Guatemala).
Xiphophorus brevis Regan, 1907: 65 (original diagnosis; Stann Creek, British
Honduras); 1906-1908: 108, pl. 14, figs. 8-9 (same material).
Xiphophorus hellerii brevis, Hubbs and Gordon, 1943: 32 (material in part; vari-
ation, dorsal rays).
?Xiphophorus rachovii Regan, 1913: 1005 (material in part, not of Regan, 1911a:
373; as synonym of X. helleri; lacking pair of black spots at caudal base;
Puerto Barrios, Guatemala).
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
Figure 22. An extremely large and deep-bodied adult male X. helleri
guentheri Jordan and Evermann (UMMZ:NYZS-GL) approximately 70 mm in
standard length, from a lily pond in Lancetilla, Honduras.
MATERIAL EXAMINED
Mexico, Veracruz
Independent Atlantic Tributary
43a. Laguna de Sontecomapan at village of Sontecomapan, 10 mi. by road E
Catemaco. UMMZ M59-24. 122 young to adults.
Mexico, Tabasco
Rio Tonala system
43b. Arroyo de La Venta at La Venta. UMMZ NYZS-GL. 12 young to adults.
Rio Grijalva system
44a. Arroyo del Azufre at Banos de Azufre, 4 mi. W Teapa. UMMZ M59-35.
15 half-grown to adult.
44b. Spring washing place at Teapa. UMMZ NYZS-GL. 101 half-grown to
adults.
45. Arroyo Cocona tributary to the Rio Puyocatengo 1 mi. N Teapa. UMMZ
NYZS-GL. 4 half-grown.
46. Hacienda de Santa Anna in ditch 39 km. from Villahermosa. UMMZ NYZS-
GL. 4 young to adults.
47. Arroyo Huasteca 28 km. from Villahermosa to Teapa. UMMZ NYZS-GL.
4 young to adults.
Mexico, Chipas; Guatemala, Peten and Alta Vera Paz.
Rio Usumacinta system
48. Upper Rio de la Pasion (= Rio Chajmayic) at Sebol. UMMZ 143770. 1
half-grown.
49. In and just about mouth of first arroyo tributary to upper Rio de la Pasion
from E below Arroyo San Simon. UMMZ 143767. 48 young to adults.
50. Arroyo de Xotal about 7 km. NW of Chicle Station on Laguna Perdida.
UMMZ 143769. 151 young to adults.
51. Rio Cenizo tributary to Rio Salinas (= Rio Negro = Rio Chisoy = Rio
Chixoy), upper Rio Usumacinta system, 30 mi. NW Coban. UMMZ 105458.
6 subadults.
52. Spring at Santa Toribia, tributary to Rio de la Pasion. UMMZ 158452. 10
half-grown to adults.
53. Upper Rio de la Pasion (= Rio Chajmayic) between Rio San Senora and
Rio Ceilia. UMMZ 143776. 21 half-grown to adults.
54. Yaxoquintela, Chiapas (probably Usumacinta drainage). UMMZ 161763.
3 subadults.
55. Laguna de Zotz W Laguna de Peten. UMMZ 143768. 33 young to adults.
1960
BULLETIN FLORIDA STATE MUSEUM
Guatemala, British Honduras
Rio Hondo system
56. Aguada at Uxactun, Guat. UMMZ 97871. 172 adults.
57. Broken ridge 4 mi. W Gallon Jug, Rio Chanchic tributary to Rio Hondo,
Orange Walk, B. Hond. UMMZ NYZS-GL. 4 half-grown.
58. Cenote Creek 8 mi. SE Gallon Jug, Orange Walk, B. Hond. UMMZ NYZS-
GL. 8 half-grown.
British Honduras
Rio Belize system
59. Rio Privacion, an upper tributary to Rio Frio, Mount Pine Ridge 12 mi. S
El Cayo, Guatemala border. UMMZ 97884. 84 half-grown to adult.
60. Rio Frio and tributaries in vicinity of San Augustine, Mt. Pine Ridge, El
Cayo District. UMMZ 159294. 424 young to adults.
61. Xunantunicl Creek, vicinity of Mayan Ruins, Benque Viejo, UMMZ 158408.
50 young to adults.
62. Arroyo Xunantunich W western branch of Rio Belize near Benque Viejo,
El Cayo District. UMMZ NYZS-GL. 1 adult male and female.
Rio Grande system
63. Coroso Creek tributary to Jacinto Creek along San Antonio Road, Toledo
District. UMMZ NYZS-GL. 10 young to adults.
Guatemala
Rio Polochic system
64. Tributary to Rio Polcchic, 14 mi. W Panzos on road to La Tinta. USNM
114256. 51 young to adult males and females.
Rio Motagua system
65. Irrigation ditch W Gualan. USNM 73971. 1 subadult.
66. Brook E ot Los Amates. USNM 73932. 10 specimens.
67. Gualan. USNM 73991. 1 specimen.
68. Los Amates. USNM 73931. 2 specimens.
69. Sulphur River, Puerto Barrios. USNM 73930. 1 specimen. This is an in-
dependent coastal tributary.
Honduras
Independent Atlantic tributaries
70. Brook at Veracruz emptying into Gulf of Honduras, Cortez. UMMZ NYZS-
GL. 70 half-grown to adults.
71. Second Rio Tulian at Tulian, Cortez (the second stream is independent and
flows into the Bay of Cortez about /4 mi. E of large stream, the first Rio
Tulian). UMMZ NYZS-GL. 23 half-grown.
Rio Mapache system
72. Rio Mapache at Masca within 40 ft. of ocean, Cortez. UMMZ NYZS-GL.
1 half-grown.
Rio Chamelecon system
73. Rio Benejo, N San Pedro Sula, Cortez. UMMZ NYZS-GL. 64 young to
adults.
74. Rio Armenta N San Pedro Sula on Chaloma Road, Cortez. UMMZ NYZS-
GL. 3 young to adults.
Rio Lancetilla system
75. Tributary of Rio Lancetilla 1 mi. S Tela, Atlantida. UMMZ NYZS-GL.
1 young.
76. Tributary of Rio Lancetilla at Lancetilla near Royal Palm Avenue, Atlantida.
UMMZ NYZS-GL. 4 half-grown.
77. Lily pond fed by creek tributary to Rio Lancetilla, Atlantida. UMMZ
NYZS-GL. 4 adults.
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
Rio Colorado-Rio Lean system
78. Tributary of Santiago branch of Rio San Alejo at San Alejo, Atlantida. Many
swordtails seen but not taken.
Rio San Juan system
79. Stream 48.5 km. W Ceiba near San Juan, Atlantida. UMMZ NYZS-GL.
1 half-grown.
80. Stream 47.3 km. W Ceiba along railroad near San Juan, Atlantida. UMMZ
NYZS-GL. 11 half-grown.
Rio Bonito system
81. Rio Bonito 10.1 km. W Ceiba, Atlantida. UMMZ NYZS-GL. 7 half-grown
to subadults.
Independent Atlantic tributaries
82. Stream 18.3 km. E Ceiba, Atlantida. UMMZ NYZS-GL. 14 half-grown.
This station may be in Sambo Creek, a small coastal river between Ceiba
and Belfate.
DIAGNOSIS. A short-headed, variably deep-bodied subspecies of
helleri. Midlateral stripe bright red (paler than the surrounding in-
tegument in alcohol), always developed as a solid band, occasionally
with dusky or brownish stripe on caudal peduncle; with one to 4 ad-
ditional solid red bands on sides above, rarely below, midlateral stripe:
axillary stripe lacking in adult males. Dorsal fin with a median and
subdistal row of red spots, or red spots ringed in black. Caudal fin
excluding "sword" long and broad; caudal fin of adult females rarely
with ventral margin of black pigment. Distal serrae on ray 4a of
gonopodium well developed, sharply or bluntly pointed, higher than
wide with a distinct notch separating base from shaft, never obso-
lescent, 3 to 6 in number. Vertebrae 28 to 30. Scales in a lateral series
25 to 29, usually 26 or 27. Dorsal fin rays 11 to 17, usually 13 to 15.
REMARKs. The types of X. guentheri (collected by Godman and
Salvin, 26 January 1864) and X. brevis (collected by Robertson, 1890),
both in the British Museum, were examined by Myron Gordon, who
reported that they correspond with the material here referred to
X. helleri guentheri. Both nominal species are represented by some
spotted individuals, although those from the Rio Chisoy (guentheri)
are more heavily spotted than the ones from British Honduras (brevis).
Regan (1907) gave additional records from Stann Creek, British
Honduras. This material was the basis for his original description of
the nominal species, X. brevis. No green swordtails have been taken
from the Rio Sibun system, British Honduras. The Sibun descends
rather steeply from the northern peaks of the Maya Mountains. Its
waters flow rapidly most of the year and during the rains become tor-
rential and flood the banks in the lowland reaches. Possibly green
swordtails have not been successfully established here. Similarly, no
BULLETIN FLORIDA STATE MUSEUM
green swordtails are known from the Rio Sarstun on the southern
frontier of British Honduras.
A black-spotted form of this subspecies was taken at stations 62
(Belize system, British Honduras) and 77 (Lancetilla system, Hon-
duras). The number of spotted individuals is probably one percent
or less of the total wild populations. The macromelanophore spots
in the Belize swordtail are large and irregularly distributed over the
side. The spotting in the Lancetilla form is arranged in regular rows,
for the most part corresponding to the lateral striping.
Xiphophorus helleri alvarezi, new subspecies
Figures 19-20, 23, 32, tables 3-5, 27-30
Figure 23. Holotype of Xiphophorus helleri alvarezi, n. ssp. (UMMZ: 177304),
an adult male 39.0 mm in standard length.
TYPES. Holotype, an adult male (UMMZ 177304) 39.0 mm. in
standard length, and allotype, an adult female (UMMZ 177305) 39.4
mm. in standard length, both from a single known locality (station 83)
below.
MATERIAL EXAMINED
Mexico, Chiapas
Rio Usumacinta system
83. Rio Santo Domingo, a tributary of the Rio Jatate, upper Rio Usumacinta
system, 90 km E Comitan; M. del Toro; August 1948. 19 adults excluding
holotype and allotype, only two of which, a male and female (UMMZ
160727) are now available; 16 of the remaining specimens are in the private
collection of Sr. Jose Alvarez del Villar, Secretaria de Marina, Mexico, D. F.,
and one in the British Musuem.
DIAGNOSIS. A long headed, slender bodied subspecies of helleri.
Midlateral stripe red (paler than the surrounding integument in al-
cohol), always developed as a solid band, occasionally with an addi-
tional red band on sides above midlateral stripe; axillary stripe lacking
in adult male. Dorsal fin with a median and subdistal row of red and
black spots, or without definitive spotting. Caudal fin excluding
"sword" long and broad; caudal fin of adult female without ventral
margin of black pigment. Distal serrae on ray 4a of gonopodium
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
small, blunt, subtriangular in outline, never obsolescent, 8 in number.
Vertebrae 29. Scales in a lateral series 26 or 27, usually 27. Dorsal
fin rays 11 or 12, usually 11.
The striking general features of this swordtail combine a slender
body, the absence of the black midlateral stripe and a long, slender,
whiplike caudal appendage.
Data available on a specimen sent by Sr. Alvarez to the British
Museum (an adult male 43.0 mm. in standard length) correspond well
with the present diagnosis: greatest depth of body 10 mm.; least depth
of caudal peduncle 8 mm.; length of caudal appendage or "sword"
50 mm.
REMARKS. As shown above, this form is readily separable from
X. h. guentheri from adjacent river basins in southern Mexico and
Guatemala by relative body measurements, dorsal fin ray and lateral
scale counts, and details at the tip of the gonopodium of adult males.
Its red rather than black lateral stripes and its relatively long caudal
fin show it to be closely allied to the southern swordtails.
The 21 specimens of this swordtail, collected about 1500 feet
above sea level, were given to Jose Alvarez del Villar of the Escuela
Nacional de Ciencias Biologicas of Mexico shortly after their capture.
Sr. Alvarez had prepared a description of these fish as a new subspecies
of X. helleri, but its publication was delayed by lack of the compara-
tive material. I gratefully acknowledge that the description given
here is based largely upon Sr. Alvarez's original manuscript. Unfor-
tunately the bulk of the material is not presently available to me, but
the four adult specimens Sr. Alvarez generously sent to the University
of Michigan confirm the new taxonomic status he proposed. The
striking differences between guentheri and alvarezi and the uniformity
of the few individuals of alvarezi at hand warrant its recognition.
This subspecies is named for Jose Alvarez del Villar who brought
the new material to my attention and generously provided many
measurements and observations.
TABLE 1
SUMMARY OF THE DISTRIBUTION AND VARIATION IN THE NUMBER OF DORSAL FIN RAYS IN Xiphophorus
7 8 9 10 11 12 13 14 15 16 17 (N) Means S.E.
X. couchianus
X. maculatus
X. variatus
X. miller
X. montezumae
X. pygmaeus
X. clemenciae
X. helleri
8 134
2 45 1281
59
3
1
53 1
84 11
67 1233
73 4
3 195
72 159
3 11
93
386 54
283 26 1
20 1 4
9 1
726 555 363 54
( 196)
(2023)
(2402)
( 80)
(508)
(257)
(24)
5 1 (1797)
TABLE 2
SUMMARY OF THE DISTRIBUTION AND VARIATION IN THE NUMBER OF SCALES IN A LATERAL SERIES IN Xiphophorus
22 23 24 25 26 27 28 29 (N) Means S.E.
