<%BANNER%>






THE ECOLOGICAL AND PHYSIOLOGICAL CONTROL

OF WATER LOSS IN SNAKES














By
THOMAS HENRY KRAKAUER















A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY











UNIVERSITY OF FLORIDA


1970








ACKNOWLEDGCEME WIS


I would liI:e to thank Drs. Brian K. McNab, Frank G. Nordlic

and Daniel A. Ealkin for help during the course of the study and

durig-i the prepe::etion of thi' mr-nuscript. Dr. John Fr. rnderso.

and Mr. Paul E. Holer helped in many ways including serving as

valuable sounding boards for some of my ideas. My wife, Janet..

helped in all thr.e *ays that wives usually do, and helped in the

preparation of the illustrations.











TABLE OF CONTENTS

Acknowledgementts ....................... ....................... ii

List of Tables................................................ iv

List of Figures................................ .. ...... ... ..... v

Ab L act .............................. ............ ....... vii

Introduction.................................... ......... 1

Materials and Methods ............................. ...... .... 3

Skin Permeability in Air .................................. 3

Skin Permeability in Aqueous jMedia........................ 7

Results .................................................. ...... 9

Discussion....................................................... 28

Cutaneous Water Loss....................................... 28

Oxygen Consumption ........................................ 34

Respiratory Uater Loss................................... 34

Partitioning of Uater Loss............................ ..... 38

Water Exchange with Aqueous Media......................... 45

Conclusions ................................................... 50

Surr-rary............................................... ... ... 52

References................................................... 54

Biographical Sketch............................................ 57


iii










LIST OF TABLES

Table 1 Cutaneous water loss................................. 24

Table 2 O;ygen consumption ................................... 25

Table 3 Respiratory water loss.............................. 26

Tab le 4 Perce.t of tuLal vwaLer los; that iS cutaneous....... 27











LIST OF FIGURES

Figure 1 Diagram of equipment ............................... 5

Figure 2 Cutaneous r'ater loss as a function of weight

i n NiJtrix fasciata pictiventris...................... 11

Figure 3 Cutano 0:s vator loss as a function of weight

in atri fasciata ccoE ressic '.i da.................... 11

Figure 4 Cutaneous water loss as a function of weight

at 250C.............................................. 13

Figure 5 Oxygen consumption as a function of weight

in Natri: fasciata pictiventris..................... 15

Figure 6 Oxygae consumption as a function of weight

in Niatrix fasciata corpressicauda ................... 15

Figure 7 Oxygen consumption as a function of weight at

250C................................................. 17

Figure 8 Respiratory water loss as a function of weight

in Nat.rix fasciata pictiventris..................... 19

Figure 9 Respiratory water loss as a function of weight

in Natrix fasciata compressicauda.................... 19

Figure 10 Respiratory water loss as a function of weight

at 250C.............................................. 21

Figure 11 Isotope fluid: rates as a function of weight

at 25 C .............................................. 23

Figure 12 The effects of body position and activity on

cutaneous after r loss................................. 33





Figure 13








Figure 14








Figure 15






Figure 16



Figure 17






Figure 18






Figure 19


Respiratory water loss per milliliter of oxygen

consumed as a function of oxygen consumption in

Natrix fasciata pictiventris and ThamnoDhis

sauritus ... ..........................

Respiratory water loss per milliliter of oxygen

consumed as a function of oxygen consumption in

Natrix fasciata compressicauda, