Estivation in the sirenid salamanders, Siren lacertina (Linnaeus) and Pseudobranchus striatus (Le Conte)


Material Information

Estivation in the sirenid salamanders, Siren lacertina (Linnaeus) and Pseudobranchus striatus (Le Conte)
Physical Description:
v, 67 leaves : ill. ; 28 cm.
Etheridge, Kay, 1954-
Publication Date:


Subjects / Keywords:
Dormancy (Biology)   ( lcsh )
Sirenidae   ( lcsh )
Salamanders   ( lcsh )
Zoology thesis Ph. D
Dissertations, Academic -- Zoology -- UF
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1986.
Bibliography: leaves 62-66.
Statement of Responsibility:
by Kay Etheridge.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000926069
oclc - 15997696
notis - AEN6729
sobekcm - AA00004857_00001
System ID:

Full Text








I would like to thank several people for their invaluable

assistance during this study. Marty Crump, my major advisor, has been

a great help throughout my studies, and her encouragement and common

sense advice have helped me over many obstacles. John Anderson, David

Evans, Harvey Lillywhite, Frank Nordlie and Michele Wheatly not only

gave me guidance and technical assistance, but also trusted me with

their equipment. Archie Carr and Walter Judd have advised me during

this study and have reviewed this dissertation. Mark Seyfried helped

me comprehend soil physics, and Lou Guillette introduced me to the art

of histology. Chip Oglesby, Allison Rogers and Sam Ward, who

volunteered their time through the undergraduate research assistant

program, were a tremendous help during one of the most labor intensive

portions of this study. Don Banknight helped supply many of the

salamanders used in this study.

I am grateful Wo many of the faculty, staff and graduate students

of the Department of Zoology for their assistance in innumerable

facets of this study. Financial support for this work was from the

Department of Zoology and a Sigma Xi Grant-in-Aid of Research.

Finally, I wish to thank my family and friends for their support

during my studies.




ABSTRACT... ....... ..................... ............ .. ............... i.


I INTRODUCTION..............................................

Background................................................ 1
Study Animals and Habitat..................................3
Objectives ............................................... 5

II THE ENERGETICS OF ESTIVATION..............................9

Introduction........................... ...................9
Materials and Methods...................................10
Results. ....... ... o...... ................ ............... 13

III WATER BALANCE DURING ESTIVATION .........................33

Introduction ............................................. 33
Materials and Methods ................................... 37
Results.................................................. 40
Discussion.................................. ............ 47

IV SUMMARY AND CONCLUSIONS................................. 56

LITERATURE CITED ............... ..... .. .... ......... ...... ........62

BIOGRAPHICAL SKETCH.................. ,............................ 67

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy




December, 1986

Chairman: Martha L. Crump
Major Department: Zoology

Siren lacertina and Pseudobranchus striatus are aquatic

salamanders that inhabit drought-prone waters in the southeastern

United States. When drought occurs, sirenids burrow into the

substratum and remain inactive until the habitat is reflooded. This

study investigated adaptations that allow these two species to survive


Siren lacertina and P. striatus were induced to estivate in the

laboratory and several parameters were measured: rates of metabolism,

blood lactate levels, ventilation rates, rates of water exchange with

soil, urinary bladder volumes and solute concentrations of plasma and

urine. In addition, lipids stores were measured in animals from

aquatic populations, and the cocoon of an estivating S. lacertina was

examined by light microscopy.

In both species resting metabolic rates are 60 to 70% lower in

estivators than in aquatic animals. At these low metabolic rates

adult S. lacertina can survive two to three years without food.

Smaller S. lacertina and P. striatus have higher mass-specific

metabolic rates during estivation, but even a 1 g sirenid can survive

for several months on stored energy. Ventilation rates decrease

during estivation, but low blood lactate levels indicate that

respiration is aerobic.

The cocoon is composed of dead squamous epithelial cells. This

layer does not block water movement and estivating sirenids exchange

water with the soil. The rate of water exchange with the soil

increases with a decrease in body size. Rate and direction of water

exchange are also related to the water potential of the soil.

Estivators maintain a positive water balance in soil with water

potential greater than -5 bars, but begin to lose water in drier soil.

Siren lacertina store dilute urine in their bladder; the mass of

this urine may be the equivalent of as much as 10% of their body

mass. Plasma concentrations of sodium, potassium and chloride are

similar in aquatic animals and estivators; however, plasma urea

concentrations are six times higher in estivators. Due to urea

accumulation, plasma osmolality is higher in estivators than aquatic

animals. This lowers the water potential of estivators and decreases

the rate at which they will lose water in very dry soil.


Amphibians and reptiles can be described as low energy systems in

comparison to birds and mammals, which could be called high energy

systems (Pough 1983). This comparison relates to the fact that lower

vertebrates are primarily ectothermic and expend little energy for

maintenance of body temperature. Therefore, reptiles and amphibians

have low rates of energy intake and expenditure relative to birds and

mammals, which fuel their high metabolic rates with high rates of

energy intake. Differences in rates of energy flow have several

consequences, one of which is the frequency at which an animal must

"refuel." High energy systems require a constant source of food; low

energy systems can rely on a more variable supply. Under adverse

conditions most ectotherms can reduce their energy costs and remain

inactive for prolonged periods until conditions improve. Some birds

and mammals can lower their energy demand for short periods (torpor)

or seasonally (hibernation), but they cannot begin to approach the

length of time that ectotherms can remain inactive. The ability of

ectotherms to remain dormant for long periods allows them

opportunities to exploit transient or ephemeral resources that would

not support a high energy endotherm.

Energy may not be the only limiting resource, however, and lack

of water can restrict the use of some environments. Of all the


tetrapod vertebrates, amphibians appear to be the most closely tied to

moist or aquatic habitats. Most amphibians have an aquatic larval

stage and skin that offers little resistance to water loss; hence,

they seem to be ill-adapted for survival in a dry environment.

Nevertheless, several species of anurans and a few salamanders do live

in xeric habitats; some examples are Ambystoma tigrinum (Delson and

Whitford 1973), Scaphiopus multiplicatus and S. couchii (Ruibal et al.

1969), Heleioporus eyreia (Bentley et al. 1958), Limnodynastes

spenceri, Neobatrachus pictus and several species of Cyclorana (Lee

and Mercer 1967). Typically, these animals are terrestrial, nocturnal

species that spend most of their adult lives in underground burrows,

surfacing to feed and breed for relatively brief periods when water is

available. In addition to behavioral adaptations these amphibians

have in common several physiological adaptations that permit this type

of life history. As is the case with other ectotherms, their

metabolic costs are low, and they can survive long periods of

estivation on stored energy (e.g., van Beurden 1980). This type of

strategy has been most thoroughly studied for several species of

Scaphiopus, the spadefoot toads (Ruibal et al. 1969; Seymour 1973c;

McClanahan et al. 1976). Most studies of amphibian estivation have

focused on terrestrial or semi-aquatic anurans. However, one aquatic

anuran, Xenopus laevis (Balinsky et al. 1967), and one family of

aquatic salamanders, the Sirenidae, are known to estivate when their

habitats dry up. Little is known concerning the effects of drought

upon aquatic amphibians. This study focused on the adaptations that

enable sirenid salamanders to survive prolonged drought.

Study Animals and Habitat

All three species in the family Sirenidae are indigenous to

shallow water habitats in the southeastern United States. The two

species investigated in this study were the greater siren, Siren

lacertina Linnaeus, which ranges from the District of Columbia south

to the tip of Florida, and the dwarf siren, Pseudobranchus striatus Le

Conte, which is found from southern South Carolina to the eastern half

of the Florida panhandle, and south to the tip of Florida (Conant

1975). The third species in the family, which is not included in this

study, is the lesser siren, S. intermedia Le Conte. These species are

similar in habitat and morphology, but are dissimilar in size; P.

striatus rarely exceeds 25 cm in length, whereas S. lacertina may

reach almost 1 m.

Little is known about the life history of these secretive,

nocturnal salamanders. Reproduction occurs in late winter or early

spring and females deposit eggs singly or in pairs on submerged

aquatic vegetation (Goin 1941; Davis and Knapp 1953; Ultsch 1973).

