The Marine Toad, Bufo marinus:
A Natural History Resume
of Native Populations
GEORGE R. ZUG
and
PATRICIA B. ZUG
m
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY ? NUMBER 284
SERIES PUBLICATIONS OF THE SMITHSONIAN INSTITUTION
Emphasis upon publication as a means of "diffusing knowledge" was expressed
by the first Secretary of the Smithsonian. In his formal plan for the Institution, Joseph
Henry outlined a program that included the following statement: "It is proposed to
publish a series of reports, giving an account of the new discoveries in science, and
of the changes made from year to year in all branches of knowledge." This theme
of basic research has been adhered to through the years by thousands of titles issued
in series publications under the Smithsonian imprint, commencing with Smithsonian
Contributions to Knowledge in 1848 and continuing with the following active series:
Smithsonian Contributions to Anthropology
Smithsonian Contributions to Astrophysics
Smithsonian Contributions to Botany
Smithsonian Contributions to the Earth Sciences
Smithsonian Contributions to the Marine Sciences
Smithsonian Contributions to Paleobiology
Smithsonian Contributions to Zoology
Smithsonian Studies in Air and Space
Smithsonian Studies in History and Technology
In these series, the Institution publishes small papers and full-scale monographs
that report the research and collections of its various museums and bureaux or of
professional colleagues in the world of science and scholarship. The publications are
distributed by mailing lists to libraries, universities, and similar institutions throughout
the world.
Papers or monographs submitted for series publication are received by the
Smithsonian Institution Press, subject to its own review for format and style, only
through departments of the various Smithsonian museums or bureaux, where the
manuscripts are given substantive review. Press requirements for manuscript and art
preparation are outlined on the inside back cover.
S. Dillon Ripley
Secretary
Smithsonian Institution
S M I T H S O N I A N C O N T R I B U T I O N S T O Z O O L O G Y ? N U M B E R 2 8 4
The Marine Toad, Bufo marinus.
A Natural History Resume
of Native Populations
George R.
and Patricia B.
SMITHSONIAN INSTITUTION PRESS
City of Washington
1979
A B S T R A C T
Zug, George R., and Patricia B. Zug. The Marine Toad, Bufo marinus: A
Natural History Resume of Native Populations. Smithsonian Contributions to
Zoology, number 284, 58 pages, 20 figures, 22 tables, 1979.?Bufo marinus is one
of the most common and widespread anurans of Central and South America,
yet little is known of its natural history in native habitats. Through a survey
of the literature and brief ecological studies in Panama, we attempted to sum-
marize the natural history of the toads. Emphasis is placed on distribution,
sexual maturity, reproductive cycles, body size and growth, population structure
and size, food and feeding behavior, activity and movements, thermal physiology,
and water balance.
In general, marine toads have one or two breeding seasons each year, appar-
ently timed so metamorphosis occurs during periods of high humidity and prey
abundance. Toadlets grow rapidly and reacn the size of sexual maturity (ap
proximately 90 mm snout-vent length) at about one year from metamorphosis,
although they may not breed their first year. Populations in semi-natural habi-
tats are of moderately high density, 50 to 150 late term juvenile to adult toads
per hectare. One-year-old toads comprise at least 50 percent of total population
with a progressive and rapid decline of the older year classes. Apparently two-
thirds of the population turns over each year. The toads are opportunistic
feeders with ants and beetles predominating in their diet. Most toads are quite
mobile, changing feeding sites regularly within a forage area of roughly 160 sq
m. A toad is generally active one evening out of every three or four; however,
weather conditions do alter activity patterns. The toads have a broad tempera-
ture tolerance but have no or little control of water loss. The dry season and
the resulting desiccation may be a major mortality factor.
OFFICIAL PUBLICATION DATE is handstamped in a limited number of initial copies and is recorded
in the Institution's annual report, SMITHSONIAN YEAR. SERIES COVER DESIGN: The coral Montastrea
cavernosa (Linnaeus).
Library of Congress Cataloging in Publication Data
Zug, George R 1938?
The marine toad, Bufo marinus.
(Smithsonian contributions to zoology ; no. 284)
Bibliography: p.
1. Bufo marinus. I. Zug, Patricia B., joint author. II. Title. III. Series: Smithsonian
Institution Smithsonian contributions to zoology ; no. 284
QL1.S54 no. 284 [QL668.E227] 59V.8 78-606171
Contents
Page
Introduction 1
Acknowledgments 1
Distribution, Habitats, and Systematics 1
General Materials and Methods 3
Sex, Size, and Growth 7
Physical Appearance 7
Sexual Maturity and Reproductive Cycle 8
Maturity 8
Sex Ratios 9
Reproduction 9
Remarks 10
Larval Development and Metamorphosis 11
Body Size 11
Sexual Differences 12
Regional Differences 14
Growth 14
Larval Growth 14
Postmetamorphic Growth 14
Sexual Differences 16
Seasonality of Growth 17
Population Structure and Size 18
Age/Size Distribution 18
Abundance 19
Postmetamorphs 19
Old Juveniles to Adults 20
Speculations on Demography 21
Age Distribution 22
Fertility 22
Recruitment and Survivorship 22
General Remarks 23
Food and Feeding Behavior 24
Prey Preference 24
Regional and Seasonal Differences 25
Size Selection 26
Feeding Behavior 26
Consumption Rates 26
Locating and Catching Prey 27
Interactions 27
Activity and Movements 28
Materials and Methods 28
Activity 28
Daily Activity 29
Sexual and Age Activity Patterns 30
iii
j v SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
Page
Seasonal Activity Patterns 30
Movement 31
Foraging Area 31
Homing 31
Physiology: Heat and. Water 32
Thermal Physiology 32
Materials and Methods 34
Results and Discussion 34
Water Economy 35
Materials and Methods 37
Results and Discussion 37
Mortality Factors 38
Predators 38
Disease and Parasites 39
Summary 39
Appendix I: Sample Data 41
Appendix II: Weight-Length Relationships of Arthropods 44
Literature Cited 54
The Marine Toad, Bufo marinus.
A Natural History Resume
of Native Populations
George R. Zug
and Patricia B. Zug
Introduction
The marine toad, Bufo marinus, is the most
widespread and common American amphibian.
Yet in spite of its commonness, we know little of
its life history in its native habitat. All major
ecological studies (see Zug, et al., 1975, for litera-
ture review) have been undertaken in areas where
it is an exotic. The goal of our study has been to
correct this oversight and to obtain comparative
ecological information from natural populations of
the marine toad. Our approach has been purposely
superficial, for we wished to survey a wide variety
of the toad's life history aspects in a short time
and hopefully stimulate other researchers with
marine toads in their backyard to study them in
detail. We also wished to bring together the scat-
tered natural history notes on this toad in its
native habitat and those physiological and be-
havioral studies that apply directly to the toad's
ecology.
Acknowledgments.?Many people have helped
and encouraged us throughout our study, and we
wish to thank them all. The staff and associates of
the Smithsonian Tropical Research Institute
(STRI), especially Stan Rand and Cathy Toft, made
our field work pleasant and successful. Wim Wolda's
George and Patricia Zug, Department of Vertebrate Zoology,
National Museum of Natural History, Smithsonian Institu-
tion, Washington, B.C. 20560.
monthly collections of toads enhanced our repro-
ductive samples. Blair Hedges and Joan Dudley
helped greatly in the dietary analysis. Fran McCul-
lough allowed us access to her growth data on cap-
tive toads, and Charles Myers lent us his size data
on Darien toads. The reviewers, A. Bennett, R.
Crombie, R. Heyer, and F. McCullough, tempered
our excesses and improved the clarity of the text.
The fieldwork was sponsored by several Smithso-
nian Institution sources: Smithsonian Research
Foundation, Environmental Science Program, Fluid
Research Fund, and Department of Vertebrate
Zoology. We gratefully acknowledge their suport.
Distribution, Habitats, and Systematics
The natural range of the marine toad extends
from approximately 27? N latitude in southern
Texas and western Mexico to 10? S latitude in
central Brazil (Figure 1). The break or transition
between B. marinus and its southern relatives?
arenarum, ictericus, paracnemis, poeppigii, and
rufus (Bertini and Cei, 1962)?lies between 10?
and 14? S and remains to be accurately delineated.
By and large, the toad is a lowland animal and
Dedicated with affection and appreciation to Dr.
Charles F. Walker, University of Michigan.
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURE 1.?Presumed natural range of Bufo ma-
rinus. (Triangle = literature records; circle =
museum specimens from Smithsonian Institu-
tion; square = University of Kansas Museum of
Natural History; hexagon = Museo de Zoologia
da Universidade de Sao Paulo.)
generally occurs below 1000 m in elevation, al-
though populations are known from above this
elevation (Table 1). The maximum elevation for
populational survival is apparently determined by
the minimum thermal tolerance limits, so that the
maximum elevation is lower at the higher lati-
tudes, e.g., 500 m in Sinaloa, Mexico (Hardy and
McDiarmid, 1969) compared to 1600 m in Vene-
zuela (Rivero, 1961).
The toad occurs in a variety of habitats (Table
2), but nowhere is it more common than in open
areas resulting from man's activity and in the
grasslands. The most commonly cited habitat is in
and around human settlements. As has been noted
(Sexton, et al., 1964; Heatwole, 1966), B. marinus
has become an associate or commensal of man and,
in forested areas, is commonly in villages and small
open areas but seldom seen elsewhere. Although
not excluded from the forest, we suspect that the
toads are largely marginal of open areas, and un-
broken forest can and does act as a barrier to dis-
persal. Lescure (1975:66) notes the toad's absence
in isolated Indian villages in Amazonia and also
proposes unbroken forest as a barrier.
B. marinus is an old species. A fossil toad from
the La Venta fauna of the late Miocene of Colom-
NUMBER 284
TABLE 1.?Selected examples of Bufo marinus occurrences
above 1000 m in natural populations
Locality
MEXICO
Tamaul ipas: Rancho del C1elo
GUATEMALA
Al ta Verapas
EL SALVADOR
La L i b e r t a d : Santa Tecla
HONDURAS
no speci f ic l o c a l i t y given
COSTA RICA
Heredia: Volcan Poas
VENEZUELA
Andes
Elevation
(m)
1150
41300
ca. 1200
1300
ca. 2100
1600
Mart in ,
Stuar t ,
Mertens
Source
1958
1950
. 1952
Meyer and Wilson, 1971
Taylor,
Rivero,
1952
1961
TABLE 2.?Selected examples of habitats occupied by Bufo
marinus populations (habitat descriptors unchanged from
original sources)
Habitat
sandy beach
along streams
clearings in cloud forest
tropical scrub forest
quasi-rainforest
broad-lea fed forest
dense scrub forest
coconut grove
rainforest
grasslands
aquatic-riparian
open broad!eaf forest
true savanna
tropical moist forest
tropical dry forest
tropical arid forest
subtropical wet forest
subtropical moist forest
subtropical dry forest
tropical rain forest
open cleared areas
Locality
Mexico
"
"
"
Guatemala
"
Honduras
Guyana
Trinidad
Source
Hardy & McDiarmid, 1969
" " "
Mar t in , 1958
Duellman, 1961
Duellman, 1965
n ii
?I n
n II
Duellman, 1963
Stuart, 1950
" "
U H
H II
Meyer & Wilson, 1971
?I I I !?
H U H
II II II
H U H
? 1 II II
Beebe, 1925
Kenny, 1969
bia (Estes and Wassersug, 1963) is indistinguishable
from modern marinus from northern South Amer-
ica. It was discovered in a floodplain deposit, which
suggest that marinus habitat preferences have
always been for open areas. When marinus origi-
nated is unknown, but we believe it arose as an
adaptation to a seasonally xeric habitat, such as
savanna or scrub forest. Cei (1972) postulates the
Guiana shield as its ancestral home from which it
radiated southward and speciated in the eastern
Andean forest (poeppigii), Chaco (paracnemis),
Brazilian planaltos (rufus), Atlantic coastal forest
(ictericus) and Patagonia (arenarum complex); more
recently (Plio-Pleistocene), it expanded northward
across the Panamanian isthmus into Central
America.
B. marinus and its relatives are considered to be
one of the older or more structurally primitive
groups of the broad-skulled toads (Martin, 1972).
They share many similarities with the valliceps
group of Central and North America. The latter
has been interpreted as an Early Tertiary expan-
sion of marinus-valliceps stock into Central
America.
Since B. marinus is an old and widespread
species, regional differentiation of populations is
not unexpected. However, there are no strong
morphological differences between populations (see
Savage, 1960, for clinal variation in larvae charac-
teristics) and apparently no differences in call. Dif-
ferences in chromosome arm ratios and lengths do
exist (Doyle and Beckert, 1969) between geographi-
cally distinct populations, so it is possible that more
than one species is hiding under the name marinus.
General Materials and Methods
The toad populations of two areas were selected
for ecological monitoring: Summit Gardens (SG)
and Barro Colorado Island (BCI). Summit Gardens
is a combined zoological and botanical garden in
the Canal Zone, Panama, approximately 7 km
southeast of Gamboa. In the center of the gardens
there are two cement-enclosed ponds used for grow-
ing water lilies. The lily ponds are irregularly
shaped with a circumference of approximately
110 m for the upper and 90 m for the lower pond.
The upper pond is in a lawnlike setting with a few
scattered trees and shrubs forming little apparent
diurnal shelter for toads. The lower pond is en-
circled by a narrow belt (5-10 m) of grass and then
a fairly complete ring of shrubs and trees. The
grass along each pond is mowed and raked regu-
larly so that there is little or no natural cover. The
ponds are linked by a small cement-bottomed
stream (approx. 100 m) that passes through a bam-
boo and tree gallery. A 10-meter-wide strip along
the edge of the ponds and their connecting streams
was designated as the study area. Nocturnal forays
showed the toads to be concentrated around the
ponds and seldom elsewhere except in the vicinity
of the zoo.
Barro Colorado Island is a biological preserve
in the Panama Canal, approximately 15 km north-
west of Gamboa. The study sites were the labora-
tory/housing (Figure 2) grounds and dock/cause-
way area (Figure 3) of the Smithsonian Tropical
Research Institute's biological station on BCI. The
laboratory/housing site has an area of approxi-
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
mately 2812 sq. m, excluding the area of the vari-
ous buildings; the area of the dock/causeway site
is approximately 888 sq. m. The two sites are nearly
75 m apart, separated by a steep slope of secondary
growth grasses and shrubs. The dock/causeway site
is for the most part a bare or low grass habitat with
numerous temporary stacks of lumber and equip-
ment; only at its eastern end is the vegetation?
grasses and cat-tails?continuous. This site is
bounded on one side by water and on the other
side by a sharp embankment on the western end
that gradually slopes into a grassy area with build-
ings on the eastern end. The laboratory/housing
site is bordered largely by seasonal rainforest.
Within this site, there are numerous ornamental
trees and shrubs; the open areas are cut infre-
NUMBER 284
FIGURE 2.?Sketch of laboratory/housing study area on BCI. (Area enclosed by forest (heavy
stippling) and/or solid black line = study area; crosshatched rectangles = various buildings;
lighter stippling = patches of ornamental shrubs; L = outside lights, usually lit for early part
of evening.)
quently so herbaceous ground cover varies from
abundant to dense. Both sites were partially lighted
from dusk (ca. 1830) to 2200 by the diffuse light
from the interior of the buildings and by scattered
outside lights. As will be seen later, the lights tend
to concentrate the toads, but toads did not con-
fine their activity to the brightly lighted areas.
In both study areas (SG, BCI), the toads were
hand caught and individually marked by toe clip-
ping (only the terminal two phalanges) following
the system of Nace et al. (1973). Snout-vent length,
body weight, and sex were recorded upon the
initial capture, and any succeeding ones at two-
week intervals. Time and site of activity were re-
corded every time a toad was observed.
Samples were gathered from five other localities
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
in order to obtain information on feeding habits
and the reproductive cycle. The Gamboa sample
(G) came from a variety of locations within the
town, viz., school ground, paddock, and canal main-
tenance yard; all locations were in close proximity
to buildings, artifically lighted for most of the
night, and with the grass and shrubs receiving fre-
quent care. The Summit Hill (SH) sample (ca. 9 km
SE of Gamboa) came from the edge of a golf
course; the site is indirectly lighted for most of the
night, and the grass is mowed frequently. The Las
Cumbres (LC) sample (ca. 18 km E of Gamboa)
came from a residential area supporting a variety
of herbaceous ground cover and indirect and direct
artificial lighting for much of the night. The
Cocole (C) sample (ca. 18 km SSE of Gamboa) came
from a shallow gravel pit; there is no artificial light-
ing and, except for an unpaved road, the site was
covered with a dense stand of 1-1.5 m high grass.
The Los Santos (LS) sample (from the Peninsula
de Azuero, ca. 150 km SW of Gamboa) came from
the border of a drying cattle tank; the tank was
surrounded by small clumps of thorny scrub and
patches of short grass and herbs; there was no arti-
ficial lighting. Appendix I summarizes the collect-
ing data.
All the specimens from the preceding five sam-
ples were preserved in 10% formalin within three
hours of capture. Stomach contents were removed
later in the laboratory, air dried, weighed, sorted,
NUMBER 284
FIGURE 3.?Sketch of dock/causeway study area, on BCI. (Wavy lines
same as Figure 2.)
water; other symbols
counted, and identified to lowest reliable taxonomic
level. Gonadal examination was also performed
later in the laboratory.
All measurements (testes) or weights (ovaries and
fat bodies) were taken from only one side, the
right. Histological sections were made from several
testes of each monthly sample. Further details on
materials and methods are presented in the follow-
ing sections, where appropriate.
Sex, Size, and Growth
PHYSICAL APPEARANCE
To pervert a phrase from Gertrude Stein, a toad
is a toad is a toad?only more so for the marine
toad. Marine toads have few admirers and are
usually described in a derogatory manner, such as
looking like mobile cow patties. Such epithets are,
however, of little help in distinguishing them from
their warty brethren. Although the recently meta-
morphosed toadlets may cause confusion, they are
unique in bearing regular rows of paravertebral
tubercles. Once the toads reach a snout-vent length
of approximately 50 mm, they can be recognized
by the following set of characters: a heavy-bodied
toad with its maximum width nearly three-fourths
of its body length; broad head with truncate snout;
bony ridges (cranial crest) on the periphery of the
head, i.e., canthal, preorbital, postorbital, and short
8 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
temporal ridges; very large ovate to triangular paro-
toid gland extending from the temporal ridge to a
level beyond the axillary region; the tympanum
distinct, about half the diameter of the eye; skin on
back and legs very warty; fingers free, toes distinctly
webbed for at least half of their length; juveniles
of both sexes and mature females with a mottled
dorsal pattern, ground color a dusky brown covered
with irregular blotches of beige and chocolate,
large chocolate scapular blotches and usually a
beige mid-dorsal stripe present; adult males uni-
color cinnamon brown and spinose, each wart, no
matter what its size, capped by one to many horny
spines.
The tadpoles of B. marinus are small (approx.
10-25 mm in total length), round-bodied creatures.
The body and tail are black or dark brown with
a distinctive pale cream stripe along the lower edge
of the caudal musculature. The tail fin is a uniform
translucent gray. The mouthparts show the typical
bufonid condition with wide oral disc laterally
papillate and strong, keratinous jaws with serrate
margins. Further details of tadpole morphology are
in Altig (1970), Breder (1946), Kenny (1969), and
Savage (1960).
SEXUAL MATURITY AND REPRODUCTIVE CYCLE
MATURITY.?Female toads from Papua New
Guinea reach sexual maturity at a snout-vent
length of 70-80 mm (Zug et al., 1975). In the Canal
Zone, Panama, females apparently do not reach
maturity until they are 90-100 mm long. Only a
single 93 mm toad showed oogenesis; all other
gravid or actively oogenetic females were greater
than 100 mm. The largest inactive oogenetic female
was 102 mm, and there were several other pre-
sumably "immature" females in the 90?100 mm
class.
