Title: Osmoregulation in crocodilians
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00066443/00001
 Material Information
Title: Osmoregulation in crocodilians
Physical Description: Book
Creator: Mazzotti, Frank J.
Dunson, William A.
Affiliation: University of Florida -- Davie, Fla. -- Department of Wildlife and Range Sciences
Pennsylvania State University -- Department of Biology
Publisher: Oxford Journals
Publication Date: 1989
General Note: Drawn from American Zoologist (currently Integrative and Comparative Biology), Vol. 29, pp. 903-920 (1989
 Record Information
Bibliographic ID: UF00066443
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.


This item has the following downloads:

OsmoCroc_AmerZool_1989 ( PDF )

Full Text

AMER. ZooL,, 29:903-920 (1989)

Osmoregulation in Crocodilians'

Department of . 'and Sciences, University ofFlorida,
Broward County Extension 3245 Ave., Davie, Florida 33314

Department of The Pennsylvania State University,
208 Muelier Lab., University Park, Pennsylvania 16802

SYNOPSIS. Recent crocodilians live primarily in freshwater habitats. However two species
(Crocod ylus acutus and C. porous) are estuarine specialists; two others (C. nilolicus and C.
johnsioni) that are primarily found in fresh water, have estuarine populations. Routes of
uptake of water and sodium include drinking, feeding and associated incidental drinking,
and integumental and buccal diffusion. Routes of loss include faeces-cloacal fluid, lingual
salt glands, integumental and buccal diffusion, and respiratory loss. The least understood
route of salt and water exchange is that of the oral and buccal '.! ., 1: which are much
more permeable to water and sodium than the general integument. The freshwater alli-
gator -' a . a manner typical foran amphibious reptile.
Body sodium turnover is low and the general integument is quite low in .. ... : ., to
sodium. Water turnover is more rapid (in terms of molar exchange) but still relatively low
for an aquatic reptile. Most water exchange occurs across the integument and buccal
epithelia. The presence : salt glands in freshwater crocodilians remains enigmatic,
as does the failure of these exocrine glands in estuarine species to respond to saline loading.
Secretion does occur after injection of the parasympathetic stimulant methacholine chlo-
ride. The "salt water crocodile" (C. porosus) possesses a suite of osmoregulatory adaptations
similar to those found in other estuarine reptiles. Water and sodium balance are maintained
primarily by an extremely low general permeability to sodium, by economies in water loss,
and by excretion of excess sodium by the lingual salt glands. Further work is needed to
examine newly hatched C porosus, and the possibility of ontogenetic change in lingual
gland function in C. aculus. The importance of incidental drinking of sea water during
feeding (recently discovered in turtles) needs to be evaluated in crocodilians. The use of
osmoregulatory data in interpretation of the evolutionary history of the genus Crocodylus
needs to be viewed with caution. The hypothesis that all species of Crocodylus originated
from the transoceanic migration of a saline-tolerant form may not be the most parsimo-
nious explanation.

Osmoregulation is % similar in
all aquatic reptiles ...... ... 1- Dunson
and Mazzotti, 1989; Taplin, I"''~ The
most pervasive aspect of crocodilian osmo-
regulation is the amphibious habit of all
extant forms, in combination with a uni-
form external :.. :i ': Even the most
aquatic species spend a great deal of time
basking on land and/or r. in subter-
ranean refugia or dens. U: omeofthe
*. forms (Williston, 1914), no extant
species is truly marine or -!--.

SFrom the Symposium on oj the Crorodilia
presented at the Annual Meeting of the American
Society of Zoologists, 27-30 December 1987, at New
Orleans, Louisiana.

Living crocodilians do differ in the use
of either : : or saline water habitats. This
has major consequences for osmoregula-
tory mechanisms because crocodilians
maintain body fluid concentrations essen-
tially the same as other vertebrates (about
one-third that of sea water). Probably an
-.--:-- important, but very poorly known,
aspect of crocodilian osmoregulation is the
exposure of many species to periodic/sea-
sonal droughts. Little information on the
. h:-:..- ... consequences of such dehy-
,. .:: i. .i exists, but they must be
considerable. The study of possible aesti-
vation in .. .. :: .... should receive a high
priority since it has great potential for fur-
thering our understanding of the evolution
of the i:. :: .. salt glands found in Croco-


TABLE 1. Potential routes ofintake and loss of water and
salts in crocodilian, .
1. ....i'.. without feeding
2. ..1 "incidental" to feeding
3. Feeding
4. Integumental '.' . .. : body surface
-buccal area
1. Faeces/urine-cloacal fluid
2. Respiratory water loss
3. Lingual salt glands
4. Integumental ,i* : : body surface
-buccal area

This review .ii summarize selected lit-
eratureon : ili: :: osmoregulation, and
compare it to other aquatic reptiles. A pri-
mary emphasis will be to document the
essential similarity in osmoregulation
among all aquatic reptiles.
Osmoregulatory data are now being used
to support evolutionary theories of the ori-
gin of the Crocodylus group. Proponents of
the "transoceanic migration hypothesis"
.:: .. that a 1::. salt-tolerant form of
Crocodylus gave rise to all extant species
(Densmore, 1983; Taplin, 1984a: Taplin
et al., 1985). An alternative explanation, as
is proposed for marine snakes and turtles,
is that a gradual :- ... ... to estuarine
transition led to marine forms (Dunson,
1986; Dunson and Mazzotti, 1 ', An-
other possible scenario could involve the
origin of lingual salt glands under the de-
hydrating conditions of periodic droughts
in freshwater systems. We will argue here
that current data are inadequate to estab-
lish any of these theories as the preferred
hypothesis. What is needed is a broader
perspective on crocodilian osmoregulation
in the context of f:: ::;, on other : :::i
and a great deal of new information spe-
cifically targeted to discriminate between
alternative hypotheses. First, however, one
must 1I: :... '.1 understand the present
data base .'- .1..':i with water and salt bal-

The osmoregulatory "problems" posed
by life in fresh or saline waters are .: - .:

related to the amounts of water and salts
that are exchanged across various body
In crocodilians, potential routes
of exchange are the same as in other aquatic
reptiles (Table 1; Dunson, 1979; Minnich,
1979). Of course, the relative importance
of each route varies with external salinity
and the specific body plan of each reptile.
For example, ( .., ,i,., are especially
prone to expose buccal areas of epithelium
lining the mouth to the exterior. This also
occurs in turtles (where the pharynx and
cloaca also may be effectively exteriorized),
but not in snakes which are tightly sealed.
Since these bucco-pharyngeal areas are
-::.. in permeability than the general
integument (Dunson, 1967; Dunson and
Mazzotti, 1988) they may play a :L::.
cant role in w. : :ii water and salt balance.
A recently discovered route of salt and
water uptake in reptiles is "incidental
drinking" (Shoemaker and "'.: 1984;
Dunson, 1 .:_i). This represents water
: along with the food, as distinct
from water swallowed without any . .
(true drinking). Although as yet directly
measured only in turtles (Dunson and Maz-
zotti, 1989), incidental drinking is clearly
a route of major significance, especially in
estuarine forms. Dunson (1985) has sug-
gested that this is the single most important
source of salt intake in estuarine-marine
reptiles, is related to the geometry of the
food consumed, and explains variations in
salt gland size among marine reptiles. Spe-
cial mechanisms probably have developed
among sea turtles (and presumably among
other marine-estuarine reptiles) to mini-
mize such sea water uptake, by expelling
most of the water swallowed with food.
Another major route of water exchange
that is poorly understood is the integu-
ment. Direct measurement of crocodilian
.i ::.. : i :' .::.. i :: has only rarely
been attempted (Dunson, 1981; Dunson
and Mazzotti, 1'' The general body
shape of crocodilians is quite similar to that
of lizards, as is the total surface area rela-
tive to mass (Table 2). It is :.. : .... ::..: that
reptiles of this body form have relatively
low total surface areas; snakes have a total
surface area that is much higher (Table 2).
However as important as body form may
be in general adaptation, it appears to have


TABLE 2. The relation between lotal exterior surface area (A in cmz) and body ma^s (M in g) in various reptile (arranged
in order of declining area of a 100 g reptile).
A for
Species Equation M 100 g Reference
Pelamis platurus (sea snake) A = 12.78M0717 347 Dunson, 1978
Nerodia aipedon (water sake) A = 14.17M067" 323 Dunson, 1978
Trionyx sp. softshelll turtle) A = 16.61M Mo' 281 Dunson, 1986
mississippiensis : .. A = 13.7M116' 261 Davis et ol., 1980
Crcodlitus sp. (crocodile) A = 11.7M"14 223 Dunson, 1982
"Standard lizard" A = 10M067 219 Minnich, 1979
Chelydra serpentina (snapping turtle) A = 8.62M\i78 196 Dunson, 1986
Pharyngeal/buccal areas inside the mouth are not included.