36
8 396 140 2
48 135
1
38
14 63 2
1 24 42
6 12 17
6 5
21 109 127
1 (
3
3
83 8
* Santa Catarina
9.24
9.32
10.88
10.01
11.66
10.84
11.33
12.77
0.04
0.01
0.02
0.01
0.03
4 0.03
0.02
0.02
couchianus*
maculatus
variatus
miller
montezumae
pygmaeus
clemenciae
helleri
8
26.03
23.25
25.90
25.87
26.67
26.45
26.46
26.85
6
6
TABLE 3
DISTRIBUTION OF NUMBER OF VERTEBRAE IN Xiphophorus
26 27 28 29 30 31
X. couchianus
Rio Santa Catarina 1 8
X. maculatus
Rio Papaloapan 9
Rio Coatzacoalcos 1 9
Rio Grijalva 3
Belize River (B. Hond.) 1 3
X. variatus
X. variatus xiphidium
Rio Corona 2 7
Rio Caballero 8
X. variatus variatus
Rio Moctezuma 8
Rio Tempoal 2 31 1
Estero Cucharas 5 27 2
Rio Cazones 23 3
X. variatus evelynae
Rio Necaxa 1 23
X. miller
Laguna Catemaco 20 2
X. montezumae
X. montezumae montezumae
Rio Tamesi 2 15
Rio Verde 11
Rio Salto de Agua 3 21 3
X. montezumae cartezi
Rio Axtla 1 18
X. pygmaeus
X. pygmaeus nigrensis
Rio Choy 6
X. pygmaeus pygmaeus
Rio Axtla 7
X. clemenciae
Rio Sarabia 1 10
X. helleri
X. helleri helleri
Rio Nautla 7 1
Rio Antigua 42 2
Rio Jamapa 10 1
X. helleri strigatus
Rio Coatzacoalcos 1 5
X. helleri guentheri
Rio de la Pasion (Guat.) 7 6
Arroyo Xotal (Guat.) 3 13 1
Laguna de Zotz (Guat.) 6
Rio Hondo (Guat.) 10 11
Rio Frio (B. Hond.) 1 8 2
Belize River (B. Hond.) 3 8
X. helleri alvarezi
Rio Santo Domingo 4
TABLE 4
VARIATION IN GREATEST RELATIVE DEPTH OF BODY IN Xiphophorus
Males Females
Min.-Max. (N) Mean S.E. Min.-Max. (N) Mean S.E.
X. couchianus
Rio San Juan near Sta.
Catarina
X. maculatus (Mexico)
Rio Papaloapan
Obispo
Papaloapan
(Guatemala)
Rio de la Pasion
Rio San Pedro de Martir
Laguna de Zotz*
Laguna de Peten
Uaxactun
X. variatus
X. variatus xiphidium
Rio Sota la Marina
Rio Pilon
Rio Purificacion and
Rio Corona
Rio Corona (Meek)
X. variatus variatus
Rio Tamesi
Rio Guayalejo
2.2-3.8
2.1-2.5
2.2-2.7
2.2-2.7
2.5
2.7-3.1
2.3-2.7
2.2-2.4
2.5-3.1
2.1-2.8
2.5-3.1
2.2-2.8
3.01 0.05
2.24 0.03
2.51 0.04
2.44
2.5
2.89
2.51
2.30
- 0.03
0.02
- 0.04
2.81 0.05
2.35 0.02
2.70 -- 0.07
2.3-3.3
2.1-2.6
2.2-3.0
2.3-2.6
2.4-2.7
2.6-3.1
2.3-2.6
2.2
2.7-3.9
2.1-3.0
2.6-3.1
(10) 2.50 0.06 2.4-2.9
2.66 0.02
2.28 0.02
2.59 0.05
0
2.42 0.01
2.54 0.04 >
2.81 0.02
2.48 0.01 >
2.2 -
C,)
3.12 0.05
2.56 0.05
2.87 0.05
2.59 0.02 -
-
TABLE 4 (continued)
Males Females
Min.-Max. (N) Mean S.E. Min.-Max. (N) Mean S.E.
Rio Panuco
Rio Valles
Rio Axtla
Rio Calaboza
Estero Cucharao
Estero Tancochin
Rio Tuxpan
Rio Cazones
Arroyo Mariandrea
Rio Tecolutla
Arroya Sin Nombre
Rio Nautla
Tlapacoyan
Maria de la Torre
X. variatus evelynae
Rio Tecolutla
Rio Necaxa
X. miller
Rio Papaloapan
Laguna Catemaco
X. montezumae
X. montezumae montezumae
Rio Tamesi
Rio Sabinas
2.1-2.6
2.7-3.4
2.1-3.9
2.1-3.5
2.4-3.5
2.1-2.3
2.9-3.5
2.4-3.0
2.6-3.1
2.4-3.1
2.9-3.3
3.1-3.7
2.32 0.03
3.05 0.07
2.54 0.06
2.31 0.02
3.05 0.05
2.20 - 0.04
3.14 0.05
2.45 0.12
2.81 0.04
2.76 0.08
3.12 0.04
3.45 -- 0.04
2.2-2.7
3.3-3.8
2.3-3.6
2.1-2.9
2.9-3.7
2.2-2.7
2.8-3.6
2.7-3.0
2.7-3.0
2.7-3.3
3.0-3.7
2.9-3.4
2.42 - 0.02 U
3.48 0.07 r
2.61 0.04
2.41 __ 0.02
3.21 0.04
2.46 0.04 M
Cj1
0
3.18 0.06
2.91 0.04
2.94 0.04 M
2.94 0.06 z
C
cj>
3.29 0.05
0
3.19 0.02
--
2.9-3.5 (13) 3.16 - 0.05 3.0-3.4
(15) 3.20 _-t 0.03
TABLE 4 (continued)
Males Females
Min.-Max. (N) Mean S.E. Min.-Max. (N) Mean S.E.
Rio Panuco
Rio Verde
Rio Salto de Agua
X. montezumae cortezi
Rio Panuco
Rio Moctezuma
Rio Axtla
Rio Calaboza
X. pygmaeus
X. pygmaeus nigrensis
Rio Panuco
Rio Choy
X. pygmaeus pygmaeus
Rio Panuco
Rio Axtla
X. clemenciae
Rio Coatzacoalcos
Sarabia
X. helleri
X. helleri helleri
Rio Nautla
Rio Chacalacas
Rio Antigua
Pond at Jalapa
Rio Bejucos
3.0-3.4
2.7-3.1
3.2-3.4
2.7-3.6
3.21 0.08
2.89 0.03
3.32 0.10
3.15 0.07
3.08 0.10
4.10 0.04
3.30 0.15
3.74 -
3.7 -
4.25 0.03
4.16 0.04
2.7-3.6
3.5-4.5
2.9-3.7
3.6-4.0
3.7
4.0-4.4
3.9-4.4
2.9-3.4
2.8-3.3
3.2-3.7
3.2-3.9
3.5
3.2-3.5
3.7-4.3
(16) 3.12 0.03
(13) 3.07 0.04
(14) 3.53 0.04
(10) 3.43 0.04 -
(1) 3.51 --
0
7) 3.35 0.04 >
4 4
(4) 4.01 -
3.0-3.2
3.0-3.4
3.6-4.0
3.0-3.7
3.06 0.03
3.27 0.06
3.77 0.02 <
3.40 0.03
_________C l
TABLE 4 (continued)
Min.-Max.
