Citation
Salinity as a limiting factor in the distribution of reptiles in Florida Bay: a theory for the estuarine origin of marine snakes and turtles

Material Information

Title:
Salinity as a limiting factor in the distribution of reptiles in Florida Bay: a theory for the estuarine origin of marine snakes and turtles
Series Title:
Bulletin of Marine Science, 44 (1) : pp. 229-244, 1989
Creator:
Dunson, William A.
Mazzotti, Frank J.
Publisher:
Bulletin of Marine Science
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
University of Florida. ( LCSH )
Biotic communities -- Florida ( LCSH )
Natural history -- Florida ( LCSH )
Long Sound ( local )
Key Largo ( local )
Salt glands ( jstor )
Salinity ( jstor )
Fresh water ( jstor )
Genre:
serial ( sobekcm )
Spatial Coverage:
North America -- United States of America -- Florida

Notes

Funding:
This collection includes items related to Florida’s environments, ecosystems, and species. It includes the subcollections of Florida Cooperative Fish and Wildlife Research Unit project documents, the Sea Grant technical series, the Florida Geological Survey series, the Coastal Engineering Department series, the Howard T. Odum Center for Wetland technical reports, and other entities devoted to the study and preservation of Florida's natural resources.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved, Board of Trustees of the University of Florida

Downloads

This item has the following downloads:


Full Text





BULLETIN OF MARINE SCIENCE, 44(1): 229-244, 1989


SALINITY AS A LIMITING FACTOR IN THE DISTRIBUTION
OF REPTILES IN FLORIDA BAY:
A THEORY FOR THE ESTUARINE ORIGIN
OF MARINE SNAKES AND TURTLES

iT '.:; ....: A. /,: : and Frank J. Mazzotti

ABSTRACT
S: is hypothesized to be the major abiotic factor limiting the colonization of Florida
Bay by estuarine This premise is supported by the small number of species of reptiles
found in the bay in comparison with fresh water, the distinct osmoregulatory specializations
ofthe :. :;.. ... :- .. : that occur there, and a remarkable cline in the ability to tolerate
sea water found among modern-day estuarine and coastal reptiles. This latter cline in os-
moregulatory abilities is believed to represent a model of the evolutionary stages through
which pelagic snakes and turtles have passed in developing adaptations for life in the open
sea. Florida Bay is an especially useful site for the study of such adaptations since it is the
only known location in this hemisphere where three :. ... i estuarine : ,:, are sym-
patric: the American crocodile (Crocodvlus acutus), the diamondback terrapin : ..
terrapin), and the mangrove snake (Nerodia clarkii compressicauda). Small : .. ...-. of
freshwater turtles and the .:: .: also occur in tidal creeks along the northern shore. Recent
advances in the study of turtles suggest that the single most important factor in determining
tolerance to high salinity is the amount ofsea ... .: .... ... : with food ingestion.
This finding needs to be extended to other reptiles to test the hypothesis that fish eaters (such
as snakes) that do not crush or bite chunks from their food have reduced incidental drinking.
This could explain how the mangrove snake can survive in estuaries without a salt gland,
whereas the sympatric terrapin possesses a sizeable lachrymal salt gland. We hypothesize
that the following represent major transitional stages in the gradual evolution of marine
snakes and turtles from freshwater ancestors: (1) utilization of behavioral osmoregulation to
avoid salinities that cannot be directly tolerated; (2) a reduction in net salt uptake and water
loss and in incidental drinking of sea water while feeding; (3) the first appearance of rudi-
mentary salt-excreting glands: (4) hypertrophy of salt glands as dictated by rates of salt uptake,
in concert with the development of a specialized external morphology suited for a pelagic
life.


Ecologists have long been interested in how species differ in their use of re-
sources. Habitat, food and time are three traditional categories of resource di-
mensions (Pianka, 1975). The habitat dimension is generally most important in
resource partitioning of reptiles, with time and food as secondary factors (Toft,
1985). The presence or absence of a species in a given habitat is generally thought
to be related to the combined effects of biotic (competition, predation) and abiotic
(temperature, salinity, humidity, oxygen, pH, etc.) factors (C ..... ::, 1975). Al-
though any single mechanism rarely acts alone, sometimes a single factor is so
dominant that the species occurrence is regulated, limited or controlled primarily
by this one parameter. Cases where abiotic influences dominate are not well
documented but seem quite obvious where there are steep gradients in physical
or chemical factors, or heterogeneous (patchy) distribution of abiotic conditions.
Freda and Dunson (1986) have documented a case of the latter for the occurrence
of amphibians in ponds that vary in pH and other aspects of water chemistry. Of
course, it is necessary to experimentally manipulate species by transplantation
and exclusion tests to determine the interactive effects of abiotic and biotic factors.
For example, a very early study showed that soil pH controlled the effect of









BULLETIN OF MARINE SCIENCE, VOL. 44, NO. 1, 1989


competitive interactions between two species of bedstraw (Tansley, 1917). The
competitive ability of plants has also been related to soil nutrient status and the
degree of habitat disturbance (Wilson and Keddy, 1986). In diatoms, changes in
levels of nutrients can reverse the outcome of competitive interactions (Tilman
et al., 1981). Rahel (1984) has demonstrated that water pH and winter .
levels control the fish assemblages found in Wisconsin lakes. Hairston (1980) and
Jaeger (1971) found that shifts in moisture and temperature conditions, which
,. ... .. 1 favored one species of plethodontid salamander over another, could
account for field distributions. A manipulative field study by Travis and Trexler
(1986) demonstrated an interaction between environmental rigor and growth
regulation by intraspecific density dependent factors in toad tadpoles.
We suggest that salinity is the most significant proximate factor limiting the
presence of : ::i in coastal habitats, such as Florida Bay. We believe that this
is a direct consequence of the abiotic harshness of the estuary (due to fluctuating
.i:..::. and the relatively poor ability of reptiles generally to tolerate high
salinities. Estuaries are rigorous in some abiotic features but they also are usually
rich in nutrients and potential food. Thus there is a selective advantage for de-
velopment of adaptations allowing reptiles to colonize saline waters. Since estu-
arine animals are characteristically limited to this narrow coastal habitat (for
example, blue crabs and oysters), one could .: .i:. : that estuarine reptiles would
also tend to become habitat :: ::: "
The specific purpose of this paper is to review evidence pertaining to the role
of salinity in limiting the distribution of estuarine reptiles within Florida Bay.
Our general objectives are threefold: (a) to reconstruct the evolutionary stages that
led from freshwater to estuarine to pelagic reptiles and to propose a new theory
for the gradual evolution of marine snakes and turtles from freshwater ancestors,
(b) to understand the types of adaptive strategies involved, especially the response
to salinity as an abiotic limiting factor, and (c) to compare and contrast pathways
and mechanisms of osmoregulation. Some significant new data concerning the
importance of water .:!:-i in growth, and of the role of "incidental drinking"
are presented here for the first time.