Sexual maturity in males is more difficult to de-
termine. The smallest male in the Canal Zone
samples with well-developed secondary sex charac-
teristics such as uniform brown dorsum and kerati-
nous thumb patches was 103 mm. Several males in
the 85-95 mm size class showed the transition pat-
tern between the juvenile/female coloration and that
of sexually mature males (captive males developed
nuptial pads one to two weeks prior to juvenile
pattern transition?F. McCullough, pers. comm.).
On BCI, the smallest calling and amplexing male
was 106 mm. Of the 31 BCI toads with distinct
male characteristics, only 3 were less than 90 mm,
and the smallest was 84 mm. Thus the circumstan-
tial evidence suggests that male toads mature at
a somewhat smaller body length (85-95 mm) than
the females. Histological sections of the testes of
several males (snout-vent lengths of 81, 84, 87, 88,
and 90 mm) show heavy concentrations of sperma-
tozoa and spermatids in the seminiferous tubules
of the largest individual. The other males show
active spermatogenesis with a predominance of pri-
mary spermatocytes in the smallest specimen and
secondary spermatocytes and spermatids in the
middle three specimens. On histological grounds,
TABLE 3.?Selected examples of sex ratios from anuran populations
Taxon
Sex ratio
females:males
Raw
252:643
275:545
130:262
209:698
120:647
34:66
20:80
36.64
7:93
80:148
17:52
55:94
.
309:169
Calculated
1:2.6
1:2.0
1:2.0
1:3.3
1:5.4
1:1.9
1:4.0
1:1.8
1:13.3
1:1.8
1:3.1
1:1.7
1:10
1.8:1
Situation Author
Bufo bufo
B. calamita
B. vailiceps
Gastrophryne carolinensis
Pseudacris tr iseriata
Rana pretiosa
mixed biotopes (Mar-Oct)
n n II
II H n
breeding pond (May-Jun)
breeding pond (Apr-Aug)
breeding pond (Jun)
abandon f ie ld (Apr-May)
flooded f i e ld (Feb)
breeding pond (Feb-Apr)
pond (summer)
Heusser, 1968
?I II
Heusser, 1969
Bla i r , 1960
?I M
H II
II II
Anderson, 1954
Whitaker, 1971
Turner, 1960a
NUMBER 284
maturity (or more specifically reproductive ability)
is attained at approximately 90 mm snout-vent
length.
SEX RATIOS.?In anurans, there usually is a pre-
ponderance of one sex over the other (Table 3).
Although the data are limited, males appear to out-
number females in terrestrial species, females out-
number males in semiaquatic species.
Even though our data for B. marinus are scanty,
different sex ratios for populations in different
habitats are apparent. Counting only the sexually
mature individuals in our samples, males outnum-
ber females at Cocole (males; females, total num-
ber?1:2.3, 10), Gamboa (1:6, 7), and Summit Gar-
den (1:3.2, 38). Females outnumber males at Barro
Colorado Island (2.4:1, 88), Las Cumbres (4.7:1,
17), Los Santos (1.5:1, 15), Summit Hill (2:1, 9),
and El Real, Darien (4.1:1, 82; data from C. W.
Myers). The most striking point of these data is
the strong tendency for one sex to be at least twice
as numerous as the other. Why the sexual imbal-
ance should be so strong is unclear, and our knowl-
edge of the different populations and their habi-
tats is too sparse to draw definite conclusions. These
data suggest, however, that the sex ratio is not
species specific, but largely determined by envi-
ronmental vagaries (abiotic as well as biotic) en-
countered by each population.
REPRODUCTION.?The inability to obtain large
monthly samples from the Canal Zone leave this
life history aspect unresolved. Our data are sugges-
tive and nothing more.
A summary of the seasonality of oogenesis for
Canal Zone toads is presented in Figure 4. Four
oogenetic states were selected (Jorgensen, 1973):
previtellogenesis, the ovaries are small and granular
in appearance; early vitellogenesis, the ovaries are
still small, although the oocytes become evident as
discrete cells; late vitellogenesis, the ovaries are
greatly enlarged and the oocytes are well yoked and
beginning to pigment; gravid, the ova are mature
and the ovaries fill the entire body cavity. These
states are equivalent to the body weight/ovary
weight proportions (Zug et al., 1975), except that
the late vitellogenic state encompasses the values
of 30 through 100.
The presence of gravid females in April and May
suggests that reproduction occurs then, during the
late dry season and the early wet season. The active
vitellogenesis in these same samples further indi-
PV EV LV MO
JAN ?
APR ? D
NOV A
DEC D
FIGURE 4.?Reproductive state of adult (>95 mm) toads in
Canal Zone. (PV = previtellogenesis, EV = early vitello-
genesis, LV = late vitellogenesis. MO = mature ova or
gravid; square = LC sample, circle = C sample, triangle =
G-SH sample.)
9
l l lJi. IL
0 .5 I 2 0 .5 I 2 0 .5 I 2 0 .5 I 2 0 5 1 2
FAT BODY WEIGHT lq)
PV EV LV
LOS
MO SANTOS
FIGURE 5.?Fat body weights of adult females in different
phases of oogenesis. (Fat body weights, in grams, are divided
into four classes: <0.5, 0.51-1.0, 1.01-2.0, > 20; Canal Zone
sample?C, G, LC, SH?is segregated into four phases of
oogenesis; see Figure 4 for explanation of abbreviations.)
cates that reproduction might continue into June
and, possibly, July. Whether reproduction occurs
at other times remains unanswered. Of the nine
mature females from Los Santos, eight are in a
previtellogenic or very early vitellogenic phase, the
ninth in mid-vitellogenesis. From this we assume
that the toads of this xeric area breed in the mid
or late wet season, i.e., July or later, for there is
10 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
o
? 4
0 .5 I 20 5 1 2 0 5 1 2 0 . 5 1
FAT BODY WEIGHT (g)
JAN APR MAY NOV
FIGURE 6.?Monthly variation in fat body weights of adult
(>90 mm) male toads from the Canal Zone. (Weight classes
same as Figure 5.)
no evidence of May breeding (beginning of the wet
season).
The fat body weights (Figure 5) suggest that the
Los Santos female toads require a period of intense
feeding in order to replenish their lipid stores be-
fore vitellogenesis can begin. The necessity of feed-
ing at the beginning of the wet season may delay
breeding until the mid or late wet season. A por-
tion of the Canal Zone toads, in contrast, have
more than adequate lipid reserves for vitellogenesis,
because the fat bodies are not entirely depleted by
the completion of vitellogenesis. The six mature
males from Los Santos also have small to non-
existent fat bodies (<0.5 g). The fat bodies of the
majority of adult males and all immature females
and males from the Canal Zone are small (Figure
6). Immature individuals would be expected to
direct all their energy into growth, whereas adult
males might be expected to use the fat bodies to
store energy for periods of low food availability or
for periods of nonfeeding during the breeding sea-
son. Although the data are meager, such storage is
not indicated.
The testis lengdi to snout-vent length ratio of
adult male toads was examined to determine if this
ratio reflected a spermatogenic cycle. For the sam-
ples from January, April, May, and November, the
ratios in the adult males (>90 mm) show no signifi-
cant difference between the months, means of 6.6,
6.2, 6.4, and 6.5 mm, respectively. Comparison of
immature to mature males show the testes of im-
matures to be proportionately smaller, although
the ranges of the two groups strongly overlap. We
can state only that testis length correlates well with
body length (N=27; Y=5.92+0.21X; r=0.73), and
upon maturity, the testes experience a growth surge.
Changes in volume or weight might reflect periods
of active spermatogenesis.
Testes from several adults were examined for
spermatogenic activity by measuring the maximum
diameter of 15 seminiferous tubules in the center of
the testis (Figure 7). There is clearly an increase in
tubule diameter from the beginning of the dry sea-
son (January) to the beginning of the wet season
(May), possibly followed by a decline of testicular
activity during the wet season. Although the num-
ber of sperm were not counted, the densest aggre-
gations in the seminiferous tubules occurred in the
April sample; the lightest aggregations in the Jan-
uary sample.
REMARKS.?The data on gonadic activity (Figures
4 and 7) suggest a seasonal reproductive cycle for
the Canal Zone toads. The major reproductive
effort appears to occur about the end of the dry
and the beginning of the wet season. In contrast,
the Los Santos toads are assumed to breed in the
middle of the wet season (July or later). A series of
65 adult females collected in October from El Real,
Darien, contained 53 individuals with enlarged
ovarian eggs, suggesting a major breeding period in
November or December (data from C. W. Myers).
However, the data are so limited that they do not
preclude sporadic breeding throughout the remain-
der of the year or, at least, another peak of breed-
ing activity. Breder's data (1946) showed spawning
3
CO
8
r -
ou
s
t r
(mm
!
en
UJ
i??
i . i
4-
3-
2-
o
o
o
8
UJ
CO
J F M A M J J A S O N O
MONTHS
FIGURE 7.?Seasonal change in testicular activity of adult
male toads from Canal Zone.
NUMBER 284 11
? ? SINALOA
? ? EL PETEN
? ? GUATEMALA
? ? CANAL ZONE
? ? I ? DARIEN
? WM TRINIDAD
J F M A M J J A S O N D
MONTHS
FIGURE 8.?Breeding activity of B. marinus within its natural
range. Data, top to bottom, are from Hardy and McDiarmid
(1969), Duellman (1963), M. Dix (pers. comm.), Breder (1946)
and our samples, Breder (1946) and C. W. Myers (pers.
comm.), Kenny (1969).
occurred in early January along the Rio Chagres,
Canal Zone. Assuming that such breeding still
occurs, we believe this represents a minor breeding
peak, because of the reduced testicular activity.
However, Breder's data demonstrate the major
fault of our reproductive data, aside from its small
size, is that we combined population that are sep-
arated by distances of 5 or more kilometers. We
realize that adjacent but separate populations may
have different breeding periodicity dependent upon
their local climatic regime and their choice of
breeding site. Hypothetically, stream breeding toad
populations might have been selected for dry season
reproduction, because of the low flow rate at that
time (analogous to Smilisca sila); whereas tempo-
rary- or shallow-pond toad populations breed in
the wet season when the ponds are full and stable.
Breder's data (1946) on tadpoles and reproduc-
tion choruses indicate an extended breeding sea-
son for the populations in the Rio Chucunaque
drainage of Darien, Panama. Data from Latin
TABLE 4.?Development of B. marinus tadpoles from hatch-
ing to metamorphosis
Locali ty
Oahu
Puerto Rico
Panama
Negros Island
Queensland
Trinidad
Venezuela
Duration
30 days
60-70 days
4-6 weeks
7-8 weeks
75-80 days
6 weeks
29-32 days (experimental)
56 days (natural )
Source
Pemberton, 1934
Se1n, 1937
Breder, 1946
Alca la , 1957
Straughn, 1966
Kenny, 1969
Durant, 1974
Durant, 1974
America are summarized in Figure 8. The figure
is based on actual spawning and/or presence of
tadpoles, and not choruses for the latter can be
misleading, e.g., Honegger's fig. 3 (1970). Most pop-
ulations seem to have two peaks of breeding
activity. In spite of statements to the contrary, e.g.,
breeding throughout the year in Nicaragua (Villa,
1972), we believe that this bimodal reproduction
is typical of the majority of native B. marinus pop
ulations.
LARVAL DEVELOPMENT AND METAMORPHOSIS.?
Like many aspects of the marine toad's life cycle,
no thorough study of its larval biology has been
done. Females deposit long strings of eggs, contain-
ing thousands of ova (Figure 9), in the shallow
water of temporary or permanent streams and
ponds. It is assumed that the female releases her
entire clutch during a single amplexus, rather than
mating with several males over several days or
weeks. The eggs hatch in 36 hours (Kenny, 1969)
to four days (Breder, 1946), and the larvae grow
and metamorphose in one to two months (Table 4).
We have no data on development time for the
Canal Zone toads. Mass metamorphosis was ob-
served twice. In late March, 1974, large number of
toadlets were metamorphosing from the few re-
maining pools in the drying streams of the Madden
Forest Reserve; no metamorphosis was observed on
BCI at that time. Mass metamorphosis occurred on
BCI in mid May 1975 but not in May 1976. In
addition to indicating an early dry season breeding
period, these data suggest an irregular develop-
mental time that may in some way be regulated by
environmental conditions. The drying pools may
have forced an earlier metamorphosis in the dry
season to increase the likelihood of survival of the
young. The ideal time of metamorphosis would
seem to be at the beginning of the wet season, e.g.,
May 1975 sample. Insect abundance is rapidly in-
creasing at this time and the high humidity permits
rapid and distant dispersal.
BODY SIZE
Bufo marinus is purported to contain the
"world's largest toad" (230 mm snout-vent length,
Surinam; Reed and Browsky, 1970). Although this
fact in and of itself has no biological significance, it
does emphasize several points. Adult marine toads
are generally large (100-150 mm) and, thus, poten-
12 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
30000
CO
LJ_
O
LU
CD
15000
6000
3000
Y=0.0006X
r = 0.54
3.497
10 14 19
SNOUT-VENT LENGTH (cm)
FIGURE 9.?Relationship between snout-vent length of gravid females and number of mature ova
in both ovaries. (Total number of ova estimated by counting ova in 0.4 to 0.8 g sample of
ovary and then extrapolating this value to total ovarian weight.)
tially one of the major nocturnal predators on
small terrestrial animals. The large size of adults
coupled with their large parotoid glands decrease
the probability of predator attacks and suggest the
likelihood of increased longevity once adulthood is
reached. The large size of females increases their
reproductive potential by enabling them to carry
more eggs. A final point, which is not documented
but a common impression held by many herpetol-
ogists, is that the marine toads of the Guiana shield
area tend to average larger than anywhere else.
For our Panamanian data, we shall examine only
two aspects of body size: the possibility of sexual
and regional differences in body size.
SEXUAL DIFFERENCES.?Only three samples, Barro
Colorado Island, Summit Garden, and El Real, are
sufficiently large to analyze for size differences in
sexually mature adults (see section on maturity,
p. 8). The weight-length regressions (Figure 10
and Table 5) are very similar for males and females
in the BCI and ER samples. At the larger body
lengths, females tend to be slightly larger, i.e.,
higher slopes or allometric constants, than males
but not greatly so. These slight differences result, at
least in part, from the wider ranges of body lengths
in the females, e.g., BCI males 90-130 mm, BCI
females 95-160 mm. The narrow size range
(90-110 mm) in the SG sample results in signifi-
cantly different slopes for the male and female
regression curves. Although it will become more
NUMBER 284 13
600
BARRO COLORADO ISLAND
40 100
SNOUT-VENT LENGTH (cm)
200
FIGURE 10.?Sexual differences in weight-length regression (power) of Barro Colorado Island and
El Real toad samples. (Crosses and solid line = females; circles and broken line = males; data
for power equations in Table 5.)
14 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
TABLE 5.?Summary of weight (Y, grams) and length (X,
millimeters) regression relationships of Bufo marinus at six
localities (A and B data derived from power regression
Y = AXB; r is correlation coefficient)
Local i ty
Barro Colorado Island
female
male
Summit Garden
female
male
Cocole
Gamboa
Summit H i l l
El Real
female
male
N
209
54
26
50
7
23
11
12
19
83
65
17
A
0.00008
0.00001
0.00004
0.00010
0.00003
0.00187
0.00013
0.00010
0.00005
0.00010
0.00010
0.00040
B
3.019
3.418
3.137
2.965
3.240
2.340
2.906
2.977
3.127
3.030
2.933
2.722
r
0.99
0.98
0.93
0.96
0.97
0.82
0.98
0.99
0.98
0.94
0.92
0.95
apparent in the later discussion on age or size dis-
tribution, females average larger than males (see
ER sample in Figure 10).
REGIONAL DIFFERENCES.?A comparison of six
sampling localities (Table 5) demonstrates no major
differences in the body weight-length regressions
between any of the localities. In all the samples,
the Y-axis intercept is nearly 0, and the slope is
nearly 3. Although the habitats of the sampling
localities may differ?strikingly so to our eyes?
these differences have caused no apparent differ-
ences in the growth of the toads, such as were seen
in the savanna and the rainforest margin toads in
Papua New Guinea (Zug et al., 1975). This uni-
formity is somewhat surprising as we would expect
populational adjustments in body dimensions rela-
tive to the harshness of its environment, particu-
larly since such adjustments have been observed
elsewhere.
GROWTH
The likelihood of death for a marine toad is
probably greatest from metamorphosis to subadult-
hood. During this period, the toadlet is a juicy
little morsel that has lost the toxicity of its larval
stage (Licht, 1967; Wassersug, 1973) and has not yet
gained the protection of large body size and well-
developed parotoid glands. The dangers of juvenile
life would place a high premium on rapid growth.
The data from other species of toads (Turner,
1960b; Clarke, 1974; Labanick and Schlueter, 1976)
show that they reach adult size within one full year
of growth (not necessarily a calendar year) after
metamorphosis. The marine toad has this capa-
bility or, at least the introduced toads in Hawaii
(Pemberton, 1934) do. Our data for the Barro Colo-
rado Island marinus indicate a similar rapid
growth.
LARVAL GROWTH.?We have no data on larval
growth and only a minuscule amount exists in the
literature. The time available for growth is vari-
able, ranging from one to nearly three months
(Table 4). Presumably the shorter the larval period,
the smaller is the metamorphosing toadlet; how-
ever, Durant's data (1974: fig. 4) show that weight
of toadlets metamorphosing at 32 days is the same
as those metamorphosing at 56 days. Straughn
(1966) shows the tadpoles developing from approxi-
mately 5 mm at hatching to 15 mm at metamor-
phosis; this yields a growth rate of approximately
0.128 mm/day.
POSTMETAMORPHIC GROWTH.?The growth of the
postmetamorphic marine toad is as poorly known
as that of the larval stages. There are a few anec-
dotal reports. Pemberton (1934) reports toadlets in
Hawaii grow from 6-12 mm to 60-75 mm in three
months and reach 90-120 mm in six months; this
would be a minimum growth rate of 0.434 mm/day.
Australian toads reach approximately 76 mm by the
beginning of their second summer and continue
growing for another three years at roughly 25
mm/year (Straughan, 1966); these data yield growth
rates of 0.139 and 0.068 mm/day, respectively. The
controls in two growth hormone experiments
(Zipser, Licht, and Bern, 1969) show growth rates
of 0.287 g/day and 0.471 g/day during eight and
six week runs, respectively; the initial weight of
each sample was 20.1 g.
Our data on growth derive from the mark-
recapture study on Barro Colorado Island; the
Summit Garden study had too few recaptures to
be reliable. Of the BCI toads, we had 47 recaptures;
20 of these are multiple recaptures. The intervals
vary from five to 402 days; the maximum time be-
tween the initial marking and the last recapture
is 760 days. Unfortunately, of the 75 postmeta-
morphics marked in May 1975, none were recov-
ered the following May, so our growth data is
largely confined to older juveniles (70-85 mm), sub-
adults, and adults.