a minor role in osmoregulation. An anal-
i-' with thermoregulation is instructive
(Dunson, 1986). As Scholander (1955) has
eloquently pointed out, it is the heat trans-
fer across a surface that is the primary
determinant of thermoregulatory :.1 .i -
tion. The same is true of crocodilian sur-
face area and ..... :': It is not so
much the :-..- --' surface area per se
as its permeability that determines the role
in osmoregulation. The buccal area of
crocodilians is a case in point. Although
much remains to be learned about the role
of this oral epithelium, it appears to be
highly permeable to water and to play a
..... role in overall exchange, despite
an obviously minor surface area in com-
parison with the general body integument
(Dunson and Mazzotti, 1988). Such a local
area of '.:. .: epithelial permeability
probably also occurs in the ocular mem-

The alligator is by far the best studied
of the freshwater crocodilians (Ellis, 198 la;
Coulson and Hernandez, 1983; Ellis and
Evans, 1984), .!:.. there are some data
also on estuarine species in fresh water (C.
acutus: Schmidt-Nielsen and .. ...' .... ,
1967; Evans and Ellis, 1977) (C. porosus:
Taplin, 1982, 1984b). There seem to be
few differences among crocodilians in their
abilities to adapt to fresh water. Thus the
discussion below will be organized into
physiological categories of salt and water
Plasma composition of several species of
crocodilians in fresh water is unremarkable
in that it resembles that of freshwater rep-
tiles generally (Dill and Edwards, 1931;

. . 1970; Minnich, 1979; Coulson
and Hernandez, 1983; Ellis and Evans,
1' i: These animals face the classic : !
lem of hyper-osmoregulation (mainte-
nance of a body fluid concentration far
above that of the environment). Thus they
would be expected to possess mechanisms
for conservation of solutes such as sodium,
and excretion of excess water.

Sodium turnover
Ellis and Evans (1'-: I; have character-
ized sodium exchange in missis-
".' Total body sodium was 82
pmoles/g wet mass, similar to that of C.
. at 75 ,moles/g wet mass (Taplin,
1- .), but higher than the value for C.
aculus (43 jpmoles/g wet mass; Evans and
Ellis, 1 '. . Ellis and Evans (1984) attrib-
uted the difference between ,.' .... and
C. acutus to = .. of the -'---...- used
to solubilize the body (either acid digestion
in the former or homogenization in the
latter). This seems unlikely to be the case
since i .:::: :. (1982) recorded a similarly
low value for a single C. acutus (39 Armoles/g
wet mass) when the animal wasi .. ..
in acid, not homogenized as was done by
Evans and Ellis (1977). It is also quite pos-
sible that differences in whole body sodium
concentration are related to differential
amounts and degrees of calcification in
bone. Dunson and Heatwole (1'-'._ doc-
umented large differences in total body
sodium of turtles that were related to shell
size and body size of the animal. Such. : .
seem to be : : .. :: : related to relative sizes
of the extracellular and intracellular spaces.
For sodium flux calculations it is impor-
tant to know how much of the total body
sodium is exchangeable. In juvenile rep-



tiles generally all of the body sodium is
exchangeable (Dunson and Heatwole,
1' However, with growth (and pre-
sumably progressive ,i ... . of the
bones) the relative amount of '. ::
able sodium in turtles diminishes (Dunson
and I eatwole, 1'- ,- This. :: : is thought
to be due to binding of sodium to bone
crystal. '1 ::: :! sodium pools account
for all ot the body sodium in ... -i .:
(Ellis and Evans, 1984) and small C. porosus
(T .. 1 ~-). It is very likely that the
proportion of sodium that is -::: --
in crocodilians diminishes with age/size,
but this is yet to be studied. Data are also
needed on the skeletal mass and .
tion of crocodilians of varying sizes. We
believe that Ellis and Evans (1984) were
incorrect in assuming that '. . E... Alli-
gator have a .. .. .. ..1. bone sodium
The rate constant ( : :: of body
sodium ex. 1::: -1 per unit time) for
sodium efflux. : .. 1 -;.. i : in fresh
water is 0.00068/hr; thus sodium efflux is
about 5.5 pmoles/100 ghr (recalculated
from Ellis and Evans, 1984, using total body
sodium rather than their estimated
exchangeable value; Table 3). This is quite
similar to sodium :t... from the fresh-
water turtle Trachemys scripta (1.6-3.3
aimoles/100 g-hr; Dunson, 1967). Sodium
efflux for C. aculus in fresh water was 2.5
Amoles/100 g-hr (Evans and Ellis, 1 .I
The value for the :: :- ii turtle, Trionlyx
(1.5 Amoles/100 g-hr; Dunson, 1979), is
somewhat lower. It is unclear if such rel-
atively minor differences are biologically
significant without further ... i:.:. : .
of :ii body sodium balance. In Alli-
gator, ::'. of the sodium efflux was from
the head, neck and i.... .... :. z4 .. from
the body, tail and hindlimbs, and only 11 %
from the cloaca (Ellis and Evans, 1: *;
This suggests that sodium conservation in
renal/cloacal excretory processes is fairly
effective, and that ( 11. :H ry epithelial dif-
fusion may be the main route of loss of
sodium. In C. acutus, of sodium efflux
in fresh water is from the head, neck aind
forelimbs, 1". !: ... the body, hindlimbs
and tail, and 1': from the cloaca (Evans
and Ellis, 1977).

....unfed C. porosu. in fresh water had
net sodium losses of 0.6-0.9 gmoles/100
g-hr; efflux was 2.4-3.0 pmoles/100 g-hr
(T !.. 1982). There was no evidence of
net sodium uptake, even when bath sodium
concentrations rose as high as 1 rmAl. Most
.) sodium loss was :: :. .... . ,ith
the remainder being i .. . ..... :: all
sodium influx was integumentary. The
rates of sodium exchange in the. ......
ians C. porosus, C. acutus and C. johnstoni
were all similar (half lives of 50-75 days;
I- .. and Ellis, 1979; T' ..: 1 "- but
higher than values for the freshwater tur-
tles Trionyx spinijerus and Trachemys script
(Dunson, 1967, 1979). Taplin (1982) sug-
gested that this is due to lower :.. .....::::
tal sodium permeability in turtles, which
seems reasonable. Whatever the explana-
tion for the. i:: .. . it .... that croc-
odilians are less efficient or .. . . .in
conservation of sodium in fresh water than
are turtles.
Several studies have ;.... .i to esti-
mate integumentary ...:...... permeability
indirectly (Bentley and 1...:: : else,
1965; Ellis and Evans, 1984), although this
is :.. ...: with difficulty. IHowever, Ellis
and Evans' (1984) in vivo estimate of body
integumental sodium efflux (0.01 moles/
cm2 hr) in Alligator is remarkably close to
direct measurements on isolated skin: ::::;
moless. ... -hr; Dunson and Mazzotti,
1988). It should be noted that this rep-
resents .:.. i:: :: ::i efflux, not net loss.
However in fresh water there may be only
an '. ; . . influx of .... ..... In Cai-
ma,, net loss of sodium across the :".--
ment (measured by holding the head out
of the bath) was the same as in Alligator
(about 0.01 pmoles/cm'2hr; Bentley and
.. ...... :i-Nielsen, 1965). However in C.
acutus the body .:: ::::: ::- had a much
lower estimated sodium efflux (0.001
Aimoles/cm2 hr; Evans and : :. 1977).
This .... that the euryhaline C. acutus
has a lower integumental ... :., perme-
:. li::. than two: :. r forms
and Caiman). An additional finding was that
some portion of the head-neck-forelimbs
must be more permeable than the general
body :i .:;--- -.1. This has now been con-
firmed by a few direct measurements of the


0n 0

R -

+1 +1

0 0

I- -
-* o

c 0

C4 Cl

C0 Cl

CC 1=
Cl 0

0 *- -l

0 0


0 0

Cn C4
co co CM o c r" 5^
I : +1I +1

<-- ^-*

co o -

+1 +1 +

Cl Cl CC CC I--
C o 0 0 C

C X:
3 C


0 r
- 0
II + 1

- CC


+I- I + + +I++ 1
In c l f GM c
cy 0 o en o

- CC C
+1 -} +1 +1 +

- cl

6D o CTI G l CC

0 -^ -50 -f -^'

l 11 -0 H c
+1 -}1 1 +: C+
--i", 0 0 ~ CC C CC^












o =

C ' ~- C




0 ^3 00

3C C N.- 'C
Q b, ".2 ~ ^


sodium permeability of epithelia from the
roof of the mouth and the -... .. -- of Alli-
gator (: ....... and Mazzotti, 1988). It
:: likely that the oral epithelia of all
crocodilians are considerably more perme-
able to sodium and water than the general
integument (see C. porosus below).
Ellis and Evans (1984) have suggested
that non-dietary sodium uptake can com-
pensate for 86% of .... ..: .*... ..: sodium
i ".. in Alligator. We consider this unsub-
stantiated. Certainly present evidence
shows that crocodilians cannot absorb net
sodium from fresh water via an active
uptake mechanism such as is known for
some freshwater turtles (i: :... .. 1' '-
We believe it is unwise to assume direct
quantitative equivalence of isotopic efflux
measurements and net loss data as Ellis and
Evans (1'. i: have done, since differential
technical artifacts are likely. We recom-
mend further measurements of isotopic
influx and efflux in the same animals, and
a detailed examination of the in vitro trans-
port characteristics of buccal epithelia.