Siren ingest large amounts of plant material along with prey items

such as crustaceans, mollusks and insect larvae, but it is not known

whether or not Siren digest cellulose (Davis and Knapp 1953; Burch and

Wood 1955; Ultsch 1973; Hanlin 1978). Plant material has not been

reported in the gut of P. striatus, but dwarf sirens do eat small

crustaceans and insect larvae (Freeman 1967). Sirenids may be among

some of the most productive salamanders; Gehlbach and Kennedy (1978)

found that S. intermedia was the dominant vertebrate species in a

Texas beaver pond, due in part to its great fecundity (151-226

eggs/female'year) and rapid growth to sexual maturity in one year.

This study was conducted from 1981 through 1986 in the vicinity

of Gainesville, Florida, where all three species occur sympatrically

in an array of aquatic habitats that vary in degree of permanency.

For example, all three species have been found both in the River Styx,

a permanent body of water, and in drought-prone ponds on Paynes

Prairie, Alachua County, Florida (A. Carr, pers. comm.). Typical

sirenid habitat in this area is characterized by shallow, soft-bottom,

weed-choked waters. One of the most common aquatic plants is the

introduced water hyacinth, Eichhornia crassipes, and sirenids are

often found in close association with the root systems of these

abundant floating plants. This type of habitat is uninhabitable for

many aquatic vertebrates because weed-choked waters are usually

hypoxic and hypercapnic (Ultsch 1976a), and are often transient

resources. Shallow aquatic habitats periodically dry out or change

with succession to a terrestrial environment, and survival of the

inhabitants is dependent upon an ability to withstand dehydrating

conditions, or an ability to move to new suitable habitat.

No evidence has been reported to suggest that sirenids migrate

overland in times of drought to seek new aquatic habitats. Instead,

they estivate during dry conditions by burrowing into the soil and

remaining inactive until the habitat is reflooded (Cockrum 1941;

Freeman 1958; Reno et al. 1972; Gehlbach et al. 1973). In the area of


this study drought can occur at any time of the year. Over a 70 year

period the summer months had the highest mean rainfall; however, in

1983 and 1984 there were summer droughts (Fig. 1-1). During this

study drought also occurred in winter and spring. Although the term

estivation was originally defined as summer inactivity (from the Latin

for summer, aestas), it also has traditional associations with

inactivity due to drought and is used in that sense here. Although

the ability to estivate permits these aquatic salamanders to colonize

potentially transient bodies of water, little is known about the

physiological mechanisms involved in estivation.

This study centers on two species of sirenid salamanders (Siren

lacertina and Pseudobranchus striatus) that serve as excellent models

for the study of estivation in amphibians; both species are aquatic,

but live in habitats susceptible to drought. Aquatic amphibians are

subjected to several stresses as their habitats dry up, including

dehydration, starvation and the accumulation of nitrogenous wastes

that are normally excreted in water. The objectives of this research

were to examine the adaptations used by sirenids to cope with these

stresses and to compare the abilities of these two species to survive

drought. These comparisons are important primarily because of the

size differences between these two species. Individuals of both

species were induced to estivate in the laboratory and, using a range

of sizes within each species, two major areas were investigated.


0 C


V)- C
S "M
*d 0 o

a) *r-4-

0'- *o

o- *r-

n -0

LL. 4-)
9- --

> *J *
C 0>

S4- 0.

.1 C 44n

4- a

0t 4. jQ

I-I- --I- -1-I II__.--I- I-

I I I I I I I I i I I

( o) "T I V NIVU :








Energetic Costs of Estivation

By measuring oxygen consumption during estivation, energy

expenditure was calculated and then compared to rates of energy

consumption in nonestivators. In addition, seasonal fluctuations in

energy stores available for use during estivation were measured by

extracting lipids from animals collected bimonthly throughout the


Water Balance During Estivation

Rates of water loss and gain in soil were measured to determine

the effect of body size and soil water content on these rates. Plasma

and urine samples were analyzed to examine the effects of estivation

on ion concentrations and nitrogenous wastes.

Data from this study were used to determine adaptations used by

sirenids to survive drought and also allowed estimation of the length

of time different size classes can survive without standing water.

Finally, the responses of these aquatic animals to drought were

compared to those of terrestrial amphibians that estivate.



Because estivating animals are inactive and do not feed, they

must rely on stored energy for an unpredictable amount of time.

Potential survival time is dependent in part upon the quantity of

stored energy available to the estivator at the onset of estivation

and the rate at which this store is depleted. In amphibians, as in

most animals, energy stored as lipid tends to vary with body size, sex

and time of year (Rose 1967; Brenner 1969, Byrne and White 1975;

Fitzpatrick 1976; Morton 1981). Resting metabolic rates in amphibians

also vary with several factors, particularly temperature and body

size. Furthermore, in estivating lungfish (Delaney et al. 1974) and

anurans (Seymour 1973c; Loveridge and Withers 1981; van Beurden 1980;

McClanahan et al. 1983) resting metabolic rates are 50-70% lower than

rates of nonestivators at the same temperatures.

In this part of the study the energetic of sirenid estivation

were investigated. Lipid content was measured in adult Siren

lacertina throughout a year to determine whether fat stores vary

seasonally, for if they do the timing of drought could affect survival

time. Metabolic rates were measured to estimate the potential amount

of time sirenids could survive before depleting stored energy and to

determine if metabolic rates could be reduced below resting levels.

In addition, the effect of body size on potential survival time was

studied by comparing the metabolic rates of adult and juvenile S.

lacertina and Pseudobranchus striatus over a range of body sizes.

Materials and Methods
Metabolic Rates

Siren lacertina and Pseudobranchus striatus were collected from

ponds, rivers and lakes around Gainesville, Florida. Large S.

lacertina were caught in a dipnet or on a hook and line using beef

heart as bait. Smaller animals were caught using a hand-held

dredge. Animals that were used for measurements of oxygen consumption

were held in water-filled aquaria for two weeks before estivation was

induced and were not fed during this period or during the experiment.

Both S. lacertina and P. striatus were induced to estivate in the

laboratory at various times throughout the year. Animals were weighed

and placed individually into containers that held soil saturated with

water. Container size and soil depth varied with animal size;

containers were large enough to allow the salamanders to extend to

their full length, and soil depth was approximately four times the

body diameter of the animal. After animals burrowed into the soil,

the soil was allowed to dry slowly. Water was added as necessary to

maintain soil water content between 6 and 12% by mass. Estivating

animals were kept at 23'C, the mean annual soil temperature at ]0 cm

depth in the area of Gainesville, Florida (based on data from the

University of Florida Agronomy Department).

Rates of oxygen consumption (V, ) were measured at one- to two-

week intervals early in estivation and at one-month intervals after 50

days in estivation. For each point during estivation, sirenids

representing a range of body size were carefully excavated from their

burrows and weighed. Metabolic measurements were made for 53

individuals, ranging in mass from 0.9 g to 1061 g. Oxygen consumption

of animals smaller than 20 g was measured using a Gilson respirometer

and standard stoichiometric techniques. Oxygen consumption of larger

animals was measured using a flow-through system. Each animal was

placed in a plastic tube slightly larger than its body dimensions.

This tube was sealed except for incurrent and excurrent air ports at

opposite ends. A pump was used to draw air sequentially through the

tube, a column of soda lime to remove C02, a column of silica gel to

remove water and an Applied Electrochemistry Oxygen Analyser Model

S-3A, to measure the fraction of oxygen in the air sample. Rates of

oxygen consumption were calculated using equations given by Withers

(1977) and were corrected to standard temperature and pressure

(STPD). All measurements were made during the animals' inactivity

periods, between 0800 and 1800 h. Handling of animals in an

estivating state had negligible effect. After the animal had spent

one to two hours in the metabolic chamber the rate of oxygen

consumption was stable enough to be measured. Rates of oxygen

consumption were measured at 23*C for all estivating sirenids. To

estimate QIO for estivators, rates of oxygen consumption were measured

at 23'C and 32C for seven S. lacertina (404 to 733 g) that had been

in estivation for more than four months.

Blood lactate levels were measured for estivating S. lacertina to

determine if anaerobic metabolism was an important component of the

energy budget of these animals. Eight adult salamanders that had been

estivating for seven or more months were excavated and anesthetized

with sodium brevitol (0.01 mg/g). Blood was collected by cardiac

puncture, using a heparinized syringe and 23 gauge needle. Whole

blood was diluted 1 to 2 in 8% perchloric acid to precipitate proteins

and then centrifuged. Samples were analyzed for lactate content using

a colorimetric assay from Sigma Diagnostics.