To examine the history of growth in BCI toads,
we have been forced to make several assumptions
NUMBER 284 15
TABLE 6.?Growth in recently metamorphosed Bufo marinus
from laboratory/housing study area on Barro Colorado Island
in May 1975 (body length in mm, rate in mm per day)
Sampling
per iod
8 May
15 May
18 May
19 May
20 May
21 May
22 May
Mean
snout-vent
length
18.2
20.9
22.7
23.9
25.2
28.0
25.6
Sample
size
9
19
12
11
12
14
8
Growth
rate
_
0.385
0.600
1.200
1 .300
2.800
-2.400
TABLE 7.?Body length to age conversion table (conversion
values based on growth rate of 0.65 mm/day for first five
weeks and 0.37 mm/day thereafter; fuller explanation given
in text)
and arbitrary decisions on the relationship between
size and age, because a complete growth series was
not obtained. Firstly, we do not know the size at
metamorphosis of the May 1974 toadlets, or how
old the toadlets were by the time they had reached
our study site (a minimum of 150 m from the
metamorphosis beach and probably considerably
longer as the toad hops). By measuring a small
sample (24) of metamorphosing toadlets, we deter-
mined the mean and modal snout-vent to be 11 mm.
By averaging the growth rate of recently metamor-
phosed toadlets (Table 6), we obtained a rate of
0.647 mm/day, and by extrapolation, we can postu-
late that the toadlets measured on May 8 are 11
days old. If they were to continue to grow at this
rate, they would reach 70 mm in 91 days (13
weeks), which is in complete agreement with Pem-
berton's findings; however, we believe this rate to
be too high for the larger BCI toads.
In the absence of data for the 30-70 mm toads,
we have two ways of estimating the growth rate:
(1) determine growth rate in captive toads; (2) use
the mean growth rate of mark/recapture BCI toads.
The mean growth rate in four captive Jamaican
toads (initial snout-vent lengths of 33, 34, 41, and
48 mm) is 0.369 mm/day; the toads were measured
at weekly intervals over a period of 155 days (data
from F. McCullough). The average growth rate
from the 76 recapture determinations is 0.373
mm/day. These two estimates are surprisingly simi-
lar considering that the toads were derived from
different populations and were subjected to quite
different environments and, as such, we believe
that 0.37 mm/day provides a reliable estimator to
Week
0
1CVJ
5
6
10
15
20
Snout-vent
length (mm)
11.0
15.5
20.0
33.5
36.1
46.5
59.5
72.5
Week
25
30
35
40
45
50
55
60
Snout-vent
length (mm)
85.5
98.5
111.5
124.5
137.5
150.5
163.5
176.5
convert known body length to age (Table 7). In the
table, we have carried the size to age conversion to
extremes, for growth can be expected to slow down
after sexual maturity and show occasional bursts
during periods of high food abundance. However,
these extremes are necessary as they provide us with
an initial age to plot our growth data.
Figure 11 summarizes the growth data for the
BCI sample. In spite of the arbitrariness of the age
assignment, the overall pattern is clear. Growth is
fairly rapid for several weeks, then it slows down,
but not drastically so, until sexual maturity is
reached. Sexually mature toads are still capable of
rapid growth, e.g., one female grew from 87 to 133
mm in ten weeks, however most toads greater than
110 mm show slower growth rates. The larger
toads tend to show a long periods of no or very
slow growth.
The variability observed in Figure 11 results
from at least three factors: (1) the arbitrariness of
age assignment, (2) seasonal differences, and (3)
sexual differences. Only the first is discussed here;
the latter two factors are investigated in subse-
quent sections. Most of the individuals upon first
capture were less than 100 mm long. On the basis
of our age conversion, this would suggest that they
are no more than 31 weeks old, which further sug-
gests that they metamorphosed in November or
later. We doubt this, because November is roughly
the last strong pulse of the wet season, insect abun-
dance is low, and conditions will continue to re-
main unfavorable for toads as the dry season ad-
vances, so a high and constant growth rate is
16 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
16
14
e
-H 12
Z 10 H
UJ
_ l
I -
UJ ?
O 6
CO
20 24 30 40 50
WEEKS SINCE METAMORPH
FIGURE 11.?Growth in Barro Colorado Island toads. (Assignment of age at first capture taken
from Table 7; intervals thereafter are actual time to next capture; each solid circle = sample of
recently metamorphosed toads (See Table 6), each line = different individual and its growth.)
unlikely. We are more inclined to believe that
these late juvenile and young adult toads are ap-
proximately one year old, having metamorphosed
at the beginning of the wet season (May or June)
?this will become more apparent when we exam-
ine the age/size distributions in a subsequent sec-
tion. This assumption leads to an interpretation
that recently postmetamorphic toads grow rapidly
for at least the first half of the wet season, growth
slows or nearly halts for the remainder of the wet
season and continues at this level through the dry
season. At the beginning of the next wet season,
the one-year-old toads have a burst of growth,
which ends quickly. Subsequent energy intake is
used either for maintenance or reproduction. Each
year as the toad gets larger and older, the growth
rate becomes lower until growth essentially stops.
This hypothesis is consistent with the lack of
growth of the 163 mm toad and the "2-year-old" 120
mm toad, and explains the bursts of growth in the
80-110 mm toads.
SEXUAL DIFFERENCES.?Since females tend to
reach larger body sizes than males, the females may
grow faster. The data in Table 8 do show a higher
NUMBER 284 17
60 70
OSIS (arbitrary age assignment)
80 90 100
growth rate for females, but not significantly so
(F=1.38, df 6/19; t=O.33O, df 25). The alternate
for reaching larger size is to grow for a longer pe-
riod of time. Our recapture data indirectly show
this, for all longer recapture records (Figure 11)
are females and, with the exception of the 163 mm
toad, all show steady growth. The slightly higher
growth rate and longer period of postmaturation
growth can thus explain the presence of the large
female size classes in marine toad populations.
SEASONALITY OF GROWTH.?The BCI toads en-
counter two distinct seasons?wet and dry?and
their activity patterns appear to change with the
season. Toads are most common at the beginning
of the wet season and abundances slowly decrease,
TABLE 8.?Sexual differences in growth rate of adult female
(snout-vent length, 90-130 mm) and adult male (85-130 mm)
toads during May 1975 on Barro Colorado Island
Sex
Females
Males
N
20
7
Mean
0
0
.570
.522
Standard
deviation
0.318
0.372
0
-0
Range
.000-1.000
.133-1.067
18 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
TABLE 9.?Seasonal differences in growth rate of juveniles and
adult females (snout-vent length, 75-130 mm) from Barro
Colorado Island
Sample
period
Nov-May
Feb-May
May-Jul
Mean
Standard
deviation Range
0.041
0.143
0.285
0.032
0.086
0.195
0.007-0.070
0.000-0.259
0.078-0.687
reaching the low point at the end of the dry season.
Insect abundances seem to follow the same pattern,
so we would expect maximum growth early in the
wet season and minimum growth during the dry.
The BCI toads do follow such a pattern of growth
(Table 9). Although the differences are not statis-
tically significant (F=3.71, df 2/15), the pattern is
clear. The highest growth rate occurs at the peak
of food abundance in the early wet season, lowest
growth rate during low food abundance during the
end of the wet season and the entire dry season.
Population Structure and Size
The close association of the marine toad with
the man-made landscape has created the false im-
pression that they are extremely abundant and
ubiquitous animals. Although they may be very
abundant, this abundance seems to occur only
where they are exotics (Pemberton, 1934; Zug et
al., 1975) and/or in close association with man
(Heatwole, 1966). We were constantly informed
throughout our study of the toad's superabun-
dance, in spite of their low visibility to us. The
data in this section and the activity section demon-
state that they are not nearly as abundant or ubiq-
uitous as supposed. In both sections, our discussion
will be confined largely to the seminatural popu-
lations of Barro Colorado Island and Summit Gar-
dens.
AGE/SIZE DISTRIBUTION
While it is possible to estimate the age of toads
(Table 7), these estimates are assumed to be accu-
rate only for toads less than 100 mm snout-vent
length, and the accuracy of prediction for these
size classes is less than desirable. Thus, our follow-
ing comments on age must be viewed as speculative.
The three censuses of the Summit Garden's pop
SNOUT-VENT LENGTH (cm)
4 5 6 7 8 9 10 II 12 13 14 15 16
females &
juveniles
males
NOV 74
MAY 76
FIGURE 12.?Size distributions of Summit Gardens toad popu-
lation. (Each class encompasses snout-vent length range of
5 mm; class limits 0.0-0.49, 0.50-0.99; juveniles and adult
females plotted above axis, adult males below.)
ulation show that it is composed predominantly of
young adults, one to two years old (Figure 12). Of
the adults, males form the majority. The popula-
tion is composed of moderate-sized individuals,
roughly half of which are late term juveniles and
half adults. Only during the May 1975 census did
we observe a very large male and female. Repro-
ductive activity was not observed, and only one
recent postmetamorphic toadlet was discovered.
There is a slight upward shift of the male size/age
classes between November 1974 and May 1975 and
a downward shift between May 1975 and May 1976.
The upward shift may reflect growth of a single
metamorphic cohort and the downward shift as the
appearance of a new metamorphic cohort; however,
the samples are too small and too widely spaced
to add much credibility to this suggestion.
The census data for the Barro Colorado Island
population is somewhat more complete. Like the
SG population, this population is composed pri-
marily of moderate-sized toads (late term juveniles
and young adults) for most of the censuses (Figure
13); however, in May 1975 a large metamorphic
cohort had joined the population. Also the BCI
population contains more large individuals, prob-
ably two or more years old. Of the sexually mature
NUMBER 284 19
SNOUT-VENT LENGTH (cm)
1 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18
APR 74
NOV 74
FEB 75
r ?
JUL75
MAY 76
FIGURE 13.?Size distributions of Barro Colorado Island toad
population. (Additional explanation in legend of Figure 12.)
toads, females predominate, nearly two and half
times as many females as males. Females also domi-
nate, nearly exclusively so, in the large size classes;
comparing the two main censuses, one male was
greater than 120 mm in 1975 and none in 1976,
in contrast to 11 females in 1975 and six in 1976.
The growth of males appears to halt or proceed
very slowly after sexual maturity, whereas the fe-
males continue to grow at a moderate to fast rate.
The most noteworthy features of Figure 13 con-
cern the recruitment and growth of single-aged
cohorts through the population. From November
1974 through July 1975, a cohort can be seen to
gradually increase in size from 65-90 mm in No-
vember, to 75-95 mm in February, to 80-105 mm
in May, and finally to 100-120 mm in July. In
May 1975, a metamorphic cohort joins the popula-
tion. This cohort is very numerous relative to the
metamorphic cohort of May 1976. The larger size
of the 1976 cohort members suggests that they had
metamorphosed three to four weeks earlier than
their sister cohort members had in the preceding
year. Also their low numbers suggest a higher mor-
tality, either of larval stages or recently metamor-
phosed toadlets. These data tend to support our
conjecture in the section on postmetamorphic
growth that the BCI toads take one full year to
reach sexual maturity. It is possible that some of
these one-year-old individuals participate in the
population's reproductive activity; however, we ex-
pect that their contribution is minimal until their
second year. The major reproductive effort would
appear to be confined to the relatively small group
of females larger than 110 mm and males larger
than 105 mm, but like many of our comments, this
is speculative.
ABUNDANCE
Population size is estimated by three different
methods: (1) Lincoln index and index with Bailey
modification, (2) Schumacher-Eschmeyer regression
method (1943), and (3) Zippin removal method
(1958). All estimates of population size from mark/
recapture studies are known to have biological
errors in their assumptions as well as sampling
errors, which often cause incorrect estimates of the
population size. We will not discuss these errors
here, since analyses of these difficulties have been
published by others, e.g., Eisenberg (1972), Etter-
shank and Ettershank (1973), and Turner (1978).
The use of three techniques provides a somewhat
independent means for the comparison of the esti-
mates (Table 10). Only the Zippin method requires
a brief explanation, since we did not remove in-
dividuals from the population. We treated the new
or unmarked toads caught each evening as the sam-
ple that was removed for that sampling period. In
all estimates, the recent postmetamorphs are an-
alyzed separately from the late-term juveniles, sub-
adults, and adults.
POSTMETAMORPHS.?During the May 1975 moni-
toring period 76 postmetamorphic toadlets were
20 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
captured on the BCI study site. As mentioned pre-
viously, the toadlets were migrating from the meta-
morphosis beach approximately 150 m away; pre-
sumably after reaching the study area, many were
staying. The estimates (Table 10) suggest a popula-
tion of approximately 200 toadlets. Although this
may be the case, our search was quite thorough, yet
we found no more than 20 individuals in any
evening, so 200 resident postmetamorphs may be
an overestimate. Since the toadlets were actively
dispersing, it seems more likely that many of them
were passing through the study site.
In May 1976, only four postmetamorphic toadlets
were captured on the BCI site. They were also
older individuals than those captured the preceding
year, and thus, many represent the resident rem-
nants of an April 1976 dispersal. The data are in-
sufficient to estimate the density of this particular
cohort.
OLD JUVENILES TO ADULTS.?The SG population
consisted solely of late-term juveniles, subadults,
and adults. The November sample reflects low
activity more than number of toads present. Both
May samples should be better estimates of popula-
tion size, since this is a period of high activity. All
three methods (Table 10) yield similar estimates
of density, approximately 54 toads per hectare in
1975 and 84 toads per hectare in 1976. The differ-
ences do support our impression that the toads
were more abundant in 1976 than in 1975. While
these densities seem to be an accurate estimate of
toad abundance around the lily ponds of the study
TABLE 10.?Summary of toad population estimates of Sum-
mit Gardens and Barro Colorado Island study sites (estimates
determined by Schumacher-Eschmeyer (1943) regression
method, Lincoln index, and Zippin (1958) removal method;
raw data for all calculations listed in Appendix I, Table B)
Samples
Summit Gardens
Nov 74
May 75
May 76
Barro Colorado
Apr 74
Nov 74
Feb 75
May 75
May 76
BCI toadlets
May 75
Sch-Esch
16
27
42
6
56
27
68
51
218
Lincoln
16
28 (22-33)
42
6
64 (60-67)
27
70 (68-71)
54 (52-56)
182 (81-300)
Zi ppin
11
24
38
8
30
_
68
53 (53-53)
79 (79-79)
area, they are misleading for the botanical garden
of Summit Gardens as a whole. The toads were
concentrated around the ponds. Elsewhere in the
garden, they were absent or density was extremely
low. No toads were ever observed during the 0.5
kilometer walk to the lower pond and the 0.1 km
walk between ponds, or in the areas adjacent to
the pond clearings. Apparently the ponds attracted
more insects, which in turn attracted the toads.
The variable abundance (Table 10) of the BCI
toads reflects the levels of toad activity during the
various sampling periods. While we will discuss
toad activity patterns later, it must be noted that
we found toads most active in May at the beginning
of the wet season, thus we will use only the May
estimates in the following discussion. Furthermore,
the high similarity of the estimates from the three
different methods and the lack of or low numbers
of unmarked toads on the last few survey evenings
indicate that we had censused nearly all members
of the population. These data yield an estimate of
184 toads per hectare in 1975 and 138 toads per
hectare in 1976. Although these data are of interest
for comparative purposes, the actual number of
toads living on the study site is the important num-
ber, for it is this number of individuals upon which
the continued survival of the BCI population de-
pends. Toads occur elsewhere on the island, but
in very low numbers. Toads are rarely seen in the
forest, and tree-fall clearings are usually small (ap-
proximately 100 sq. m) and shortlived. We would
hazard an estimate that marine toad density is no
more than 1 toad per hectare for the remainder of
the island. Even at the biological station, the dis-
tribution of toads is not uniform.
The BCI study site consists of the laboratory-
housing area separated (approximately 100 m) from
the dock area by a steep slope covered by low
grasses, herbs, shrubs, and a few buildings. This
area was thoroughly searched several times during
each monitoring period; no toads were ever found,
so it was excluded from the regular nightly censuses
and from the density determinations. Toads must
occur on this slope, at least temporarily, for we
have recorded switches in feeding sites of toads
from the dock to the lab and vice versa. On the
study sites, toads tended to congregate in lighted
areas to feed on the attracted insects; however not
all regularly lighted areas were used and preferences
shifted from visit to visit.
NUMBER 284 21
13
X 10
I?
o
z
UJ
_J
\- 7
55-
o
CO
3 -
2 -
I -+
10 30 50 100
WEEKS SINCE METAMORPHOSIS
150 180
FIGURE 14.?Hypothetical growth curve for a cohort of Barro Colorado toads metamorphosing
early in the wet season. (Closed bar = wet season; open bar = dry season; upper curve = female
growth; lower curve = male growth; point of convergence = sexual maturation; further details
provided in text.)
Abundance for the laboratory area was nearly
identical in May 1975 and May 1976, 34 and 33
individuals, respectively. In the dock area, 34 toads
were captured in 1975 and 20 in 1976. Thus, the
larger population of 1975 resulted entirely from an
increased abundance at the dock area. Even though
there is movement between the two areas, the data
suggest a higher stability of the laboratory area sub-
population. This suggestion is reinforced further
by the recapture data of November 1974 taken
during a period of low activity; 18 toads were
found in the laboratory area and 10 in the dock
area. Perhaps the stability of abundance in the
laboratory reflects the carrying capacity of this
area. In fact, the narrow range of the November
1974, May 1975, and May 1976 estimates (51-68)
suggests a stability for the BCI population. By com-
bining the areas of the study sites and the slope,
the entire clearing of the biological station is esti-
mated to occupy nearly a full hectare, which results
in a density of 51-68 toads per hectare. So the
marine toads may be considered abundant now on
BCI, certainly much more so than in 1928-30, when
Dunn (1931) saw only three toads during his 57
days on the island.
SPECULATION ON DEMOGRAPHY
Our data are too incomplete to yield an accurate
picture of marinus demography. Nonetheless, we
22 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
wish to extract as much demographic information
from our data as possible in order to provide future
workers with a bit of comparative data and, more
importantly, to expose the numerous gaps in the
knowledge of marine toad population dynamics.
Firstly we wish to remind researchers of Cagle's
(1956) and Turner's (1962) summaries of the pop-
ulation parameters that must be measured in order
to gain a full understanding of amphibian popula-
tion dynamics.
AGE DISTRIBUTION.?In Table 7, we developed a
hypothetical conversion table for estimating a
toad's age from its body length. The conversion
values are based on the assumption of constancy
of growth rate once the postmetamorphic growth
surge is passed. As we have shown, growth rate is
not constant; there are seasonal changes, sexual
differences, and age (or body size) differences. We
earlier speculated on how these differences might
modify the relationship of body length to age and
wish to carry this speculation one step further and
suggest ages for the various size classes.
At metamorphosis, the toadlets will range in
body length from 9-13 mm. Growth will be rapid
for the first five weeks, assuming that metamor-
phosis occurs near the beginning of the wet season;
the toadlets should triple their size (27-36 mm).
With increasing size and decreasing food abun-
dance as the wet season advances, the juveniles
grow slowly and reach a length of 60-80 mm by
the beginning of the dry season. During the dry
season, the growth rate is reduced further, so by
the end of their first year, the cohort's size range
is 75-95 mm. A few of the cohort have reached the
size of sexual maturity, but probably not the hor-
monal state of maturity. The same pattern of
growth, although not the same rate, is shown dur-
ing their second year, at the end of which the
females are 100-120 mm long and the males 90-110
mm, because the latter's growth rate is sharply
reduced with the attainment of sexual maturity.
This is the first year when all surviving members
of the cohort should be able to reproduce. The
third year shows the same pattern as the preceding
years with an additional reduction of growth, par-
ticularly in the males. Males at the end of three
years are 97-117 mm long, females 112-132. Beyond
this point, male growth essentially stops, and, if
Straughan's (1966) data can be applied to BCI toads,
the females have two more years of active, albeit
reduced, growth. (See Figure 14 for diagramatic
illustration of cohort growth.)