Rio Jamapa
Rio Chico at Cordoba
X. helleri strigatus
Rio Papaloapan
Motzorongo
El Hule
Obispo
Achatal
Otopa
Rio Coatzacoalcos
X. helleri guentheri
Rio Grijalva
Rio Usumacinta
Rio de la Pasion
Laguna de Zotz
Rio Xotal
Rio Hondo (Uaxactun)
Rio Belize
Rio Privacion
Rio Frio
Rio Polochic
Rio Motagua (Guatemala)
Independent Atlantic
tributaries (Honduras)
X. helleri alvarezi
Rio Usumacinta
Rio Santo Domingo
* Mostly small specimens
Males
(N)
3.6-4.4
3.0 -
2.7-3.4
2.7-3.4
2.9-3.8
2.9-4.0
2.9-4.1
3.0-3.3
3.0-3.5
3.2-3.5
2.7-3.6
2.7-3.4
3.3-3.6
3.4-3.9
2.8-3.1
2.6-3.4
2.7-3.2
4.0
Mean S.E.
3.87 0.05
3.0 -
2.97 +- 0.05
3.06 + 0.02
3.38 0.05
3.43 0.05
3.25 0.04
3.18 0.03
3.24 + 0.05
3.37 -
3.17 0.04
3.00 + 0.04
3.46 0.03
3.63 0.03
2.93 0.04
3.08 -
2.95 -
4.0 -
Min.-Max.
3.1-3.9
2.6-3.1
2.6-3.1
2.6-3.1
3.2-4.0
2.6-3.4
2.9-3.1
2.9-3.2
2.6-2.9
2.5-2.9
3.1-3.5
3.1-3.5
2.5-2.7
2.6-3.4
2.4-2.8
4.2
Females
(N)
(15)
(13)
(11)
( 6)
(4)
(17)
Mean S.E.
3.41 0.05
0
Cir
2.94 0.04 2
2.78 0.04
2.80 0.06 ^
:3.58 -
2.96 0.03 O
2.55 0.05
( 1) 4.2
TABLE 5
VARIATION IN LEAST RELATIVE DEPTH OF CAUDAL PEDUNCLE IN Xiphophorus
Mean S.E. Min.-Max.
Females
(N)
Mean S.E.
Rio San Juan near Sta.
Catarina
X. maculatus (Mexico)
Rio Papaloapan
Obispo
Cosamaloapan
Papaloapan
(Guatemala)
Rio de la Pasion
Rio San Pedro de Martir
Laguna de Zotz*
Laguna de Peten
Uaxactun
X. variatus
X. variatus xiphidium
Rio Sota la Marina
Rio Pilon
Rio Purificacion and
Rio Corona
Rio Corona
X. variatus variatus
Rio Tamesi
Rio Guayalejo
4.7-6.7
3.1-4.9
3.5-3.8
3.9-4.2
3.4-4.6
3.7
4.1-4.9
3.6-4.4
3.4-3.9
5.46 -_- 0.12
3.74 0.08
3.65 0.09
4.03 0.03
3.88 0.07
3.7 -
4.45 0.05
3.99 0.08
3.05 -
4.5-5.8
4.1-6.1
4.3-5.4
5.23 0.13
5.02 0.03
4.83 0.12
4.9-6.4
3.5-5.5
3.8-4.4
3.9-5.0
3.8-4.3
3.8-4.3
4.3-5.1
3.6-4.4
3.9
5.58 0.09 w
4.08 0.05
4.10 0.07
4.41 0.07 t
0
4.08 0.02 5
4.04 0.07 >
4.61 0.04 3
4.04 0.04 >
3.9 --
5.83 0.06
5.55 0.12
5.10 0.11
6.18 0.09
ca
5.1-6.4
4.9-6.6
4.7-6.3
5.5-7.0 (10) 5.91 --t 0.15 5.4-7.3
X. couchianus
Min.-Max.
Males
(N)
TABLE 5 (continued)
Males Females
Min.-Max. (N) Mean S.E. Min.-Max. (N) Mean S.E.
Rio Panuco
Rio Valles
Rio Axtla
Rio Calaboza
Estero Cucharas
Estero Tancochin*
Rio Tuxpan
Rio Cazones
Arroyo Mariandrea
Rio Tecolutla
Arroyo Sin Nombre
Rio Nautla
Tiapacoyan
Maria de la Torre
X. variatus evelynae
Rio Tecolutla
Rio Necaxa
X. miller
Rio Papaloapan
Laguna Catemaco
X. montezumae
X. montezumae montezumae
Rio Tamesi
Rio Sabinas
Rio Panuco
Rio Verde
Rio Salto de Agua
5.0-6.8
5.0-6.1
4.6-6.6
4.7-5.8
4.5-6.3
4.8-6.3
4.9-5.7
4.2-5.3
4.6-5.4
4.4-5.9
4.7-5.9
5.7-6.7
4.3-5.4
4.6-5.0
3.8-5.0
5.79 0.09
5.51 0.12
5.45 0.15
5.13 0.04
5.29 0.09
5.52 0.26
5.35 0.09
4.83 + 0.13
4.84 0.08
4.89 0.16
5.19 0.08
6.24 0.05
4.71 0.09
4.80 0.08
4.35 0.07
5.4-6.8
4.9-6.9
5.2-7.0
5.0-6.9
4.8-6.1
5.5-6.7
5.4-6.7
4.9-5.9
5.0-5.5
5.2-6.0
4.8-5.9
5.6-6.6
5.1-6.1
4.9-6.3
4.8-5.6
(35) 6.07 0.06 4
(9) 6.12 0.25 0
(71) 6.04 0.10 n
(50) 5.74 0.06 z
(31) 5.53 0.06 3
(12) 5.99 0.09
(16) 5.92 0.10
(10) 5.37 0.11
(10) 5.30 0.05
(10) 5.50 0.09
(14) 5.21 0.07
(20) 6.04 - 0.05
C!
Mn
5.62 0.07
5.68 0.07
5.21 0.08 Cn
TABLE 5 (continued)
Mean S.E. Min.-Max.
X. Montezumae cortezi
Rio Panuco
Rio Moctezuma
Rio Axtla
Rio Calaboza
X. pygmaeus
X. pygmaeus nigrensis
Rio Panuco
Rio Choy
X. pygmaeus pygmaeus
Rio Panuco
Rio Axtla
X. clemenciae
Rio Coatzacoalcos
Sarabia
X. helleri
X. helleri helleri
Rio Nautla
Rio Chachalacas
Rio Antigua
Point at Jalapa
Rio Bejucos
Rio Jamapa
Rio Chico, Cordoba
X. helleri strigatus
4.4-5.0
4.1-5.5
3.7-4.8
5.1-6.5
4.4-5.0
4.9-5.2
4.7
5.6-6.5
5.5-6.8
5.0-5.8
4.54 0.11
4.74 0.10
4.28 0.12
(17) 5.87 0.09
4.8-5.7
5.2-6.1
5.3
5.37 0.05
5.49 0.11
5.30
4.8-5.6
6.7-7.1
4.70 0.10
5.91 0.04
6.04 0.08
5.51 0.16
5.09 0.10
0
6.99
-
5.62 0.06
5.33 0.09
6.23 0.04
5.72 0.04
5.71 0.11 0
5.5-5.8
5.0-5.5
5.8-6.5
5.3-6.0
4.9-6.7
Min.-Max.