HABITAT UTILIZATION BY REPTILES IN FLORIDA BAY
For our purposes the aquatic habitats of Florida Bay may be roughly divided
into four categories: (a) open waters of the main portion of Florida Bay, (b) lagoons
within islands, (c) enclosed bays and ponds, and (d) creeks along the northern
shore. With the exception of the sea turtles, for which no data are available, all
reptiles of Florida Bay are closely associated with land masses (islands or the
mainland). However, several species regularly use all four of the above categories
of aquatic habitat in Florida Bay.
The ecology of the crocodile (Crocodylus acutus) has been described by Mazzotti
(1983) and Kushlan and Mazzotti (in press). Most crocodiles spend the majority
of their time in fall and winter in the "back country" along the northern shores
of the bay. This area is a maze of mangrove swamps, small bays and creek systems.
Movement to exposed shoreline areas and islands occurs during the breeding
season in the spring and summer. This seems especially characteristic of females
that often nest on elevated beach berms. These seasonal movements result in use
of areas with different salinities. Although the mean salinity for all observations
(Fig. 1) was 1 i it was 10 + .. in the nonbreeding season compared to 18 +
10%o in the breeding season. A large proportion of the sightings was in nearly
fresh water. It is unclear to what degree these data represent a "preference" for









DUNSON AND MAZZOTTI: REPTILES OF FLORIDA BAY


Figure 1. Distribution of the crocodile (C. acutus) at all seasons along the southernmost mainland
of Florida and the northern shore of Florida Bay. Each point represents one or more sightings between
1976-1981. Stippling indicates dry land or mangrove areas (after Mazzotti, 1983).


low salinities, or for the habitat type itself. It is likely that crocodiles made con-
siderably more use of the Florida Bay islands and adjacent Key Largo prior to
extensive disturbance by humans (Ogden, 1978; Kushlan and Mazzotti, in press).
They do not breed currently in the Lower Florida Keys (Jacobsen, 1983), and it
is problematic whether they have ever done so on a regular basis.
There is a significant difference in the distribution of alligators (Alligator mis-
sissippiensis) and crocodiles in the Florida Bay area (Ogden, 1978; Mazzotti, 1983).
Alligators and crocodiles overlap here in their nonbreeding ranges, but not in
breeding areas. Most of the alligators seen within the range of the crocodile were
nonbreeding-sized males, and virtually all sightings were at salinities below 15%o.
Although there is no evidence of any direct competition with crocodiles, alligators
do occur elsewhere in quite saline areas of the Lower Keys (Jacobsen, 1983).
However, unlike crocodiles, Lower Keys alligators nest in elevated hammock
areas near pools of fresh or brackish water.
Diamondback terrapins (Malaclemys terrapin) are widespread in Florida Bay.
They are most commonly seen associated with shallow lagoons inside some of
the islands, where large numbers bury themselves in the soft sediments (Wood,
1981). It is assumed that they feed in bay waters near the islands, as well as in
the lagoons, but very little is known in detail of their habits. They have not been
observed in the fresh and brackish waters of the enclosed bays and creeks of the
northern shore. In contrast, the mangrove snake (Nerodia clarkii compressicauda)
has been found in island lagoons, in waterways such as Davis Creek, in interior
sloughs just north of Long Sound, and in roadside canals east of highway US 1
at mile marker 119. This latter location is fresh water, although the canal on the
eastern side of the highway embankment has a further intrusion of saline water
than on the west (Fig. 2). Of the five Nerodia fasciata-like snakes caught at this







BULLETIN OF MARINE SCIENCE, VOL. 44. NO. 1, 1989


169 172


177 180


190 193


1 I I I i I i i I i i l l I


EAST SIDE


JEWFISH
CREEK
1


t I
L. BLACKWATER CANAL
SOUND C111
I I


36-


WEST SIDE


I I I I I I I I I I I I I I I


105 106 107 108 109


110 111 112 113 114 115
MILE MARKER


116 117 118 119 120


Figure 2. Changes in salinity on either side of highway U.S. I between Florida City (inland to the
north) and Key Largo. Four nearly simultaneous measurements were made on the east and west sides
of the embankment for days in October 1979, and January, February, and March 1981. Note that the
east side is usually more saline; there is a southwards shifting of the salinity front on the west side.
The mile (km) markers indicate the distance between Key West (at 0) and Florida City.








DUNSON AND MAZZOTTI: REPTILES OF FLORIDA BAY


Figure 3. The southern tip of the mainland of Florida showing collection sites for turtles in habitats
that may occasionally be brackish. P = Pseudemys nelsoni. T = Trionyx ferox. K = Kinosternon
baurii palmarum. Numbers indicate turtles captured or sighted.


site, two were N. c. compressicauda (the estuarine species) in appearance, two
were typical N. f pictiventris (the freshwater species), and one was an apparent
hybrid. In contrast, on the western side of the highway, only typical freshwater
phenotypes and large numbers of green water snakes (Nerodia cyclopion) were
caught. Despite many hours spent at night in crocodile habitat along the northern
shore of Florida Bay, only the true mangrove snake was found there. Indeed there
is a remarkable scarcity of water snakes generally along the shores of Florida Bay.
Extensive surveys for turtles were carried out in the Lower Taylor River, in
Snook Creek, in creeks and along highway US 1 north of Long Sound (Dunson
and Seidel, 1986). Softshell (Trionyx ferox) and striped mud turtles (Kinosternon
baurii palmarum) were common along the highway near mile marker 119 in fresh
water (Fig. 3). In the sometimes brackish creeks and shallow bays north of Joe
Bay and Long Sound, there were only two species of turtles present in small
numbers, T. ferox and the redbellied turtle (Pseudemys nelsoni). The paucity of
turtles in this superficially suitable coastal location is an enigma and in marked
contrast to the Virginia coast (Dunson, 1986).
There are at least three possible explanations for the scarcity of freshwater
reptiles in coastal areas of extreme south Florida: (a) an extremely wide seasonal