This hypothetical growth sequence converts the
May 1975 size distribution (Figure 13) to the fol-
lowing age distribution: a postmetamorphic cohort
(73 individuals), approximately one month old; two
individuals, approximately three months old, which
may represent a cohort that metamorphosed early
in the dry season or very rapidly growing toadlets
of the postmetamorphic cohort; a one-year-old
cohort of 49 members; a two-year-old cohort of six
female and five males; two females of a three-year-
old cohort and two males of three- or more year-
old cohort; a four- or more year-old cohort of six
females. The May 1976 age distribution is similar
but much reduced in number of individuals: post-
metamorphic cohort (4 individuals), one-year-old
cohort (32), two-year-old cohort (6 males, 6 females),
three-year-old cohort (4 females), four-year-old
cohort (2 females).
FERTILITY.?Once again our data are incomplete
for determining the frequency of reproduction and
the age-specific fertility. A rough estimate of clutch
size can be obtained from the egg number-body
length regression curve (Figure 9) when a female's
body length is known. However, the curve is based
on too few data points to indicate if there is peak
in the fertility or reproductive potential. Further-
more, the use of a size to age conversion table
assumes that individuals of the same body length
are also of the same age; this is certainly not true.
We have assumed that the BCI toads have a
single breeding season each year and that a female
breeds every year. Marine toads are opportunistic
elsewhere (Figure 8; Honegger, 1970: fig. 3), and
they also may be so on BCI, although this is doubt-
ful. If the population has more than one breeding
season each year, an individual female might also.
If the population possesses a single breeding season,
this does not necessarily imply that a female breeds
every year.
RECRUITMENT AND SURVIVORSHIP.?There is abso-
lutely no data on the survivorship in the larval
stage of the marine toad. We have assumed it to be
reduced as compared to other anuran species,
owing to the toxicity of the egg coat, distasteful-
ness, and schooling behavior of the tadpoles. Data
on the fertility of an egg mass, egg predators, and
hatching success are required as well as data on
tadpole survivorship.
NUMBER 284 23
TABLE 11.?Survivorship matrix for Barro Colorado Island
toad population (right diagonal side of matrix contains num-
ber of newly caught and marked individuals and number sub-
sequently recaptured; left diagonal side shows percent sur-
vival)
Sample
period
Apr
Nov
Feb
May
May
Apr
6
67%
17%
1974
Nov
0
30
13%
0%
Feb
0
0
13
77%
8%
1975
May
4
4
10
52
8%
1976
May
1
0
1
4
44
We continue to postulate that the highest mor-
tality or, at least, the highest mortality per unit
time occurs during and immediately following
metamorphosis. In addition to death due to drown-
ing and embryological abnormalities, the toadlets
tend to form large aggregations near the site of
metamorphosis and this probably results in a high
incidence of predation. Another source of mortality
may be dehydration and starvation if metamor-
phosis occurs during a dry spell.
As the toadlets continue to grow, mortality by
predation should decrease. Environmental mor-
tality would also decrease if these early growth
stages coincide with the favorable conditions of the
early wet season. The next period of major mor-
tality is probably the toads' first dry season. Lack
of adequate shelter would result in death by desic-
cation, and an extended dry season with its asso-
ciated low food abundance could lead to starvation
of those individuals with low energy reserves at
the beginning of the dry season. After reaching
maturity the toads should experience low mortality,
although they might die from the same environ-
mental stresses they experienced as juveniles.
In order to test these assumptions, we need a
more detailed monitoring of marine toad popula-
tions. Of the eggs laid, what percentage hatch?
How great or small is larval mortality? Is this
mortality due primarily to predation or environ-
mental stress? What is the rate of successful meta-
morphosis and what is the daily or weekly survival
rate of the postmetamorphic toadlets? How many
toadlets survive their first year? What is the year-
to-year survival rate of adults and is there a sexual
difference?
We can provide only estimates for the latter few
questions. Our estimates must be considered re-
cruitment values rather than survivorship, since we
have no concrete estimate of the egg or larval pop
ulation. Seventy-six postmetamorphic toads joined
the population in 1975; none were recaptured in
1976. In 1976, only four toadlets were seen to enter
the population. Table 11 provides an estimate of
late-term juvenile and adult survivorship. The an-
nual survivorship of this age or size class is sur-
prisingly low, less than 17 percent in most cases.
Only the April 1974 group show a significantly
higher survivorship, probably because of the small
sample size. It is also interesting to note that the
four survivors in 1975 appeared to be just termi-
nating their reproductive activity in 1974. The
longest-lived individual of this group was a male
with a 92 mm snout-vent length at first capture and
two years later was only 114 mm; this suggests a
minimum age of three years for the toad in 1976
and adds a credibility to our hypothetical growth
curve (Figure 14). Because of the disparity in sam-
ple sizes, an averaging of the survival rates would
be meaningless; we can only postulate that annual
survivorship of the adult toads is about 10 percent.
Whether 90 percent of the toads actually die re-
mains unanswered. Some may be moving into the
surrounding forest and still may contribute repro-
ductively to the population.
The BCI population gained 52 and 44 new
"adult" members in 1975 and 1976 respectively, or
a recruitment rate of 74 and 85 percent. These
values will also act as an estimate of the annual
turn-over rate. They are rather high for an animal
with a potential life span of 16 years (Pemberton,
1949).
An estimate of survival rate for marine toads
from egg to adulthood can be made by assuming
an initial deposition of 10,000 eggs and 50 individ-
uals surviving to sexual maturity. This yields an
estimate of 0.5% survival, which must be consid-
ered a maximum survival rate since a reproductive
effort of only 10,000 eggs is low for a single marinus
female.
GENERAL REMARKS
Bufo marinus is an extremely adaptable anuran,
as its extensive natural distribution and occurrence
24 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
in a wide range of tropical and subtropical habitats
demonstrate. This adaptability will undoubtedly be
reflected in many divergent population structures
and sizes, each an adaptation to environmental
constraints of a particular habitat. A brief compari-
son of the BCI and SG populations show such dif-
ferences. The SG population is substantially smaller
in abundance (28 in 1975; 42 in 1976) and density
(54/hectare; 84/hectare) as compared to the BCI
abundance (70; 54) and density (184; 138). The
BCI population was larger in 1975 than 1976,
whereas SG population showed the reverse trend.
The population structure of both population is rel-
atively stable from year to year but quite different
in organization. On BCI, at least four age classes
are present. The SG population appears to contain
only one, or at the most two age classes, thereby
suggesting the SG population is a migrant one that
recruits new individuals each year from an un-
known source area. We observed no reproductive
activity in the lily ponds, and the density of croco-
dilians in the pond may prevent successful repro-
duction there. In contrast, the BCI toads breed in
Gatun Lake adjacent to their year-round home and
their young move into the area occupied by their
parents.
Food and Feeding Behavior
The marine toad is an opportunist and will ap-
parently eat almost every animate object it can
catch. Its diet has been analyzed numerous times
(e.g., Dexter, 1932; Hinckley, 1963; Illingworth,
1941; Pippet, 1975; Wolcott, 1937; Zug et al., 1975)
in exotic areas, and although terrestrial arthropods
form the bulk of its diet, snails, earthworms, and
small vertebrates are also engulfed. They tend to
be sloppy eaters, since rocks, bits of plants, etc. are
frequently found in their digestive tracts. At least
in one population (Zug et al., 1975), plant matter
appears to have been intentionally ingested. They
also have adapted well to urban life and will eat
the canned food set outside for dogs and cats
(Alexander, 1965).
The feeding habits of the marine toads are herein
based on an examination of stomach contents, ex-
cept for the fecal samples of BCI toads. The toads
were preserved in 10% formalin within three hours
of capture. Later in the laboratory, the stomach
was slit open and the contents removed and stored
separately. The contents were air dried and weighed
in toto; plant matter and miscellaneous objects
(stones, pieces of glass, hair, etc.) were removed and
weighed individually. The weight of these two
classes of items usually accounted for less than 5%
of the stomach content and, as such, are considered
as accidentally ingested items. The animal matter
was sorted and identified to familial level or higher.
The number of items of each group were counted
and an estimate of body length of each item was
made.
The feeding success of an animal is usually de-
termined by the actual weight (wet or dry) and/or
volume of its stomach contents or the number of
prey items present. Although these values are use-
ful, we believe that an animal's feeding success can
be more accurately portrayed by determining the
actual biomass of the prey items in its stomach. We
determined the weight-to-length power regression
equations for the common prey items (Appendix II)
and then used these equations to calculate the bio-
mass of the ingested prey in each toad. Although
this method is more laborious, it provides a more
accurate estimate of the various prey groups' con-
tribution to the toad's energy budget. The weight
or volume of the ingested prey is greatly distorted
by differential digestion rates of the different prey,
e.g., several hours after ingestion, most of a soft-
bodied prey item will be digested whereas much of
a hard-shelled prey item will remain. Similarly,
number of prey is misleading; energetically one
cicada may equal a dozen ants. It is for these rea-
sons that we have chosen to use biomass of ingested
prey rather than the other parameters. We also
wish to encourage other researchers to do likewise
and have, therefore, included our arthropod weight-
length data in Appendix 3.
PREY PREFERENCE
The occurrence of prey types for our samples are
summarized in Appendix I, Table A. As noted
earlier for the introduced populations, marine toads
feed primarily on terrestrial arthropods. Ants and
beetles are the most numerous remains in the stom-
ach. Of the ants, Atta, the leaf-cutting ants, are com-
mon. The presence of these spiny ants, as well as
other distasteful and toxic arthropods, demonstrates
a lack of prey discrimination by the toads and a
high level of diet flexibility to adjust their ingestion
NUMBER 284 25
COCOLE GAMBOA
LOS SANTOS SUMMIT HILL
FIGURE 15.?Prey preference of marine toads from four localities. (Proportions calculated from
prey biomasses and only those prey representing more than 2.7%?10 or more degrees of circle?
are plotted separately.)
to any prey that is available. Grant (1948) also
noted the utter disregard of this species for noxious,
biting, or stinging prey.
REGIONAL AND SEASONAL DIFFERENCES.?Since
toads are opportunists, their diets reflect the re-
gional and seasonal abundance of prey. Our data
are insufficient to recognize any seasonal trends, but
do provide us with the impression that in most
localities (excluding the urban sample from Gam-
boa) ants are the staple items for most of the year
26 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
and only with the increase in prey abundance early
in the wet season do other prey become dominant
dietary items.
The diagrams of prey occurrences (Figure 15)
are based largely on samples collected in May, ex-
cept for the Gamboa sample, which was collected
in November. Ants dominate in both of the savanna
samples (Cocole and Los Santos) with beetles the
next common prey items. These two localities are
the most "natural" sites and, thus, the most accu-
rate indicators of the toad's diet before becoming a
commensal of man. The Summit Hill sample
demonstrates the dietary habits of toads depending
exclusively on the insect attraction ability of street
or parking-lot lights. The Gamboa sample shows
greater dietary diversity and reflects the diversity
of feeding sites in an urban environment. The
greatest diversity occurs in the BCI toads, which
use the lighted areas as their major feeding stations;
these areas attract insects from the surrounding
rainforest, secondary growth, buildings, and the
lake.
SIZE SELECTION.?Marine toads appear to be as
non-selective in their choice of prey size as
they are in their choice of prey items. They show
no tendency to increase prey size with an increase
in their snout-vent length (Table 12). The absence
of such a trend was tested further by examining
the linear regression of modal prey size to toad
head width in the four samples. Even though the
slopes of all regressions are positive, they are low,
as are the linear correlation coefficients (Cocole,
0.49; Gamboa, 0.13; Los Santos, 0.03; Summit Hill,
0.25). If selectivity exists for late-term juveniles,
subadults, and adults, it is for them to select prey
in the 5.1-10.0 mm (0.01-0.1 g) size class. Com-
bining all sample localities, 70% of toads had their
modal prey length fall in this size class, and the
percentage is slightly higher if mean prey length is
used. We suspect that this results from the inter-
action of two factors: (1) large toads tend to ignore
prey less than 2.0 mm long; and (2) the most abun-
dant prey items are in the 5.1-10.0 mm range.
Heatwole and Heatwole (1968) showed that the
degree of physiological satiation modifies a Fowler's
toad (Bufo fowleri) choice of prey size. The upper
size-threshold of acceptable prey decreases as the
toad becomes increasingly satiated. The marine
toad may show a similar feeding response.
FEEDING BEHAVIOR
In spite of the marine toads' commonness, toler-
ance of human observers, and adaptability to lab-
oratory conditions, there are few studies of its
feeding behavior and physiology. Observations and
experiments on aspects of feeding, such as consump-
tion or ingestion rates, digestion rates, and prey
selection, are relatively easy to perform, and we had
planned to include them in our study. However,
largely through the lack of sufficient time we were
forced to abandon these experiments. The follow-
ing sections summarize a few of our casual observa-
tions, the published observations on feeding, and,
as usual, our speculative impressions.
CONSUMPTION RATES.?The rate of ingestion can
be expected to be a function of food availability.
If given an opportunity, marine toads are first-rate
gluttons. On BCI during periods of high prey
abundance, the toads would glut themselves on
insects, so much so that they would crackle when
picked up from the insect exoskeletons in their
stomachs. Such satiation is apparently not a rare
event, for 40% of the toads examined contained
over 50 prey items in their stomachs. Usually one
prey species (either an ant or a beetle) formed the
TABLE 12.?Frequency of different sized prey in different sized marine toads from Gamboa (prey
size classes selected to portray relative live weight; using equation W = O.OOOIIA length classes
approximately equivalent to <0.0001, 0.0001-0.001, 0.001-0.01, 0.01-0.1, 0.1-1.0, and >1.0
in grams, respectively)
Snout-vent
length
(mm)
64
86
92
105
110
Head
width
(mm)
26
33
34
37
40
<1.0 mm
0
0
0
2
0
1.1-2.0
0
0
0
3
0
2.1-5
1
29
0
0
5
Prey classes
.0 5.1-10.0
0
15
18
48
101
10.1-22.0
1
11
18
2
2
?22.0 mm
0
0
0
0
0
NUMBER 284 27
majority of the prey items. Another 10% of the
toads had empty stomachs. Since they were taken
concurrently with the toads with high prey counts,
it is obvious that they had not fed on the evening
of capture and perhaps not for several preceding
evenings. As we will show later, an individual is
not active every evening even during periods of
optimal prey abundance, suggesting that toads feed
maximally when active and then retire for several
days to assimilate their food.
Prey abundance is likely reflected in the biomass
estimates of the stomach contents. The Cocole
sample had an average of 1.1961 g of prey per
toad (range, 0.1359-2.3661), the Gamboa sample
2.3222 g (0.3472-6.7532), the Los Santos sample
0.8955 g (.0418-3.5533), and the Summit Hill sam-
ple 8.8589 g (0.1533-35.3671). The two "natural"
populations (Cocole and Los Santos) possess the
lowest stomach content biomass and are the only
ones with members with empty stomachs; two and
three individuals, respectively. The toads of these
two populations occur in savanna-like habitats and
depend upon the natural movement and distribu-
tion of prey in their feeding territories, whereas
the Gamboa and Summit Hill toads utilize the
artificial aggregation of prey around lights, thereby
elevating the number of prey available to them.
Although we have no data on frequency of feed-
ing, it seems likely that toads of the "artificial"
populations ingest more prey per unit time than
those of the "natural" populations. Thus, the con-
sumption rate of the former would be greater than
that of the latter. It would be of interest to dis-
cover if the latter populations compensate for a
lower consumption rate by increasing the duration
of their feeding activity.
The closest we can estimate consumption rate
is to use the daily intake value proposed by Heat-
wole and Heatwole (1968). This value is simply
the number of prey items in a toad's stomach, and
is based on the assumption that toads only feed
once a day and that food completes its passage
through the stomach before the next feeding ses-
sion begins. Based on this assumption, the May
samples from Cocole, Los Santos, and Summit
Hill and the November sample from Gamboa have
mean daily prey intake rate of 45.6, 30.0, 50.6, and
50.3, respectively, or combined 43.2 prey/toad.
Although numerous objections can be enumerated
against the validity of these estimates, until some-
one actually records the prey capture rates of wild
toads, this estimate provides a rough index to a
marine toad's consumption rate and a measure of
feeding activity/energetics during each feeding
bout.
LOCATING AND CATCHING PREY.?Vision appears
to be the primary sense for the detection and
capture of prey by toads (Heatwole and Heatwole,
1968). Marine toads certainly seem to rely heavily
on moving prey; however, the presence of snails
in their diet suggests that they require only slight
movement for detection. In addition, some marine
toads feed on stationary food (Alexander, 1965).
The latter is clearly a learned behavior, possibly
originating from the presence of moving prey on
the stationary food. If olfaction is used, its role
remains undiscovered.
Hearing has definitely been shown to be utilized
by the marine toad in prey location (Jaeger, 1976).
Jaeger observed an adult toad locate a pool by
following the calls of male Physalaemus pustulosus
and later eating two of the pustulosus. We ob-
served a similar event that suggests marine toads
recognize the various sounds made by their prey.
Some BCI students were dissecting an ant nest in
an area usually inhabited by at most a single
active toad. As soon as the students left, toads be-
ban to arrive at this site of high insect activity.
Eventually four toads were feeding on the ants,
apparently attracted to the site by the noise of the
ants attempting to repair their nest.
Ingle (1971, 1976) used marine toads in his ex-
periments on spatial vision in anurans. Like most
frogs, the marine toad possesses a snap zone that
is essentially a circular zone with the toad on the
edge of the circle, and any moving object within
the circle causes either an orientation movement
and/or a snap response. When tested with stimulus
pairs, the marine toad will (1) immediately snap
at one of the retreating stimuli, (2) follow with
its head directed at the leading stimuli, and (3)
duck approaching paired stimuli whether large
or small but ignore a single small stimulus.
INTERACTIONS.?We did not observe any antago-
nistic interactions between toads using the same
feeding sources. Tht toads seemed to distribute
themselves so they had non-overlapping snap zones.
In this way, they equally shared the resource area,
but not to the extreme of a ring of five toads al-
ternately feeding on emerging termites (Fellows,
28 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
1969). In laboratory experiments, R. Boice and
C. Boice (1970) discovered that marine toads es-
tablished intraspecific feeding heirarchies; heir-
archial position was positively correlated to body
length and was established by tongue striking (see
C. Boice and R. Boice, 1970, for Bufo feeding and
aggressive postures).
Activity and Movements
With the exception of tadpoles and recently
metamorphosed toadlets, marine toads are pre-
dominantly nocturnal. They appear shortly after
dusk and feed for a few hours before retiring.
Usually several nights will pass before a toad re-
appears again at its feeding station (Brattstrom,
1962:176), thus periods of inactivity exceed those
of activity.
In the following discussion, "activity" is divided
into two discrete categories, activity and movement.
Activity is the presence of a toad above ground
and engaged in an overt behavior, e.g., feeding,
calling, locomotion. We will examine activity pat-
terns of individuals and the population as a whole.
Movement is the presence of an active toad in dif-
ferent locations. Our data on movement examine
the nightly, weekly, and seasonal shifts in the
centers of activity of individuals.
MATERIALS AND METHODS
The data on activity and movements are derived
from the Barro Colorado Islands site and primarily
from the May 1975 and May 1976 monitoring
periods. However from the beginning of our study,
we recorded the individuals active each evening
and their locations. Beginning in May 1975, we
attempted to make hourly surveys and record
which toads were active and where. An evening
survey was begun at dusk, approximately 1850,
and concluded at 2330. The survey of the entire
laboratory site took 45 minutes and the dock site
20 minutes under normal circumstances. When
toad activity was high or when new toads had to
be measured and marked or other projects required
attention, the survey became irregular so that the
census hours were not sampled equally. Usually
the laboratory site was surveyed three times each
evening and the dock site two times.