Males
(N)
Females
(N)
Mean S.E.
TABLE 5 (continued)
Males Females
Min.-Max. (N) Mean S.E. Min.-Max. (N) Mean S.E.
Rio Papaloapan
Motzorongo
El Hule
Obispo
Achatal
Olapa
Rio Coatzacoalcos
X. helleri guentheri
Rio Grijalva
Rio Usumacinta
Rio de la Pasion
Laguna de Zotz
Rio Xotal
Rio Hondo (Uaxactun)
Belize River, Brit. Hond.
Rio Privacion
Rio Frio
Rio Polochic
Rio Motagua (Guatemala)
Independent Atlantic
tributaries (Honduras)
X. helleri alvarezi
Rio Usumacinta
Rio Santo Domingo
4.4
4.0-4.8
4.3-5.0
4.3-5.7
4.3-6.0
4.3-5.4
4.8-5.1
4.9-5.2
4.8-5.1
4.3-5.2
4.1-4.8
4.9-5.9
5.0-6.2
4.4-5.0
4.6-5.4
4.1-4.9
6.0
4.4 -
4.57 0.04
4.64 0.03
4.85 0.09
5.13 0.11
4.80 0.07
4.87 0.04
5.06 0.03
4.97 -
4.69 0.05
4.46 0.04
5.26 0.10
5.43 0.06
4.67 0.06
4.92 -
4.49 -
6.0 -
4.6-5.4
4.3-5.4
4.3-5.1
5.2-6.0
4.9-6.3
4.9-5.5
5.0-5.3
4.4-5.2
4.2-5.0
5.1-6.3
5.3-6.1
4.8-5.2
4.3-6.0
4.6-5.1
6.9
5.14 0.06 0
4.72 0.05 M
4.75 0.07
5.75 -
5.37 0.06
(10) 5.82 0.11 ,
(20) 5.52 0.05 I
9) 4.99 0.06 2
7) 5.04 0.14 g
8) 4.82 0.07 C
4.82
6.9 -
*Mostly small specimens
*-ri
BULLETIN FLORIDA STATE MUSEUM
TABLE 6
DISTRIBUTION AND VARIATION IN THE NUMBER OF DORSAL FIN RAYS IN
Xiphophorus couchianus
8 9 10 11 (N) Mean S.E.
Rio Grande (Mexico)
Santa Catarina
4 mi. west of Monterrey
Monterrey
7 113 20 (140)
15 4 1 (20)
1 6 29 (36)
TABLE 7
DISTRIBUTION AND VARIATION IN THE NUMBER OF DORSAL FIN RAYS IN
Xiphophorus maculatus
7 8 9 10 11 (N) Mean S.E.
Rio Jamapa
Rio Papaloapan
Cosamaloapan 1867
Obispo 1902
Papaloapan 1932-1939
Rio Coatzacoalcos
Almagrez
Jesus Carranza
Minatitlan
Rio Uxpanapa
Coatzacoalcos
Rio Tonala
Huapacal
Tres Bocas
La Renal
La Venta
Rio Grijalva
1 27 71 1 (100) 9.72 -- 0.05
4
16
34
2 60
23
1 10 278
2 89
2 138
1 9
8
10
3 124
(13) 9.85 0.19
(63) 9.76 0.06
(112) 9.73 0.05
( 69)
( 47)
(310)
( 98)
(151)
(11)
(10)
(11)
(140)
5 148 26 (179) 9.12 0.02
Rio Usumacinta
Rio de la Pasion
Rio San Pedro de Martir
Laguna de Zotz
Laguna de Eckivil
Laguna de Peten
Rio Hondo, Uaxactun
Brook no. 1
Brook no. 2
Belize River
13
17
37
1
15
2
2 18
(71)
(23)
3 (200)
( 1)
58)
( 3)
21)
9.82 0.05
9.26 0.09
9.83 0.03
9
9.69 0.07
9.33
8.96 0.08
( 7) 9.00 0.21
(325) 9.27 - 0.03
9.09 0.04
9.30 0.13
9.78 0.08
Vol. 5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
TABLE 8
DISTRIBUTION AND VARIATION IN THE NUMBER OF SCALES IN
Xiphophorus maculatus
A LATERAL SERIES IN
22 23 24 25 (N) Mean S.E.
Rio Jamapa
Rio Papaloapan (Papaloapan)
Rio Coatzacoalcos
Almagrez
Jesus Carranza
Minatitlan
Rio Uxpanapa
Coatzacoalcos
Rio Tonala (La Venta)
Rio Grijalva
Brook no. 1
Belize River
36
2 39
1 37
3 40
36
35
33
33
2 40
15
52
(50)
(50)
(45)
(54)
(50)
(50)
(50)
(50)
(50)
(22)
2 (75)
23.28 0.06
23.14 0.06
23.13
23.15
23.28
23.30 --
23.34
23.34
23.12 -
23.32
23.33
BULLETIN FLORIDA STATE MUSEUM
TABLE 9
DISTRIBUTION AND VARIATION IN THE NUMBER OF DORSAL FIN RAYS IN
Xiphophorus variatus
9 10 11 12 13 14 (N) Mean S.E.
X. variatus xiphidium
Rio Sota la Marina
San Carlos
Rio Santa Lucia 2
Rio Purificacion 38
Rio Corona 13
Rio San Marcus 1
X. variatus variatus
Rio Tamesi
Rio Panuco
Inland 2
Coastal
Estero Cucharas
Estero Tancochin
Rio Tuxpan 1
Rio Cazones
Arroyo Mariandrea
Tihuatlan
Rio Tecolutla
Arroyo Sin Nombre
Rio Nautla
Tlapacoyan 1
Maria de la Torre 1
X. variatus evelynae
Rio Tecolutla
Rio Necaxa
( 3)
(46)
(124)
(235)
( 40)
7 65 20 1 ( 93)
42 247 28 (319)
9 200 31 2 (292)
9 163 39 1 (212)
4 86 7 (97)
7 38 (46)
30 91
1 33 41
1 64 45
12 1 (134) 11.88 0.05
( 75) 11.53 0.06
(110)
(159)
(290)
95 61
76 209
9 80 37 1 (127) 12.24 0.05
10.0 -
10.02 0.05
9.71 0.04
10.00 0.02
10.23 0.08
11.16 0.06
10.94 0.03
10.92 0.03
11.15 0.03
11.03 0.03
10.80 0.07
11.40 0.05
10.40 0.04
10.76 0.03
Vol. 5
4
5
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
TABLE 10
DISTRIBUTION AND VARIATION IN THE NUMBER OF SCALES IN A LATERAL SERIES IN
Xiphophorus variatus
24 25 26 27 28 (N) Mean S.E.