BULLETIN OF MARINE SCIENCE, VOL 44. NO. 1,989


variation in ....;, that is too rigorous for most species, (b) a limitation in
reproduction by the absence of elevated nest sites except in areas of highest
salinities, and (c) a limitation of growth by low food abundance linked to low
primary productivity. Whether one or a combination of these and other factors
are involved can not be determined at this time. However, it is certainly possible
to test these hypotheses by appropriate experimental designs. For example, reptiles
could be placed in enclosures and fed artificially to determine whether .....
alone is a i::::::::. factor. Mounds could also be constructed to facilitate nesting
in areas : ....:. lacking suitable sites.
SALINITY TOLERANCE OF CROCODILES AND ALLIGATORS
The American crocodile (C. acutus) has its last stronghold in the Florida Bay
area. Quite possibly the core of its breeding range has always been in extreme
south Florida (Ogden, 1978; Mazzotti, 1983; Kushlan and Mazzotti, in press). As
described above, it is believed that this species once used the entire bay area, but
has now been driven to the northern shore by the intensive development of Key
Largo and heavy use of adjacent waters. The rarity and endangered status of C.
acutus have made studies of :I.::i tolerance and osmoregulation very difficult.
Thus one might be tempted to assume similarity with the much better studied
estuarine crocodile (C. porosus) of Australia (see T .: ...: et al., 1985, for a review).
' :... i it may eventually prove to be true that both species osmoregulate in similar
ways, it is premature to make this judgment at present. Taplin et al. (1985) argue
that C. porosus can be relatively independent of :. i. drinking water, even as
hatchlings. Yet there is little known about osmoregulation of C. .... .. ..:::.
the period immediately after hatching. C aculus :. i increase their tolerance
to sea water exposure during the first 3-4 months of rapid, ... :1. (Mazzotti and
Dunson, 1984). Thus, the apparent osmoregulatory difference between C. porosus
and C. acutus may be due to a comparison between animals of different ages or
sizes, and inherent salinity tolerance. While it is known that subadult C. acutus
can live for extended periods in 1,I -.. sea water on a diet of fish (Dunson, 1970;
Ellis, 1981) this certainly is not the case for hatchlings (Dunson, 1982; Mazzotti
et al., 1986). Neither Evans and Ellis (1977) nor Dunson (1982) were able to
induce small C. aculus to elevate their level of .-.. .... efflux high enough to
indicate the presence of a salt gland. Taplin et al. (1 ': demonstrated that
subadult C. aculus do possess the same type of methacholine-induced hyperos-
motic :.. :..:: gland secretion now known from many C. :/:.. However, it is
possible that this represents a pharmacological artifact, since salt loading fails to
induce such secretion. Only further experimentation can resolve this issue. It is
clear, however, that hatchling C. acutus, which appear so .. ;.-:.:: to high
salinities (Dunson, 1982), can survive under certain conditions by behavioral
osmoregulation ,i.il: .::: and Dunson, 1984; Mazzotti et al., 1986). Just as in
terrapins (Dunson, 1985), periodic drinking of fresh to brackish water during rain
appears to account for growth of hatchling C. acutus in the wild at salinities that
kill them in the laboratory. Field trials of this hypothesis are needed (perhaps in
enclosures where rainfall could be excluded), although laboratory simulations
strongly support it. Despite some controversy about mechanisms of osmoregu-
lation in crocodiles, it appears clear that Florida Bay animals are able to reproduce
successfully in areas of low and ':.l.: salinities. Relative growth rates may be
related to the availability of rainfall in high salinity sites, since optimum growth
occurs at about 9%o sea water (Dunson, 1982). It is likely that a most important
feature of the habitat of young crocodiles is its capacity to collect rainfall, at least
for a short period, to allow drinking.










DUNSON AND MAZZOTI:F REPTILES 01 I FORIDA BAY


As described above, alligators occur along the northern shores of Florida Bay
at salinities of 15I% and less :i'' : .r., 1983). In the Lower Keys they are wide-
spread, although primarily found in fresh and low salinity ponds and ditches
(Jacobsen, 1983). Extensive alteration of the habitat on Big Pine Key has increased
the availability of such low salinity areas, and this island probably harbors the
most alligators in the Lower Keys. Nearby Little Pine Key has not been so dis-
turbed and some alligators occur in natural brackish ponds. Since aquatic food
in such interior, often fishless, wetlands is scarce to nonexistent for a large predator,
feeding must either occur on land or in the sea. The latter appears more likely
and would explain sightings made in saline canals and mangrove swamps. During
observations of estuarine alligators in North Carolina, Birkhead and Bennett
(1981) observed spatial shifts of the population following blockage and diversion
of the headwaters of a stream. The alligators moved to an area adjacent to a canal
receiving the diverted flow of fresh water. This allowed them to feed in the highly
productive estuarine habitat while maintaining access to a source of fresh water.
These field data suggest that alligators are less tolerant of saline water than C.
acutus, but there are few published data on this question. No lingual "salt glands"
occur in alligators (Taplin et al., 1982). Lauren (1985) concluded that juvenile
alligators were unable to survive chronic exposure to salinities greater than 1 '
in the laboratory. However, rates of mass loss in 100% sea water of unfed alligators
are quite similar to those of C. acutus, as are several measures of sodium and
water turnover (Mazzotti and Dunson, 1984). There is, however, a major behav-
ioral difference between alligators and C. acutus. Alligators will drink saline water
while C. acutus does not. Two alligators died in 35%o and two even in 9%o, after
drinking (L. Brandt and F. Mazzotti, unpubl.). A halophilic, gram negative bacteria
(Vibrio '. '.:. not previously known to be pathogenic to alligators was
isolated from their viscera. It appears that the alligator may be slightly less tolerant
of high salinities than C. acutus and that this difference :.. : '..::. an important
means of habitat separation in areas of sympatry.
SALINITY TOLERANCE OF TURTLES
The only turtle anywhere in the world known to be entirely restricted to estuaries
is the diamondback terrapin (Malaclernmys terrapin). No differences have been
found in salinity tolerance of terrapins from Florida Bay and Virginia (Dunson,
1985). This species has striking physiological and behavioral adaptations for
adjusting to salinity change. Perhaps the most intriguing of these are the presence
of a lachrymal salt gland that is only activated during prolonged dehydration, the
storage of extracellular fluid in the interstitial space, and the -.. .1: .., of rain
water. Salinity has a pronounced effect on the growth of hatchlings from Florida
Bay (Dunson, 1985). There is an optimum at about ''. sea water; no .:."
occurs above about 2 l%1 sea water. This poses an interesting paradox since mean
salinities near nest sites are often far in excess of this latter figure. How then can
the hatchling terrapins survive? In the laboratory they can grow in 35%c sea water
if given one drink of fresh water every 2 weeks. Thus in nature they must utilize
periodic rainfall to rehydrate. To drink rain water or at least brackish water, they
may need to be located in lagoons inside islands which serve as a catchment basin,
allowing drinking of fresh water as it runs off the elevated ground. Considerable
amounts of sea water are taken in by terrapins due to incidental drinking while
.. ... (Table 1). To test how the uptake of 9/% sea water stimulates growth,
hatchling terrapins were grown as before (Dunson, 1985) but in pure sodium
chloride solutions, rather than sea water. The results (Table 2) clearly show that
the growth optimum is associated with uptake of sodium chloride, since maximum










BULLETIN OF MARINE SCIENCE, VOL, 44. NO. 1, 1989


Table 1. The uptake ofsodium and water via incidental swallowing of sea water during feeding. 24Na
influx was measured over 3 h (unfed) and then I h after a meal [mean + SD .'

Water ingesion predicted from
Wet body Water Na influx. pmoles/100 g h Na uptake, mt/100 g-h
mass salinity
Species g Unfed Fed Unfed Fed
Estuarine diamond- 39 4 6 35 10.9 5.1 1,043* + 433 0.02 0.01 2.1* + 0.8
back terrapin, Ma-
laclemys terrapin
Freshwater snapping 86 + 45 17.5 6.7 2.6 698t + 244 0.08 + 0.03 2.7t + 0.9
turtle, Chelydra (6)
serpentina
Fed chopped squid,
t Fed chopped minnows.



growth occurred near 125 mM sea water is about 120 mM sodium). The
effects of substituting other anions and cations and a nonelectrolyte of equivalent
osmotic pressure (glycerol) are shown in Table 3. In 9/% sea water growth (3.3-
3.5%/day) was significantly higher than in other solutes. Neither chloride nor
sodium in combination with other ions, nor glycerol (Table 3) had any effect on
growth above that of fresh water (Table 2; Kruskal-Wallis, P > 0.21). Thus, it
appears that a certain amount of sodium chloride enhances growth, but too much
inhibits it. The exact reason for this remains obscure at present, although it must
involve mechanisms for regulation of extracellular fluid. One experimental ap-
proach would be to examine the turnover of sodium in the body at various
salinities, and under different states of body fluid hydration. A promising start
has been made in the recent discovery that shell size and degree of calcification
is important in determining sodium concentration of freshwater turtles (Dunson
and Heatwole, 1986). In addition, the proportion of sodium that is exchangeable
declines with increasing size. This is a significant change in the previous view
that reptilian body sodium is essentially all exchangeable. It appears that these
principles apply to estuarine terrapins also, since in a preliminary comparison of
two sizes (both small) there were striking differences in total sodium concentration,
with a lesser but distinct effect on exchangeable sodium as well (Table 4). These
previously undetected body size effects must now be taken into consideration
when analyzing fluxes of sodium, and in any comparisons within or between
species.