Ventilation rate was measured in 15 animals at several intervals

during estivation. If an animal was positioned in a flow-through

chamber such that its mouth was near the excurrent port, ventilation

could be detected by a rapid and extreme change in the fraction of.

oxygen in the excurrent air. This could be measured accurately only

for large animals.

Lipid Stores

In 1984 eight to ten adult S. lacertina were collected bimonthly

from Orange Lake, Alachua County, Florida, and frozen for later

analysis. Orange Lake is a relatively permanent body of water;

therefore, lipid stores of these specimens were not affected by

periodic droughts. In March of 1986 nine juvenile S. lacertina (3.6

to 25.5 g body mass) were collected and frozen. Unfortunately, it was

impossible to collect significant numbers of juveniles year round.

Animals were later thawed, weighed, macerated in a tissue homogenizer

with 70% ethanol and dried to a constant mass. The dry tissue was

weighed and each sample was extracted several times with petroleum

ether at 35C to remove all lipids (Martof 1969). Samples were dried

once more and reweighed. Total body lipid content was calculated for

each animal by subtracting lean dry mass from total dry mass.


Sirenids that were induced to estivate behaved similarly

regardless of size, species or time of year. When placed in saturated

soil, they burrowed into the substratum using their snout as a

shovel. As long as the soil was saturated with water, each animal

kept its mouth in contact with the surface, but as the soil dried the

salamanders retreated into the burrow formed by their body

movements. In some cases an opening was maintained with the surface,

but in several containers the opening collapsed and was filled as the

soil dried. Animals survived for several months in soil regardless of

whether burrows were open or closed. When they were removed for

metabolic measurements, the salamanders often made a sudden jerky

movement, emitted a yelping sound and ventilated by gulping three or

four times. All animals became still soon after handling ceased.

During estivation the gills atrophied and the entire body of an

estivator became covered with layers of dead epithelial cells.

Metabolic Rates

The mass-specific rate of oxygen consumption (V, /M)

was inversely correlated with body size (y = -11.37x + 50.4,

r2 = 0.7992; Fig. 2-1). Mass-specific metabolic rate


Q. C
0 (U

o*- S


4 .)

0 *c

cS O


OB c

-C L
*- 4S


*4- **-*
0C *0 0


o **

C U-

0 C

eo CM
-4 -

c *^ <







* 0

0 0 0 0 0 0

(46/I0 In) NOlldVlNSNOO ?0 IO 31tlVt

(V0 /M) decreased with time in estivation for sirenids of all sizes

(Figs. 2-2, 2-3, and 2-4). Time of year had no effect on metabolic

rate during estivation. The effect of sex on metabolic rate is

unknown because it was difficult to determine the sex of individuals

externally. Larval S. lacertina (under 2 g mass) had rates of oxygen

consumption that appeared similar to rates of P. striatus of similar

body size (Fig. 2-2); however, the sample sizes were too small for

statistical comparison. By day 60 of estivation, the V0 /M for these
smallest sirenids had decreased to 24.5 1.2 ul 02/g*h (mean S.E.),

a reduction of 56% from the mean rate of 53.6 2.4 ul 02/g'h at day

18. A similar decrease in V2 /M was seen in the other size groups of
estivating sirenids. Juvenile sirenids (6-110 g) measured near day 60

of estivation had a mean V02/M (18.0 3.0 ul 02/g'h) that was 59.7%
lower than the mean rate during the first two weeks of estivation

(44.7 5.8 ul 02/g'h). Mean V0 /M of adult S. lacertina measured
over the same two month interval decreased 63.9% (from 30.5 5.9 il

02/g*h to 11.0 0.7 ul 02/g'h). In all size classes the greatest
decline in Vo /M occurred during the first two months of estivation

and leveled off with only a slight decrease thereafter.

The QIo (mean S.E.) for V0 /M in estivating S. lacertina was
2.1 0.1 for seven animals measured at 23C and 32C. Whole blood

lactate levels in estivating S. lacertina were low (0.58 0.13
mmol/l; mean i S.E.), indicating that metabolism in these animals was
predominantly aerobic. Ventilation rate decreased from 5 to 12

breaths per hour during the first week of estivation, to one or less

breaths per hour by day 300.

r- o












0 0


c e



* **


.0 0O


(q.61/ O Irn)
NOIldlInSNO0O 0 IO 31-VU













o l


I- r















* 0

0 0

(q.B/5O Irf) NOlldyinSNOO 30 0o 3.VaIv












o1 0



0) (
o <







0 0

* *







0 0 0 0 0 0
( O 1 NOL O
(4-5/4O Irn) NOIldvmnSNOO 30 JO 31VH

Lipid Stores

Lipid stores varied seasonally in adult S. lacertina (Fig.

2-5). Females had the largest percentage of dry mass as lipid in the

fall whereas fat stores of males peaked in the summer. Both sexes had

the lowest total body lipid content from January through March. The

total body lipid content expressed as a percentage of dry mass for all

adult animals examined (n = 56) was 7.6 1.0 (mean S.E.). Mean dry

mass in these animals was 20.4% of total wet mass; therefore, average

lipid content was 1.6% of wet mass. Juvenile S. lacertina collected

in March had a total body lipid content of 5.4 1.8% (mean S.E.);

this amount was not significantly different from that of adults

collected during the same month (2.7 0.6%; t = 1.51, p > .05).


Sirenid salamanders conserve energy during drought-triggered

estivation by becoming inactive and by substantially reducing their

metabolic rate. Aquatic Siren and Pseudobranchus are capable of

trimodal respiration, exchanging gases across branchial, cutaneous and

pulmonary respiratory surfaces. The partitioning of gas exchange in

aquatic sirenids is dependent upon body size, ambient gas tensions,

temperature and the activity level of the animals (Guimond and

Hutchison 1973; Ultsch 1976b; Wakeman and Ultsch 1976). In Florida,

typical sirenid habitats are shallow, plant-choked waters that are

often hypoxic and hypercapic. Under these conditions, sirenids rely

largely upon their lungs for both 02 uptake and removal of


-0 e0

0 0
- LE
(1) m
0 r (

0 0

O- s


*- L3 i
0 0

e- e o,


C .- W
o L

- r- 0

u- +C


e ei

- ^a
Ii )

e <






(ssvII Alia -o %) saidii Aaoa -vioi.




- 0
z ~


CO2 (Guimond and Hutchison 1973; Ultsch 1976a). The lungs of S.

lacertina are nearly two-thirds of the body length and are septate and

infolded, thus providing a large respiratory surface area (Guimond and

Hutchison 1973). These effective lungs provide for gas exchange when

sirenids retreat into the substratum during drought. While the soil

is still saturated with water, these salamanders must breathe at the

surface, but as the soil dries the animals can retreat deeper into the

substratum. Because the gills atrophy during estivation and the layer

of dead epithelial cells covering the body probably impedes cutaneous

respiration, the lungs may become the primary site of gas exchange.

In very porous soil, such as sand or sandy loam, interstitial spaces

allow sufficient gas diffusion such that an opening to the surface

need not be maintained. Seymour (1973b) found that gas concentrations

in sandy soil next to buried spadefoot toads were close to atmospheric

levels even at a depth of 70 cm. Sirenids estivating in the

laboratory at 23'C did not need an opening to their burrows; however,

at higher temperatures or in less porous soil an opening to the

surface may be needed for gas exchange.

Estivating sirenids move very little unless they are disturbed by

handling. Most estivators moved suddenly when touched and then became

motionless again, but few attempted to crawl away. These animals

appeared to be in a state of light torpor similar to that described

for the estivating lungfish Protopterus aethiopicus (Delaney et al.

1974) and for estivating mammals (Hudson and Bartholomew 1964). This

ability to respond to external stimuli during estivation could aid in

avoidance of predation, particularly because sirenids often couple

their sudden jerky movements with a startling yelping noise.

The rates of oxygen consumption measured at the onset of

estivation are close to the resting rates measured by Ultsch (1976a)

for aquatic S. lacertina and P. striatus of the same body size. The

metabolic rates of sirenids of all sizes decline gradually during the

first two months of estivation and then reach a plateau that is 60 to

70% lower than the rate at the onset of estivation. Gehlbach et al.

(1973) measured V0 /M at 25"C in aquatic and estivating S. intermedia

from Texas, but they reported rates that were nearly three times as

high as those of estivators measured in this study and aquatic animals

measured by Ultsch (1976a). It is possible that these S. intermedia

were not at rest when rates of oxygen consumption were measured.