X. variatus xiphidium
Rio Soto la Marina
Rio Corona
X. variatus variatus
Rio Tamesi
Rio Panuco (Inland)
Estero Cucharas
Rio Cazones
Arroyo Mariandrea
Tihuatlan
Rio Tecolutla
Arroyo Sin Nombre
Rio Nautla
Tlapacoyan
Maria de la Torre
X. variatus evelynae
Rio Tecolutla
Rio Necaxa
1 8 25 4 (38) 25.84 0.09
1
24
5 27
3 11 10
1 1 3
( 1) 26
16 1 (41) 26.44 0.09
11 (43) 26.14 0.09
1 (25) 25.36 0.16
( 5) 25.4
3 6 11 2 (22) 25.55 0.19
6 (15) 25.40 0.14
1 1 (30) 25.77 0.09
7 3 (10) 26.30 0.14
TABLE 11
VARIATION IN RELATIVE LENGTH
OF HEAD IN Xiphophorus variatus
Mean S.E. Min.-Max.
X. variatus xiphidium
Rio Sota la Marina
Rio Pilon (Meek)
Rio Purificacion and Rio
Corona
Rio Corona (Meek)
X. variatus variatus
Rio Tamesi
Rio Guayalejo
Rio Panuco
Rio Valles
Rio Axtla
Rio Calaboza
Estero Cucharas
Estero Tancochin **
Rio Tuxpam
Rio Cazones
Arroyo Mariandrea
Rio Tecolutla
Arroyo Sin Nombre
Rio Nautla
Tiapacoyan
Maria de la Torre
X. variatus evelynae
Rio Tecolutla
Rio Necaxa
The length of the head from
the standard length.
** Mostly small specimens.
3.1-3.7
3.0-3.7
2.8-3.7
3.4-3.7
3.3-4.1
3.3-3.6
3.2-3.8
3.2-3.8
3.1-3.8
3.2-3.7
3.3-3.8
3.1-3.6
3.4-4.0
3.4-3.6
3.1-3.5
3.37 0.05
3.34 0.03
3.37 0.09
3.51 0.03
3.61 0.04
3.48 0.03
3.46 0.05
3.50 0.02
3.34 0.03
3.42 0.10
3.57 0.05
3.35 0.05
3.56 0.05
3.54 0.03
3.25 0.03
3.0-3.9
3.2-4.1
2.9-3.8
3.2-3.9
3.3-3.9
3.3-3.6
3.2-4.1
3.3-3.9
3.1-4.0
3.3-3.8
3.5-3.7
3.4-3.9
3.4-3.8
3.5-3.9
3.1-3.6
3.41 0.04
(62) 3.56 0.04
(14) 3.36 0.06
-I
(31) 3.55 0.03
(35) 3.59 0.03 0
(9) 3.49 0.03
(71) 3.57 0.05 >
(50) 3.53 -- 0.02
(31) 3.40 0.04 >
(12) 3.56 0.05
(16) 3.64 0.02 c
(10) 3.57 -+ 0.05
(10) 3.62 0.04
(10) 3.67 _ 0.04
(14) 3.29 0.04 o
the tip of the lower jaw to the end of the opercular membrane was arithmetically divided into
Min.-Max.
Males
(N)
Females
(N)
Mean S.E.
TABLE 12
VARIATION IN RELATIVE LENGTH OF CAUDAL PEDUNCLE IN Xipihophorus variatus variatus
AND Xiphophorus variatus evelynae
Min.-Max.
X. variatus variatus
Rio Panuco
Rio Axtla
Rio Cazones
Arroyo Mariandrea
Rio Tecolutla
Arroyo Sin Nombre
Rio Nautla
Tiapacoyan
Maria de la Torre
X. variatus evelynae
Rio Tecolutla
Rio Necaxa
2.07-2.25
2.06-2.24
2.09-2.31
1.96-2.20
2.03-2.22
2.08-2.62
Males
(N)
Mean S.E.
2.14 0.01
2.15 0.01
2.20 0.03
2.14 0.02
2.14 0.06
(13) 2.29 0.04
Min.-Max.
Females
(N)
2.24-2.56
2.27-2.65
2.19-2.60
2.30-2.77
2.40-2.50
2.54-2.87
Mean S.E. C
con
q
2.43 0.03
2.50 0.02
2.46 0.04 C
2.48 0.04
2.44 0.01
2.69 0.03 c
TABLE 13
VARIATION IN RELATIVE LENGTH OF CAUDAL FIN IN Xiphophorus variatus variatus
AND Xiphophorus variatus evelynae
Males Females
Min.-Max. (N) Mean S.E. Min.-Max. (N) Mean S.E.
X. variatus variatus
Rio Panuco
Rio Axtla 3.1-3.7 (10) 3.36 0.08 3.6-4.3 ( 9) 3.79 0.07
Rio Cazones
Arroyo Mariandrea 2.9-3.5 ( 9) 3.10 0.07 2.9-3.5 (16) 3.21 0.04
Rio Tecolutla
Arroyo Sin Nombre 3.1-3.5 ( 9) 3.29 0.06 3.4-4.0 (10) 3.61 0.07
Rio Nautla
Tlapacoyan 3.0-3.2 (10) 3.16 0.02 3.1-3.5 (10) 3.34 0.04
Maria de la Torre 3.1-3.5 (10) 3.24 0.04 3.2-3.8 (10) 3.48 0.06
X. variatus evelynae
Rio Tecolutla
Rio Necaxa 2.7-2.9 (13) 2.84 0.02 2.6-3.0 (12) 2.70 0.03
TABLE 14
VARIATION IN THREE PROPORTIONAL MEASUREMENTS IN Xiphophorus miller FROM LAGUNA CATEMACO
(Rio PAPALOAPAN SYSTEM)
Males Females
Min.-Max. (N) Mean S.E. Min.-Max. (N) Mean S.E.
Length of Head 3.2-3.6 (38) 3.42 0.02 3.4-3.6 (20) 3.49 0.02
Length of Caudal Peduncle 2.0-2.2 (38) 2.13 0.01 2.3-2.5 (20) 2.38 0.01
Length of Caudal Fin 3.2-4.0 (38) 3.65 0.03 3.5-4.1 (20) 3.85 0.03
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
TABLE 15
DISTRIBUTION AND VARIATION OF THE NUMBER OF DORSAL FIN RAYS IN
Xiphophorus montezumae
10 11 12 13 14 (N) Mean S.E.
X. montezumae montezumae
Rio Tamesi 20 19 1 ( 40) 11.53 0.09
Rio Panuco
Rio Verde at Rascon 13 19 9 ( 41) 11.90 0.11
Rio Salto de Agua 1 15 12 ( 28) 11.40 0.10
X. montezumae cortezi
Rio Panuco
Rio Valles at Valles 1 2 ( 3) 11.7
Rio Moctezuma 79 88 1 (169) 11.53 0.04
Rio Axtla 1 82 154 13 (251) 11.71 0.04
Rio Calaboza 2 1 2 1 ( 4) 13.0
TABLE 16
DISTRIBUTION AND VARIATION IN THE NUMBER OF SCALES IN A LATERAL SERIES IN
Xiphophorus montezumae
25 26 27 28 (N) Mean S.E.