Table 2. The effect of water NaCI concentration on the 2 week growth rate of Malaclemys terrapin
fed chopped minnows at 28C. Mean t SD (N = 6). Hatched from eggs collected in Accomac County,
VA. Growth rates followed by different letters are significantly different (Kruskal-Wallis multiple
comparison, P < 0.05)

Water [NaCI], mM .. .: J Significance code Initial mass (g)
0 0.87 0.40 a 16.9 + 1.3
0 1.39 + 0.35 a, b 13.9 + 1.4
50 1.59 + 0.52 a,b,c 14.3 1.1
100 2.06 + 0.51 b,c, d, f 23.7 + 3.8
125 2.61 + 0.59 c,e,g 19.4 1.8
150 1.38 + 0.46 a, f 30.0 _t 4.7
175 1.97 + 0.54 b, e, f, h 17.1 + 1.9
250 1.48 + 0.65 a, d, g, h 31.6 7.1










DUNSON AND MAZZOITI: REPITILES OF FLORIDA BAY


Table 3. The comparative effect of different water solutes of approximately equal osmotic pressure
on the 2 week growth rate of Malaclemys terrapin fed chopped minnows at 28C. Mean SD (N =
6). Hatched from eggs collected in Accomac County, VA. Growth rates followed by different letters
are significantly different (Kruskal-Wallis multiple comparison, P < 0.05)

Water Significance code Initial mass (g)
25% sea water 3.28 1.08 a 11.4 + 0.8
3.47 1.17 a 15.2 + 1.3
125 mM KCI 1.04 + 0.51 b 22.7 + 2.1
62.5 mM NaSO, 1.03 0.24 b 29.7 3.7
250 mMolal glycerol 0.93 + 0.80 b 17.0 + 1.7
1.28 + 0.68 b 13.1 + 2.1



As discussed above (Fig. 3) turtles of primary : ::. origin make relatively
little use of the estuarine habitats of Florida Bay. The less common of the two
species, the softshell (T7 .. ,I : was quite intolerant of saline exposure (Dun-
son, 1981; Dunson and Seidel, 1986). In contrast, the more common redbellied
turtle ( ... nelsoni) was rather tolerant (at least in adults), showing low
rates of unfed mass loss in .. sea water. However,this species is herbivorous
and would be expected to undergo a large uptake of salt due to incidental drinking
while feeding in sea water. Since it apparently does not possess a salt gland
(Dunson and Seidel, 1986), it is probably mainly restricted to waters that are low
in salinity. As is the case for coastal .: in Virginia (Dunson, 1986), P.
nelsoni is ideally suited to undergo a gradual selective enhancement of .:....:
tolerance. It may, indeed, already be superior in this respect to inland populations,
and this possibility needs to be tested with a broad range of size classes. A similar
test on i-' '. .. from Virginia and New Jersey showed that coastal hatchlings
had a higher salinity of optimal growth (Dunson, 1986). A third turtle (Kinosternon
baurii .. .'.. .- that was found just north of the estuarine influence of Florida
Bay (Fig. 3) was quite intolerant of sea water exposure (Dunson and Seidel, 1986).
It may, thus, be prevented from colonizing coastal waters.

SALINITY TOLERANCE OF SNAKES

The only aquatic snake that we have found in Florida Bay or in the immediate
vicinity is the mangrove snake, recently designated by Lawson (1985) as Nerodia
clarkii compressicauda. The mangrove snake offers a most interesting view of
. .:.. marine adaptations since it lives in highly saline habitats, yet apparently
has no salt gland. It has a great degree of resistance to dehydration due to a low
net uptake of salts and a low net loss of water. Eight snakes from Chokoloskee
Island survived an average of 46 days in 100% sea water, with no source of fresh

Table 4. The effect ofsize on cation .... .... ..... .: : .... : .:, ofdiamondback terrapins
of equal ages (Malaclemys terrapin). Mean + SD, N = 3 for each group. Kept in 9%c sea water for 3
months, and then in fresh water for 2 days ...."= grams wet mass, :.. = grams dry mass)

Na K
Exchangeable Na
Wet mass (g) omol/(gwm) gmol/(gdm) pmol/(gwm) pmol/(gdm) as % total


8.7 + 1.0
78.2 + 15.0
P value (i test)


27.3 + 1.2 187 + 33 105 20
34.3 + 3.1 147 + 8 81 + 10
0.010 0.056 0.068


95.3 + 4.5 647 + 108
73.7 + 4.9 319 + 42
0.003 0.004


237









BULLETIN OF MARINE SCIENCE, VOL. 44, NO. 1, 1989


water (Dunson, 1980). Five mangrove snakes from Florida Bay had normal rep-
tilian plasma concentrations (153 mM sodium, 371 mOsmolal) (Dunson, 1980).
Salt loading of the Chokoloskee snakes caused them to raise their body sodium
efflux from 1.2 to 11.7 Amoles/100 g-h. This seemed to suggest that an extremely
small salt gland might be present. However, when the experiment was repeated
with three Florida Bay snakes kept for 66 days in 100% sea water, sodium efflux
did not exceed 4 ymoles/100 g-h i : :: and Dunson, 1984). Since these latter
animals were severely dehydrated and hypernatremic, a salt gland should have
been activated if one were present. Dehydrated mangrove snakes drink fresh water
avidly when dehydrated (Miller, 1985) and undoubtedly :::::: rain and brackish
water as a source of free water, as do terrapins and American crocodiles.

STEPS IN THE EVOLUTION OF MARINE SNAKES AND TURTLES
From the case histories of Florida Bay species discussed here it should be
apparent that estuarine reptiles differ considerably in their degrees of adaptation
for life in salt water. When one considers the worldwide range of such variation,
it is hard to escape the conclusion that each species represents a single link along
an evolutionary chain stretching from fresh water to the open sea. It is our con-
tention that one can reconstruct a general evolutionary sequence extending from
the primitive to the most derived condition. The strongest evidence for the validity
of this hypothesized evolutionary pathway comes from the taxonomic breadth of
the data base. Present-day crocodilians, snakes, and turtles undergoing such adap-
tive change seem to pass through similar stages. This suggests that the diverse
ancient groups of marine reptiles may have done the same, even though evolution
proceeded along parallel pathways. Of course, we can never be sure about the
presence of structures, such as salt glands, in fossil forms. We can only assume
that since no modern reptiles can produce a ureteral urine hyperosmotic to plasma,
fossil reptiles were similarly constructed, and would have needed an extrarenal
mechanism of salt excretion and water conservation similar to that of modern
forms.
We suggest the following four major steps in the evolution of salinity tolerance
in aquatic snakes and turtles.
1. A Primary Reliance on Behavioral Osmoregulation.--The simplest form of
behavioral osmoregulation is to keep the mouth closed and avoid any drinking
of sea water. That this requires a special type of neurological control is evident
from the inappropriate drinking by freshwater reptiles when they are dehydrated
in saline water (Dunson, 1980; L. Brandt and F. Mazzotti, unpubl.). Reptiles can
avoid severe salinity stress by seeking out a less rigorous microclimate. This
assumes that less saline conditions are available, at least periodically. For example,
the Key mud turtle (Kinosternon b. baurii) leaves its island ponds when salinities
rise above about 15%0 (Dunson, 1981). Since freshwater mud turtles generally
respond to pond drying by !...... in terrestrial retreats, they already possess
a mechanism for behaviorally avoiding stressful salinities. It is :. : that selection
would gradually favor an increase in physiological aspects of salinity tolerance,
allowing turtles to continue to feed in more saline ponds. Another type of be-
havioral osmoregulation can occur in tidal streams that are low in salinity at ebb
tide, and much higher at flood tide. Snapping turtles (( ... .' ...: serpentina) can
thus utilize habitats that are periodically saline or fresh (Dunson, 1986). In this
circumstance the animal has to have sufficient physiological tolerance to survive
in sea water during the period around high tide, but it always has the option of
moving upstream to a less saline area. These two examples must represent the









Dt \SON AND MAZZOTTI: REPTILES OF FLIORIDA BAY


primitive or less derived condition, since the populations involved are hardly if
at all distinct from the presumed parental freshwater stocks. However, the value
of some behavioral mechanisms of osmoregulation does not necessarily diminish
with the appearance of :-...::. : physiological changes. For example, the dia-
mondback terrapin (Malaclemys terrapin) retains an important behavioral mech-
anism, the drinking of rain water (Dunson, 1985). The retention of such useful,
but primitive traits, at a specialized grade of development may reflect differences
in energetic costs of various mechanisms. If a terrapin can drink fresh rain water,
it presumably gains free water at a considerably lower ".. : than if it drank sea
water and extracted free water by excreting the salt through the salt gland. Of
course, terrapins live in some environments where such "low-cost" water is only
sporadically available and they must have specialized physiological mechanisms
to cope with the extreme situation.

2. Acquisition of /': .. ........ .:' .*. ... ** Especially Reduction in Net Salt
U:- .:'. Net Water Loss and Incidental 7'... :.. -As soon as mechanisms of
behavioral osmoregulation allow the colonization of coastal/estuarine habitats at
the ecotone between fresh and saline water, an opportunity is created for the
selection of physiological traits favoring increased salinity tolerance. Such selec-
tion :.:.- :: leads to the formation of a saline ecotype (Weider and Hebert,
1987), which might become reproductively isolated and evolve into a new species.
In theory one should look for the earliest stages of physiological specialization
for salinity tolerance in coastal populations of reptiles that are morphologically
indistinguishable from freshwater populations. An extensive analysis of just such
a situation has been carried out on tide marsh snapping turtles (Chelvdra; Dunson,
1986). The most significant difference in salinity tolerance observed was an upward
shift in the optimum ....: for growth of the hatchlings. .... ,, ...... turtles have
a higher rate of water turnover than seems ideal for estuarine life, but their large
size and periodic access to fresh water may compensate for this. The young are
much more susceptible to ,: 'i.. toxicity and it appears that selection has op-
erated to increase the salinity at which optimal: ... i. occurs. However, the most
serious problem that snapping turtles face in sea water is a large amount of
incidental drinking (sea water swallowed along with the food; Table 1). Unfed
snapping turtles maintain a fairly low rate of sodium uptake from 50% sea water
(17 -~ '. When fed they take in a volume of saline water similar to that of terrapins,
yet they lack the salt gland to excrete excess NaCl. Turtles that crush or bite their
prey into chunks are clearly at a greater risk of imbibing salt in this way. They
probably regurgitate most of the water swallowed along with their food, but there
is a certain amount that remains in the gut. In contrast, snakes which do not chew
their food can swallow fish prey with a very minimal incidental uptake of salts.
This may be the explanation for the ability of the mangrove snake (N. clarkii
compressicauda), which lacks a salt gland (Dunson, 1980), to live in Florida Bay.
Freshwater reptiles entering sea water face the double threat of net salt uptake
and net water loss. Most freshwater forms already have a relatively low integu-
mentary .:::: : : to sodium. Water permeability is more variable and some
extremely aquatic species have high rates of water exchange across the skin (Stokes
and Dunson, 1982; Dunson, 1986). In general, however, integumentary perme-
ability does not seem to pose a major problem in tolerating saline waters. Instead
it is the net uptake of salts via the head/mouth, distinct from actual drinking.
The exact source of such sodium influx, whether by diffusion across the oral
mucosa, or the eyes, or by tiny amounts of drinking, is not clear. It is apparent,
however, that the head is the locus of uptake since 24Na uptake declines to nearly