During the April 1974, November 1974, and May
1975 surveys each toad had to be captured to iden-
tify it by its clipped toes. We were unsuccessful in
marking the toads with nail polish and butyrate
dope paint in April and November. The markings
either rubbed off or stimulated the toad to shed
its skin. In May 1975, the toads were marked with
silver nitrate branding (Thomas, 1975), but this
technique proved inadequate for marine toads.
The brands did not take well on the toad's skin,
the dark brand was difficult to decipher on the
dark background coloration of adult females and
juveniles, and the marks disappeared when the
skin was shed. In 1976, we used colored waistbands
(Emlen, 1968) with good success. The bands nearly
eliminated our handling of toads once the bands
were fitted. Aside from the occasional loss of a
band, the disadvantage of bands is an obscuring
of the colored marking by mud from the toad's
rest burrow or shed skin. We recommend the bands
for short term use only, because approximately 25
percent of the toads were developing skin ulcers
from waistband abrasion when we removed the
bands at the conclusion of our study.
For analysis of toad activity, we have used the
presence/absence data from all four monitoring
periods, with a heavy reliance on the May 1975
and May 1976 censuses. The movement data are
derived largely from the May 1976 census. Forag-
ing area or home range is calculated only for those
toads that were seen at locations at least 3 m apart.
After each hourly survey, the position of each toad
was plotted on a site map similar to those in
Figures 2 and 3. To determine forage area, an
overlay was placed on an accurately scaled site
map and the toads' locations during the different
censuses were plotted. All points were arbitrarily
enclosed in an irregular polygon. The area of this
polygon, excluding the area of enclosed buildings,
is our estimate of the toad's forage area. We believe
that in all cases it is an underestimate. The length
of the forage area is the maximum length of the
polygon. The maximum distance moved is the
longest distance moved during a single evening,
not successive evenings.
ACTIVITY
Bufo activity in the temperate zone is affected
by many factors; temperature, light intensity, moon
NUMBER 284 29
APR 74
NOV 74
D D O D D D
MAY 75 88
R R R R R R R R R R R R R
MAY 76
D D D D D D R D R R
D D D R R D D D D R R D R D D R
FIGURE 16.?Daily activity patterns of Barro Colorado Island
toads during four monitoring periods from 1974 to 1976.
(Open circle = dock site toads; closed circle = laboratory
site toads; each circle represents one toad, each column one
day; D = sampling day with no rain, R = sampling clay
with rain; blank columns in 1975 and 1976 histograms =
no census days.)
phase, weather, age, and season (for examples, see
Martof, 1962; Fitzgerald and Bider, 1974a, b, c).
Marine toad activity is presumably also modified
by similar factors; however few studies have been
made to identify the factors and their effects. Zug,
et al. (1975) showed activity was greatly depressed
during the Papuan dry season, and Brattstrom
(1962) noted an approximate three-day activity
cycle for BCI marine toads in the middle of the
wet season.
DAILY ACTIVITY.?Only a fraction of the total
toad population is active, even under the most
ideal conditions. During the May 1975 and May
1976 censuses, the wet season had begun; rain fell
frequently?on approximately 25 percent of the
census days?relative humidity was always 75 per-
cent after sunset, and insect abundance was high.
Yet on the average, only 31 percent of the popula-
tion was active each evening, and 50 percent was
the highest level of activity observed. The relative
level of activity is summarized in Figure 16 for
the dock and laboratory sites. The activity of the
toads in the two areas, while similar, does not
show an identical pattern. During periods of alter-
nating dry and rainy days, rain tends to stimulate
activity on the day of the rain and the following
day. As the number of consecutive dry days in-
creases, the activity of the population declines.
While there is no stability of activity levels, the
dock population seems to have less extreme fluc-
tuation in number of toads active. This may be
due, in part, to lower human activity and con-
stancy of light at the dock in the evening.
The activity patterns of individual toads are quite
variable. Only one individual, the largest adult
female in 1976, was active almost every night. The
majority show either a pattern of activity for two
to three nights in succession followed by several
nights of absence; others show a single evening
activity alternating with irregular periods of in-
activity and an occasional bout of activity on two
or three successive nights. A few toads were active
for only one or two nights during an entire two-
week observation period. The toads of the labora-
TABLE 13.?Summary of mean number of evenings a toad is
active on Barro Colorado Island (sample size is enclosed in
parentheses; juveniles are toads less than 90 mm and greater
than 60 mm snout-vent length)
Locality
Laboratory
Juveniles
Femal es
Males
All
Dock
Juveniles
Femal es
Males
All
Number of
survey days
Apr 74
0
1.5(2)
0
1.5(2)
0
0
3.2(4)
3.2(4)
6
Nov 74
1.3(11)
1.0(5)
1.0(2)
1.2(18)
1.5(6)
4.0(1)
1.5(2)
1.7(9)
12
May 75
2.2(11)
3.4(16)
2.8(6)
2.8(33)
3.3(6)
3.9(14)
2.9(12)
3.2(32)
10
May 76
2.2(8)
4.3(14)
4.2(4)
3.6(26)
5.0(2)
5.5(15)
5.5(4)
5.2(21)
14
30 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
TABLE 14.?Mean number of subadult and adult toads active per hour in Barro Colorado Island
study site (first number in each column is mean number of toads, number in parentheses is
number of sampling periods)
Sampling
period 1900-2000 2000-2100 2100-2200 2200-2300 2300-2400
Laboratory
May 1975
May 1976
Dock
May 1975
May 1976
6.1(7) 7.2(5) 4(4) 2.7(4)
4.5(13) 3.2(8) 2.5(14) 2(8) 0.8(4)
12.3(6) 5.7(3)
5(1) 7(11) 5.7(6) 5.3(8) 3.2(3)
tory site are active on the average about one day
out of every four, those from the dock site about
one day out of every three (Table 13). This gen-
eral pattern is probably characteristic of all popu-
lations.
When active, a toad is seldom active for the
entire night. Most toads appear and stay at a feed-
ing site for one to two hours and then disappear
for the night and usually for several successive
nights. The short duration of each night's activity
of the BCI toads is probably shared only with
other populations that depend upon lights as insect
attractants. Presumably toads from more "natural"
areas are active longer in order to catch sufficient
prey. Furthermore, toads from such areas may also
have to be active more frequently.
The greatest number of active toads occurs one
to two hours after dusk (Table 14). From 2000 on,
there is a steady decline in numbers, and by 2400
(midnight), few if any toads are found. This pattern
conforms to that of most other nocturnal ecto-
therms, and the few mornings we checked for pre-
TABLE 15.?Summary of mean number of toads active each
evening and frequency of wet days on Barro Colorado Island
during four monitoring periods
Locali ty
Laboratory
Dock
Total
Percent of
days with
rain
Apr
0.
2.
2.
0
74
5
2
7
Nov
1.
1 .
3.
100
74
8
4
2
May
9
11
20
20
75
.7
.1
.8
May
6
8
15
29
76
.9
.2
.1
dawn activity showed none. It is important to note
that the early evening activity peak coincides with
an increase in oxygen consumption and locomotor
activity during photoperiod experiments (figs. 1-3
and 12 in Hutchison and Kohl, 1971). Is the toad's
metabolic rate increasing in preparation for feeding
or simply a reflection of increased locomotor ac-
tivity?
Park et al. (1940) provide an enigmatic report of
exclusively diurnal activity in BCI marine toads.
Both their field observations and experimental data
show a preponderance of diurnal activity. We sus-
pect that they confused Bufo typhonius with B.
marinus, because they state that the toads were
abundant along the forest trail during the day.
B. typhonius is a forest toad, generally very abun-
dant, and exclusively diurnal; B. marinus is not.
SEXUAL AND AGE ACTIVITY PATTERNS.?Adult fe-
males show a slight tendency to be active more
frequently than either juveniles or adult males
(Table 13). The differences are slight. Since adult
males grow more slowly than either females or
juveniles, they might be expected to feed less fre-
quently, and thus show less activity. The converse
would be expected of juveniles, because they possess
the highest growth rate. The only time when male
activity exceeded that of females was in April when
the males were half-heartedly calling by the lake,
while the females high on the hillside ignored them.
SEASONAL ACTIVITY PATTERNS.?The first two sur-
veys^ ?one at the height of the dry season and the
other at the height of the wet season?demon-
strated that toad activity was strongly influenced
by seasonal weather conditions. At both times, an
average of only three toads were present each
evening (Table 15 and Figure 16). Toad activity at
NUMBER 284 31
the dock site was four times that of the laboratory
site in the dry season, probably in part because of
the greater humidity at the lake margin. At the
height of the wet season, the activity at the two sites
was nearly identical. The same situation occurred
at the beginning of the wet season (May 1975 and
May 1976), although the number of active toads
was five to ten times greater. In addition to more
toads being active each evening during the wet
season, each toad tended to be active during more
evenings (Table 13).
These activity patterns are predictable consider-
ing the marine toad's moisture and energy require-
ments. In the dry season, prey abundance is low to
moderate, but the aridity of air and soil imposes
rigid restraints on the length and frequency of
activity in order to avoid excess water loss. At the
height of the wet season, water loss is inconsequen-
tial, but prey abundance and activity are low. So
the toads must await short dry spells with increased
prey activity, if they are to balance energy expen-
diture of feeding with energy intake. Maximum
activity at the beginning of the wet seasons cor-
relates well with the high humidity and high prey
abundance and activity. It is at this time that the
toads can maximize their feeding efforts to grow
and lay in energy stores for the much leaner periods
of the wet and dry seasons.
MOVEMENT
The marine toad's wide distribution in tropical
America indicates a propensity for long distance
dispersal, hence a well-developed colonizing ability.
Our observations suggest an initial dispersal of the
postmetamorphic toadlets. They travel several hun-
dred meters before they are consistently found in
a single location. Whether the subadults and adults
have another dispersal phase and what percentage
of the population participates in such a dispersal
remains unknown.
FORAGING AREA.?There is no evidence of terri-
tories (defended areas) for marine toads in the wild.
However, when they share a feeding site, they do
not form closely packed aggregations, but rather
maintain a regular spacing between each other.
R. Boice and C. Boice (1970) report the establish-
ment of feeding hierarchies in a laboratory situa-
tion, so it is not unlikely that in the wild the spac-
TABLE 16.?Summary of forage area/home range dimensions
and maximum observed movement during a single evening
in Barro Colorado toads (sample sizes in parentheses)
Locality
Laboratory
Juveniles (2)
Females (10)
Males (3)
All (15)
Dock
Juveniles (1)
Females (11)
Males (3)
All (15)
Forage
area
(rn2)
95.1
169.2
167.8
159.1
213.4
155.4
166.3
161.4
Maximum
length of
area (m)
32.5
41.5
46.7
41.4
30.5
31.5
29.7
31.1
Maximum
length of
movement (m)
16.4(2)
22.5(7)
21.2(9)
_
9.4(4)
11.9(3)
10.5(7)
ing results from similar interactions with the larger
toads obtaining the better feeding stations.
Most toads (80% of our recapture sample)
changed their feeding sites either during a single
evening or during successive evenings of activity.
The area encompassing these different feeding sites
can be labeled a foraging area or a home range.
We prefer to use forage area for our data since it
is based on the distribution of feeding sites.
The foraging area for the BCI toads is approxi-
mately 160 square meters (Table 16). There are no
apparent area differences between the dock and
laboratory toads or between adult males and fe-
males; the juvenile sample is too small to be con-
clusive. The maximum length of the forage area
(Table 16) does show a pronounced difference be-
tween dock and laboratory toads. At the dock, the
length of the forage area is shorter than that at the
laboratory site. We doubt that this difference is
biologically significant and probably results from
topography of the site and our method of delineat-
ing forage areas. The topography of the sites and
greater irregularity of lighting at the laboratory
may also account for the greater distances traveled
in a single evening.
HOMING.?In 1959, Brattstrom (1962) tested the
"homing" ability of 15 marine toads at the BCI
laboratory site. Seventy-three percent of his experi-
mental toads return to their feeding stations within
seven days at which time the experiment was ter-
minated. He concluded that of those that were
recovered they had 100 percent accuracy in locating
the correct feeding station and that homing was
32 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
probably due to use of visual cues and retention of
previously learned topographic cues. Although we
do not fault his results or conclusions, we do not
believe that the experiment was a true test of
homing ability. Foremost, Brattstrom was assuming
that the toads had a single feeding station and that
seven days was sufficiently long for all the toads of
his sample to be active again. As we demonstrated
earlier, marine toads are not site specific, but occupy
a large forage area (mean area of 160 sq. m).
Furthermore, he noted that a toad is active on the
average of every 3.6 nights, which is similar to our
findings, but we have also shown that it may be
absent from a specific site for more than seven days.
Most of Brattstrom's toads were released within
their foraging area and were simply returning to a
preferred feeding site within that area. Some of the
toads were probably outside of their current forag-
ing area, but probably not outside of an area with
which they were familiar.
During our study, we observed movement of
toads between the dock and laboratory sites and
changes in forage areas. Two toads moved from the
laboratory site to the dock between monitoring
periods and one moved the opposite direction.
During monitoring periods, we recorded three move-
ments from dock to laboratory and one complete
round trip. Even though these movements are in
the minority, they do show that the BCI toads do
have larger "home" ranges than that implied by a
foraging area determined by a few nights of research.
Preferred feeding station shifts were common
during the May 1975 and May 1976 monitoring
periods. For example, the largest female in 1976
spent the first ten days with its activity center
around the Physalaemus pond with occasional mid-
evening excursions to the dining room bug light.
It then shifted its feeding station exclusively to the
dining room. Other toads showed similar shifts in
feeding station, frequently from one end of the
laboratory or dock sites to the other.
These movements strongly indicate that the toads
are familiar with most of the open (nonforested)
area around the STRI facility. Whether they could
return to this area if moved a half kilometer or
more into the forest remains an unknown. Within
the open area, we agree with Brattstrom that visual
recognition of landmarks serves as the main orienta-
tion cues. However, Williams (1967) has shown that
marine toads can use spatial cues (muscular imprint-
ing) to compensate for incorrect visual cues in
determining their direction of travel. Thus, there
is no reason to assume that they use a single set of
cues in moving about their foraging areas.
Physiology: Heat and Water
The geographic distribution, habitat selection,
and activity pattern of B. marinus appear to be
influenced by the toad's temperature tolerance or
preference and water economy. In this section, we
summarize the published data on B. marinus ther-
mal and water balance physiology and report on
a few physiological tests that we performed.
THERMAL PHYSIOLOGY
The marine toad has a wide temperature toler-
ance range. Native juvenile and adult toads have
a critical thermal minimum (CTMin) of 10?-12? C
and a critical thermal maxium (CTMax) of 41?-
42? C (Table 17). Tadpoles have a CTMax of
41.5?-42.5? C. The CTMin has not been deter-
mined for tadpoles, but Stuart (1951) reported tad-
poles and egg masses in mountain streams and
pools at 13.5? C. The CTMax and CTMin tem-
peratures reported here are ecological-behavioral
death points and not actual death points. At these
temperatures, the toads have lost the ability to
use coordinated movements to escape the adverse
temperatures (see Brattstrom (1968) for a discus-
sion of behavioral death versus lethal temperature).
Postmetamorphic toads are primarily nocturnal
animals that seek shelter in burrows or semibur-
rows during the day. Thus, they seldom, if ever,
encounter temperatures that would exceed their
CTMax, and it seems unlikely that their geo-
graphic distribution, choice of habitat, or activity
patterns are regulated by high temperatures. How-
ever, these traits may be influenced by low tem-
peratures. Stuart (1951) found Guatemalan toads
only below elevations of less than 1500 m, which
coincides with the isopleth for the mean minimum
temperature of 15? C. Similarly, the northern limit
of the toad's range in Texas coincides with the
15? C isopleth for the January mean minimum
daily temperature. These mean minimum tem-
peratures are 3 to 5 degrees higher than those of
the CTMin. This should not be unexpected, for
the former measures population survival over a
NUMBER 284 33
TABLE 17.?Summary of temperature tolerance of Bufo marinus
CTMin
CO
Adults
10-12
11
10
6.1
Tadpoles
13.5
CTMax
41-42
35.0
30.5
39.5
36.5?
39.3
41.8
40.0
40.8
38.9
40.0
41.5
41.6
42.5
Acclimation
Temp
10-12
41-42
7
7
27
15
26
38
27
37
10
25
NA
26-29
13.5
21
23
25.6
Time
(days)
NA
NA
1
2
7
14+
14+
14+
4
4
4+
7
0
3+
30+
7
NA
NA
Locali ty
Guatemala
Guatemala
Panama
Panama
Panama
Panama
Panama
Panama
Panama
Florida
Florida
Florida
Queensland
Queensland
St . Croix
Guatemala
Puerto Rico
Florida
Florida
Source
Stuart, 1951
Stuart, 1951
Brattstrom & Lawrence, 1962
Brattstrom & Lawrence, 1962
Brattstrom & Lawrence, 1962
Brattstrom, 1968
Brattstrom, 1968
Brattstrom, 1968
Brattstrom, 1968
Krakauer, 1970
Krakauer, 1970
Krakauer, 1970
Johnson, 1972
Johnson, 1972
McManus & Nell i s , 1976
Stuart, 1951
Heatwole et a l . , 1968
Krakauer, 1970
Krakauer, 1970
long period of the time, and the latter measures
individual survival for a brief moment under arti-
ficial conditions. Toads from the periphery of the
species range should be more hardy or low tem-
perature tolerant, since they are from a stock that
has evolved under wider temperature fluctuation
and periodic freezing temperatures and are also
acclimated to lower temperatures during the cold
part of the year. A comparison of low temperature
survival in Floridian toads (Krakauer, 1970; accli-
mated at 27? C for 4 days) and Panamanian toads
(Brattstrom, 1961; acclimated at 27? C for 3 days)
supports this point. All Panamanian toads were
dead after 48 hours at 7? C, yet the Floridian toads
were normal after 72 hours at 7? C and only after
96 hours was half of the sample dead.
In contrast to the postmetamorphic individuals,
the CTMin is of little importance to the eggs and
larvae. Admittedly, low temperature can be just
as hazardous to the premetamorphic stage as to
the postmetamorphic stage; however, the premeta-
morphic stage exists for a brief period of time?
seldom more than two months?so that their pres-
ence can be scheduled to avoid low temperature.
Thus, the CTMax may determine where the pre-
metamorphic stages can live, particularly so since
they are deposited and grow in shallow, commonly
unshaded bodies of water. The CTMax of tadpoles
is a degree higher than that of postmetamorphic
individuals (Table 17). Even though this difference
may not be significant due to differences in experi-
mental techniques, tadpoles probably encounter
higher temperature more frequently because of
their diurnal habits and the shallowness and open-
ness of their habitat, although they will tend to
aggregate in cooler areas of a pond rather than
the hot water at its margin (Mares, 1972). High
temperatures, as long as they are not too high, are
advantageous for the tadpoles, since the rate of
development and growth is directly proportional
to temperature. In many cases, B. marinus lays its
eggs in temporary pools and tadpoles must com-
plete their growth and development before the
pool dries. It seems likely that the eggs or embryos
of the marine toad are less temperature tolerant
than the tadpoles, since this occurs in other frogs
(Zweifel, 1977). Three species of Sonoran desert
toads have a maximum tolerance of 33?-34? C for
34 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
the survival and growth of their embryos (Zweifel,
1968). B. marinus embryos probably do not exceed
this tolerance limit.