Water turnover
: i il Alligator have a body water con-
centration (7 wet mass, Ellis, 1981a)
fairly typical of reptiles : ... ... 1 '". ';.
Thorson (1968) reported only minor vari-
ation in the total body water of four ... :
of crocodilians (means of 72.9-73.1% wet
mass; :-.. of 71.0-74.4). Other reptiles
(turtles) are known to show major variation
in this parameter with.--.-.- in relative
bone mass related to ...*.. .. .. and age
differences (Dunson and Heatwole, 1986).
Body fat also influences body water con-
centration (Minnich, 1979). In contrast to
the small amount of sodium : i. from
Alligator, water efflux is substantial (about
0.3-0.9 ml/100 g-hr, Table 3). Variation
in overall water efflux is considerable, but
does not seem to be clearly related to exter-
nal ..' It is likely that water efflux is
inversely related to body size in Alligator
although there are too few data for a def-
inite conclusion (Table 3). : i i (1981a) cal-
culated that of unidirectional water
efflux is integumentary, 2 respiratory,
and : cloacal. Dunson and Mazzotti
(1988) similarly found that a large pro-

portion of water turnover in .".. occurs
through the integument. They also made
direct measurements of in vitro ....
tal water permeability for the hrst time,
:. .:t-.-. the previously ..: 1 but
unproven, large difference in permeability
of oral epithelia and general body integ-
ument. The exact rate of water efflux across
the body integument of Alligator can now
be directly compared with that of a wide
variety of reptiles (Table 4). The results
are somewhat .:.:. I. ::. in that the skin of
S.'. ..'. is considerably less permeable to
water than that of many other freshwater
." .. =- It is, however, similar to that of
the -... . ..... snake .. ' .' T hus
despite the relative significance of the
integument in water balance of in
comparison with other routes of water loss,
water turnover in this .:.. '! : croco-
dilian is less than that of highly aquatic
S :: ..: This seems in contra-
diction with the external morphology of
Which is i .. . for life in
water, but it may reflect the fact that a
great deal of time is actually spent on land.
Small unfed C. porosus in fresh water
exchange most of their water across the
... ..... (about ..' of either unidirec-
tional influx or efflux; Taplin, 1I ':
Integumental influx is I..1:'! : :-. than
efflux, but net water intake via this route
was only one-sixth the amount drunk. Buc-
cal exchange was not separated from that
of the : -- - ; .-..- ..;-- In contrast to
a net loss of sodium, water balance is main-
tained. Since cloacal water losses greatly
exceed net :' .. .... ..-y uptake, drink-
is necessary to maintain a steady state.
Taplin (1' -.; concluded that the most
estuarine crocodilian (C. is no less
SIT. : : in osmoregulating in fresh water
than are the freshwater specialists C. john-
stoni and Alligator.

water loss
Part of the discussion above has focused
on exchanges of water across the integu-
ment of i. while it is submerged in
fresh water. Yet it is well known that this
species (as all crocodilians) spends consid-
erable time out of water either basking
and/or resting on land. Under these cir-


TABLE 4. Permeability of i oated rptilian /n.Ak to water when the outside olulion i.
Ringer' o lution (DunAon, 1978).'

.cea water and the inude i., retilian

Habitat and species eiiiux Reference
Fresh water-highly aquatic
Trionyx muticus softshelll turtle) 1,049 Dunson, 1986
Regina scptenivitiata (queen snake) 573b Stokes and Dunson, 1982
Che'lydma .erpenlina (snapping turtle) 303 Dunson, 1986
Fresh water--amphibious
Nerodia .ipedon (northern water snake) 99 Dunson, 1978
Nerodia fasciata picliventrs (banded water snake) 70 Dunson, 1978
Kinostfrnon subrubrum (mud turtle) 70 Dunson, 1986
iissiipieisis * 46 Dunson and Mazzotti, 1988
Agkisirodon 'piscivor us (cottonmouth snake) 20-501 Dunson and Freda, 1985
Euneclis notaeus (yellow anaconda) 28h ILjungman and Dunson, 1983
Cerberus rhynehopsn (dog-faced water snake) 36 Dunson, 1978
N'rodio fasc ata rompresicauda (mangrove snake) 21 Dunson. 1978
Marine-amphibious to pelagic
Various species of sea snakes 40-1331 Dunson and Stokes, 1983
SEfflux in pmole/cm' hr.
b Efflux measured between fresh water (inside) and I M NaCI (outside).

cumstances evaporative water loss occurs
not only from the respiratory tract but
across the integument as well. As in other
reptiles, these :.. ...... -..-y evaporative
losses would be expected to be much lower
quantitatively than the water to water dif-
fusion rates (Tercafs and Schoffeniels,
1' -;. Yet in the intact animal, rates of
mass (water) loss of ... C. aculus im-
mersed in sea water (1... initial mass/
day) are actually quite similar to the total
evaporative loss in air at 96...' relative
......'. "... (1.3-1.'. /day) (Dunson, I
The only detailed study of evaporative
water loss in a crocodilian is that of Davis
et al. I(1 .:, on,' I f.. ofloss were
directly related to temperature and
inversely related to body size. r. -atory
water loss at . was 0.12 pl/g-hr; cuta-
neous water loss was 0.14 gil/g-hr. 1:
ratio of :r-atory to cutaneous loss
increased with decreasing temperature.
: .. total evaporative water loss rate of
animals -. -h:- 2.5-6.6 kg was 0.61
mass/day (at ::, dry air) of which the
cutaneous component was 54 Skin resis-
tance was 55 sec/cm, a value within the
range : :i; for other : I :1 Davis et
al. (1- :':; concluded that the rate of evap-
orative water loss of Alligator places it in a

transitional state between aquatic and ter-
restrial reptiles. This is most . :.
since measurement of the water flux
across Alligator integument led Dunson and
Mazzotti (1 : to a very similar conclu-

.' '/ cloacal excretion
In '- -'- Alligator in fresh water, only
11 of sodium efflux and 2 of water
i.. are cloacal (Ellis, 1981a; i-ii. and
Evans, 1984). Alt': ::. : the kidney and
cloaca have only a minor role in sodium
chloride excretion, they are the primary
route of excretion of nitrogenous wastes.
Coulson and Hernandez (1983) have
reviewed kidney :.. : of Alligator in
great detail. Hydrated unfed Alligator
excretes over half the urinary nit-r "
ammonium bicarbonate; :
tion uric acid is the primary means of nitro-
gen excretion. The five most common
c .-,,-,-,:..-r: of urine of unfed hydrated
Alligator were (in 1 l.: order) ammo-
nia, bicarbonate, uric acid, creatinine and
pi..-.-ph r- It appears at present that all
crocodilians are similarly ammonotelic-uri-
but detailed interspecific compar-
isons of renal and cloacal function are lack-



Even in the estuarine C. acutus the kid-
neys do not markedly regulate water and
salt output. The osmotic urine/plasma ratio
varied only between 0.7 and 0.9 when croc-
odiles were hydrated or dehydrated
(Schmidt-Nielsen and Skadhauge, 1967).
However a "downstream" regulation of
water and salt absorption occurs in the
cloaca. Sodium chloride was almost com-
pletely absorbed in the cloaca of hydrated
C. acutus but was less completely absorbed
after salt loading (Schmidt-Nielsen and
Skadhauge, 1967). The precipitated uric
acid in crocodilian cloacal fluid may con-
tain significant amounts of some electro-
lytes bound as urate salts. Minnich (1972,
1979) has discussed the importance of this
process in terrestrial reptiles faced with
problems of water shortage. It is not yet
clear whether urate salts are an important
route of cloacal excretion in crocodilians.
Sodium and potassium contents of cloacal
uric acid solids in C. acutus and C. porosus
were not affected by water salinity; only
potassium was present at significant levels
(393-671 jmoles/g dry mass) (Dunson,
1982). As Skadhauge (1977) has suggested,
the crucial role of the cloaca in modifying
crocodilian ureteral urine needs to be fur-
ther examined with perfusion techniques.
In contrast to fresh water, crocodilian
physiological reactions to saline water vary
among different species. Thus the follow-
ing discussion will consider each species