Even though ventilation rates decreased dramatically as

estivation proceeds, S. lacertina relied primarily on aerobic

pathways. Mean blood lactate levels for estivating S. lacertina (0.58

0.13 mmol/l) were similar to those of Scaphiopus hammondii resting

on the surface (Seymour 1973c).

Several species of frogs and a species of lungfish that estivate

in response to drought show a decrease in metabolic rate similar to

that of sirenids (Table 2-1). The rate of this decline varies among

these species. In the frog Lepidobatrachus llanensis V2 /M drops

nearly 80% after only seven days of dormancy (McClanahan et al. 1983),

whereas a drop of 82% is evident after 25 days in dormant Scaphiopus

couchii (Seymour 1973b). These differences may reflect different

environmental conditions in the habitats where these species occur, or

Oxygen Consumption Rates of Estivating Amphibians
and Lungfish (Protopterus) at 23-25C

Minimum V /M
Resting V0 /M during Dor ancy
Species (Mass) ("1 02/g'h (u1 02/g'h) Reference

Lepi doba trachus
lanensis 97 20 McClanahan et al.
(62-115 g) 1983

adspersus 37 10 Loveridge and
(10-1030 g) Withers 1980

couchi 51 11 Seymour 1973a
(R=20 g)

platycephalus 156 45 van Beurden 1980
(x=14 g)

aethiopicus 25 5 Delaney et al. 1974
(4100 kg)

differences in the way that estivation/dormancy was induced in these

separate studies. Rate of dehydration, for example, could affect the

time frame of a change in metabolic rate. Unfortunately, the

physiological mechanisms involved in the reduction of metabolic rate in

these dormant ectotherms are unknown. Although starvation alone has

been shown to reduce the metabolic rate of many vertebrates including

reptiles (Benedict 1932; Belkin 1965), lungfish (Delaney et al. 1974)

and frogs (Hill 1911), the pathways that mediate this effect are

unclear. Even less is known about the direct effects of desiccation

upon metabolism. Changes in the metabolic rate of estivating sirenids

may be related to thyroid function. Siren lacertina that had been

estivating for eight months had circulating levels of thyroxine that

were significantly lower than the plasma thyroxine levels of active

animals (Etheridge, unpublished data). However, no experimental

evidence exists to indicate that the reduction in metabolic rate seen

in estivators is thyroid mediated.

One common thread appears in the comparison of V0 /M of
estivating amphibians and lungfish; the minimum V0 /M during dormancy

for each species is similar (Table 2-1). When the data for each

species are adjusted for body size and temperature, the

minimum V0 /M of estivators is between 10 and 20 jl 02/g'h for each
2 .
species. Resting VO /M for the same group of species varies much

more, ranging from 37 to 150 ul 02/g'h. This consistency of the

lowest Vo /M may indicate that these rates represent the minimum cost

of existence for these ectotherms. In these animals resting metabolic

rates appear to be the "idling speeds" when resources (i.e., food and

water) are abundant. In animals that become dormant these "idling

speeds" can be reset 60-80% lower.

The amount of time an animal can survive without food is

dependent upon its rate of energy consumption and upon the amount of

stored energy the animal can utilize. For estivating sirenids the

most important determinants of metabolic rate are body size, length of

time in estivation and body temperature. The energy stores of each

animal will vary with time of year and possibly with body size.

The number of days estivating sirenids could survive without food

was estimated by calculating energy stores and rates of energy

consumption for animals of various sizes and lipid content (Table

2-2). The higher mass-specific metabolic rate of the smaller

salamanders results in a more rapid depletion of stored energy than in

the larger animals. A 1 g animal would deplete its energy reserves

1.5 times as fast as a 500 g animal even if both began estivation with

the same body composition (Table 2-2). Increasing environmental

temperatures would cause more rapid depletion of energy stores;

conversely, lower burrow temperatures would prolong the length of time

an animal could estivate. The Q10 of estivating S. lacertina (Q01

2.1) is essentially the same as that reported by Guimond and Hutchison

(1973) for aquatic S. lacertina (Q10 = 2.2).

Differences in total body lipid stores at the onset of estivation

would also contribute to differences in potential survival time. The

greater total caloric content of a 500 g S. lacertina with lipid

stores totaling 20% of its dry mass (the mean lipid content for

females in fall) would allow it to survive 40 days longer without food




a) -

C *-
e 0
>j i

*r- > a) Cn

40 >

> <0
o 0

iS <
) 0


0 >0





U Q.




+J >'

0 10 CM
d- Co 0

a) 4- 4-
Sr- f0

0 0 -4

a) 4- 4-

0 0 -4

40 CM

a m




0 CoO

0 f )
cc M


in r -

C .0 U)
-4 (M (-)




0 >9

S 4-0

C 4

a) 01 )


01 0i r
o3 to

L u


0 C0

a) 0


0 Q.

0 C 9.

(%< a)a
0 *0

S. a)

a) s.

10 M >

0 US.

>9 a)e
S. -c o
O 9 0

> o

.W L.



4- t-9.

L 4-
.- NJ


U. r-


N 0a

r- 0
0 a n

Es 0

)0 *-

E 0


C Oo


N *

.0 -0

a) L.

*0 3

N *

.- -

o 010


* o

0. M

t 0


P- *I
4- 0B
+- *I
a) I-

0 *

*<- -4
*9- S.
9- a)
C e
0 0)
a) >
E o

> 4.)>
0 0
> M


0 as

CO a)*r
CO 4-a

S 0 *3

0 0- >
C s-
C m S

*9- -0-0

S. > V0
(U -; *U

C.) 0O**
03 C
r- 4-

C- "S c..


O- 0
<.3 $- VI

O 10

0 0
r- O

QI- Cc
et a
r 00

3o cQ 0
U -c

) C

0. COC
>9 00






L1 C


C4 C






























D s

4 <

IC 0
a a



e 01
*<- .

. ^

















0 .

e L
o a
} .C'
E +-

.0 u -I

W 4-

than a 500 g estivator with lipid stores comprising only 3% of its dry

mass (the mean lipid content for summer animals). A very large S.

lacertina with a large quantity of stored lipid, such as the 1125 g

female with 43% body lipid (Table 2-2), potentially could survive two

to three years without food. Even a 1 g sirenid could survive for

several months (Table 2-2) on stored energy, and in most cases Florida

droughts last only a few weeks or months. Hence, during an average

drought, most of the sirenids in a population may be able to survive

by estivating. In a prolonged drought, however, the continuation of a

population of sirenids at a given site may be dependent upon the

survival of a few large individuals with greater than average lipid




When an amphibian remains inactive in a burrow energy stores

decline because energy cannot be replenished until the animal can

emerge and feed. Body water fluxes are more complex, because water

can be gained or lost at several sites (Fig. 3-1). In amphibians

water can be lost by evaporation from the respiratory surfaces, via

urine, or across the skin. Most vertebrates must eat or drink to

replenish body water, for their only source of water gain during

prolonged inactivity is the very slight amount of metabolic water that

is formed when foodstuffs are oxidized. Amphibians, on the other

hand, can take up water across their permeable skin (for review see

Duellman and Trueb 1986) and many can reabsorb water from their

urinary bladder (e.g., Ruibal 1962; Main and Bentley 1964; Bentley

1966). In addition, amphibians and lungfish that estivate burrow into

the soil to avoid desiccating surface conditions and employ various

strategies to maintain water balance.

One strategy involves increasing resistance to water loss as much

as possible by forming an impermeable "cocoon." In estivating African

lungfish (Protopterus) this cocoon is composed of dried mucus (Kitzan

and Sweeney 1968), but in most estivating amphibians studied thus far

electron microscopy has revealed the cocoon to be composed of multiple








0 = C
) 0o L


0 0
o CO C

Co v,

layers of unshed epithelial cells (Lee and Mercer 1967; McClanahan et

al. 1976; Loveridge and Craye 1979; Ruibal and Hillman 1981). In two

anurans, Pyxicephalus adspersus (Loveridge and Withers 1980) and

Lepidobatrachus llanensis (McClanahan et al. 1976), the cocoon of

estivators reduces water loss so effectively that these animals can

survive for several months in dry soil with little change in

concentrations of their internal solutes.

A second strategy for water balance during estivation has been

described for spadefoot toads. These animals maintain a favorable

water potential gradient with moist soil such that water is absorbed

from the soil across the skin. Scaphiopus hammondii can decrease

their body water potential to levels as low as -15 bars and thus

absorb water from any soil with a higher water potential (Ruibal et

al. 1969). Spadefoot toads maintain these low water potentials

primarily by accumulating urea in their plasma, thereby greatly

increasing their osmotic potential without disrupting ion balance

(McClanahan 1967, 1972; Shoemaker et al. 1969).

The objectives of this portion of the study were to investigate

adaptations that allow S. lacertina and P. striatus to survive the

dehydrating conditions of drought and to compare these adaptations to

those of other estivating amphibians. One specific area investigated

was the way in which these animals maintain water balance during

estivation, that is, whether they become water-resistant or exchange

water with the soil. The contribution of the urinary bladder to water

balance and the effect of estivation on plasma and urine solute

concentrations were determined. In addition, the effects of body size

and soil water potential on potential survival time were studied.

Materials and Methods

Siren lacertina and P. striatus were collected and induced to

estivate in the laboratory as previously described. These estivators

were maintained in moist soil (6 to 12% water content) so that plasma

parameters would reflect a hydrated state. Blood and urine samples

were collected from three S. lacertina on day 210 of estivation and

from five S. lacertina on day 260 of estivation. Concurrently, blood

samples were taken from 13 freshly caught, aquatic S. lacertina. An

attempt was made to collect urine from the aquatic animals as well.

Animals were maintained at 23C for all experiments.

Animals were anesthetized with sodium brevitol (0.01 mg/g) and

blood was collected by cardiac puncture using a heparinized syringe

and a 23 gauge needle. Each blood sample was centrifuged, the

hematocrit determined and the plasma frozen for later analysis. Urine

was collected by palpating the area of the urinary bladder; these

samples were frozen immediately. To estimate the volume of urine held

in the bladder of each estivator, animals were weighed before and

after urine was expressed.

Plasma and urine samples later were thawed for the following

analyses. Sodium and potassium concentrations ([Na+], [K+]; meq/1)

were determined for 5 pl samples using an Instrumentation Laboratory

443 Flame Photometer. Urea and ammonia concentrations ([NH ], [NH ];

mmol/l) were determined using the colorimetric Berthelot method (Sigma

test kit 640). Chloride concentration ([C1-]; meq/1) was determined

for 10 ul samples using a Radiometer CMT 10 Chloride Titrator. A

Wescor 5100 B Vapor Pressure Osmometer was used to determine the total

osmotic concentration (mOsm/kg) of 5 pl samples. A one-tailed

Students t-test was used to compare sample means of estivators and

active animals for each parameter.

Rates of water loss and gain were measured for estivating S.

lacertina (3-550 g body mass) and estivating P. striatus (less than

2 g) in soils of varying water content. The soil used for these

experiments was obtained from typical sirenid habitat in Alachua

County, Florida. Mechanical analysis of this soil established the

composition as 92.66% sand (primarily 0.25-0.50 mm particle size),

4.52% silt and 2.82% clay. This soil was mixed thoroughly and

filtered through a screen (1 mm mesh) to remove debris. A

characteristic moisture curve for this specific soil was constructed

by beginning with saturated soil and then measuring water potential

and water content at several intervals as the soil dried slowly in

room air. Water potential (- bars) was measured with a ceramic tip

tensiometer. Soil water content was determined by weighing the

sample, drying it to a constant mass and then reweighing it. The

percent of mass lost during drying represented the percent of water

content by mass. Once the characteristic moisture curve was

established the soil water potential of any sample of this soil could

be determined by calculating percent water content.

Large plastic containers (50 x 39 x 20 cm) were set up to hold

soils of three different moisture contents (0.7%, 5.0% and 5.0%

water). Air dry soil (0.7% water) was used to simulate the most

dehydrating possible field conditions. Preliminary observations

indicated that estivators maintained mass in soil with greater than

7.0% water content; therefore, slightly drier soils (5.0% and 6.0%

water content) were used to establish the point at which estivators

began to lose water to soil. Each container was tightly covered and

the water content of the soil was checked periodically to insure

stability. Salamanders estivating for seven months or more were used

for this experiment. Estivators had been maintained in moist soil

(10-12% water content) to insure maximum hydration before these

experiments. They were then exposed to the most dehydrating soil

(0.7% water content). After this they were exposed to the soils of

different water content in the following order: 5.0%, 0.7% and

6.0%. Each animal was weighed and then buried in the soil with only

its head exposed. Respiratory water loss was assumed minimal because

estivators ventilate only once or twice each hour. After two hours

the animal was excavated, any soil adhering was carefully brushed off,

and the animal was reweighed. All mass changes were recorded as water

lost or gained (mg/g body mass hour). Data from animals that moved

or urinated (as indicated by any moisture around the cloaca) during

the experiment were discarded. The protocol was designed to maximize

the rates of water exchange by putting hydrated salamanders into

dehydrating soil for measurements of water loss, then moving these

dehydrated animals into soil with a higher water potential to

determine rates of water uptake. For example, animals were placed in

tne driest soil (0.7% water) to dehydrate them prior to exposure to

the soil with 6% water content.

The tail tip of an estivating S. lacertina (day 210) was removed

and examined to determine the structure of the cocoon. The tissue was

fixed in 10% neutral buffered formalin, transferred to 70% ethanol and

then prepared for paraffin embedding. The tissue was sectioned at

10 1m, stained with Alcian Blue and counter-stained with Hematoxylin

and Eosin (Humason 1972) for examination by light microscopy. Because

Alcian Blue stains mucopolysaccharides, it was used to indicate

whether mucus was a major component of the cocoon. Hematoxylin stains

nuclear material; hence, it was used to determine the presence or

absence of cells in the cocoon layer.


The body position of estivating sirenids is such that most of the

surface area of the animal is in direct contact with the surrounding

soil. Siren lacertina do not coil, but maintain a somewhat S-shaped

configuration. In some individuals the tail was curled forward with

the tip touching the body. Estivating P. striatus were usually curled

into a figure eight with the head resting near the tail.

Blood parameters of S. lacertina estivating 210 days were not

significantly different from those of animals estivating 260 days;

therefore, these data were pooled for comparison with the aquatic

animals. Plasma ion concentrations ([Na"], [K+1, [Cl-]) were not

significantly different between the two groups (Table 1). However,

plasma urea concentration was significantly higher in the estivators


C 0


o -



0 S-
a- C
S- EU-

I o

:3 -
'- a,

0 -


e <


I- 0


r- 0

o o"
C -

r- -





*Oe C
0 --*

















*r- -Y


4I Co
C- -4
! s It
*.- -
a, Q

'* a





4-) r

oI o

U U1

o a






+ c
o) e-

*- C

V) ,

* 0



(t = 9.39, p < .001) as was the total plasma osmotic concentration (t

= 3.89, p < .005). The mean hematocrit was higher in estivators (33.9

3.9%) than in active animals (26.1 2.7%), but the difference was

not statistically significant (t = 1.55, p > .05).

No urine could be expressed from any active Siren despite

numerous attempts. Estivating sirenids were never observed to void

urine unless disturbed by handling. Urine collected from 10

estivating S. lacertina (mean mass of 479.7 41.2 g) made up 5.3

0.8% (mean i S.E.) of the mass of these animals. The maximum amount

of stored urine measured was 10.1% of body mass. Urine collected from

estivators was hyposmotic in comparison to the plasma (Table 3-1).

Concentrations of Na+ and Cl- were significantly lower in urine than

in plasma (t = 18.4, p < .001; t = 2.21, p < 0.05), whereas the

concentration of K+ was slightly higher in urine (t = 2.53,

p < 0.025). Urea concentration was not significantly higher in the

urine (t = 1.51, p > 0.05). Levels of ammonia were very low in the

plasma of both aquatic (2.4 0.2 mmol/l) and estivating (4.2 0.9

mmol/l) S. lacertina. Ammonia concentration in the urine of

estivators was ten times that in the plasma (45.7 t 10.2 mmol/l).

Soil water potential of the sand used in these experiments

remained high (between 0 and -1 bars) when soil water content was

greater than 8% (Fig. 3-2). At soil water contents below 6% the water

potential of this soil begins to increase exponentially with slight

decreases in water content. The rate of water exchange with the soil

was related to body size of the salamanders. This relationship was

strongest in the driest soil (Fig. 3-3), where it can be related as a

FIG. 3-2. Characteristic moisture curve for soil used in water
exchange experiments.




1 10-

- I



I--I-I I i
2 4 6 8 10 12


FIG. 3-3. Rates of water loss for estivating S. lacertina in soil
with 0.7% water content (closed circles) and in soil with
5.0% water content (open circles).




co 16
1 12-

m 8-



A w v

MASS (g)

* *

0 0

100 200 300 400 500 600
100 200 300 400 500 600

power function (y = 215x-0.64, r2 = 0.96). Two 1 g P. striatus placed

in this same soil (0.7% water content) lost water at rates of 51 and

92 mg/g'h. Six of 14 animals maintained their original mass in the

soil with 5.0% water content; the other eight lost mass, but at a much

lower rate than in the drier soil (Fig. 3-3). Similarly, 4 of 13 S.

lacertina maintained their weight in the soil with 6.0% water content,

and the remainder gained water from this soil (Fig. 3-4). Thus, a

positive water balance is achieved by estivators in soil with water

potential greater than -5 bars (6% water content), whereas estivators

in drier soil (i.e., 5% H20 content, less than -5 bars) may lose water

(Fig. 3-4).

Light microscopy revealed the dry outer layer or "cocoon" of

estivating S. lacertina to be composed primarily of dead, squamous

epithelial cells. This layer did not stain for mucopolysaccharides as

it would if mucus were the primary component, but remained brown in

color. In a few animals that were allowed to dehydrate in dry soil,

the cocoon appeared to be thicker than in hydrated estivators.


The long cylindrical body shape of sirenid salamanders results in

a relatively high surface to volume ratio. While this may be

advantageous for cutaneous respiration in water, it is disadvantageous

during a drought because it creates a large surface area for water

loss. Siren lacertina are fairly stout animals with a thick, muscular

body wall and tightly wrapped skin, and they are incapable of coiling

up to reduce surface area. Pseudobranchus striatus are more slender

FIG. 3-4. Rates of water exchange with the soil for estivating S.
lacertina in soil with 5.0% water content (closed circles)
and in soil with 6.0% water content (open circles).




0 0
o o

0- ----- ,"- -1 ,-- ----

E -1-






100 200 300 400 500 600
MASS (g)

and can coil somewhat, but most of the surface area of both species is

in direct contact with soil during estivation. Consequently, water

exchange with the soil is the most important factor in water balance

of estivating sirenids. The rate of water exchange with soil is a

function of several parameters, primarily surface area, resistance and

the difference between the water potential of the soil and that of the


The water potential of an estivating animal is related to the

osmotic potential of its body fluids. An amphibian with high plasma

solute concentrations has lower water potential (- bars) than an

animal with lower plasma solute concentrations, and all else being

equal more water will move across the skin of the first animal. The

water potential of estivating S. lacertina is increased by urea

accumulated in the plasma. Elevated plasma urea levels in estivators

accounted for the difference between total plasma osmolality of

estivators and aquatic animals. On the other hand, plasma ion

concentrations appear to be closely regulated. This strategy of

decreasing water potential by storing urea has been demonstrated in

several species of amphibians, including Ambystoma tigrinum (Delson

and Whitford 1973), Xenopus laevis (Balinsky et al. 1967), Bufo

viridis (Degani et al. 1984), and several species of Scaphipus

(McClanahan 1967; Shoemaker et al. 1969; Jones 1980). In these

species plasma ion concentrations increased much less (no more than

two times hydrated levels) than plasma urea levels (up to 30-fold

increase). The difference between the two variables of soil water

potential and water potential of the animal not only determines the

direction of water exchange (water moves from an area of higher to

lower water potential), but the rate at which the exchange occurs.

Sirenids lost water in soil with water potentials lower than -5 bars

(5.0% water content), but they lost water more rapidly in the soil

with 0.7% water content (water potential less than -15 bars). The

rate of water loss was greatly affected by the third variable, surface

area of the estivator. Small S. lacertina and P. striatus dehydrated

rapidly due to their high surface to volume ratio. The relationship

of body size to rate of water exchange could be seen most clearly in

the driest soil (0.7% water), where hydric properties of the soil play

the major role in determining water loss. In soils with 5.0 or 6.0%

water content the correlation between rate of water exchange and body

size is less evident. This increased variability in rates indicates

that water exchange in soils of higher water potential may be mediated

more by hydric properties of the animal than by those of the soil.

Variability in rates of water exchange may reflect differences in

water potential of individual animals and/or differences in resistance

to water flux.

Resistance is the fourth variable in determining rates of water

exchange and is the most difficult to measure. Resistance as defined

here is the capacity of the outer layer of the animal to retard water

flux. This involves many factors such as the degree of vasodilation

or vasoconstriction, hormonally mediated processes and physical

properties of the outer layer. A salient characteristic of estivating

amphibians is the outer covering of the animal referred to as the

cocoon, and this is usually the focus of any investigation of

resistance to water loss. The cocoon of sirenids examined in this

study appears to be layers of stratum corneum. Examination by light

microscopy revealed flattened, nucleated cells. Furthermore, these

layers do not stain for mucopolysaccharides as they would if mucus

were the main component. These findings differ from those of Reno et

al. (1972) who described the cocoon of estivating S. intermedia as

layers of dried mucus with no cellular integrity. For sirenids in

this study, the cocoon layer could be peeled off in sheets and

preserved intact, indicating that it is much like a typical shed skin

of any amphibian. McClanahan et al. (1983) described cocoon formation

in Lepidobatrachus llanensis as a normal shedding cycle in which the

layers do not detach from the animals. Water loss in air in cocooned

L. llamensis was reduced as much as 70% after one week of estivation

(McClanahan et al. 1976) and was further reduced with each additional

layer of stratum corneum added to the cocoon of these frogs during

estivation (McClanahan et al. 1983). Unfortunately, the degree to

which the cocoon retards water flux in soil has not been measured for

any estivator. These types of measurements are difficult to make, and

there is the further complication of trying to separate the

contribution of the cocoon from the other factors involved in

resistance. Using an in vitro technique, Reno et al. (1972) measured

the effect of the cocoon of S. intermedia upon water uptake from agar

blocks and found that the cocoon alone decreased water uptake by

31%. Dehydrated Siren did appear to have a thicker cocoon layer than

hydrated estivators, and it may be that resistance is varied by

increasing the rate of shedding; however, this warrants further study.

Amphibians from different environments may respond differently to

drought depending upon the type of soil in which they burrow. The

strategy employed by sirenids, spadefoot toads and others of storing

urea and absorbing water from the soil may be useful only in sandy

soils, where large interstitial spaces cause high water potentials

even at low moisture contents. Frogs that become almost waterproof,

such as P. adspersus and L. llanensis, may live where soils have a

high clay content. Clay soils have small pore size and consequently

low water potential even at high moisture content (Noble 1974). Thus

amphibians burrowed in this type of soil may not be able to reduce

their water potential below that of the soil; they therefore avoid

dehydration by becoming as impermeable to water loss as possible. The

possibility that amphibians can vary resistance according to the water

potential of the soil in which they burrow warrants additional study.

Several estivating amphibians store large quantities of dilute

urine when in positive water balance and later reabsorb water from the

bladder to rehydrate body fluids during drought (e.g., Ruibal 1962;

Bentley 1966). In some fossorial desert frogs, bladder volumes may

account for as much as 50% of the animals' body mass (Bentley 1966).

Siren lacertina can reabsorb water from its bladder (Bentley 1973).

No urine could be collected from aquatic animals, possibly because

active sirenids excrete urine continuously. Urine expressed from

hydrated estivators averaged 5.3% of body mass, and a maximum of 10.0%

of body mass was measured for one animal. These values are lower than

those reported for most other estivating amphibians, but are higher

than the values of I and 2% reported for the aquatic amphibians

Xenopus laevis (Bentley 1966) and Triturus cristatus (Bentley and

Heller 1964). Sirenids may be morphologically constrained to low

bladder volumes; they are long and cylindrical and their tight skin

and thick body wall do not allow for a very distensible body cavity

like that of most frogs. However, even a water reservoir of 5 to 10%

of body weight would extend survival time.

The length of time estivating sirenids could survive without

reaching lethal dehydration levels is dependent primarily upon body

size and soil water potential. In very dry soil (0.7% water content)

a 20 g S. lacertina losing water at a constant rate would lose 50% of

its body mass as water in 16 h. This level of water loss is lethal

for most amphibians (Ray 1958). A 500 g animal under the same

conditions would last 125 h. In soil with a water potential

approaching -5 bars (5.0% water content for the soil used in these

experiments) the 20 g animal would reach 50% dehydration in 272 h (11

days); the 500 g animal would last 694 h (29 days). Under field

conditions the rates of water loss would decrease as the soil around

the animals gained water (and increased in water potential) and as the

animals' water potential decreased due to dehydration. Thus actual

survival times would be longer, although a larger animal would still

survive longer than a smaller one in the same soil. Gehlbach et al.

(1973) reported that S. intermedia lost 26-28% of their body mass

during 16 weeks of estivation; however, these animals were in soil

that was allowed to dry out. Sirenids estivating in soil with water

potential equal to or greater than their body water potential would

not lose water to soil. Under such conditions the only mass loss

would be due to metabolized tissue, and sirenids could survive as long

as energy stores persisted. Under field conditions it may be rare for

water to be the limiting factor in survival during estivation. In

Florida brief showers occur even during severe droughts, and although

these showers do not reflood the habitat, they may suffice to moisten

the soil such that sirenids can maintain a positive water balance.

Furthermore, water potential can vary greatly from site to site

depending upon such factors as soil type, plant cover and proximity to

the water table. For example, sandy soil within a meter of the water

table would remain virtually saturated due to capillary action, even

during a drought. Sirenids in such a place would not be in danger of

dehydration and could serve to re-establish populations at sites that

did reach critical water tensions during drought.


Throughout the southeastern coastal plain of the United States

there is an abundance of shallow water habitat suitable for sirenid

salamanders. This is often a transient resource, however, because

drought is common in this region. Sirenid salamanders, although

aquatic, are well-adapted for survival during drought and thus can

exploit impermanent bodies of water. As their habitat dries out,

sirenid salamanders burrow into the substratum and remain inactive

until the area is reflooded. Survival during estivation is due to

their abilities to live for prolonged periods on stored energy and to

maintain water balance in drying soil.

Energy is unlikely to be the limiting factor in sirenid survival

during estivation except in rare cases. The metabolic rates of

estivating S. lacertina and P. striatus are reduced 60 to 70% below

rates of resting aquatic sirenids. Coupled with substantial energy

stores, these low rates of energy consumption would permit large S.

lacertina to survive for two to three years without food and would

allow even a 1 g animal to survive for several months. In a prolonged

drought the largest animals with the greatest energy stores may be the

only survivors, but these could serve to found a new population.

Water is more likely to be limiting to survival than energy, but

under usual field conditions lethal dehydration of estivating sirenids

is probably rare. The cocoon of S. lacertina, composed of dead

epithelial cells, may retard water exchange (Reno et al. 1972), but it

does not make estivators impermeable to water loss. Sirenid

salamanders exchange water with soil at a rate that shows a strong

inverse correlation with body size, indicating that relative surface

area is a major factor in water balance. The other major determinant

of the rate and direction of water flux is the magnitude of the

difference in water potential of the soil and water potential of the

estiva tor.

Water potential of an animal buried in soil is dependent

primarily upon the concentration of internal solutes in the animal's

body fluids. Siren lacertina that had been estivating seven or more

months had a mean plasma osmotic concentration of 264.2 1 9.8 mOsm/kg,

which is the equivalent of -5.8 bars of water potential. Most animals

maintained a positive water balance in soil with water potential

greater than -5.7 bars, but began to lose water in soil with water

potential lower than this. Aquatic S. lacertina had a lower mean

plasma osmotic concentration (229.8 11.7 mOsm/kg), the equivalent of

-5.0 bars of water potential, and would lose water in soil with a

higher content than would estivators. This difference in osmotic

concentration is due to the higher plasma urea concentration of the

estivating S. lacertina. The salamanders used in this experiment had

been kept in moist soil and were hydrated; it is possible that plasma

urea concentrations can be increased in estivators as they become

dehydrated, decreasing their water potential even further. This has

been well-documented in species such as Scaphiopus couchii

(McClanahan 1972), which can actively increase urea concentrations in

body fluids in response to decreasing soil water potential.

Water potential of soil is dependent upon the soil structure and

composition (i.e., percentages of sand, silt and clay) and upon the

water content (Fig. 4-1). Clay has smaller pore sizes than sand,

higher matric potentials, and consequently, much higher water

potential than sand at any given water content. Thus, a siren

estivating in clay would be in a potentially more stressful hydric

environment than one in sand. It may be that sirenid salamanders can

inhabit drought-prone waters only in areas of sandy soil. This could

help to explain the geographic distribution of sirenids, which live

primarily in the coastal plain, where soils tend to be sandy. Even

within a large area such as the coastal plain soil type varies, and

there may be areas within this mosaic that are unsuitable for sirenid

estivation. Hence, sirenid distribution may be affected by soil type

as well as water availability.

Although the physiological mechanisms associated with estivation

seem to be evolutionarily conservative among amphibians, many of these

adaptations are not present in other vertebrates. Indeed, in many

ways amphibians such as sirenid salamanders are uniquely suited to

survival during drought. Rates of energy consumption are low in

amphibians, particularly salamanders, even in comparison to reptiles

(Pough 1983). Other vertebrates must rehydrate by drinking or eating;

amphibians can absorb water from moist soil and store excess water in

the bladder for later use. Lastly, the ability to tolerate high

concentrations of urea in body tissues has not been demonstrated for

FIG. 4-1. Hypothetical water potentials of different soil types at
varying water content. (See Noble 1974 for a review of
soil water potentials.)

<- clay

< 10





10 20 30 40 50 60


any of the higher vertebrates. Ironically, the highly permeable skin

of amphibians, which would appear to be a disadvantage in xeric

conditions, is actually highly advantageous (e.g., Bentley 1966;

McClanahan 1972). Even though amphibians cannot maintain prolonged

surface activity during drought, they can survive for months or years

in the absence of food and free water, a feat well beyond the

abilities of any endotherm. In the case of sirenid salamanders, the

ability to survive drought by estivating allows them to exploit

potentially transient aquatic habitats that would be uninhabitable for

many species.


Balinsky, J.B., E.L. Choritz, C.G.L. Coe, and G.S. Van Der Schans.
1967. Amino acid metabolism and urea synthesis in naturally
aestivating Xenopus laevis. Comp. Biochem. Physiol. 22:59-68.

Belkin, D.A. 1965. Reduction of metabolic rate in response to
starvation in the turtle Sternothaerus minor. Copeia 1965:367-

Benedict, F.G. 1932. The physiology of large reptiles. Carnegie
Inst. Washington Publ. 425.

Bentley, P.J. 1966. The physiology of the urinary bladder of
amphibia. Biol. Rev. 41:275-316.

Bentley, P.J. 1973. Osmoregulation in the aquatic urodeles Amphiuma
means (the congo eel) and Siren lacertina (the mud eel). Effects
of vasotocin. Gen. and Comp. Endocrinology 20:386-391.

Bentley, P.J., and H. Heller. 1964. The action of neurohypophysical
hormones on the water and sodium metabolism of urodele
amphibians. J. Physiol. 171:434-453.

Bentley, P.J., A.K. Lee, and A.R. Main. 1958. Comparison of
dehydration and hydration of two genera of frogs (Heleioporus and
Neobatrachus) that live in areas of varying aridity. J. Exp.
Zool. 35:677-684.

Brenner, F.J. 1969. The role of temperature and fat deposition in
hibernation and reproduction in two species of frogs.
Herpetologica 25:105-113.

Burch, P.R., and J.T. Wood. 1955. The salamander Siren lacertina
feeding on clams and snails. Copeia 1955:255-756.

Byrne, J.J., and R.J. White. 1975. Cyclic changes in liver and
muscle glycogen tissue lipid and blood glucose in a naturally
occurring population of Rana catesbeiana. Comp. Biochem.
Physiol. 50A:709-715.

Cockrum, L. 1941. Notes Siren intermedia. Copeia 1941:265.

Conant, R. 1975. A field guide to reptiles and amphibians of eastern
and central North America. Houghton Mifflin Co., Boston.

Davis, W.B., and F.T. Knapp. 1953. Notes on the salamander Siren
intermedia. Copeia 1953:119-121.

Oegani, G., N. Silanikove, and A. Shkolnik. 1984. Adaptation of
green toad (3ufo viridis) to terrestrial life by urea
accumulation. Comp. Biochem. Physiol. 77A:585-587.

Delaney, R.G., S. Lahiri, and A.P. Fishman. 1974. Aestivation of the
African lungfish Protopterus aethiopicus: Cardiovascular and
respiratory functions. J. Exp. Biol. 61:111-128.

Delson, J., and W.G. Whitford. 1973. Adaptations of the tiger
salamander, Ambystoma tigrinum, to arid habitats. Comp. Biochem.
Physiol. 46A:631-638.

Duellman, W.E., and L. Trueb. 1986. Biology of amphibians. McGraw-
Hill, Inc., New York.

Fitzpatrick, L.C. 1976. Life history patterns of storage and
utilization of lipids for energy in amphibians. Amer. Zool.

Freeman, J.R. 1958. Burrowing in the salamanders Pseudobranchus
striatus and Siren lacertina. Herpetologica 14:130.

Freeman, J.R. 1967. Feeding behavior of the narrow-striped dwarf
siren Pseudobranchus striatus axanthus. Herpetologica 23:313-

Gehlbach, F.R., R. Gordon, and J.B. Jordan. 1973. Aestivation of the
salamander, Siren intermedia. Am. Mid. Nat. 89:455-463.

Gehlbach, F.R., and S.E. Kennedy. 1978. Population ecology of a
highly productive aquatic salamander (Siren intermedia).
Southwestern Naturalist 23:423-430.

Goin, C.J. 1941. The striped siren, Pseudobranchus striatus (Le
Conte). M.A. thesis, University of Florida, Gainesville.

Guimond, R.W., and V.H. Hutchison. 1973. Trimodal gas exchange in
the large aquatic salamander, Siren lacertina. Comp. Biochem.
Physiol. 46A:249-268.

Hanlin, H.G., Jr. 1978. Food habits of the greater siren, Siren
lacertina, in Alabama coastal plain pond. Copeia 1978:358-359.

Hill, A.V. 1911. The total energy exchanges of intact cold-blooded
animals at rest. J. Physiol. 43:379-394.

Hudson, J.W., and G.A. Bartholomew. 1964. Terrestrial animals in dry
heat: Aestivators. Pages 541-550 in D.B. Dill, E.F. Adolph and
C.G. Wilbur, eds. Handbook of physilogy. Williams and Wilkins
Co., Baltimore.

Humason, G.L. 1972. Animal tissue techniques. W.H. Freeman and Co.,
San Francisco.

Jones, R.M. 1980. Metabolic consequences of accelerated urea
synthesis during seasonal dormancy of spadefoot toads, Scaphiopus
couchi and Scaphiopus multiplicatus. J. Exp. Zool. 212:255-267.

Kitzan, S.M., and P.R. Sweeny. 1968. A light and electron microscope
study of the structure of Protopterus annectens epidermis. I.
Mucous production. Canad. J. Zool. 46:767-773.

Lee, A.K., and E.H. Mercer. 1967. Cocoon surrounding desert-dwelling
frogs. Science 157:87-88.

Loveridge, J.P., and G. Craye. 1979. Cocoon formation in two species
of Southern African frogs. South Africa J. Sci. 75:18-20.

Loveridge, J.P., and P.C. Withers. 1981. Metabolism and water
balance of active and cocooned African bull frogs Pyxicephalus
adspersus. Physiol. Zool. 54:203-214.

Main, A.R., and P.J. Bentley. 1964. Water relations of Australian
burrowing frogs and tree frogs. Ecology 45:379-382.

Martof, B.S. 1969. Prolonged inanition in Siren lacertina. Copeia

McClanahan, L. 1967. Adaptations of the spadefoot toad, Scaphiopus
couchi, to desert environments. Comp. Biochem. Physiol. 20:73-

McClanahan, L., Jr. 1972. Changes in body fluids of burrowed
spadefoot toads as a function of soil water potential. Copeia

McClanahan, L., R. Ruibal, and V.H. Shoemaker. 1983. Rate of cocoon
formation and its physiological correlates in a ceratophryd
frog. Physiol. Zool. 56:430-435.

McClanahan, L.L., Jr., V.H. Shoemaker, and R. Ruibal. 1976.
Structure and function of the cocoon of a ceratophryd frog.
Copeia 1976:179-185.

Morton, M.L. 1981. Seasonal changes in total body lipid and liver
weight in the Yosemite toad. Copeia 1981:234-238.

Noble, P.S. 1974. Introduction to biophysical plant physiology.
W.H. Freeman, San Francisco.

Pough, F.H. 1983. Amphibians and reptiles as low-energy systems.
Pages 141-188 in W.P. Aspey and S.I. Lustick, eds. Behavioral
energetic. OTVo State University Press, Columbus, Ohio.

Ray, C. 1958. Vital limits of desiccation in salamanders. Ecology

Reno, H.W., F.R. Gehlbach, and R.A. Turner. 1972. Skin and
aestivational cocoon of the aquatic amphibian, Siren
intermedia. Copeia 1972:625-631.

Rose, F.L. 1967. Seasonal changes in lipid levels of the salamander
Amphiuma means. Copeia 1967:662-666.

Ruibal, R. 1962. The adaptive value of bladder water in the toad,
Bufo cognatus. Physiol. Zool. 35:218-223.

Ruibal, R., and S.S. Hillman. 1981. Cocoon structure and function in
the burrowing hylid frog, Pternohyla fodiens. J. Herpetology

Ruibal, R., L. Tevis, Jr., and V. Roig. 1969. The terrestrial
ecology of the spadefoot toad Scaphiopus hammondii. Copeia

Seymour, R.S. 1973a. Energy metabolism of dormant spadefoot toads
(Scaphiopus). Copeia 1973:435-446.

Seymour, R.S. 1973b. Gas exchange in spadefoot toads beneath the
ground. Copeia 1973:452-460.

Seymour, R.S. 1973c. Physiological correlates of forced activity and
burrowing in the spadefoot toad, Scaphiopus hammondii. Copeia

Shoemaker, V.H., L.L. McClanahan, Jr., and R. Ruibal. 1969. Seasonal
changes in body fluids in a field population of spadefoot
toads. Copeia 1969:585-591.

Ultsch, G.R. 1973. Observations on the life history of Siren
lacertina. Herpetologica 29:304-305.

Ultsch, G.R. 1976a. Ecophysiological studies of some metabolic and
respiratory adaptations of sirenid salamanders. Pages 287-312 in
G.M. Hughes, ed. Respiration of amphibious vertebrates.
Academic Press, London.

Ultsch, G.R. 1976b. Respiratory surface area as a factor controlling
the standard rate of 0O consumption of aquatic salamanders.
Resp. Physiol. 26:357-369.

van Beurden, E.K. 1980. Energy metabolism of dormant Australian
water-holding frogs (Cyclorana platycephalus). Copeia 1980:787-

Wakeman, J.R, and G.R. Ultsch. 1975. The effects of dissolved 02 and
CO on metabolism and gas-exchange partitioning in aquatic
saTamanders. Physiol. Zool. 48:348-359.

Withers, P.C. 1977. Measurement of V V CO and evaporative
water loss with a flow through mask2 J. Apl Physiol. 42:120-


Kay Etheridge was born in Clarksdale, Mississippi, in 1954. She

grew up in Alabama and attended Auburn University where she received

her B.S. in zoology in 1975 and her M.S. in zoology in 1980. Not yet

satiated with higher education, she migrated still farther south to

the University of Florida where she pursued her Ph.D. During this

pursuit she also worked with manatees and became interested in

tropical biology during a course in Costa Rica. Her primary research

interests involve the physiological ecology of reptiles and

amphibians. Her spare time is devoted largely to the practice of

martial arts and the search for the ultimate chocolate ice cream.

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Martha L. Crump, Chairman
Associate Professor of Zoology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Jn F. Anderson
Associate Professor of Zoology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Walter Judd
Associate Professor of Botany

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Harvey B. Lil white
Professor of Zoology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

M chelle Wheatly C(/
Assistant Professor of Zoology

This dissertation was submitted to the Graduate Faculty of the
Department of Zoology in the College of Liberal Arts and Sciences and
to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.

December, 1986 _____ _te__a_______
Dean, Graduate School

II l l 111111 1111 II I II Ill 1 111 1111 l ll 1 1111 I
3 1262 08554 1448