Brattstrom's (1968) summary of thermal acclima-
tion in anurans shows that B. marinus requires
three days to completely acclimate to higher tem-
peratures. We question the biological significance
of such data for a tropical anuran such as the
marine toad. The survival value of complete accli-
mation seems minimal or non-existent for a toad
living in a relatively stable thermal environment
of the lowland tropics. Even toads in fluctuating
thermal environments experience temperature
change gradually and, thus, are probably con-
stantly acclimatizing to their environment. The
latter is certainly indicated by Johnson's (1972)
observations on daily variation in B. marinus'
CTMax. We are of the opinion that thermal accli-
mation is of less importance for survival of a toad
at temperature extremes than it is for the toad
to constantly adjust its biochemical and physio-
logical processes in order to optimize these
processes.
Several aspects of the marine toad's thermal
physiology remain undetermined, e.g., temperature
preference of pre- and postmetamorphic stages,
developmental and growth rates under different
temperature regimes, thermoregulation and daily
temperature regimes. The following experiment
is our attempt to provide a partial answer for the
last physiological parameter.
MATERIALS AND METHODS.?The body tempera-
tures of three toads from Summit Hill, Canal Zone,
Panama, were monitored for 24 hours. The toads
were not fed for five days prior to testing and were
fully hydrated by sitting in 1-2 cm of water for 24
hr prior to testing. Each toad was tested individu-
ally on consecutive days in a glass aquarium (150X
37X31 cm) in an open air, screen-enclosed labora-
tory. Core body temperature (Tb), air temperature
(Ta), and substrate temperature (Ts) were recorded
hourly, except for the period 0100-0500, by a YSI
tele-thermometer. Th was obtained by a small flexi-
ble vinyl probe (#402) inserted into the cloaca and
rectum for approximately one quarter of the toad's
body length; the probe was held in place by bands
of dermicel tape around the inguinal and the axil-
lary regions. Ta was obtained from an air tem-
perature probe (#405) suspended in the middle of
the aquarium, 10 cm from the bottom. T s was
3 5
I 2 4 6 8 10 12 14 16 18 20 22 24
TIME(hr)
N= I 1 0 1 0 3 1 3 2 3 2 3 2 4 4 4 3 3 3 3 3 3 3 3
FICURE 17.?Daily temperature regime of unrestrained Bufo
marinus. (Circle = mean body temperature; broken line =
mean air temperature; solid line = mean substrate tempera-
ture; bottom row along abcissa lists sample size by hour.)
obtained from a surface probe (#409) taped to the
bottom of the aquarium. Air could freely circulate
through the aquarium and half of the aquarium
received sunlight from approximately 1415 to 1530.
Relative humidity fluctuated between 65 and 85
percent (see Figure 18 for average daily regime).
RESULTS AND DISCUSSION.?As expected, the
marine toad's body temperature are closely tied to
the fluctuation in ambient temperatures (Figure
17); however, the daily fluctuation of body tem-
perature is less than that of ambient temperatures
(Ta and Ts). Body temperatures ranged from 25?
to 30? C with a daily mean of 26.4? C. Whether
this mean should be called the preferred body
temperature (Brattstrom, 1970) and whether the
greater stability of Tb as compared to Ta and T s
should be called thermoregulation are debatable.
Nonetheless the toads' body temperature fluctua-
tions are less than those of the environment. During
the night, Tb is nearly equivalent to Ta and Ts,
but during the day, Tb stays below Ta and Ts. The
lower temperature is obtained through evaporative
cooling from the body surface. Since evaporative
water loss can be controlled only to a limited degree,
the cooling is a passive physiological process. Even
though it is passive, it is continually experienced
by the toad and has probably been combined with
active physiological and behavioral aspects to
NUMBER 284 35
dampen Tb fluctuation. The actual fluctuation of
Tb is presumably even less than portrayed by this
simple test. The toad is buried or semiburied during
the day so the daytime temperatures of its shelter do
not become as high. We would postulate that only
during the dry season in open habitats could the
temperature of the shelter become high enough to
endanger the toad, and it is likely that by that point
the toad would be subjected to near lethal dehy-
dration.
WATER ECONOMY
Approximately 76 percent of the marine toad's
body weight is water (78.0%?Shoemaker, 1965;
74.1%?Thorson, 1964; 77.2%?Krakauer, 1970).
The difference between Thorson's value and those
of Shoemaker and Krakauer may indicate a differ-
ence in experimental method or a relationship
between total body water and size and/or sex. Thor-
son's sample certainly contained larger individuals
(137-450 g) than those of the latter two researchers
(100-200 g and 63-129 g, respectively), and this
size difference may reflect sexual difference, i.e.,
females tend to be heavier. Certainly, the robust-
ness and, hence, the weight of the skeleton is di-
rectly proportional to size. However, the effect of
increasing skeleton weight on total body weight
may be small in comparison to the influence ot
lipid concentration on body weight (Krakauer,
1970).
Of this total body water (= standard body weight
minus body weight when dried to a constant weight;
WS-WD), a toad can lose approximately 52 per-
cent (Krakauer, 1970) before it reaches its lethal
limits and dies (Figure 18). Relative to the toad's
standard bodv weight (Ws), the toad can lose nearly
40 percent of its Ws before death occurs. No matter
whether the lethal limit for total body water or for
Ws is used, the actual amount of water lost can be
somewhat higher under natural conditions than in
the laboratory, since the urinary bladders are emp-
tied prior to the laboratory tests. It has been known
(Steen, 1929) that the urinarv bladders of frogs act
as a water reservoir from which water can be re-
sorbed during dehydration stress. Although the
storage capacity of the marine toad's bladder is un-
known, the capacity is probably near 30 percent
of gross body weight (Wo) like that reported for
Bufo cognatus (Ruibal, 1962). If only 10 percent of
this bladder water (relative to gross body weight)
is available for resorption, this will enable the
toad to tolerate a 45 percent Ws loss before the
lethal limit is reached (Figure 18). Although the
lethal limits are high and indicate a definite adap-
tion to xeric habitats, other toads possess greater
tolerance to desiccation (fig. 4 in Krakauer, 1970),
and we suspect that desiccation is an important
mortality factor for marine toad populations in the
seasonal tropics.
In addition to the high tolerance to dehydration,
marine toads have other physiological and behav-
ioral means for reducing water loss and regaining
water. Presumably the main water loss occurs
across the skin. Toads reduce this loss by retreating
to sheltered areas during the day or other periods of
low relative humidity, and by assuming a crouched
posture that reduces the amount of skin surface
area exposed to the air. Further reduction is ob-
tained by decreasing the evaporative gradient at
the skin-air interface. This does not appear to be
an active physiological process, but a physical one
in which the skin loses moisture faster by evapora-
tion than it can be replaced by passive diffusion
from the inside. This diffusion lag creates a drier
skin and an associated reduction of the evaporative
gradient, thereby reducing the rate of water loss
(Machin, 1969). Another physical barrier to water
loss is the layer of mucopolvsaccharides in the ex-
ternal part of the dermis. This layer of ground sub-
stance is thicker in B. marinus than any other
amphibian studied (Elkan, 1968). Physiologically,
the marine toad shares the "water balance response"
with other frogs. As a frog or toad dehydrates, the
increasing concentration of plasma stimulates the
release of neurohypophyseal hormones that, in
turn, stimulate the resorption of water from the
bladder and increases the cutaneous absorption of
water from the surrounding water or soil (see
Bentley, 1971 for a summary of the "water balance
response;" Shoemaker, 1964; and Shoemaker and
Waring, 1968, for its specific action in B. marinus).
In addition to the active absorption of free water,
the external sculpturing of the toad's skin provides
a series of interconnecting channels that move water
from the venter upward along the sides to the back
thereby reducing cutaneous dehydration (Lillywhite
and Licht, 1974).
The marine toad absorbs water through its skin
(Shoemaker, 1964) when partially immersed in
36 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
UJ
o
en
u.
O
LJ
o
Q_
o
u
160-
140-
120-
100-
8 0 -
6 0 -
120 160
ACTUAL BODY WEIGHT (g)
200
FIGURE 18.?Influence of body weight on water economy in marine toad. (Lines based on total
water content of 76 percent of standard body weight (Wg), lethal limit (WD) of 52% of total
water content, gross body weight (WQ) with bladder water (10% of Wg) and dry weight (W<i) of
toad.)
water. It is likely that it also possesses the ability
to absorb water from damp soil like other Bufo
(Fair, 1970; Walker and Whitford, 1970). Although
it is often stated (e.g., Fair, 1970) that a water
balance deficit is required to initiate cutaneous
absorption, Shoemaker's data (fig. 1) indicates that,
at least for B. marinus, water exchange can occur
with no deficit. The rate of water uptake does
increase as the deficit increases, to about 15 percent,
at which time the rate levels off. The maximum rate
is about 1 ml/(100 g X hr). Presumably the marine
toad absorbs most of the water through a ventral
pelvic patch of highly vascularized skin (surface
area of this patch, y=O.13x +5.65; y=cm2, x=stan-
dard weight in g; Fair, 1970). This patch shows the
greatest change in osmotic permeability in response
to the water response hormone, vasotocin (Bentley
and Main, 1972) and has the most extensive vascu-
larization in the marine toad of all frogs studied
(Roth, 1973). Hydration occurs faster in free water
than in saturated soil (Walker and Whitford, 1970).
For information on salinity tolerance, which is
inappropriate to our discussion here, Ely (1944) and
Krakauer (1970) are recommended.
NUMBER 284 37
Since we believe dehydration to be a significant
mortality factor in marine toad populations, we
wished to determine rates of dehydration under
more natural conditions, as suggested by Seymour
and Lee (1974). The following experiments provide
such data.
MATERIALS AND METHODS.?The dehydration ex-
periments were performed in a screen-enclosed
laboratory so that the test animals were subjected
to the same temperature and humidity fluctuations
as those toads outside the laboratory. Each toad
was placed in a separate hardware-cloth cage, and
the cage was then placed on a dry sand platform.
This permits the toad to adpress its entire ventral
surface to the substrate, yet does not allow absorp-
tion of water. Body weight and relative humidity
were recorded every two hours, except for the
period from 0100-0500, for one day for each toad.
The relative humidity was measured by a sling
psychrometer swung one meter from the sand plat-
form. The average daily temperature regime is
illustrated in Figure 17. The toads were not fed
for at least three days and were fully hydrated by
placement in 1-2 cm of water for a minimum of
12 hours prior to testing. All toads were tested with
full urinary bladders.
An attempt to test the dehydration rates of the
resident Barro Colorado Island toads in a manner
similar to that proposed by Seymour and Lee
(1974) was unsuccessful. Toads weighed and re-
leased upon emergence could not be located the
following morning. The toads caught immediately
upon emergence and placed in hardware-cloth
cages were extremely excitable and would release
copious amounts of bladder water during the initial
capture and with each subsequent weigh-in. Also,
they repeatedly defecated during the test period so
that this test of dehydration had to be abandoned.
RESULTS AND DISCUSSION.?As the relative humid-
ity decreases and temperature increases the rate of
water loss (here, as in previous physiological studies,
we are assuming that weight loss is equivalent to
water loss) increases. This is strictly a physical
event, for any such increase in the evaporation
gradient increases evaporation rate. However, Fig-
ure 19 also shows that toads can modify the rate of
water loss. While toads B and C were inactive and
maintained a crouched or low sitting posture
throughout most of the test, toad A showed bursts
of activity and was often found sitting upright, both
of which actions would increase water loss by expos-
ing more skin surface and elevating respiratory
water loss.
The rate of water loss is clearly related to body
size (Figures 19 and 20; Table 18). Larger toads
lose more water and at a faster rate. However, pre-
senting the rate as g/hr or ml/hr is misleading as
far as an individual toad's survival is concerned,
for the standard formula ml/(100g X hr) or percent
body weight loss shows an inverse relationship
between water loss and size. It is this relationship
that is important to the survival of the toad in
nature. By referring to Figure 18 and using the rate
loss of Table 18, a 22 g toad has only 22 hours
before it reaches its lethal water loss, a 110 g toad
48 hours, and a 228 g toad 70 hours.
22
, -65
MIDNIGHT
FIGURE 19.?Daily water loss in marine toads of different body sizes. (Initial body weights: A ??
228 g, B = 110 g, C = 22 g; broken line = bihourly averages of relative humidity for four days
of testing.)
38 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
?
4 0 ?
o
UJ
^ ^ 0 . 5 -
o
?Ti 0 3 -
1 5 10 15 20 25
SNOUT-VENT LENGTH (cm)
FIGURE 20.?Rate of water loss relative to body size during
one day, with toads subjected to natural fluctuations of
temperature and humidity. (Circle = g/hr, Y = 0.057 X
0-57.V square = percent loss, Y = 136.71 X -o.42?, r = 0.99.)
The presence of bladder water is particularly
critical for the survival of the toads. In addition to
increasing the survival time of a dehydrating toad
by increasing the amount of expendable body water,
it enables the toad to maintain a more normal ionic
balance of its plasma. During dehydration, toads
with empty bladders have the sodium concentration
of their plasma increased, while toads with filled
bladders resorb bladder water and maintain a
normal sodium concentration (Shoemaker, 1964).
Presumably those toads with increased sodium con-
centration will be less resistant to parasite load,
starvation, etc., and might die from these other
causes in conjunction with dehydration.
As mentioned previously, we believe the dry
season to be the time and dehydration the major
cause of mortality in the Barro Colorado toad
population. This is based largely on the negative
evidence of low annual recapture rate and reduced
population activity during the dry season. The
BCI season usually extends from late December to
late April, with March being the driest month
TABLE 18.?Data summary of dehydration experiments on
Panamanian Bufo marinus (results for a 24-hour period)
Initial
body
weight
21.8
27.6
47.6
50.7
66.8
70.4
109.0
173.9
227.6
Weight
loss
8.6
8.7
11.6
13.4
15.1
16.1
21.8
25.9
30.8
Percent
body weight
loss (24 hr)
39.4
31.5
24.4
26.4
22.6
22.9
19.8
14.9
13.5
Weight
loss per
hour (g/hr)
0.358
0.363
0.483
0.558
0.629
0.671
0.908
1.079
1.283
Origin
of toad
Suninit H i l l
?Summit H i l l
Summit H111
Summit H i l l
Summit H111
?Summit H i l l
Gamboa
Summit H i l l
Gamboa
* The weight loss f o r these toads was obtained during the
thermal physiology t e s t .
(usually less than 5 cm of rainfall). The forest floor
is completely dried out and cracked; the soil of
the clearing is likely even drier and would provide
few moist, sheltered retreats. Since the juveniles
possess less tolerance to dehydration, they should
show the highest mortality, like that found in Bufo
debilis (Creusere and Whitford, 1976). Subadults
and adults probably compete for sheltered retreats,
with those in marginal shelters subject to mortality
if the dry season is extended or particularly severe.
Even for those toads not subjected to near lethal
dehydration, the dry season may restrict activity so
that the toads are incapable of catching sufficient
food to permit survival through the dry season.
Mortality Factors
Mortality usually results from the interaction of
both abiotic and biotic factors. While we are here
concerned with the major biotic factors, diseases,
parasites, and predators, we wish to emphasize our
belief that climate, particularly the dry season may
be a strong mortality factor in the post-larval stages.
The effect of climate is, at least, two pronged: (1)
dehydration owing to low moisture levels in air and
soil; (2) starvation owing to low prey abundance
and activity. Prey availability also appears to be low
in the latter part of the wet season with a slight
increase at the beginning of the dry season followed
by the lowest levels.
The following discussion deals exclusively with
postmetamorphic stages. There are no data on the
intensity of predation on marine toad egg masses
and tadpoles. It is generally assumed to be low
because of the toxicity of the egg jelly coat (Licht,
1967 and 1968) and the distastefulness of the tad-
poles (Wassersug, 1971); however, predation during
these life stages may be quite high owing to the
abundance of invertebrate and vertebrate predators
in and around tropical ponds (Durant, 1974; Heyer,
et al., 1975).
PREDATORS
Predators of introduced marine toads have been
summarized in Zug et al. (1975) and Covacevich
and Archer (1975). Reports on natural predators
are limited. Pope (1955) mentions Caiman latirostris,
and Allen and Neill (1956) list a Helicops. We ob-
served a Leptodeira annulata eating a 20 mm toadlet.
NUMBER 284 39
We suspect a large variety of mammals, birds, rep-
tiles, and amphibians, including adult B. marinus,
eat the recently metamorphosed toadlets and juve-
niles. It is our contention that few predators eat the
subadult and adult toads. In addition to their large
size, the toxic skin secretion, particularly from the
parotoid gland, acts as an effective predator de-
terrent. The structure and histochemistry of the
parotoid gland is analyzed by Hostetler and Can-
non (1974). Allen and Neill (1956) describe the
clinical effects of the toxin on man and provide a
short bibliography on this topic. Low (1972) sum-
marizes parotoid gland secretions stressing their
diversity and use in Bufo systematics. Licht and
Low (1968) tested the effects of marinus parotoid
secretions on the cardiac response of several snakes
and discovered that the secretions caused lethal
cardiac tetany in nontoad-eating species and no
cardiac disturbances in toad-eating species. Presum-
ably some birds and mammals could turn over the
adult toads and eat the viscera, thereby avoiding
the toxic skin secretions.
DISEASE AND PARASITES
Some of the parasitic or disease organisms of B.
marinus are listed in Table 19. Presently, there are
no data on whether they act as significant mortality
factors in the toads. During the dry season, almost
TABLE 19.?Parasites of Bufo marinus
Parasites Source
Bacteria
Salmonella (7 species)
Protozoans
Cytoamoeba bactipera
Ochotereneiia c f . digicauda
Tritrichomonas batrachoruw
Helminths
Creptotrema lynchi
Cyciindrotaenia americana
Glyptheimins robustus
Gorgoderina d1 aster
3ph
Rha
20
1otaenia bonariensis
bdias fuei ieborni
R. sphaerocephaia
Arthropods
Ambiyomma dissimi le
A. humerale
A. rotund urn
A", testudinis
Kourany, e t a l . , 1970
Lehmann, 1966
Mar inke l ie , 1970
Har inke i l e . 1968
Brooks, 1976
Brooks, 1976
Brooks, 1976
Brooks, 1976
Brooks, 1976
Kloss, 1971
Kloss, 1971
Jakowska, 1972
MaHnke l le , 1970
MaMnJcelle, 1970
Lehmann, e t a l . , 1969
every BCI toad has ticks, Amblyomma. Although
the ticks occur widely on the body, they appear to
be concentrated on the head, particularly on or
adjacent to the parotoid glands. When a tick falls
off, it leaves behind a small lesion. We never ob-
served any of these lesions to become infected or to
retard the normal activities of the toad.
We observed no heavy helminth infestations in
any of the toads we dissected for food and repro-
ductive data. In fact, we were surprised at the low
incidence of intestinal nematodes, and only in one
toad, did we observe a large hepatic infection. In
this Los Santos toad, the liver was composed largely
of cysts containing egg-laden trematodes.
Rather than being a primary mortality factor
parasitic infections, we suspect, reduce the toad's
resistance and make it more susceptible to death
by starvation or dehydration.
Summary
While this report is an attempt at a synthesis of
the marine toad's natural history, we are well
aware of the inadequacy of our present and pre-
viously published data. Nonetheless, we believe a
survey is needed at this time to encourage others to
study the biology of the marine toad. The main
reason we chose to build so many windmills of
conjecture on feeble bases of facts is the hope that
our readers will be tempted to topple these wind-
mills.
In Panama, male marine toads reach sexual
maturity at a snout-vent length of 85-95 mm,
females at 90-100 mm. Most toads appear to reach
80-90 mm one year from metamorphosis, although
it seems doubtful that the one year olds are capable
of breeding at this time. Growth is rapid during
the first year, approximately 0.37 mm per day and
slows down at sexual maturity, much more so in the
males than in females. The females appear to grow
for at least three years after maturity, the males for
one year and during this year their growth rate is
less than the females. As a consequence females
attain larger body sizes than males.