There seems little doubt that Alligator is
a freshwater species, with only a limited
capacity to utilize estuarine habitats. It lacks
the lingual salt glands that are found in
Crocodylus (Taplin et al., 1982). Yet Alli-
gator is commonly seen in coastal areas and
has reproducing populations in the Lower
Florida Keys (Jacobsen, 1983) and on
islands off the Georgia coast (Tamarack,
1988). Although data are limited, it is quite
possible that recruitment in the Keys is
mainly or entirely limited to areas of per-
manent fresh water, especially where mos-

quito control ditches have been dug into
the fresh ground water lens. It also seems
likely that the adults would have to return
to fresh water periodically to drink. Clearly
there is a need for a detailed study of island
Alligator to determine whether they possess
some physiological mechanisms for toler-
ating saline water or simply use behavioral
means of osmoregulation. If they are to
exist on islands where fresh water is
extremely limited and there is very little
for adults to eat in temporary freshwater
ponds, they must forage in the sea (Tama-
rack, 1988).
Although we will have to await further
studies on island Alligator, a most unique
field study in North Carolina illustrates the
behavior of Alligator when exposed to a
dramatic increase in salinity. Birkhead and
Bennett (1981) documented the move-
ments of Alligator in a tidal creek before
and after diversion of the freshwater head-
waters during construction of nuclear
power plant. Use by Alligator of the lower
reaches of the creek (mean salinity 20 ppt)
virtually ceased after the diversion.
Although animals were not marked, it
appeared that the population shifted to the
diversion canal where they once again had
access to fresh water. Thus although a rig-
orously controlled study remains to be con-
ducted, it seems quite likely that Alligator
is capable of behavioral osmoregulation. It
may use the highly productive coastal
marshes for feeding as long as it has access
to fresh water periodically. However the
limited physiological tolerance of Alligator
to high salinities makes this a risky under-
taking if brackish water of a low enough
salinity is not available. Chabreck (1971)
studied Alligator in nearly fresh (0.6 ppt)
and saline (3-16 ppt) marshes along the
Louisiana coast. Although the saline
marshes had twice as much potential food,
stomachs of animals in the freshwater
marsh contained more than six times as
much food as those from an adjacent saline
marsh. Chabreck (1971) suggested that
such a difference in food intake would result
in diminished growth in saline marshes. His
wild-caught animals did not in fact dem-
onstrate any such obvious effect on body



condition, I!.'. ...!: their previous move-
ments were not monitored.
Laboratory experiments by Mazzotti and
Dunson (1 : :; ..... that Alligator do
not gain mass in 35 ppt sea water even
when fed; they do gain mass in 4 ppt. The
ability of Alligator to survive for consider-
able periods in saline water seems to be due
to low rates of water loss and sodium .' .'
(Table 3). i . the inability of Alli-
gator to excrete excess sodium chloride lim-
its its stay in highly saline habitats. Lauren
(1985) found that ,': (mean mass 381
g) stopped -:.- at salinities of 10 ppt and
above, and lost considerable amounts of
mass. He measured plasma and urine
osmotic pressure at weekly intervals for
four weeks. In fresh water and 5 ppt (where
Sil: occurred but was low, about 0. /
day), urine and plasma concentrations did
not increase; at 10 ppt and above they did
increase 'c ... iy to levels above 100
mM sodium. It is not clear whether the
increased plasma osmotic pressure in Alli-
gator was due to net sodium chloride intake,
net water loss, or both (Lauren, 1985).
Plasma chloride rose : ... about 120 mM
in fresh water to above 180 mM ::: four
weeks at 20 ppt. Plasma sodium underwent
a lesser increase, from below 140 mM to
about 180 mM1. The faster rise in chloride
than sodium is .. i::: At 10 ppt and
above, plasma and urinary uric acid and
plasma corticosterone increased. The stress
of saline exposure at 10 ppt and above was
associated with cessation of feeding. Some
limited homeostatic . .. such as
increases in uric acid production and clo-
acal fluid sodium occur, but are presum-
ably insufficient to counter net water loss
and net sodium chloride uptake. The
S: ; to feed at 10 ppt, a salinity which
is only slightly hyper-osmotic to extracel-
lular fluids, may be related to ... .. . .' '.,
levels of salt uptake via incidental drinking
(Dunson, 1 : ). However the ability of
these'. animals (381 g) to cease feed-
ing and wait out their fate in saline solu-
tions ::, :.. a .. H: :. :: degree of phys-
iological tolerance to allow behavior to get
them away from potentially lethal *.....
ties. Hatchlings are assumed to be even less

tolerant; it would indeed be interesting to
examine the i. : of body size on salinity
tolerance across the entire size : ::. of the
species. An examination of possible inter-
population differences in tolerance would
also be useful to see if a tolerant ecotype
has arisen in coastal habitat (as in turtles,
see Dunson, 1 : ). Lauren (1'- ) sug-
gested that the integumental water per-
meability of ..... is too high to allow
much :i::. tolerance in the absence of a
salt gland. Ellis and Evans (1'.' ; similarly
concluded that integumental water per-
meability of Alligator is much higher than
that of C. aculus. In the absence of direct
measurements on C. acutus we are reluc-
tant to come to a conclusion. Instead we
point out that the mangrove snake lives
only in saline estuarine areas, can tolerate
long periods of immersion in 35 ppt sea
water, apparently lacks a salt gland, has a
higher mass-relative surface area, and yet
has about one-half the integumental water
S .... .i..i.:, of Alligator (Table 4). This
tends to corroborate the idea that the over-
all water permeability (especially buccal) of
'. .. is too high to allow long term sur-
vival at '.1 '. ::,

Crocodylus aculus
In terms of breeding ..L.* in Florida,
this species is completely estuarine. C. acu-
tus makes nests on islands, exposed shore-
lines, and creekbanks along the southern-
most mainland (Mazzotti, 1'' .- i 1 ... . .
adults may spend :.. .... .' periods in the
non-breeding season in totally fresh water
of interior wetlands. This predilection for
coastal habitats in many parts of its range,
and the presence of lingual salt --:r. in
subadults (Taplin el al., 1-' ;, : : : C.
acutus as one of the two crocodilians which
are estuarine ... (along with C. poro-
The earliest detailed study of.
ulation in C. acutus is that of Schmidt-Niel-
sen and Skadhauge (1967) who ..... !. '
examined the renal/cloacal system. They
found that the renal tubules had little abil-
ity to regulate the osmolality or electrolyte
composition of the urine. Even after a salt
load, the osmotic pressure of the urine

F. J. MAZZOTTI AND W. A. i:.. *

never exceeded that of the plasma.
Although amphibious, the crocodile
retained the ammono-uricotelic system of
nitrogenous excretion characteristic of its
terrestrial ancestors.
The first study of salinity tolerance of C.
aculus by Dunson (1970) pointed out dif-
ferences in the responses of large and small
animals. A 3.4 kg crocodile was kept in 35
ppt sea water for five months and contin-
ued to feed avidly on trout. Yet unfed
hatchlings lost an average of 1.' -. mass/
day and seemed quite intolerant of expo-
sure to sea water. Ellis (1981b) extended
these experiments with very similar results.
Relative mass loss of unfed animals was
inversely related to body mass (also con-
firmed by Mazzotti and Dunson, I'*
": :: C. aculus from high salinity
sites had plasma sodium concentrations
(141-174 '*. typical of reptiles generally.
Crocodiles .: . .! with sodium chloride
loads did not show any externally obvious
secretions from the : : Tap-
-. and Grigg(1981) and Taplin et a/. (1- .
subsequently discovered that the salt glands
are in the tongue and, as far as is known,
can only be stimulated by the injection of
the acetyhl- :- mimic methacholine
chloride. Avian, reptilian, and elasmo-
branch salt glands respond to salt injec-
tions as well as to methacholine chloride
(Peaker and : :: 1' ). The glands in
c . . ... are '" .. .. ..1. present and
their unique i ..i... to secrete after salt
loading may : : :.. our technical inabil-
ity to r' natural. :.:... leading
to gland secretion.
A series of investigators have examined
...... and water turnover in C. acutus in
:i::: water : '. modern r. I .. .
techniques. Evans and Ellis (1977) mea-
sured sodium efflux in ... 1. .1 ... in
S'. sea water (about 9 ppt). Sodium efflux
was quite low (5.6 Amoles/100 g-hr),
although these animals were not salt loaded
or dehydrated. However even in 35 ppt sea
water, sodium effluxes remained low (0.8,
2.3, 3.8 imoles/100 g-hr; Table 3). The
smallest salt gland known in snakes or tur-
tles secretes at a rate of about 15 moles/
100 g'hr (Cerberus; Dunson and Dunson,
1979). The rate of lingual gland secretion
directly from the tongue of C. aculus

(injected with methacholine chloride) is
13.5 moles/100g-hr ,.: !*: etal., 1 '..
'Ihus measurements of sodium efflux from
C. aculus in sea water to date do not support
the notion that a salt gland is f--. .-.--.:
under : .. dehydrating conditions.
Two measurements of sodium influx in sea
water were nearly identical (11.3, 11.5
Amoles/100 g-hr; 'able 3), and were
'-. -.--:- than efflux. This .. : that small
C. acutus undergo a sizeable net uptake of
sodium in sea water. This needs to be con-
firmed by actual measurements of body
sodium, and extended over a wide range
of body sizes. It seems clear that hatchling
C. arutus are quite intolerant of exposure
to 35 ppt, and that larger animals are much
less so. i .. .. is little chance that the intol-
erance of hatchlings is based on some tech-
nical flaw (i.e., small animals are more ner-
vous and simply do poorly in captivity),
since these baby crocodiles feed vora-
ciously and seem to adapt readily to exper-
imental protocols. The same may not be
true of C. porosus as very little work has
been done on :-... :' ', and they are said
to be difficult to rear in captivity.
Hatchling C. aculus :: : :: !i. do well
in natural situations of high .: ..... (Dun-
son, 1' Mazzotti et al., 1 .., P. Moler,
unpublished observations). Yet in labora-
tory simulations, growth cannot occur in
such highly saline conditions (Dunson,
1982; Mazzotti and Dunson, 1: ,. Instead
very small C. aculus are thought to depend
on periodic .. of rain water to rehy-
drate, if salinities are above 20 ppt. Such
a strategy was .-;.: --Led also for hatchling
C. porosus by .... ... (1 .,, and for
other estuarine reptiles (Dunson, 1970,
1 '' ) and a .... .:: ii:: I rat (Dun-
son and Lazell, 1982). Mazzotti and Dun-
son (1 '. suggested that the major adap-
tations of C. aculus to : : water also
include the ability to grow very i: .
during the wet season to a size much more
tolerant of the high salinities likely to be
encountered during the dry season.

Crocodylusjohnstoni and
Crocodylus niloticus
Lingual salt glands are .:
members of the ..i.: C
examined (Taplin el al., 1982,

.,: in all
1985). This


includes some entire species, and inland
populations of other species that never have
any contact with estuarine conditions. Cur-
rently the one documented function of salt
glands is extracloacal sodium chloride
excretion. Since freshwater crocodiles do
not have a need to excrete excess salt, the
origin and : i of the lingual salt glands
in freshwater crocodiles remain obscure at
present. C. johnstoni is primarily a fresh-
water form, but one unusual population
that lives in saline water is known from the
Limmen T.:;.. I system of northern
Australia. Taplin et al. (1 *-. ) compared
lingual gland secretions (stimulated by
methacholine chloride injection) and other
aspects of osmoregulation in fresh and
saline water populations. On the whole,
body salt and water levels did not differ
between the populations. Total body water
was the same; exchangeable body .i .
and ..' ... sodium and chloride concen-
trations were somewhat higher in the saline
group. There was a much more :111::.
difference in lingual gland excretion rates:
saline exposed animals secreted about eight
times as fast as those in fresh water (Table
5). This indicates that a crocodilian gen-
erally considered to be entirely of fresh-
water habits can adapt to saline environ-
ments, with a concomitant increase in the
secretary capacity of the '1:: :: salt glands.
It would be i. :':i interesting to deter-
mine whether this is a plastic response
brought on by ...... to saline waters or
whether the Limmen i'.. .: population is
an ecotype with a fixed, genetic, : I- --
mination to develop more active glands.
The occurrence of local estuarine popu-
lations in otherwise freshwater crocodiles
may also occur in other i :. such as C.
niloticus and C. palustris (Taplin et al., 1982;
Taplin and Loveridge, 1988).
U: -. i C. niloticus in sea water lost mass
at a rate of 1. /day (Taplin and Love-
ridge, 1988). Plasma sodium chloride, but
not potassium, was markedly elevated after
134 hr. Cloacal :1 i sodium concentration
increased only to about 10 mM. When
excretion of the lingual glands was blocked
by glue ... :. .1 to the tongue, the increase
in : :.: :1: body sodium was 67%
(over 92 hr), in contrast to a value of '"
in the controls. This seems to be clear evi-

dence that the lingual glands have an
excretory :..... :.. .. However over this
short period, homeostasis obviously was not
achieved by the control animals (which were
reared in fresh water and given an acute
exposure to sea water).
The extent to which species of primarily
freshwater crocodilians inhabit estuaries
obviously deserves further study. The phe-
nomenon may be more widespread than is
currently recognized. If so, this would have
an important bearing on theories of the
evolutionary history of the genus Croco-
dylus. The possession of lingual salt : 1 -...'
in inland c ... :,:: would be easier to
understand if there are coastal euryhaline
populations of the same species. Such evi-
dence would not, however, reveal which
population was ancestral to the other.

Crocodylus porosus
This species has been extensively ...
in northern .. :. i by G. G. L. Tap-
lin and their associates ('.i: : :: 1978;
Grigg etal., 1': 1' : Grigg, 1981; Tap-
lin and C. 1981; Taplin, 1982, 1 ,
b, c, 1'" I : i.. amount of infor-
mation amassed is :.: :' especially in
regard to data collected from animals under
field conditions. We will focus our atten-
tion on a recent paper that summarizes the
best available data on water and sodium
budgets (Taplin, 1985).
An idealized sodium and water budget
for a 250 g, unfed C. porosus in sea water
is presented in Table 6. Current evidence
indicates that C. porosus can remain in ;
itive *"' '': balance without .:.... or
drinking fresh or brackish water. Sodium
balance was not, however, achieved in all
flux ... ;'--... For example in one test
in sea water, influx exceeded efflux by 4
ymoles/100 g-hr; exchangeable sodium
increased by 1i (Tables 5, 6 in T.:.:::
1': '.;. In the most convincing experiment
(Table 2 in Taplin, 1'= '.), sodium efflux
-!ir-!-> exceeded influx and the exchange-
able sodium pool remained constant. Of
course, ::.. . in efflux rate may be
linked to effects of acclimation and the state
of body fluid hydration. An additional fac-
tor is the amount of: :: ii:: stress involved
in such i..... --- which may interfere
with normal maintenance of ion and water


TABLE 5. Secretory characteristics of salt glands of some marine and estuaine reptiles./


Acclimotion Na excretion,
conditions, Body pmolcs/
I abit.a ppt mass, g 100 g-hr

Turtles i : ... salt gland)
Malademi s terrapin E, S 0 469; 297 0.4 288 Dunson, 1970; Robinson
Maladremys terrapin E, S 35 216; 234 43 682 and Dunson, 1976
Mlaaclemys terrapin E, S In egg 8 65 Dunson, 1985
Chelonia nmdas M 35 68 134 685 Holmes and McBean,
Chelonia nodas M, S 35 24 590-950 644 Marshall, Cooper, and
Lizards (nasal salt Saddlier, unpublished
Varanus emimremax E, S D 150-294 34 686 Dunson, 1974
Amhlyrhy nchus M, S 35 80-428 146-255 <1,434 Dunson, 1969

Snakes (posterior sublingual salt : :.
Aipysurus fuscus M, S 35
HIldrophis elegan- M, S 35
Acalyptophis peronii M, S 35
Astrolia .iokesii M, S 35
AipySurus lamevi M, S 35
Lapomin hardwichii M, S 35
Aipysurus eydouxii M, S 35
Snake ,. .. . salt
Cerberus rhynchops M, E 35
Crocodilians ....... salt gland)
Crocod'lus acutu.' E 0
CrIcodylus johnstoni V 0
Crocodilus johnstoni E 10-2
Crocodylus porosus E > 12
Crocodylus porous E >12


Dunson and Dunson,
Dunson and Dunson,
Dunson and Dunson,
Dunson and Dunson,
)unson and Dunson,
Dunson and Dunson,
Dunson and Dunson,

74-221 16 414 Dunson and Dunson, 1979





Taplin et aL, 1982
Taplin el al., 1982
Taplin et al., 1985
Taplin 1985
Taplin 1985

I M = marine; E = estuarine; F
b Ci instead of Na.

= fresh water; S = salt loads; ) :

balance. Acclimation or i .....:..: effects
may also be involved in the large variation
in measured water fluxes. There is no doubt
that crocodiles in sea water suffer a severe
net water loss; however, the overall size of
the water fluxes varied between :
captured ..:: 1/100 g-hr) and captive-
held crocodiles (1,006 pl/100 g-hr) (Tap-
lin, 1 .
The partitioning of sodium and water
influx and efflux in C. porosus in sea water
(Table 6) has been established by flux mea-
surements of crocodiles with cloacal bags,
with the tongue surface sealed with glue,
with the mouth closed, and with a divided
chamber separating the water surrounding
the head and the body (Taplin, 1' ';
These in vivo methods have revealed a little
appreciated fact of crocodilian physiology,
that the oral (i : :i) epithelium is quite
permeable to sodium and water. Indeed

.* .: *. and water intake are primarily
through this cephalic route, with the gen-
eral :'. --: the second most impor-
tant route (Table 6). A : ... i proportion
of sodium ( :^.'. than water intake '-'
occurs in this fashion. The cloaca and
drinking are .. ., .:....... in uptake of
unfed crocodiles. In fed animals there
would :::. 1 : : .:ii be a 1:: -- amount of
"incidental drinking" as occurs in turtles
S.... . 1 . In contrast to uptake,
routes of loss of sodium and water are
. .. .. .. (Table 6). A majority ('- ; of
sodium loss occurs via the lingual salt
glands, although a surprising amount passes
across the cephalic epithelia (1 ; The
general integument and cloaca are insig-
nificant as routes of sodium loss. In con-
trast, water efflux primarily occurs across
the general: .:.. ::::- : : )and :
epithelia ).