X. montezumae montezumae
Rio Tamesi 1 6 25 1 (33) 26.79 0.10
Rio Panuco
Rio Verde 6 15 (21) 27.70 0.10
Rio Salto de Agua 8 16 4 (28) 26.80 0.10
X. montezumae cortezi
Rio Panuco (Rio Axtla) 18 17 2 (37) 26.57 0.11
TABLE 17
VARIATION IN RELATIVE LENGTH OF HEAD IN Xiphophorus montezumae
Mean S.E. Min.-Max.
Females
(N)
Mean S.E.
X. montezumae montezumae
Rio Tamesi
Rio Sabinas
Rio Panuco
Rio Verde
Rio Salto de Agua
X. montezumae cortezi
Rio Panuco
Rio Axtla
3.5-3.9
4.0-4.4
3.6-4.1
3.4-4.0
3.82 + 0.03
4.17 0.07
3.85 0.03
3.62 0.04
3.7-3.9 (15)
4.2-4.5 (16)
3.6-4.0 (13)
3.4-3.7 (10)
TABLE 18
VARIATION IN RELATIVE LENGTH OF CAUDAL PEDUNCLE IN Xiphophorus montezumae
Min.-Max.
X. montezumae montezumae
Rio Tamesi
Rio Sabinas
Rio Panuco
Rio Verde
Rio Salto de Agua
X. montezumae cortezi
Rio Panuco
Rio Axtla
Males
(N)
1.9-2.0
2.0-2.5
Mean S.E.
1.99 0.01
2.13 0.03
Min.-Max.
Females
(N)
2.2-2.6
2.3-2.6
3.76 0.01 m
4.30 0.02
3.82 0.04
3.53 0.02 o
cjn
Mean S.E. 0
2.41 0.02
2.44 0.01
1.9-2.2 (14) 2.09 0.02 2.3-2.6
Min.-Max.
Males
(N)
(10) 2.47 -t 0.02
ROSEN: FISHES OF THE GENUS XIPHOPHORUS
TABLE 19
VARIATION IN RELATIVE LENGTH OF CAUDAL APPENDAGE IN ADULT MALES OF
Xiphophorus montezumae
Min.-Max.
X. montezumae montezumae
Rio Tamesi
Rio Sabinas
Rio Panuco
Rio Verde
Rio Salto de Agua
X. montezumae cartezi
Rio Panuco
Rio Axtla
1.5-2.6
1.0-1.7
1.2-1.9
1.6-2.6
(N) Mean S.E.
2.02 0.10
1.37 -
1.65 0.12
2.01 + 0.04
TABLE 20
DISTRIBUTION AND VARIATION OF THE NUMBER OF DORSAL FIN RAYS IN
Xiphophorus pygmaeus
9 10 11 12 1,3 14 (N) Mean S.E.
X. pygmaeus nigrensis
Rio Panuco
Rio Choy
X. pygmaeus pygmaeus
Rio Panuco
Rio Axtla
4 15 1 4 ( 24) 12.21 0.19
1 72 155 5
(233) 10.70 0.03
TABLE 21
DISTRIBUTION AND VARIATION OF THE NUMBER OF SCALES IN LATERAL SERIES IN
Xiphophorus pygmaeus
25 26 27 28 (N) Mean S.E.
X. pygmaeus nigrensis
Rio Panuco
Rio Choy
X. pygmaeus pygmaeus
Rio Panuco
Rio Axtla
1 1 17 3 (22) 27.00 0.13
(16) 25.69 0.12
1960
5 11
TABLE 22
VARIATION IN RELATIVE LENGTH OF HEAD IN Xiphophorus pygmaeus
Males Females
Min.-Max. (N) Mean S.E. Min.-Max. (N) Mean S.E.
X. pygmaeus nigrensis
Rio Panuco
Rio Choy 3.6-4.0 ( 9) 3.81 0.04 3.4-3.7 ( 7) 3.55 0.03
X. pygmaeus pygmaeus
Rio Panuco
Rio Axtla 3.4-4.0 (17) 3.77 0.05 3.6-4.2 ( 4) 3.89 0.12
TABLE 23
VARIATION IN RELATIVE LENGTH OF CAUDAL PEDUNCLE IN Xiphophorus pygmaeus
Males Females
Min.-Max. (N) Mean S.E. Min.-Max. (N) Mean S.E.
X. pygmaeus nigrensis
Rio Panuco
Rio Choy 2.0-2.3 (11) 2.15 0.03 2.3-2.6 ( 6) 2.49 0.05
X. pygmaeus pygmaeus
Rio Panuco
Rio Axtla 2.1-2.3 (16) 2.19 0.02 2.6-2.7 ( 4) 2.67 -
TABLE 24
VARIATION IN TWO PROPORTIONAL MEASUREMENTS IN MALES OF
Xiphophorus clemenciae FROM SARABIA (RIO COATZACOALCOS SYSTEM)*
Min.-Max. (N) Mean S.E.
Length of Head 3.7-4.4 ( 6) 4.02 0.12
Length of Caudal Peduncle 2.0-2.3 ( 6) 2.12 0.06
Only three of the males are adult, ranging in size from 29.0 to 35.5 mm. in
standard length. The remaining three are in various stages of development, from
very early to late differentiation.
TABLE 25
GEOGRAPHICAL POPULATIONS OF Xiphophorus helleri MAINTAINED UNDER
LABORATORY CONDITIONS, AND NUMBERS OF INTRASPECIFIC
SWORDTAIL HYBRIDS PRODUCED
Period Maintained
Subspecies in Laboratory
X. h. helleri
Rio Jamapa (Veracruz, Mexico) 26 years
X. h. strigatus
Rio Papaloapan (Veracruz, Mexico) 19 years
Rio Coatzacoalcos (Veracruz, Mexico) 10 years
X. h. guentheri
Rio Grijalva (Chiapas, Mexico) 4 years
Rio Belize (British Honduras) 9 years
Rio Lancetilla (Honduras) 6 years
Types of Hybrid Produced Number of Adult Hybrids
Rio Jamapa 9 x Rio Papaloapan S F1 = 4
Rio Jamapa 9 x Rio Coatzacoalcos 8 F1 = 50; F2 = 18
Rio Jamapa 9 x Rio Lancetilla & Fi = 22; F2 = 35
Rio Coatzacoalcos 9 x Rio Belize S F1 = 35
Rio Papaloapan 9 x (Rio Jamapa-Rio Papaloapan F2) Backcross = 20
Rio Papaloapan 9 x Rio Belize 3 F1 108
Rio Grijalva 9 x Rio Lancetilla 8 F, = 20
Rio Belize 9 x Rio Lancetilla 3 F2 = 40; F2 = 28
TABLE 26
VARIATION IN THE NUMBER OF DORSAL FIN RAYS IN Xiphophorus helleri FROM
DIFFERENT NATURAL POPULATIONS RAISED UNDER SIMILAR CONDITIONS
IN THE LABORATORY
11 12 13 14 15 (N) Mean S.E.