BULLETIN OF MARINE SCIENCE, VOL. 44, NO. 1, 1989


zero when the head is physically restrained out of the water (Dunson and Rob-
inson, 1976; Robinson and Dunson, 1976; Dunson, 1 : The same is not true
of water exchange, most of which occurs across the skin (Dunson and Robinson,
1976; Robinson and Dunson, 1976; Dunson, 1978; Dunson and Dunson, 1979).
An energetic argument might be applied here, as with behavioral osmoregulation.
Natural selection could initially favor a maximal reduction in the net uptake of
salts and net loss of water, without recourse to active mechanisms via a salt gland.
The mangrove water snake illustrates this well (Dunson, 1980) as does the terrapin,
which is so impermeable that it may be in 35%o sea water for many months before
it must ..:.. switch on its salt gland to halt the process of dehydration
(Dunson and Dunson, 1975).
3. First / of Salt Glands. -The salt gland is clearly only one part of a
multifaceted network of behavioral, :f ... 1. I. .1 and morphological adaptations
engaged in control of salt tolerance. Its role has sometimes been overemphasized
in a simplistic portrayal of salt balance in marine reptiles. Estuarine snakes such
as the Lower Florida Keys population of N. c. compressicauda thrive in highly
saline habitats without a salt gland. Apparently in some locales, Crocodylus acutus
can do : -. i .. ... = and Dunson, 1984; Gaby et al., 1985; Mazzotti et al.,
1986). It would be very interesting to compare rates of incidental drinking in
these forms. Indeed, some of the most highly derived marine snakes of the Hy-
drophis group have only tiny salt glands (Dunson and Dunson, 1974). Yet for
estuarine reptiles feeding on crushed or bitten food, and for all truly marine : i
(in spite of food type), a salt gland is invariably present.
The type case for a transitional estuarine/marine form with a tiny salt gland is
the dog-faced water snake, Cerberus rynchops (Dunson and Dunson, 1'' ,: This
species occurs mainly in mangrove areas, which vary from fresh to hypersaline
water. It is highly resistant to dehydration, and rehydrates by drinking fresh water
when available. After prolonged dehydration, a minute premaxillary salt gland is
activated, allowing for the smallest rate of extracloacal sodium excretion known
for any salt gland (about 15 pmoles/100 g-h). It is hypothesized that such a tiny
gland could be critically important during extended periods of fasting and/or
during droughts when no fresh drinking water is available. This species is widely
distributed on islands in the western Pacific. It is known to feed on fish in coastal
waters in :1 : : (Jayne et al., 1988) and seems to have a life history transitional
between that of mangrove and pelagic snakes. Although Cerberus is in a separate
family from that of the true sea snakes, it is an excellent model of a prototype of
a pelagic sea snake. Another member of the Cerberus group (Fordonia leucobalia)
feeds exclusively on crabs and possesses an enormously enlarged premaxillary
gland (A. Savitsky, unpubl.). If this turns out to be a salt gland, as in Cerberus,
it will be a strong confirmation of the hypothesized relation between incidental
drinking and salt gland size (Dunson, 1985). ** .. feeds on crustaceans after
crushing them with its robust teeth, a scenario bound to increase incidental drink-
ing over that of its fish-eating relative.
4. : ... of Larger Salt Glands and External '.' '.. .! C.' for
a Pelagic Life.-Once a salt gland had been developed, it could be subject to
avenues of selection to increase its size in relation to changes in patterns of salt
and water exchange. There are five nonhomologous salt glands in marine reptiles,
so there have obviously been repeated incidences of independent origin among
modern-day species (Dunson, 1976; 1979; 1984). The largest glands are found in
the marine iguana (an algae eater) and the leatherback turtle (feeds on jellyfish),
both subject to high amounts of incidental drinking. The smallest glands are found










DIJNSON AND MAZZOI I I REPTILES OF FLORII)A BAY


in species of sea snakes i.;'..: --.. Laticauda) that feed on large fish (eels). It is
likely that consumption of larger fish not only diminishes the incidental ingestion
of sea water per meal, but also decreases the number of meals. Presence of enlarged
salt glands is generally, but not always, associated with the acquisition of a suite
of morphological characteristics associated with pelagic life (dorsal :: : i flat-
tened tail, reduction in size of ventral scutes). The estuarine .::* .:. .i:i
(;. .* .; snake may represent a case where a large gland evolved prior to as-
sumption of life in the open sea. However, most sea snakes feed on fish, except
for several that specialize on fish eggs. As would be predicted from the incidental
drinking theory, an egg-eater (Aipysurus eydouxii) has the i'.:: rate of salt gland
excretion yet known for a sea snake (Dunson and Dunson, 1974).
EXCEPTIONS TO THE ESTUARINE ORIGIN OF MARINE REPTILES
Living pelagic ..: :- are entirely snakes and turtles. In both of these groups
salt glands occur only in estuarine and marine forms. No terrestrial species possess
this organ of extrarenal excretion. In turtles lachrymal salt glands are known in
the diamondback terrapin (Malaclemys, family Emydidae) and in the sea turtles
(families Cheloniidae and Dermochelyidae, which probably have a common ma-
rine ancestor). Among snakes there are posterior sublingual salt glands in the
Hydrophiidae and Acrochordidae, and premaxillary salt glands in the Homolop-
sidae (Colubridae). Thus it is quite logical that salt glands in these groups evolved
in estuarine situations. In contrast, many families of terrestrial lizards possess
nasal salt glands, that apparently developed in xeric situations (Dunson, 1976).
There are no truly pelagic lizards, only coastal forms that feed in mangroves, the
intertidal zone, or in the upper part of the subtidal. The Galapagos marine iguana
(Amblvrhynchus) is the most adapted for a marine existence, yet it .- -.::11 has
relatively few modifications from its terrestrial ancestors (Dawson et al., 1 '
Terrestrial iguanas on the Ecuadorian mainland have large salt glands, although
considerable hypertrophy has occurred in the marine iguana (Dunson, 1969). This
"preadaptation" for coastal life due to the presence of functional salt glands is
especially clear in the Australian mangrove monitor lizard (Varanus semiremax),
which lives in freshwater as well as coastal areas (Dunson, 1974). It is clearly
derived from a terrestrial varanid stock, that has nasal salt glands (Dunson, 1976).
Thus, it seems highly likely that estuarine and coastal lizards are derived directly
from terrestrial forms, with no intervening aquatic freshwater ancestor. This may,
however, explain why there are no modern seagoing lizards. They have not de-
veloped the suite of adaptations necessary for a fully aquatic existence in fresh
water either. It may be more likely for a highly aquatic freshwater form to evolve
marine adaptations, than for the entire set of adaptations to arise in the direct
terrestrial to marine transition.
The situation in crocodilians is much less clear, but it is known that all members
of the family Crocodylidae possess :....-: .: "salt glands" that can be stimulated
by methacholine injection (Taplin et al., 1985). Since this includes many species
that are essentially : ..i:: .. :: : forms, the suggestion has been made that this group
originated in the sea and then radiated into fresh water (Taplin et al., 1985). While
this is an intriguing idea, there are several problems with it. First, there is still
some debate about the functional status of the lingual glands in crocodiles. In C.
porosus the evidence is good that they represent a small but significant contributor
to extrarenal salt excretion. However, there remains the enigma of the failure of
the glands to respond to salt injections as do all other known salt glands. There
is also no evidence yet that the glands are functional in wild C. acutus (Dunson,
1982; Mazzotti and Dunson, 1984) although they do secrete in response to metha-










IBiiI FTIN OF MARINE SCIFNCF V01_ 44, NO, 1, 1989


choline :..- 1:. .. in the laboratory (Taplin et al., 1982). Since the :.... : glands
are so small it is possible that they originated in an ancestral, inland, freshwater
species. This idea has not been given the attention it deserves since it seems
preposterous for a :: i:... ::.: animal to need a salt gland. However, many croc-
odiles occur in areas that undergo seasonal drought and the need for salt excretion
and water conservation under such dehydrating conditions is obvious. Further
information is needed on the function of the lingual glands in freshwater croco-
diles, especially during periods of aestivation. It would also be quite useful to
have more information on the degrees of genetic relatedness among the species
of Crocodylus. Although it is certainly not impossible that "reverse evolution"
of marine to freshwater forms occurred in Crocodylus, the matter requires con-
siderably more data for adequate analysis. In the Philippine lake-locked sea snake
Hlydrophis semperi it is clear that a very recent derivation from the marine H.
cyanocinctus has occurred. Indeed I. semperi has retained a functional salt gland
despite i- .: ., in a freshwater lake (Dunson and Dunson, 1974). On Rennell Island
(Solomons Group), sea snakes of the genus Laticauda have colonized the brackish
water Lake Tegano, illustrating again how a secondarily freshwater sea snake fauna
can develop (Wolff, 1969).

ACKNOWLEDGMENTS
Supported by National Science Foundation grant BSR-8212623 to W.A.D. and a grant from Florida
Power and Light Co. to F.J.M. '-' :- ..." thank the National Park Service for financial and logistical
support over the past 10 years. Necropsies and bacterial identification were .. .: by J. Harkness,
D.V.M.H. :,,: and J. Minnich critically reviewed the manuscript.