For a hypothetical C. porosus of 250 g
mass unfed in sea water (Table 6), the lin-
gual glands contributed .'. (10.5 ,moles/
I .". g'hr) of the total sodium efflux. For
secretions collected directly from the
-.-... after .... ' injection, the
rates were considerably higher, and
inversely proportional to body mass iap-
lin, 1985). The rates of sodium secretion
for crocodiles weighing 100 g, --. .: g and
50 kg, respectively, are 74, 56, and 11
,emoles/100 g'hr. The reason for the
higher relative rate of secretion in smaller
crocodiles is unknown, but may be related
to allometric effects on rates of water loss
and other factors relating to the imposed
sodium load. This may be a general feature
of the physiology of marine reptiles because
relative salt gland mass (G as % wet body
mass) shows an inverse relation to body
mass (M in g) in the sea snake Hydrophis
ornaus (G = 0.1637M 1'470; Dunson, 1-'
The maximum rate of lingual gland secre-
tion stimulated by methacholine in a 250
g C. porosus is considerably higher than the
efflux rate measured by radiosodium loss
(56 vs. 10.5 imoles/100 g hr). The differ-
ence probably does not i. :... .. an excess
of secretary capacity because laboratory
studies of sodium influx and efflux in fed
crocodiles have not yet been performed. It
is quite likely that drinking of sea water
incidental to r 11::. will require a rate of
excretion considerably in excess of the
unfed efflux rate. In the marine iguana
(Amblyrhynchus), Shoemaker and Nagy
(1984) estimated that of sodium influx
came from ingestion of sea water inciden-
tal to : i...:. Dunson and Mazzotti (1': ';
reported a 97-fold increase in sodium influx
when diamondback terrapins (Malaclemys)
were fed in sea water. Crocodiles are likely
to undergo a lesser but still substantial
increase, because they swallow their food
whole instead of biting it into chunks as do
turtles (Dunson, 1985). Indeed Grigg et al.
(1 ': :.; have measured field sodium effluxes
in C. porosus that almost exactly match the
drug-stimulated maximal rate of .. :
gland secretion. This latter study of field
turnover of water and sodium in C. porosus
represents the second such study of any
aquatic reptile. The first field measure-

TABLE 6. An1 estimated sodium and water budget (unfed)
lor a 250 g Crocodylus porosus in 35 ppi sea water at

Route of change
Salt glands
Net change


15.3 (80%,)

3.6 (20-%)
18.9 (100%)

10.5 (55%)

8.0 (42%)

0.2 (1%)
0.4 (2%)
19.1 (100%)


96 (40%)
240 (100%)

17 (6%)

102 (36%)

153 (55%)
8 (3%)
280 (100%)

Fluxes are in units of Mmoles/100 g-hr (sodium)
or tIl/100 g-hr (water) with percentage of total ex-
change in parenthsees. Modified from Taplin (1984r,

ments of aquatic water fluxes were made
on the marine ; b--...- by Shoemaker and
Nagy (1 i; Despite major problems in
interpretation ofaphagia apparently caused
by I .:. i:: : stress, and :. : 0il or low
overall growth rates, the measurements of
fluxes in free-ranging crocodiles contrib-
ute to our ...... : .... ... of the magnitude
of naturally imposed salinity stress.
The greatest single unresolved issue in
the osmoregulation of C. porosus is the sta-
tus of newly hatched neonates. For C. acu-
tus this /size is the most intolerant of
high .: ... (as discussed above). Appar-
ently the only study to date on salinity tol-
erance of neonates of C. porosus is that of
'.' .. ....:: (1978). He found that recent
hatchlings undergo drastic mass loss after
8-12 days of exposure to salinities of 15
ppt or .. :. r-. Yet these neonates showed
no behavioral preference for different
salinities. He suggested that drinking of
low salinity water after rainfall ...: '.. be
crucial to survival of these young croco-


diles. Indeed there is a virtual absence of
very small C. porosus from : : : rivers
of the northern coast of Australia (Messel
et al., 1'' : 1). Grigg et al. (1-:-:: dem-
onstrated survival and growth of 6-10 small
C. porosus (160-270 g) in the Tomkinson
River of northern Australia. : :. ::- were
25-36 ppt; there was rainfall on only three
days and no seeps of fresh water were
found. Over periods of 16-17 days or 4
mo, growth occurred but was quite .
(mean of 0.44 g/day). Grigg et al. (1 : ::.
concluded that there was no evidence of
stress in this habitat. Considering the very
low rate of .: h of these animals com-
pared to other crocodilians (Mazzotti,
1983), this is not a convincing statement.
Indeed the fact that some rainfall did occur
leaves the possibility open that drinking of
brackish water occurred. Even more sig-
nificantly, recent :. ':1. (about 70-75
g) were not studied during the period of
early growth when they ..-:: : ir have
the least tolerance for exposure to saline
water. Mazzotti and Dunson (1" have
hypothesized that for neonatal C. aculus,
drinking of i .: .: i available rainwa-
ter and very rapid growth over a 2-4 mo
period to a more ..... -tolerant size are
important osmoregulatory strategies. Of
course ..: crocodiles also have to be
sufficiently impermeable to resist dehydra-
tion between episodic rainfall events. It is
clear that there is a strong need for rig-
orous tests of the effects of .i:..'. on
growth ofC. porosus :i.: with neonates
(newly . : and .- .--..- to larger
sizes. Ideally these experiments should be
carried out in the laboratory and in ..
enclosures to test for any i .:. of han-
: : stress. The value of such an approach
can be '. .. ... by the demonstration that
S .. diamondback t.. '' ': ..'.'-
,mys), which possess a salt gland .....: in
maximum secretary capacity to that of C.
porosus, have an optimal growth in water
of about 9 ppt (Dunson, 1 .. ). Growth can
occur at a lower rate in 35 ppt, if fresh or
brackish water is supplied periodically,
mimicking rainfall events. In Malaclemys,
the ability to grow at constantly i.' '. salin-
ities also increases rapidly with body size/


Densmore (1983) was the first to propose
that the current .- .: ...:..,.. of Crocodylus
is the result of a relatively recent (post-
i .. or about 4-5 million years) trans-
oceanic ...:.-: *' -- of a sea water-tolerant
form ancestral to all living species. Taplin
and Grigg's (1981) discovery i :.:: :: : salt
glands in C. porosus provided circumstan-
tial proof that such an event may have
occurred. Taplin (1'"I- and Taplin et al.
(1 .' have elaborated considerably on this
idea and :.: ... 1 further evidence by the
presence of .. : salt :... in all seven
species of Crocodylinae tested to date
(stimulated with methacholine chloride
injections). The presence of salt '.: .. in
many strictly freshwater species is para-
doxical and the transoceanic migration
hypothesis (T( : '.. ::: provides an appealing
solution. We would like to point out alter-
native hypotheses which have merit and
discuss ways in which further tests can be
conducted. We also will consider untested
...... inherent to the TOMH. Our
goal is not so much to attempt to disprove
or dispute the TOMH as it is to, :.. :...
the maintenance of an open mind on the
subject until alternatives are carefully con-
sidered. Evolutionary hypotheses should be
rigorously and experimentally evaluated
from all perspectives despite their super-
ficial attractiveness in reconciling disparate
One of the most important assumptions
of the TOMII is that conclusions about the
protein molecular clock must be valid
(Densmore, 1983). Since .i;. .. in
proteins .::: : species of Crocodylus are
quite small, the time since they diverged
must also be small. However '- -..:... of
such a "clock" depends on comparisons
with protein i::^ ... of other vertebrate
groups and their known fossil histories. The
rate of change in proteins might be slower
in crocodiles than in other vertebrates,
i .- : a conservative ::: : 1 (all
crocodiles are basically the same in exter-
nal structure). No slowing of the protein
clock appears to have occurred between
the !: :. ..-.. of the two species of Alli-
gator (Densmore, 1983). However A. mis-