In all adult populations studied, there is a pre-
ponderance of one sex over the other, usually at
least two times as many of one sex as the opposite.
In half of the samples females outnumbered males,
the reverse situation in the other half. Reproduc-
40 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
tion appears to be cyclic with one or two egg laying
periods each year. The male's spermatogenic cycle
peaks in April and declines to a low in January.
Breeding period appears to be timed so that
metamorphosis occurs during periods of high
humidity and prey abundance, primarily at the
beginning of the wet season.
Marine toad populations consist largely of young
adults (one- to two-year-olds), with a small minority
of the older individuals. Approximately two-thirds
of the population is new each year; whether this
turnover is due to emigration or mortality is un-
known. The density of subadults and adults is
variable: 50 to 70 toads per hectare at Summit
Gardens, 140-180 at Barro Colorado Island. The
actual population size is certainly much smaller,
about 60 toads at the BCI site, and fairly constant
from year to year.
The toads are feeding opportunists and will eat
almost any animate object they can catch, particu-
larly those prey in the 5.1-10.0 mm size class for
all adult-sized toads. Ants and beetles tend to be
the most numerous prey items and also the prey
contributing the most biomass to the toad's diet.
During high humidity and prey abundance, a
toad is active one out of every three or four eve-
nings. About a fourth of the population will be
active during a single evening. Peak activity occurs
from one to three hours after dusk. Most toads are
active for only a couple of hours before disappear-
ing from their feeding sites.
Toads have several feeding sites within a larger
forage area (average of 160 sq m) and regularly
move from one to another. The change of feeding
site usually occurs with the beginning of a new
activity period, rather than during a single eve-
ning. Both activity and movement are influenced
by weather conditions. Long periods of dry or rainy
weather suppress activity.
Marine toads are eurythermal and apparently do
not thermoregulate. Their temperatures follow the
fluctuation of the ambient temperatures; the two
are nearly equivalent at night, but during the day,
body temperature is lower than ambient. Water
balance seems to be a more critical factor for sur-
vival in Panamanian toads. While they can tolerate
a 53% loss in total body water before dying, they
are not physiologically efficient in slowing or re-
ducing water loss. A recently metamorphosed toad-
let will reach its lethal limit within one day if
moisture is not available, an adult toad within two
to three days. This indicates that the dry season
may be a critical period for survival.
Mortality is, however, probably highest owing to
predation. Although no data are available, the
period from metamorphosis to maturity seems to
be the time of highest mortality. Vertebrates are
likely the major predators.
Appendix I
Sample Data
TABLE A.?Research use of marine toad samples examined in this study
L o c a l i t y and date
CANAL ZONE
Barro Colorado Is land
2-8 Apr. 1974
20 Nov. - l Dec. 1974
6-22 May 1975
10-25 May 1976
Cocole
26 Nov. 1974
10 May 1975
Gamboa
28 Nov. 1974
11 May 1976
Summit Gardens
26-29 Nov. 1974
9-14 May 1975
14 & 20 May 1976
Summit H i l l
28 Nov. 1974
9 May 1975
14 May 1976
LOS SANTOS
Los Santos
12 May 1975
PANAMA
Las Cumbres
17 Dec. 1974
17 J a n . 1975
17-19 A p r . 1975
17-19 May 1975
17-19 Jun . 1975
Ecological Food Physiological
N monitoring habits Reproduction test
6 +
29 +
141 +
55 +
1 + +
10 + +
10 + +
2 + +
11 +
23 +
24 +
2 + +
9 + +
8 + +
15 + +
2 +
2 +
4 +
7 +
3 +
41
42 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
TABLE B.?Mark/recapture data from Barro Colorado Island and Summit Gardens sites used to
estimate population size (U = unmarked, R = recapture of marked, T = total individuals
captured?used in Schumacher-Eschmeyer estimate; 1st = marked and released in first sampling
period, 2nd = total capture in second sample, MR = marked in first and recaptured in second
period?data used for Lincoln index population estimate; position in date row indicates level
at which summation of column U stopped; C = consecutive number of new individuals caught?
data used for Zippin method of population estimate; MT = recentlv metamorphosed toadlets,
less than 35 mm; where column is incomplete, data row again designates level of summation;
all but last column of data based on old juveniles, subadults, and adults, i.e., toads greater than
60 mm)
Date
Summit Gardens
26 Nov 74
27
28
29
9 May 75
10
14
14 May 76
Barro Colorado
3 Apr 74
4
5
6
7
8
20 Nov 74
21
22
23
24
25
28
29
30
1 Dec 74
13 Feb 75
26
6 May 75
7
8
U
4
3
1
3
15
5
4
15
9
4
2
0
0
0
0
2
0
3
0
3
0
7
0
12
3
9
4
21
8
8
R
0
1
1
3
0
10
6
0
5
0
2
0
1
4
3
0
0
0
0
3
0
0
1
4
2
0
2
0
11
10
T
4
4
2
6
15
15
10
15
14
4
4
0
1
4
3
2
0
3
0
6
0
7
1
16
5
9
6
21
19
18
1st
8
15
20
15
6
15
27
9
2nd
6
15
10
14
4
16
5
6
MR
3
10
6
5
4
4
2
2
C
4
3
1
3
15
5
4
15
9
4
2
17
12
3
9
4
MT Date
9 May 75
14
15
18
19
20
21
22
10 May 76
11
12
13
15
16
17
18
19
21
22
23
24
25
BCI toadlets
8 May 75
15
18
19
20
21
22
U
2
2
7
6
3
6
2
3
7
9
4
10
5
5
2
1
3
1
2
3
1
0
9
19
16
11
10
11
3
R
3
3
10
13
11
26
27
32
0
4
6
5
12
9
11
13
20
12
18
22
15
16
0
0
4
0
2
3
5
T
5
5
17
19
14
32
29
35
7
13
10
15
17
14
13
14
23
13
20
25
16
16
8
19
12
11
12
14
8
1st
54
57
63
65
47
49
52
53
27
50
62
76
2nd
14
32
29
35
20
25
16
16
12
12
14
8
MR
11
26
27
32
18
22
15
16
4
2
3
5
C
39
18
11
16
14
10
3
4
5
1
9
19
16
11
10
11
3
MT
30
16
7
NUMBER 284 43
TABLE C.?Identity of prey and their frequency in stomach contents of Panamanian marine toads
(first four localities based on stomach contents only; numbers in parentheses are numbers of
empty stomachs and number of stomachs with prey, respectively; Barro Colorado Island data
based on stomach contents and on eight fecal samples, three of which are aggregates; because
fecal samples were included, precise numbers of prey are not listed for BCI
Prey
Arthropoda
Diplopoda (millipedes)
Arachnida
Phalangida (daddy-longlegs)
Araneida (spiders)
Insecta
Odonata (dragonflies)
Orthoptera
Acrididae (grasshoppers)
Tettigoniidae (katydids)
Gryllidae (crickets)
Blattidae (roaches)
Isoptera (termites)
Dermaptera (earwigs)
Henri ptera
oblong bugs (e .g . , reduviids)
obovate bugs (e .g . , pentatomids)
Homoptera
Cicadidae (cicadas)
Neuroptera (lacewings)
Coleoptera
Carabidae (ground beetles)
Curculionidae (weevils)
Scarabaeidae (scarabs)
oblong beetles ( e . g . , elaterids)
round beetles (e .g . , chrysomelids)
Lepidoptera (moths)
Hymenoptera
Formicidae (ants)
winged hymenopterans
Annelida (earthworms)
Mollusca
Gastropoda (snails)
Cocole
(2/9)
1
6
1
5
2
9
Gamboa
(0/10)
5
2
2
1
3
4
1
1
7
3
4
1
8
5
Summit H i l l
(0/11)
2
1
1
1
1
1
3
5
3
9
8
2
1
6
8
1
Los Santos BCI
(3/12) (1/1)
+
+
1
1
+
+
+
+
+
+
+
1 +
3 +
3 +
1
1
10 +
2 +
1 +
Appendix II
Weight-Length Relationships of Arthropods
After examining the stomach contents of a small sample
of Panamanian marine toads and identifying the arthropod
remains, we made an effort to obtain the lengths and weights
of living individuals from each group of arthropods repre-
sented in the stomachs. The groups presented here and
identified from the stomach contents are identified to the
familial level or to a higher taxonomic level. In a few
cases, such as beetles and bugs, members of the same order
may be analyzed in different groups, if they have different
body shapes, because shape greatly influences the weight-
length relationships. Although the constituents of each
group may appear to be somewhat arbitrary, they represent
the highest level of determination that can be accurately
made from the bolus found in a toad's stomach. Unless an
animal is clearly a specialized feeder, it seems wasteful of a
researcher's time to attempt specific determination and leads
to a false sense of accuracy.
The regression lines were calculated and plotted on an
HP 98 calculator/plotter with the "Family Regression" pro-
gram. Each graph contains a power equation for the plotted
curve and its correlation coefficient (r). Although the correla-
tion coefficient of the parabolic curve was occasionally higher
than that of the power curve, the former overestimates the
weights of the smaller size class so we consistently used the
power equation in order to obtain the greatest accuracy
throughout our calculations of arthropod live weight or
biomass in each toad's stomach.
The illustration and common name on each graph is
sufficient to identify the arthropods represented, except for
the subdivisions of the beetles and bugs. The obovate bug
sample is composed largely of pentatomids, the oblong bug
of pyrrhocorids and reduviids. The round beetles are prin-
cipally chrysomelids, but include other beetles that have a
circular outline. The oblong beetles include mainly ceram-
bycids, elaterids, and lampyrids. The scarab beetles are
exclusively of that family, and predominantly of a broad
elliptical outline. The winged hymenopterans include both
bee- and wasp-shaped insects. All the arthropods were col-
lected in the vicinity of the STRI compound on Barro
Colorado Island, Panama.
0.50r
SPIDERS
0 2.0 4.0 10.0
LENGTH
20.0
NUMBER 284 45
4.0
0 5.5 11.0 27.5
LENGTH (MM)
55.0
46 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
0.55
0 6.0 IP.O 30.0
LENGTH (MMi
60.0
NUMBER 284 47
0.05
0.025
0.01
0.005
Y*0.0000001X
r-0.95
5.2910
0.05
 r
O 2.0 4.0 10.0
0.025
0.01
0.005
0
r=o.oooi x
r^O.91
3.1168
EARWIGS
0 1.0 2.0 5.0
L ENOTH <MM)
20.0
TERMITES
10.0
48 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
0.6
0.3
K.
0.12
0.06
0.5
Y* 0.0001 X
r -- 0.98
3.0042
O 25 5.0 12.5
ki
0.25
O.I
0.05
0
Y* 0.0001 X
r=0.98
2S443
0 3.0 6.0 15.0
LENGTH iMM>
25.0
30.0
NUMBER 284 49
0.2
O.I
0.04
0.02
3.02 73
Y = 0.0001 X
r - 0.95
CAR A BID
BEETLES
3.0 5.0 7.0 13.0 23.0
, 5
0. 75
0.3
0.15
0
Y-0.0001X
2.7918
SCARAB
BEETLES
O 3.0 6.0 15.0
LENGTH (MM)
30.0
50 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
0 .08
0.04
0.016
Q008
0
2.75
Y = 0.000/X
/--- 0.97
3.2705
O 1.05 2.1 5.25
ki
1.3 75
0.5 5
0.275
0
Y =0.0001X
r=0.95
2.5921
ROUND BEETLES
10.5
OBLONG
BEETLES
0 43 9.0 22.5 45.0
LENGTH
NUMBER 284
1.5
51
0.75
<0
0.3
^ 0.15
<3
Y-aoooi x
r* 0.99
2.7 721
0 3.0 6.0 15.0
2.5
1.25
0.5
0.25
0
Y- 0.0001 X
r=0.96
2.8014
0 5.6 11.2 28.0
LENGTH
3 0.0
56 .0
52 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
0.25
0.125
0.25 0 2.5 5JO 12.5
5
0.125
0.05
0.025
0
Y - 0.0001 X
r = 0.94
2.8198
0 3.5 7.0 17.5
LENGTH
25.0
WINGED
HYMEN OP TERA NS
35.0
NUMBER 284 53
3.0
k.
0.6
0.3
0
Y= 0.0001 X
r--0.97
2.4864
O 9.0 18.0 45.0
LENGTH (MM)
90.0
Literature Cited
Alcala, Angel C.
1957. Philippine Notes on the Ecology of the Giant Ma-
rine Toad. Silliman Journal, 4:90-96.
Alexander, Taylor R.
1965. Observations on the Feeding Behavior of Bufo mari-
nus (Linne.). Herpetologica, 20(4):255-259.
Allen, E. Ross, and Wilfred T. Neill
1956. Effect of Marine Toad Toxins on Man. Herpeto-
logica, 12(2):150-151.
Altig, Ronald
1970. A Key to the Tadpoles of the Continental United
States and Canada. Herpetologica, 26(2):180-207.
Anderson, Paul K.
1954. Studies in the Ecology of the Narrow-Mouthed
Toad, Microhyla carolinensis carolinensis. Tulane
Studies in Zoology, 2(2): 15-46.
Beebe, William
1925. Studies of a Tropical Rain Forest: One Quarter of
a Square Mile of Jungle at Kartabo, British Guiana.
Zoologica, 6(1):1-193.
Bentley, P. J.
1971. Endocrines and Osmoregulation: A Comparative
Account of Water and Salt in Vertebrates. 300 pages.
Berlin: Springer-Verlag.
Bentley, P. J., and A. R. Main
1972. Zonal Differences in Permability of the Skin of
Some Anuran Amphibia. American Journal of
Physiology, 2:361-363.
Bertini, F., and J. M. Cei
1962. Seroprotein Patterns in the Bufo marinus Complex.
Herpetologica, 17(4):231-238.
Blair, W. Frank
1960. A Breeding Population of the Mexican Toad (Bufo
valliceps) in Relation to its Environment. Ecology,
41(1):165-174.
Boice, Carol, and Robert Boice
1970. Precedence in Feeding and Mounting in Captive
Toads. Psychonomic Science, 20(5):259-260.
Boice, Robert, and Carol Boice
1970. Interspecific Competition in Captive Bufo marinus
and Bufo americanus Toads. Journal of Biological
Psychology. 12(l):32-36.
Brattstrom, Bayard H.
1961. Thermoregulation in Tropical Amphibians. Year
Book of the American Philosophical Society, 1960:
284-287.
1962. Homing in the Giant Toad, Bufo marinus. Herpeto-
logica, 18(3): 176-180.
1968. Thermal Acclimation in Anuran Amphibians as a
Function of Latitude and Altitude. Comparative
Biochemistry and Physiology, 24(1):93?111.
1970. Amphibians. In G. Causey Whitlow, editor, Com-
parative Physiology of Thermoregulation, volume I:
Invertebrates and Nonmammalian Vertebrates, pages
135-166. New York: Academic Press, Inc.
Brattstrom, Bayard H., and Penny Lawrence
1962. The Rate of Thermal Acclimation in Anuran Am-
phibians. Physiological Zoology, 35(2): 148-156.
Breder, C. M.
1946. Amphibians and Reptiles of the Rio Chucunaque
Drainage, Darien, Panama, with Notes on Their
Life Histories and Habits. Bulletin of the American
Museum of Natural History, 86(8):375-435.
Brooks, Daniel R.
1976. Five Species of Platyhelminths from Bufo marinus
L. (Anura: Bufonidae) in Colombia with Descrip-
tions of Creptotrema lynchi sp. n. (Digenea: Allo-
creadiidae) and Glypthelmins robustus sp n. (Di-
genea: Macroderoididae). Journal of Parasitology,
62(3):429-433.
Cagle, Fred R.
1956. An Outline for the Study of an Amphibian Life
History. Tulane Studies in Zoology, 4(3):79?110.
Cei, Jose M.
1972. Bufo of South America. In W. Frank Blair, editor.
Evolution in the Genus Bufo, pages 82-101. Austin:
University of Texas Press.
Clarke, Raymond D.
1974. Postmetamorphic Growth Rates in a Natural Popu-
lation of Fowler's Toad, Bufo woodhousei fowleri.
Canadian Journal of Zoology, 52(12):1489-1498.
Covacevich, J., and M. Archer
1975. The Distribution of the Cane Toad, Bufo marinus,
in Australia and Its Effects on Indigenous Verte-
brates. Memoirs of the Queensland Museum, 17(2):
305-310.
Creusere, F. Michael, and Walter G. Whitford
1976. Ecological Relationships in a Desert Anuran Com-
munity. Herpetologica, 32(1):7?18.
Dexter, Raquel R.
1932. The Food Habits of the Imported Toad, Bufo ma-
rinus, in the Sugar Cane Sections of Porto Rico.
Bulletin of the Fourth Congress of the International
Society of Sugar Cane Technologists, (74):l-6.
Doyle, Walter, and William H. Beckert
1969. Anuran Chromosome Variation. Journal of Cell
Biology, 43:32a.
Duellman, William E.
1961. The Amphibians and Reptiles of Michoacan, Mex-
ico. University of Kansas Museum of Natural His-
tory Publication, 15(1): 1-148.
1963. Amphibians and Reptiles of the Rainforests of
Southern El Peten, Guatemala. University of Kansas
Museum of Natural History Publication, 15(5):205-
249.
54
NUMBER 284 55
1965. Amphibians and Reptiles from the Yucatan Penin-
sula, Mexico. University of Kansas Museum of
Natural History Publication, 15(12):577-614.
Dunn, Emmett R.
1931. The Amphibians of Barro Colorado Island. Occa-
sional Papers of the Boston Society of Natural His-
tory, 5:403-421.
Durant, Pedro
1974. Factors Which Affect the Development of Two Spe-
des of Tropical Tadpoles. Sociedad Venzolana de
Ciencias Naturales, 128:21-32.
Eisenberg, Robert M.
1972. A Critical Look at the Use of Lincoln Index-Type
Models for Estimating Population Density. Texas
Journal of Science, 23(4):511-517.
Elkan, E.
1968. Mucopolysaccharides in the Anuran Defense against
Desiccation. Journal of Zoology, London, 155(1):19?
53.
Ely, Charles A.
1944. Development of Bufo marinus Larvae in Dilute Sea
Water. Copeia, 1944(4):256.
Emlen, Stephen T.
1968. A Technique for Marking Anuran Amphibians for
Behavioral Studies. Herpetologica, 24(2): 172-173.
Estes, Richard, and Richard Wassersug
1963. A Miocene Toad from Colombia, South America.
Breviora, 193:1-13.
Ettershank, G., and Daphne L. Ettershank
1973. A Computer Simulation Study of Mark-Recapture
Methods in Ecology, 1: Means and Empirical
Standard Deviation from Four Models. Proceedings
of the Royal Society of Victoria, 86(1):85-110.
Fair, J. W.
1970. Comparative Rates of Rehydration from Soil in Two
Species of Toads, Bufo boreas and Bufo punctatus.
Comparative Biochemistry and Physiology, 34:281-
287.
Fellows, A. G.
1969. Toads and Termites. Victorian Naturalist, 86:136
Fitzgerald, G. J., and J. R. Bider
1974a. Evidence of a Relationship between Age and Activ-
ity in the Toad Bufo americanus. Canadian Field-
Naturalist, 88:499-501.
1974b. Influence of Moon Phase and Weather Factors on
Locomotory Activity in Bufo americanus. Oikos, 25:
338-340.
1974c. Seasonal Activity of the Toad Bufo americanus in
Southern Quebec as Revealed by a Sand-Transect
Technique. Canadian Journal of Zoology, 52(l):l-5.
Grant, Chapman
1946. Selection between Armed and Unarmed Arthropods
as Food by Various Animals. Journal of Entomology
and Zoology, 40(3):66.
Hardy, Lawrence M., and Roy W. McDiarmid
1969. The Amphibians and Reptiles of Sinaloa, Mexico.
University of Kansas Museum of Natural History
Publication, 18(3):39-252.
Heatwole, Harold
1966. The Effect of Man on Distribution of Some Reptiles
and Amphibians in Eastern Panama. Herpetologica,
22(l):55-59.
Heatwole, Harold, Sheila Blasini de Austin, and Rita Herrero
1968. Heat Tolerances of Tadpoles of Two Species of
Tropical Anurans. Comparative Biochemistry and
Physiology, 27(3):8O7-815.
Heatwole, Harold, and Audry Heatwole
1968. Motivational Aspects of Feeding Behavior in Toads.
Copeia, 1968(4):692-698.
Heusser, H.
1968. Die Lebenweise der Erdkrote. Bufo bufo (L). Der
Magenfullungsgrad in Abhangigkeit von Jagdstim-
mung und Wetter. Sitzungsberichte der Gesellschaft
Naturforschender Freunde zu Berlin, new series,
8(2): 148-156.
Heusser, H., and K. Meisterhans
1969. Zur Populationsdynamik der Kreuzkrdte, Bufo
calamita I.aur. Vierteljahrsschrift der Naturforschen-
den Gesellschaft in Zurich, 114(3):269-277.
Heyer, W. Ronald, Roy W. McDiarmid, and Diana L. Weig-
mann
1975. Tadpoles, Predation and Pond Habitats in the
Tropics. Biotropica, 7(2):100-lll.
Hinckley, Alden D.
1963. Diet of the Giant Toad, Bufo marinus (L), in Fiji.
Herpetologica, 18(4):253-259.
Honegger, Ren? E.
1970. Eine Kr6te erobert die Welt. Natur und Museum,
100(10):447-453.
Hostetler, Jeptha R., and M. Samuel Cannon
1974. The Anatomy of the Parotoid Gland in Bufonidae
with Some Histochemical Findings, I: Bufo mari-
nus. Journal of Morphology, 142:225-240.
Hutchinson, Victor H., and Margaret A. Kohl
1971. The Effect of Photoperiod on Daily Rhythms of
Oxygen Consumption in the Tropical Toad, Bufo
marinus. Zeitschrift fur vergleichende Physiologie,
75:367-382.
Illingworth, J. F.
1941. Feeding Habits of Bufo marinus. Proceedings of
the Hawaiian Entomological Society, 11(1):51.
Ingle, David
1971. Prey-Catching Behavior of Anurans toward Moving
and Stationary Objects. Vision Research, supple-
ment 3:447-456.
1976. Spatial Vision in Anurans. In (Catherine V. Fite,
editor, The Amphibian Visual System, A Multidis-
ciplinary Approach, pages 119-140. New York: Aca-
demic Press.
Jaeger, Robert G.
1976. A Possible Prey-Call Window in Anuran Auditory
Perception. Copeia, 1976(4):833-834.
Jakowska, Sophie
1972. Lesions Produced by Ticks, Amblyomma dissimile,
in Bufo marinus Toads from the Dominican Re-
public. American Zoologist, 12(4):abst. 726.
Johnson, Clifford Ray
1972. Thermal Relations and Daily Variation in the
Thermal Tolerance in Bufo marinus. Journal of
Herpetology, 6(l):35-38.
56 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
Jorgensen, C. Barker
1973. Mechanisms Regulating Ovarian Function in Am-
phibians (Toads). In Hannah Peters, editor, The
Development and Maturation of the Ovary and its
Functions, pages 133-151. Amsterdam: Excerpta
Medica.
Kenny, Julian S.
1969. The Amphibia of Trinidad. Studies of the Fauna
of Curacao and other Caribbean Islands, 29(108): 1-
78.
Kloss, G. R.
1971. Alguns Rhabdias (Nematoda) de Bufo no Brasil.
Papeis Avulsos de Zoologia, 24(1): 1-52.
Kourany, Miguel, Charles W. Myers, and Curt R. Schneider
1970. Panamanian Amphibians and Reptiles as Carriers of
Salmonella. American Journal of Tropical Medicine
and Hygiene, 19(4):632-638.
Krakauer, Thomas
1970. Tolerance Limits of the Toad, Bufo marinus, in
South Florida. Comparative Biochemistry and Physi-
ology, 33(1): 15-26.
Labanick, George M., and Raymond A. Schlueter
1976. Growth Rates of Recently Transformed Bufo wood-
housei fowleri. Copeia, 1976(4):824-826.
Lehmann, Donald L.
1966. Two Blood Parasites of Peruvian Amphibia. Jour-
nal of Parasitology, 52(3):613.
Lehmann, H. D., B. Roth, and C. C. Schneider
1969. Die Zecke Amblyomma testudinis (Concil, 1877), ihre
Entwicklung und ihre Wirkung auf dem Wirt.
Zeitschrift fur Tropenmedizin und Parasitologie,
20:247-259.
Lescure, Jean
1975. Observations ecologiques siir les amphibiens dans
l'amazonie du nord-ouest. Leur place dans l'environ-
ment humain. In P. Centlivres, J. Gasche, and A.
Lourteig, editors, Culture sur Brulis et Evolution
du Milieu Forestier en Amazonie du Nord-Ouest,
pages 65-69. Geneve: Claude Swary.
Licht, Lawrence E.
1967. Death Following Possible Ingestion of Toad Eggs.
Toxicon, 5(2):141-142.
1968. Unpalatability and Toxicity of Toad Eggs. Herpe-
tologica, 24(2):93-98.
Licht, Lawrence E., and Bobbi Low
1968. Cardiac Response of Snakes after Ingestion of Toad
Parotoid Venom. Copeia, 1968(3):547-551.
Lillywhite, Harvey B., and Paul Licht
1974. Movement of Water over Toad Skin: Functional
Role of Epidermal Sculpturing. Copeia, 1974(1):165-
171.
Low, Bobbi S.
1972. Evidence from Parotoid-Gland Secretions. In W.
Frank Blair, editor. Evolution in the Genus Bufo,
pages 244-264. Austin: University of Texas Press.
Machin, J.
1969. Passive Water Movements through Skin of the Toad
Bufo marinus in Air and in Water. American Jour-
nal of Physiology, 216(6):1562-1568.
Mares, Michael A.
1972. Notes on Bufo marinus Tadpole Aggregations. Texas
Journal of Science, 23(3):433-435.
Marinkelle, Corneles J.
1968. Tritrichomonas batrachorum Perty from Some Co-
lombian Toads. Mittelung der Instituto de Colombo-
Alemdn Investigaciones Cientificas, 2:29-31.
1970. Observaciones sobre la periodicidad de las micro-
filarias de Ochoterenella en Bufo marinus de Colom-
bia. Revista de Biologia Tropical, (1968) 16(2):I45-
152.
Martin, Paul S.
1958. A Biogeography of Reptiles and Amphibians in the
Gomez Farias Region, Tamaulipas, Mexico. Mis-
cellaneous Publication, Museum of Zoology, Uni-
versity of Michigan, (101): 1-102.
Martin, Robert F.
1972. Evidence of Osteology. In W. Frank Blair, editor,
Evolution in the Genus Bufo, pages 37-70. Austin:
University of Texas Press.
Martof, Bernard S.
1962. The behavior of Fowler's Toad under Various Con-
ditions of Light and Temperature. Physiological
Zoology, 35(l):38-46.
McManus, John J., and David W. Nellis
1976. The Critical Thermal Maximum of the Marine
Toad, Bufo marinus. Caribbean Journal of Science
(June 1975) 15(l-2):67-70.
Mertens, Robert
1952. Die Amphibien und Reptilien von El Salvador, auf
Grund der Reisen von R. Mertens und A. Zilch.
Abhandlungen der Senckenbergischen Naturforschen-
den Gesellschaft, 487:1-120.
Meyer, John R., and Larry David Wilson
1971. A Distributional Checklist of Amphibians of Hon-
duras. Contributions in Science (218):l-47.
Nace, George W., Christina M. Richards, and Gretchen M.
Hazen
1973. Information Control in the Amphibian Facility:
The Use of R. pipens Disruptive Patterning for In-
dividual Identification and Genetic Studies. Ameri-
can Zoologist, 13(1):115-135.
Park, Orlando, Albert Barden, and Eliot Williams
1940. Studies in Nocturnal Ecology, 9: Further Analysis
of Activity of Panama Rainforest Animals. Ecology,
21(2):122-134.
Pemberton, C. E.
1934. Local Investigations on the Introduced Tropical
American Toad Bufo marinus. Hawaiian Planters'
Record, 38(3): 186-192.
1949. Longevity of the Tropical American Toad, Bufo
marinus L. Science, 110:512.
Pippet, J. R.
1975. The Marine Toad, Bufo marinus, in Papua New
Guinea. Papua New Guinea Agricultural Journal,
26(l):23-30.
Pope, Clifford H.
1955. The Reptile World. 325 pages. New York: Alfred A.
Knopf.
NUMBER 284 57
Reed, Charles A., and Richard Borowsky
1970. The "Worlds Largest Toad" and Other Herpeto-
logical Specimens from Southern Surinam. Studies
on the Fauna of Surinam and other Guyanas, 12(50):
159-171.
Rivero, Juan A.
1961. Salientia of Venezuela. Bulletin of the Museum of
Comparative Zoology, Harvard College, 126(1): 1-207.
Roth, Jan J.
1973. Vascular Supply to the Vental Pelvic Region of
Anurans as Related to Water Balance. Journal of
Morphology, 140(4):443?460.
Ruibal, Rodolfo
1962. The Adaptive Value of Bladder Water in the Toad,
Bufo cognatus. Physiological Zoology, 35(3):218-223.
Savage, Jay M.
1960. Geographic Variation in the Tadpoles of the Toad,
Bufo marinus. Copeia, 1960(3):233-236.
Schumacher, F. X., and R. W. Eschmeyer
1943. The Estimate of Fish Population in Lakes or Ponds.
Journal of the Tennessee Academy of Science, 18:
228-249.
Sein, Francisco
1937. The Development of the Giant Surinam Toad Bufo
marinus L. Journal of Agriculture, University of
Puerto Rico, 21(l):77-78.
Sexton, Owen J., Harold Heatwole, and Dennis Knight
1964. Correlation of Microdistribution of Some Panaman-
ian Reptiles and Amphibians with Structural Orga-
nization of the Habitat. Caribbean Journal of
Science, 4(l):261-295.
Seymour, R. S., and A. K. Lee
1974. Physiological Adaptation of Anuran Amphibians to
Aridity: Australian Prospects. The Australian Zoolo-
gist, 18(2):53-63.
Shoemaker, Vaughn H.
1964. The Effects of Dehydration on Electrolyte Concen-
trations in a Toad, Bufo marinus. Comparative Bio-
chemistry and Physiology, 13:261-271.
1965. The Stimulus for the Water-Balance Response to
Dehydration in Toads. Comparative Biochemistry
and Physiology, 15(2):81-88.
Shoemaker, V. H., and H. Waring
1968. Effect of Hypothalmic Lesions on Water-Balance of
a Toad (Bufo marinus). Comparative Biochemistry
and Physiology, 24(l):47-54.
Steen, W. B.
1929. On the Permeability of the Frog's Bladder to Water.
Anatomical Record, 43:215-220.
Straughn, I. R.
1966. The Natural History of the "Cane Toad" in Queens-
land. Australian Museum Magazine, 15:230-232.
Stuart, L. C.
1950. A Geographic Study of the Herpetofauna of Alta
Verapaz, Guatemala. Contributions from the Labora-
tory of Vertebrate Zoology, (45):l-77.
1951. The Distributional Implications of Temperature
Tolerances and Hemoglobin Values in the Toads
Bufo marinus (Linnaeus) and Bufo bocourti Brocchi.
Copeia, 1951(3):220-229.
Taylor, Edward H.
1952. A Review of the Frogs and Toads of Costa Rica.
University of Kansas Science Bulletin, 35(l):577-942.
Thomas, Allen E.
1975. Marking Anurans with Silver Nitrate. Herpetological
Review, 6(1):12.
Thorson, Thomas B.
1964. The Partitioning of Body Water in Amphibia.
Physiological Zoology, 37(4):395-399.
Turner, Frederick B.
1960a. Population Structure and Dynamics of the Western
Spotted Frog, Rana p. pretiosa Baird and Girard,
in Yellowstone Park, Wyoming. Ecological Mono-
graphs, 30:251-278.
1960b. Postmetamorphic Growth in Anurans. American
Midland Naturalist, 64(2): 327-338.
1962. The Demography of Frogs and Toads. Quarterly Re-
view of Biology, 37(4):303-314.
1978. The Dynamics of Populations of Squamates, Croco-
dilians and Rhynchocephalians. In C. Gans and T. S.
Parson, editors, The Biology of the Reptilia, 7:157-
264. London: Academic Press.
Villa, Jaime
1972. Anfibios de Nicaragua. 216 pages. Managua: Insti-
tuto Geografico Nacional y Banco Central de Nica-
ragua.
Walker, Richard F., and Walter G. Whitford
1970. Soil Water Absorption Capabilities in Selected Spe-
cies of Anurans. Herpetologica, 26(4):411-418.
Wassersug, Richard
1971. On the Comparative Palatability of Some Dry-
Season Tadpoles from Costa Rica. American Mid-
land Naturalist, 86(1): 101-109.
1973. Aspects of Social Behavior in Anuran Larvae. In
James L. Vial, editor, Evolutionary Biology of the
Anurans, pages 273-297. Columbia: University of
Missouri Press.
Whitaker, John O., Jr.
1971. A Study of the Western Chorus Frog, Pseudacris
triseriata, in Vigo County, Indiana. Journal of Her-
petology, 5(3-4):127-150.
Williams, John T., Jr.
1967. A Test for Dominance of Cues during Maze Learn-
ing by Toads. Psychonomic Science, 9(5):259-260.
Wolcott, George N.
1937. What the Giant Surinam Toad, Bufo marinus L.,
Is Eating Now in Puerto Rico. Journal of Agricul-
ture of the University of Puerto Rico, 21(l):79-84.
Zippin, Calvin
1958. The Removal Method of Population Estimation.
Journal of Wildlife Management, 22(l):82-90.
Zipser, Robert D., Paul Licht, and Howard A. Bern
1969. Comparative Effects of Mammalian Prolactin and
Growth Hormone on Growth in the Toads Bufo
boreas and Bufo marinus. General and Comparative
Endocrinology, 13:382-391.
58 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
Zug, George R., Eric Lindgren, and John R. Pippet
1975. Distribution and Ecology of the Marine Toad, Bufo
marinus, in Papua New Guinea. Pacific Science,
29(l):51-50.
Zweifel, Richard G.
1968. Reproductive Biology of Anurans of the Arid South-
west, with Emphasis on Adaptation of Embryos to
Temperature. Bulletin of the American Museum of
Natural History, 140(1): 1-64.
1977. Upper Thermal Tolerances of Anuran Embryos in
Relation to Stage of Development and Breeding
Habits. American Museum Novitates, 2617:1-21.


REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION
Manuscripts intended for series publication receive substantive review within their
originating Smithsonian museums or offices and are submitted to the Smithsonian Institution
Press with approval of the appropriate museum authority on Form SI-36. Requests for
special treatment?use of color, foldouts, casebound covers, etc.?require, on the same
form, the added approval of designated committees or museum directors.
Review of manuscripts and art by the Press for requirements of series format and style,
completeness and clarity of copy, and arrangement of all material, as outlined below, will
govern, within the judgment of the Press, acceptance or rejection of the manuscripts and art.
Copy must be typewritten, double-spaced, on one side of standard white bond paper,
with 1 % " margins, submitted as ribbon copy (not carbon or xerox), in loose sheets (not
stapled or bound), and accompanied by original art. Minimum acceptable length is 30 pages.
Front matter (preceding the text) should include: title page with only title and author
and no other information, abstract page with author/title/series/etc., following the establish-
ed format, table of contents with indents reflecting the heads and structure of the paper.
First page of text should carry the title and author at the top of the page and an unnum-
bered footnote at the bottom consisting of author's name and professional mailing address.
Center heads of whatever level should be typed with initial caps of major words, with
extra space above and below the head, but with no other preparation (such as all caps or
underline). Run-in paragraph heads should use period/dashes or colons as necessary.
Tabulations within text (lists of data, often in parallel columns) can be typed on the text
page where they occur, but they should not contain rules or formal, numbered table heads.
Formal tables (numbered, with table heads, boxheads, stubs, rules) should be sub-
mitted as camera copy, but the author must contact the series section of the Press for edito-
rial attention and preparation assistance before final typing of this matter.
Taxonomic keys in natural history papers should use the alined-couplet form in the
zoology and paleobiology series and the multi-level indent form in the botany series. If
cross-referencing is required between key and text, do not include page references within the
key, but number the keyed-out taxa with their corresponding heads in the text.
Synonymy in the zoology and paleobiology series must use the short form (taxon,
author, yeanpage), with a full reference at the end of the paper under "Literature Cited."
For the botany series, the long form (taxon, author, abbreviated journal or book title, volume,
page, year, with no reference in the "Literature Cited") is optional.
Footnotes, when few in number, whether annotative or bibliographic, should be typed
at the bottom of the text page on which the reference occurs. Extensive notes must appear at
the end of the text in a notes section. If bibliographic footnotes are required, use the short
form (author/brief title/page) with the full reference in the bibliography.
Text-reference system (author/year/page within the text, with the full reference in a
"Literature Cited" at the end of the text) must be used in place of bibliographic footnotes in
all scientific series and is strongly recommended in the history and technology series:
"(Jones, 1910:122)" or ".. . . Jones (1910:122)."
Bibliography, depending upon use, is termed "References," "Selected References," or
"Literature Cited." Spell out book, journal, and article titles, using initial caps in all major
words. For capitalization of titles in foreign languages, follow the national practice of each
language. Underline (for italics) book and journal titles. Use the colon-parentheses system
for volume/number/page citations: "10(2):5-9." For alinement and arrangement of
elements, follow the format of the series for which the manuscript is intended.
Legends for illustrations must not be attached to the art nor included within the text but
must be submitted at the end of the manuscript?with as many legends typed, double-
spaced, to a page as convenient.
Illustrations must not be included within the manuscript but must be submitted sepa-
rately as original art (not copies). All illustrations (photographs, line drawings, maps, etc.)
can be intermixed throughout the printed text. They should be termed Figures and should
be numbered consecutively. If several "figures" are treated as components of a single larger
figure, they should be designated by lowercase italic letters (underlined in copy) on the illus-
tration, in the legend, and in text references: 'Figure 9b/ ' If illustrations are intended to be
printed separately on coated stock following the text, they should be termed Plates and any
components should be lettered as in figures: "Plate 9b_." Keys to any symbols within an
illustration should appear on the art and not in the legend.
A few points of style: (1) Do not use periods after such abbreviations as "mm, ft,
yds, USNM. NNF. AM, BC." (2) Use hyphens in spelledout fractions: "two-thirds." (3)
Spell out numbers "one" through "nine" in expository text, but use numerals in all other
cases if possible. (4) Use the metric system of measurement, where possible, instead of
the English system. (5) Use the decimal system, where possible, in place of fractions.
(6) Use day/month/year sequence for dates: "9 April 1976." (7) For months in tabular list-
ings or data sections, use three-letter abbreviations with no periods: "Jan. Mar. Jun," etc.
Arrange and paginate sequentially EVERY sheet of manuscript?including ALL front
matter and ALL legends, etc.. at the back of the text?in the following order: (1) title page,
(2) abstract, (3) table of contents, (4) foreword and/or preface, (5) text, (6) appendixes.
(7) notes, (8) glossary. (9) bibliography, (10) index, (11) legends
i