LITERATURE :,
Birkhead, W. S. and C. R. Bennett. 1981. Observations ofa small population ofestuarine-inhabiting
near Southport, North Carolina. Brimleyana 6: 111-117.
Connell, J. H. 1975. Some mechanisms producing structure in natural communities: a model and
evidence from field experiments. Pages 460-490 in M. L. Cody and J. M. Diamond, eds. Ecology
and evolution of communities. Harvard Univ. Press, Cambridge, Massachusetts.
Dawson, W. R., G. A. Bartholomew and A. F. Bennett. 1977. A reappraisal of the aquatic special-
izations of the :..- marine iguana (Amblvrhynchus cristatus). Evolution 31: 891-897.
Dunson, M. K. and W. A. Dunson. 1975. The relation between plasma Na concentration and salt
gland Na-K ATPase content in the diamondback terrapin and the yellow-bellied sea snake. J.
Comp. Physiol. 101: 89-97.
Dunson, W. A. 1969. Electrolyte excretion by the salt gland of the Galapagos marine iguana. Amer.
J. Physiol. 216: 995-1002.
S1970. Some aspects of electrolyte and water balance in three estuarine reptiles, the dia-
mondback terrapin, American and 'salt water' crocodiles. Comp. Biochem. Physiol. 32: 161-174.
1974. Salt gland secretion in a mangrove monitor lizard. Comp. Biochem. ri, :..i 47(4A):
1245-1255.
1976. Salt glands in reptiles. : 413-445 in C. Gans and W. R. Dawson, eds. Biology of
the Reptilia. -i. :..i A. Vol. 5. Academic Press, New York.
1978. Role of the skin in sodium and water exchange of aquatic snakes placed in sea water.
Amer. J. -. 235: R151-159.
S1979. Control mechanisms in reptiles. Pages 273-322 in R. Gilles, ed. Mechanisms of
osmoregulation in animals. Wiley Interscience, New York.
1980. The relation of sodium and water balance to survival in sea water of estuarine and
fresh-water races of the snakes Nerodia fasciata, N. sipedon and N. valida. Copeia 1980:
268-280.
1981. Behavioral osmoregulation in the Key Mud Turtle, Kinosternon b. baurii. J. Herpetol.
15: 163-173.
1982. .... relations of crocodiles in Florida Bay. Copeia 1982: 374-385.
1984. The contrasting roles of the salt glands, the integument and behavior in osmoregulation
of marine and estuarine reptiles. Pages 107-129 in A. Pequeux, R. Gilles and L. Bolis, eds.
Osmoregulation in estuarine and marine animals, Vol. 9. Lecture notes on coastal and estuarine
studies. .... -Verlag, New York.










IBiiI FTIN OF MARINE SCIFNCF V01_ 44, NO, 1, 1989


choline :..- 1:. .. in the laboratory (Taplin et al., 1982). Since the :.... : glands
are so small it is possible that they originated in an ancestral, inland, freshwater
species. This idea has not been given the attention it deserves since it seems
preposterous for a :: i:... ::.: animal to need a salt gland. However, many croc-
odiles occur in areas that undergo seasonal drought and the need for salt excretion
and water conservation under such dehydrating conditions is obvious. Further
information is needed on the function of the lingual glands in freshwater croco-
diles, especially during periods of aestivation. It would also be quite useful to
have more information on the degrees of genetic relatedness among the species
of Crocodylus. Although it is certainly not impossible that "reverse evolution"
of marine to freshwater forms occurred in Crocodylus, the matter requires con-
siderably more data for adequate analysis. In the Philippine lake-locked sea snake
Hlydrophis semperi it is clear that a very recent derivation from the marine H.
cyanocinctus has occurred. Indeed I. semperi has retained a functional salt gland
despite i- .: ., in a freshwater lake (Dunson and Dunson, 1974). On Rennell Island
(Solomons Group), sea snakes of the genus Laticauda have colonized the brackish
water Lake Tegano, illustrating again how a secondarily freshwater sea snake fauna
can develop (Wolff, 1969).

ACKNOWLEDGMENTS
Supported by National Science Foundation grant BSR-8212623 to W.A.D. and a grant from Florida
Power and Light Co. to F.J.M. '-' :- ..." thank the National Park Service for financial and logistical
support over the past 10 years. Necropsies and bacterial identification were .. .: by J. Harkness,
D.V.M.H. :,,: and J. Minnich critically reviewed the manuscript.

LITERATURE :,
Birkhead, W. S. and C. R. Bennett. 1981. Observations ofa small population ofestuarine-inhabiting
near Southport, North Carolina. Brimleyana 6: 111-117.
Connell, J. H. 1975. Some mechanisms producing structure in natural communities: a model and
evidence from field experiments. Pages 460-490 in M. L. Cody and J. M. Diamond, eds. Ecology
and evolution of communities. Harvard Univ. Press, Cambridge, Massachusetts.
Dawson, W. R., G. A. Bartholomew and A. F. Bennett. 1977. A reappraisal of the aquatic special-
izations of the :..- marine iguana (Amblvrhynchus cristatus). Evolution 31: 891-897.
Dunson, M. K. and W. A. Dunson. 1975. The relation between plasma Na concentration and salt
gland Na-K ATPase content in the diamondback terrapin and the yellow-bellied sea snake. J.
Comp. Physiol. 101: 89-97.
Dunson, W. A. 1969. Electrolyte excretion by the salt gland of the Galapagos marine iguana. Amer.
J. Physiol. 216: 995-1002.
S1970. Some aspects of electrolyte and water balance in three estuarine reptiles, the dia-
mondback terrapin, American and 'salt water' crocodiles. Comp. Biochem. Physiol. 32: 161-174.
1974. Salt gland secretion in a mangrove monitor lizard. Comp. Biochem. ri, :..i 47(4A):
1245-1255.
1976. Salt glands in reptiles. : 413-445 in C. Gans and W. R. Dawson, eds. Biology of
the Reptilia. -i. :..i A. Vol. 5. Academic Press, New York.
1978. Role of the skin in sodium and water exchange of aquatic snakes placed in sea water.
Amer. J. -. 235: R151-159.
S1979. Control mechanisms in reptiles. Pages 273-322 in R. Gilles, ed. Mechanisms of
osmoregulation in animals. Wiley Interscience, New York.
1980. The relation of sodium and water balance to survival in sea water of estuarine and
fresh-water races of the snakes Nerodia fasciata, N. sipedon and N. valida. Copeia 1980:
268-280.
1981. Behavioral osmoregulation in the Key Mud Turtle, Kinosternon b. baurii. J. Herpetol.
15: 163-173.
1982. .... relations of crocodiles in Florida Bay. Copeia 1982: 374-385.
1984. The contrasting roles of the salt glands, the integument and behavior in osmoregulation
of marine and estuarine reptiles. Pages 107-129 in A. Pequeux, R. Gilles and L. Bolis, eds.
Osmoregulation in estuarine and marine animals, Vol. 9. Lecture notes on coastal and estuarine
studies. .... -Verlag, New York.











UIINSON AND)5 MNLOIII REMIII-ES OF FLORIDA BAY


1985. Effects of water salinity and food salt content on growth and sodium efflux of :,: :..
diamondback terrapins (Malaclemys). : : Zool. 58(6): 736-747.
.1986. Estuarine populations of the snapping turtle (. as a model for the evolution
of marine adaptations in reptiles. Copcia 1986: 741-756.
-- and M. K. Dunson. 1974. Interspecific differences in fluid concentration and secretion rate
of sea snake salt glands. Amer. J. Physiol. 227: 430-438.
- and 1979. A possible new salt gland in a marine homaloposid snake (Cerberus
rhynchops). Copeia 1979: 661-672.
-- and H. Heatwole. 1986. Effect of relative shell size in turtles on water and electrolyte com-
position. Amer. J. Physiol. 250 (Reg. Int. Comp. Physiol. 19): R1133-1137.
and G. D. Robinson. 1976. Sea snake skin: permeable to water but not to sodium. J. Comp.
108: 303-311.
-- and M. Seidel. 1986. Salinity tolerance ofestuarine and insular Emydid turtles (Pseudemvs
nelsoni and Trachemys decussata). J. Herpetol. 20: 235-243.
Ellis, T. M. 1981. Tolerance of sea water by the American crocodile, Crocodylus aculus. J. Herpetol.
15: 187-192.
Evans, D. H. and T. M. Ellis. 1977. Sodium balance in the hatchling American ( .... .:,: Crocodylus
acutus. Comp. Biochem. Physiol. 58A: 159-162.
Freda, J. and W. A. Dunson. 1986. Effects of low pH and other chemical variables on the local
distribution of amphibians. Copeia 1986: 454-466.
Gaby. R., M. P. McMahon, F. J. Mazzotti, W. N. Gillies and J. R. Wilcox. 1985. Ecology of a
population of Crocodylus acutus at a power plant site in Florida. J. Herpetol. 19: 189-198.
Hairston, N. G. 1980. Evolution under interspecific competition: field experiments on terrestrial
salamanders. Evolution 34: 409-420.
Jacobsen, T. 1983. Crocodilians and islands: status of the American alligator and the American
crocodile in the Lower Florida Keys. Fla. Field Nat. 11(1): 1-24.
Jaeger, R. G. 1971. Moisture as a factor influencing the distributions of two species of terrestrial
salamanders. Oecologia 6: 191-207.
Jayne, B. C., H. K. Voris and K. B. Heang. 1988. The diet, feeding behavior, growth and numbers
of a population of Cerberus rynchops (Serpentes: Homolopsinae) in Malaysia. Fieldiana: Zool.,
1394: 1-15.
Kushlan, J. and F. J. Mazzotti. In press. Population ecology of the American crocodile: historic and
present North American distribution. J. Herpetol.
Lauren, D. J. 1985. The effect of chronic saline exposure on the electrolyte balance, nitrogen me-
tabolism, and corticosterone titer in the American alligator, mississippiensis. Comp.
Biochem. Physiol. 81A(2): 217-223.
Lawson. R. 1985. Molecular studies 'ii ...... .: :... snakes. Ph.D. Thesis, Louisiana State Univ.,
Baton Rouge, Louisiana. 183 pp.
Mazzotti, F. J. 1983. The ecology of Crocodylus acutus in Florida. Ph.D. Thesis, The Pennsylvania
State University, University Park, i.. .. 161 pp.
-- and W. A. Dunson. 1984. Adaptations of Crocodylus acutus and i for life in saline
water. Comp. Biochem. Physiol. 79A(4): 641-646.
B. Bohnsack, M. P. McMahon and J. R. Wilcox. 1986. Field and laboratory observations
on the effects of high temperature and salinity on .. :: ,,.. Crocodylus aculus. Herpetologica
42(2): 191-196.
Miller, D. E. 1985. Rain water drinking by the mangrove water snake, Nerodia fasciata compres-
sicauda. Herp. Review 16(3): 71.
Ogden, J. 1978. Status and nesting biology of the American crocodile, Crocodylus acutus i*' :
Crocodilidae) in Florida. J. Herpetol. 12: 183-196.
Pianka, E. R. 1975. Niche relations of desert lizards. Pages 292-314 in M. L. Cody and J. M.
Diamond, eds. Ecology and evolution of communities. Harvard Univ. Press, Cambridge, Mas-
sachusetts.
Rahcl, F. J. 1984. ', ..... : assemblages along a bog lake successional gradient. Ecology
65: 1276-1289.
Robinson, G. D. and W. A. Dunson. 1976. Water and sodium balance in the estuarine diamondback
terrapin (IMalaclemys). J. Comp. Physiol. 105: 129-152.
Stokes, G. D. and W. A. Dunson. 1982. :- .*** and channel structure of reptilian skin. Amer.
J. Physiol. 242: F681-689.
Tansley, A. G. 1917. On competition between Galium saxatile L. (G. hercynicum Weig.) and Galium
syh'estre Poll. (G. asperum Schreb.) on different types of soil. J. Ecol. 5: 173-179.
.:... L. E.. G. C. Grigg and L. Beard. 1985. Salt gland function in fresh water crocodiles: evidence
for a marine phase in eusuchian evolution? Pages 403-410 in G. Grigg, R. Shine and H. Ehmann,
cds. Biology of Australasian frogs and : : Royal Zool. Soc. New South Wales, Australia.











244 BULLETIN OF MARINE SCIENCE, VOL. 44, NO. 1, 1989

P. Harlow, T. M. Ellis and W. A. Dunson. 1982. Lingual salt glands in Crocodylus
acutus and C. johnstoni and their absence from Alligator mississipiensis and Caiman crocodilus.
J. Comp. Physiol. 149: 43-47.
Tilman, D., M. Mattson and S. Langer. 1981. Competition and nutrient kinetics along a temperature
gradient: an experimental test of a mechanistic approach to niche theory. Limnol. Oceanogr. 26:
1020-1033.
Toft, C. A. 1985. Resource partitioning in amphibians and reptiles. Copeia 1985: 1-21.
Travis, J. and J. C. Trexler. 1986. Interactions among factors affecting growth, development and
survival in experimental populations ofBufo terrestris (Anura: Bufonidae). Oecologia (Berlin) 69:
110-116.
Weider, L. J. and P. D. N. Hebert. 1987. Ecological and physiological differentiation among low-
arctic clones of Daphnia pulex. Ecology 68: 188-198.
Wilson, S. D. and P. A. Keddy. 1986. Species competitive ability and position along a natural stress/
disturbance gradient. Ecology 67: 1236-1242.
Wolff, T. 1969. The fauna of Rennell and Bellona, Solomon Islands, Phil. Trans. Roy. Soc. B255:
321-343.
Wood, R. 1981. The mysterious mangrove terrapin. Fla. Naturalist, July-September, pp. 6-7.

DATE ACCEPTED: November 24, 1987.

ADDRESSES: (W.A.D.) Department of Biology, The Pennsylvania State University, University Park,
Pennsylvania, 16802; (F.J.M.) Department of Wildlife and Range Sciences, University of Florida,
Broward County Extension Service, 3245 College Avenue, Davie, Florida 33314.