sissippiensis has been .::ed to have a
lower than .. .: heterozygosity :: : 1-
0.022) and either a low (0.06) or normal
(0.15) proportion of polym(c: 1. loci
(Gartside et al., 1977; Adams et al., 1980).
This suggests a certain caution in inter-
pretation of allozyme data from crocodil-
ians. Another ... '.-1. that has yet to be
suggested is that the common ancestor of
modern Crocodylus species passed through
a genetic "bottleneck" (reduction in pop-
ulation size to a very low level) that greatly
reduced genetic diversity. Further exami-
nation of the :. i relatedness of Croc-
ody)lus -" by new techniques (such as
mitochondrial DNA) should help to resolve
these issues.
A second important assumption of the
TOMH is that the common ancestor of all
Crocodylus species was capable of transoce-
anic migration and : : i.... : on all
continents. While C. porosus is undoubtedly
highly salt-tolerant and has been found at
sea on occasion, it is more -r-:.;--* i;)
where it has not bee. i ::: :
example it breeds in the T.. i .. Islands, and
-. New Guinea, but not the eastern
Carolines such as in Ponape, or in Oceania.
Even ..... ... the highly aquatic sea snakes
(about - 1 which : 1:::. in the
western Pacific), only one species (Pelamis
platurus) has successfully colonized the New
World; this must have occurred only since
the Pliocene when the Central American
isthmus last emerged to block ... ::.. : ...
into the Atlantic Ocean. Thus movement
across the immense distances of the eastern
Pacific -. . is not easy, even for .. .. :
sea snakes. The hypothesized ancestral sea-
-.---,-- crocodile would have an even
greater problem in that it must have been
oviparous and there are apparently no
islands suitable as breeding way-stations in
Oceania. If one attempts to : .. what
a true sea-going .. 'ii:. might be like,
there are some ready examples in the ; :i
record (Williston, 1914: Neill, 1971). The
Teleosauridae are considered to be coastal
crocodiles, whereas the Metriorhynchidae
were i '... modified pelagic forms. Nei-
ther of these families are considered
directly ancestral to Crocodylus but they
illustrate the potentialities of the crocodil-

ian body form for evolution towards .. 1 .. :
life. Geosaurus, a metriorhynchid from the
upper Jurassic, was a fish eater (slender
snout) with paddle-like limbs and an elon-
gate tail with a fin on the end. Such a highly
modified animal must have been totally
:.:...:'. except at. ing time. It must
also have had a 1.: ..... .... .. mecha-
nism for extracloacal salt excretion. This
type of creature would have been able to
cross the open ocean easily, in comparison
with modern crocodiles. It might be con-
sidered too specialized morphologically to
revert to the general crocodilian form.
Howeever a marine form less I : :
than Geosaurus but more .. .. 1. than
C. porosus may have done so and been the
ancestor of modern Crocodylus.
A major alternative explanation of the
presence of lingual salt glands in fresh-
water crocodiles is that the !. .1 are
needed during seasonal droughts (Taplin
et al., 1982). Very little is known about ter-
restrial dormancy or aestivation in croco-
diles . that it may be an .....
S.. 'in many I such as C. johnstoni.
Thus one can postulate that 1::. .' salt
glands evolved first as a means of excreting
sodium chloride and .... . water in
terrestrially c 1. 1..: .. conditions.
Secretion rates and numbers of ..
tubule openings are lower in the fresh-
water than in estuarine forms (Taplin et
al., 1982, 1985). A : .. selection for
1 : '. : secretion rates could have occurred
in estuarine populations. This scenario is
very similar to that .. ..... for the evo-
lution of salt glands in the marine '
i: ... terrestrial ancestors with salt glands
(Dunson and Mazzotti, 1989). It is also
analogous to the estuarine evolutionary
. "' ....... for the origin of salt
glands in snakes and turtles (Dunson and
Mazzotti, 1:-.:. There is a very obvious
way of ..:- the "dehydration-aestiva-
tion hypothesis" for the ... '. of crocodile
salt glands. One : : 1 needs to studycom-
pletely freshwater c( .... ... to deter-
mine what use if any is made of the glands
under dehydrating conditions. If the glands
are functional in ... ;. .
i1. :.-..: this alternative hypothesis gains
credence. If the -.1 :...i. tobedegen-


rate or :I.1- with no apparent func-
tion, then the TOMH gains additional sup-
Another .. . way to approach this
issue is to consider patterns of evolution in
other aquatic vertebrates. The TOMH
: :;. argues that crocodilians are sec-
ondarily .. : . species. In classic zoo-
geographic terms, such a secondary form
would be chiefly found in fresh water, but
would retain some salt tolerance (Darling-
ton, 1957). Killfish of the genus Fundulus
illustrate a fairly i .: .' pattern of this sort
(Griffith, 1974). There are 27 species in
mainland North America of which eight
are completely '.. i. .:. ., ten are brack-
ish, eight are principally freshwater but also
occur in coastal brackish areas, and one is
an inland saline lake form. The primarily
estuarine species are tolerant of salinities
in excess of 70 ppt. The inland species are
less tolerant (generally with lethal levels
below 29 ppt), but much more tolerant than
typical freshwater fish. In some ways this
resembles the situation with crocodiles, in
that the freshwater forms have a reduced
S'. capacity to excrete salts. In
other ways, especially the presence of more
S. .. than estuarine forms and the
generally low level of pelagic 1 .:: in
C. porosus and C. aculus, these two examples
are not analogous. Thus this comparison
could favor the hypothesis that the evo-
lution of lingual salt glands occurred in the
amphibious-freshwater species, not in the
estuarine-marine forms.
In conclusion we propose that, until new
data are obtained, there is at present no
clear preponderance of evidence to decide
between the alternative hypotheses pre-
sented above. We believe that a rigorous
experimental examination of all possible
ideas is necessary to realistically appraise
the available ..... and to search for
other .1 ... :.

Financial support was provided by Flor-
ida Power and Light Co. (F.J.M.) and by
NSF grant BSR-8212623 (W.A.D.). This is
:- Series No. '"- i of the Institute of
Food and Agricultural Sciences, Univer-
sity of Florida, Gainesville, Florida.

Adams, S. E., M. H. Smith, and R. Baccus. 1980.
Biochemical variation in the American .
Herpetologica 36:289-296.
Bentley, P. J. and K. Schmidt-Nielsen. 1965. Per-
meability to water and sodium of the crocodilian,
Caiman slerops. J. Cell. Comp. Physiol. 66:303-
Birkhead, W. S. and C. R. Bennett. 1981. Observa-
tions of a small population of cstuarine-inhabit-
ing .: .. . near' i ;. North Carolina.
Brimleyana 6:111-117.
Chabreck, R. H. 1971. The foods and feeding habits
of from fresh and saline environments
in Louisiana. Proc. 25th Ann. Conf. SE Assoc.
Game Fish Comm. Pp. 117-124.
Coulson, R. A. and T. Hernandez. 1983.
metabolism. Studies on chemical reactions in vivo.
Conp. Biochem. Physiol. 74B:1-182.
Darlington, P. J., Jr. 1957. The geo-
graphical distribution of ani & Soils,
New York.
Davis, J. E., J. R. Spotila, and W. C. Schefler. 1980.
w ater loss from the American alli-
gator, .' mi sissippiensis: The relative
importance of respiratory and cutaneous com-
ponents and the regulatory role of the skin. Comp.
Biochem. Physiol. 67A:439-446.
Densmore. L. D., Ill. 1983. Biochemical and immu-
nological systematics of the order Crocodilia. In
M, Hecht, B. Wallace, and G. Prance (eds.), Ez,o-
lutionar, biology, Vol. 16, pp. 397-465. Plenum
Publ. Corp., New York.
Dessauer, H. C. 1970. Blood chemistry of .
: : evoluitioary aspects. In C. Gans
and T. S. Parsons (eds.), -.' of the Rep/iiia,
Morphology C, Vol. 3, pp. 1-72. Academic Press,
New York.
Dill, D. B. and H. T. Edwards. 1931. Physicochemical
properties of crocodile blood (Crocodils aculus,
Cuvier). J. Biol. Chem. 90:515-530.
Dunson, W. A. 1967. Sodium fluxes in fresh-water
turtles. J. Exp. Zool. 165:171-182.
Dunson, W. A. 1969. Electrolyte excretion by the
salt gland of the Galapagos marine iguana, Amer.
J. Physiol. 216:995-1002.
Dunson, W. A. 1970. 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.
Dunson, W. A. 1974. Salt gland secretion in a man-
grove monitor lizard. Comp. Biochem. Physiol.
47(4A): 1245-1255.
Dunson, W. A. 1978. Role of the skin in sodium and
water exchange of aquatic snakes placed in sea
water. Amer. J. Physiol. 235:RR151-R 159.
Dunson, W. A. 1979. Control mechanisms in rep-
tiles. In R. Gilles (ed.), 1Mechanisms o osmoregula-
tion in animal/, pp. 273-322. Wiley Interscience,
New York.
Dunson, W. A. 1982. Salinity relations of crocodiles
in Florida Bay. Copeia 1982(2):374-385.
Dunson, W. A. 1985. Effects of water salinity and
food salt content on growth and sodium efflux of