Rio Coatzacoalcos stock 1 12 3 1 (17) 12.24 -- 0.16
Belize River stock 2 18 (20) 12.90 0.07
Honduras stock 3 7 1 (11) 13.82 0.18
DISTRIBUTION AND VARIATION IN
TABLE 27
THE NUMBER OF DORSAL FIN RAYS IN Xiphophorus helleri 0
11 12 13: 14 15 16 17 (N) Mean S.E.
X. helleri helleri
Rio Nautla
Rio Chachalacas
Encero
Plan del Rio
Rio Antigua
Jalapa
Rio Bejucos
Rio Jamapa
L'Encero
Cordova
Boca del Rio
X. helleri strigatus
Rio Papaloapan
Orizaba
Otopa
Motzorongo
El Hule
Papaloapan
Obispo
Achotal
Rio Coatzacoalcos
Almagres
Jesus Carranza
Minatitlan
5 2
4 82 7
6 6
22 98
8 91
4 52 6
12 134 40
3 2
5 8
1 2
1
2 45
4 104
16
2 6
9
8 23
4
7) 12.28 0.22
93) 12.03 0.04
12) 12.50 0.15
(121) 11.83 0.04
(110) 12.03 0.04
( 62) 12.03 0.05
(187) 12.16 0.04
( 5) 11.40 0.26
14)
11)
14)
79)
(193)
(27)
( 8)
(19)
( 33)
( 6)
11.71
12.64
13.00
12.38
12.44
12.52
11.75
12.58 0.14
11.82 0.09
12.33 0.24
TABLE 27 (continued)
DiSTRIBUTION AND VARIATION IN THE NUMBER OF DORSAL FIN HAYS IN Xiphophorus helleri
11 12 13 14 15 16 17 (N) Mean S.E.
X. helleri guentheri
Rio Tonala
Rio Grijalva
Rio Usumacinta
Upper Rio Usumnacinta
Rio de la Pasion
Arroyo Xotal
Laguna de Zotz
Rio Hondo, Uaxactun
Belize River
Benque Viejo
Rio Privacion
Rio Frio
Rio Grande (Brit. Hond.)
Rio Motagua
Rio Chamelecon
Rio Lancetilla
Rio San Juan
Rio Bonito
Estero Salado
X. helleri alvarezi
Rio Santo Domingo
4 6 2
14 20 2
1 12
16 67
1 3 67
1
1 3 2
5 33
3
9
48 1
7 2
25 1
2 2
3 3 1
9 2
12) 13.83 -t 0.21
( 7) 13.72 0.11
( 6)
67)
(130)
(24)
(167)
( 22)
( 83)
(120)
( 10)
6)
64)
7)
( 1)
7)
1 ( 12)
12.67
13.46
13.92
14.00
13.72
13.36
12.81
13.38
14.10
12.17
13.34
13.85
15
14.71
15.33
0.13
- 0.04
0.05
0.10
0.34
0.08
0.34
0.29
0.19
( 21) 11.23 0.10
16 5
TABLE 28
DISTRIBUTION AND VARIATION IN THE NUMBER OF SCALES IN A LATERAL SERIES IN Xiphophorus helleri
25 26 27 28 29 (N) Mean S.E.
X. helleri helleri
Rio Nautla
Rio Antigua
Jalapa
Bejucos
Rio Jamapa
L'Encero
Boca del Rio
X. helleri strigatus
Rio Papaloapan (Papaloapan)
Rio Coatzacoalcos
Almagres
Jesus Carranza
Minatitlan
X. helleri guentheri
Rio Tonala
Rio Grijalva
Rio Usumacinta
Rio del la Pasion
Rio Hondo, Uaxactun
Belize River
Benque Viejo
Rio Frio
Rio Grande (Brit. Hond.)
Rio Motagua
Rio Chamelecon
Rio Lancetilla
Rio San Juan
Rio Bonito
Estero Salado
X. helleri alvarezi
Rio Santo Domingo
3 7
4 4
7 (59)
( 4)
2
6 7
1 7
10
1 5
3
6 26
1
5
3
( 2)
(17)
(15)
(11)
( 7)
(6)
2 (60)
4 (7)
(1)
(6)
(10)
27.85 0.14
27.70 0.16
27.50 0.20
27.97 0.07
26.8
27.0
26.94 0.16
27.21 0.13
27.3 -
26.25 0.32
26.09 0.10
26.0
25.88 0.19
26.40
26.09
26.00
26.50
26.43
27.28
27.0
26.17
26.70
- 0.17
0.09
0.22
0.24
0.09
0.33
0.21
- 0.16
5 15 1 (21) 26.86 0.14
5 15
1 (21) 26.86 _ 0.14
Subspecies and
Drainage
h. helleri
Rio Nautla
Rio Chachalacas
Rio Jamapa
h. strigatus
Rio Papaloapan
Rio Coatzacoalcos
h. guentheri
Rio Griialva
Rio Usumacinta
Rio Hondo
Belize River
Rio Frio
Rio Polochic
Independent Atlantic
tributaries (Hond.)
h. alvarezi
Rio Santo Domingo
TABLE 29
DISTRIBUTION AND VARIATION OF RELATIVE LENGTH OF CAUDAL FIN IN Xiphophorus helleri
Class Intervals
2.6 .7 .8 .9 3.0 .1 .2 .3 .4 .5 .6 .7 .8 .9 4.0 .1 .2 .3 .4 .5 (N) Mean S.E.
1 1 2 2 1 1 ( 8) 3.47 + 0.06
1 3 9 4 8 9 7 4 1 1 2 (49) 3.64 0.03
1 3 10 13 14 14 10 12 12 13 7 3 1 2 (117) 3.80 0.03
2 5 3 6 7 4 8 2 1
1 5 9 11 9 2 6 4 2 1
2 1 5 2
1 8 4 8 9 2 2 1 1
3 8 3 2
1 1 1 2 2 2 2
1 1 6 2 2 2 5 1
2 6 9 1
2 3 6
( 38) 3.19 0.03
50) 3.18 0.01
10) 2.87 0.04
( 36) 3.01 0.03
1( i) 2.92 0.02
11) 3.15 0.06
21) 3.19 0.05
18) 2.75 0.02
12) 2.87 0.04
( 1) 2.9
TABLE 30
FREQUENCY DISTRIBUTION OF GONOPODIAL SEGMENTS IN Xiphophorus helleri
Distal serrae (ray 4p)
2 3 4 5
h. helleri
Rio Nautla
Rio Antigua
Rio Jamapa
h. strigatus
Rio Papaloapan
Rio Coatzacoalcos
h. guentheri
Rio Grijalva
Rio Hondo
Rio Frio
La Lima, Honduras
2
1 7
5 11
7 9
1 9 9 1
6 10
11
4 5
Subterminal segments (ray 3)
60 1 7 0
2
4 6
10 8
2 3 8
1 7 5 6
3
4 3
6
o 10u x
r
1-
1 -
z
r
C
1 Ct-
0
H
5 3 4 1
4
1 H
I^
h. alvarezi
Rio Santo Domingo
(Rio Usumacinta system)
Subspecies
|