: ... .. diamondback terrapins (Malaciicmys).
Physiol. Zool. 58(6):736-747.
Dunson, W. A. 1986. Estuarine populations of the
snapping turtle (Chelydra) as a model for the evo-
lution of marine adaptations in reptiles. Copeia
Dunson. W. A. and M. K. Dunson. 1974. Interspe-
cific differences in fluid concentration and secre-
tion rate of sea snake salt glands. Amer. J. Phys-
iol. 227:430-438.
Dunson, W. A. and M. K. Dunson. 1979. A possible
new salt gland in a marine homalopsid snake (Cer-
berus rhynchops). Copeia 1979(4):661-672.
Dunson, W. A. and J. Freda. 1985. Water perme-
ability of the skin of the amphibious snake Agkis-
trodon piscivorus. J. Herpetol. 19(1):661-672.
Dunson, W. A. and H. Heatwole. 1986. The effect
of relative shell size in turtles on water and elec-
trolyte composition. Amer. J. Physiol. ** :
Int. ... Physiol. 19):R1133-R1137.
Dunson, W. A. and J. D Lazell, Jr. 1982. Urinary
concentrating capacity of Rattus ratius and other
mammals from the Lower Florida Keys. Comp.
Biochem. Physiol. 71 A: 17-21.
Dunson, W. A.and F.J. Mazzotti. 1988. Some aspects
of water and sodium exchange of freshwater
crocodiles in fresh water and sea water: Role of
the integument. Comp. Biochem. Physiol. 90A:
Dunson, W. A. and F. Mazzotti. 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. Bull. Mar. Sci. 44:229-
Duns0on, W. A. and G. D. Stokes. 1983. Asymmet-
rical diffusion of sodium and water through the
skin of sea snakes. Physiol. Zool. 56:106-111.
Ellis, T. M. 1981a. - - .. in the American
: .... Masters Thesis, Univ. of Miami, Coral
Gables, Florida.
Ellis, TI. M. 1981b. Tolerance of sea water by the
American crocodile, Crocodylus aculus. J. Herpe-
tol. 15(2):187-192.
Ellis, T. M. and D. H. Evans. 1984. Sodium balance
in the American alligator. J. Exp. Zool. 231:325-
Evans, D. H. and T. M. Ellis. 1977. Sodium balance
in the hatchling American crocodile, Crocoidus
aritus. Comp. Biochem. Physiol. 58A:159-162.
Gartside, D. F., H. C. Dessauer, and T.Joanen. 1977.
Genic homozygosity in an ancient : : (Alli-
gator missisippiensis). Biochem. Genetics 15:655-
Griffith, R. W. 1974. Environment and salinity tol-
erance in the genus Fundulus. .. 1974(2):
Grigg, G. C. 1981. Plasma homeostasis and cloacal
urine composition in Crocodylus porosus caught
along a salinity gradient. J. Comp. Physiol. 144:
Grigg, G. C., L. E. Taplin, B. Green, and P. Harlow.
1986. Sodium and water fluxes in free-living
Crocodylu poroCus in marine and brackish condi-
tions. Physiol. Zool. 59(2):240-253.
G. C., L. E. Taplin, P. Harlow, and J. '

1980. Survival and ,. :, of :.,,. :...:. Croco-
dyluis porosus in salt water without access to fresh
drinking water. Oecologia 47:264-266.
Holmes, W. N. and R. .. McBean. 1964. Some aspects
of electrolyte excretion in the green turtle, Che-
lonia mydas mydas. J. Exp. Biol. 41:81-90.
Jacobsen, T. 1983. Crocodilians and islands: Status
of the American .:i ... .. the American croc-
odile in the Lower Florida Keys. Fla. Field Nat-
uralist 11(1):1-24.
Lauren,D.J. 1985. The effect of chronic saline expo-
sure on the electrolyte balance, nitrogen metab-
olism, and corticosterone titer in the American
alligator, missssiippienss. Comp. Bio-
chem. Physiol. 81A(2):217-223.
Ljungman, 1. N. and W. A. Dunson. 1983. Integ-
umentary water and sodium :.. .. ', of the
yellow anaconda, Eunectes motaeuo. Comp. Bio-
chem. Physiol. 76A(1):51-53.
Magnusson, W. E. 1978. Nesting ecology of Croco-
dyius porosus Schneider, in Arnhem I.and, Aus-
tralia. Ph.D. Diss., Univ. of Sydney, Sydney, Aus-
Mazzotti, F.J. 1983. The ecology of Crocodyus aculus
in Florida. Ph.D. Diss., Pennsylvania State Univ.,
University Park, Pennsylvania.
Mazzotti, F. J., B. Bohnsack, M. P. McMahon andJ.
R. Wilcox. 1986. Field and laboratory obser-
vations on the effects of high temperature and
salinity on hatchling Crocodylus aculus. '
logica 42(2):191-196.
Mazzotti, F, J. and W. A. Dunson. 1984. Adaptations
of Crocodylus acutus and for life in saline
water. Comp. Biochem. Physiol. 79A:641-646.
Messel, H., A. G. Wells, W. J. Green, et al. 1979-81.
Surneys o tidal river systems in the Northern Territory
of Australia and their crocodile populations. Series
of 17 monographs. '" .. Press, Sydney,
Minnich,J. E. 1972. Excretion of rate salts by rep-
tiles. Comp. Biochem. Physiol. 41A:535-549.
Minnich, J. E. 1979. Reptiles. In G. M. 0. Maloiy
(ed.), Comparative physiology oI osmoregulation in
animals, Vol. 1, Chap. 7, pp. 391-641. Academic
Press, New York.
Neill, W. T. 1971. The last of the ruling reptiles. Alli-
gators, crocodiles and their kin. Columbia University
Press, New York.
Peaker, M. andJ. L. Linzell. 1975. lands, in birds
and reptiles. Monographs of the Physiol. Soc. No.
32. Cambridge University Press, Cambridge,
Robinson. G. D. and W. A. Dunson. 1976. Water
and sodium balance in the estuarine diamond-
back terrapin (Malaclemys).J. Comp. Physiol. 105:
Schmidt-Nielsen, B. and E. 1967. Func-
tion of the excretory system of the crocodile
(Crocodylus aculus). Amer. J. Physiol. 212:973-
Scholander, P. F. 1955. Evolution of climatic adap-
tation in homeotherms. Evolution 9:15-16.
Shoemaker, V. H. and K. A. Nagy. 1984. ...
elation in the Galapagos marine iguana, Amhly-
rhync/hus cristaits Physiol. Zool. 57(3):291-300.


S E. 1977. Excretion in lower vertebrates:
Function of gut, cloaca, and bladder in .... ...
the composition of urine. Fed. Proc. 36(11):2487-
Stokes, G. D. and W. A. Dunson. 1982. Permeability
and channel structure of reptilian skin. Amer. J.
Physiol. 242:F681-F689.
Tamarack, J. L. 1988. Georgia's coastal island alli-
gators, variations in habitat and prey availability.
Proc. 1.1.C.N. Crocodile Specialist Group, Quito,
Ecuador. 1986. (In press)
Taplin, .. E. 1982. . .. in the estuarine
crocodile. Ph.D. Diss., IUniv. of Sydney, Sydney
Taplin, L. E. 1984a. Evolution and zoogeography of
crocodilians: A new look at an ancient order. In
M. Archer and G. ... ': eographlaiid
-,ivoution in Australasiai-animals in space and time,
pp. 361-370. Htesperian Press. Perth, Australia.
Taplin, I. E. 1984b. Homeostasis of plasma electro-
lytes, sodium and water pools in the estuarine
crocodile, Crocodyl'u porosus, from fresh, saline
and hypersaline waters. Oecologia 63:63-70.
Taplin, L. E. 1984c. ;i ...: of fresh water but not
seawater by the estuarine crocodile, Cro(odylus
porous. Comp. Biochem. Physiol. 77A:763-767.
Taplin, L. E. 1985. Sodium and water budgets of the
fasted estuarine crocodile, Crocodylus porosis, in
sea water. J. ..... Physiol. 155B:501-513.
Taplin, L. E. 1988. Osmoregulation in crocodilians.
Biol. Rev. 03:333-377.

Taplin, L. E. and G. C. - 1981. Salt glands in
the tongue of the estuarine crocodile. Science
Taplin, L. E., G. C. Grigg, and L. Beard. 1985. Salt
gland function in fresh water crocodiles: Evi-
dence for a marine phase in eusuchian evolution?
In G. C. Grigg, R. Shine, and H. Ehmann (eds.),
S of Australasian frogs and reptiles, pp. 403-
410. Surrey Beatly and Sons, Sydney, Australia.
Taplin, L. E., G. C. Grigg, P. Harlow, T. M. Ellis.
and W. A. Dunson. 1982. Lingual salt glands in
Crtocoiylu, ac(uluv and C. johnitoni, and their absence
from e mississippiensis and Caiman crocodil-
us. J. .... Physiol. 149:43-47.
Taplin, I.. E. and J. P. Loveridge. 1988. Nile croc-
odiles, Crocodylus niloticus and estuarine croco-
diles, Crocod)lus porosus show similar osmoregu-
latory responses on exposure to seawater. Comp.
Biochem. Physiol. 89A:443-448.
Tercafs, R. R. and E. Schofleniels. 1965. Pheno-
menes de permeabilite au niveau de la peau des
reptiles. Ann. Roy. Soc. Zool. Belg. Brux. 96:9-
Thorson, T. B. 1968. Body fluid partitioning in Rep-
tilia. Copeia 1968(3):592-601.
Williston, S. W. 1914. Water reptiles of the pfas and
pre,int. University of Press, Chicago, Illi-

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs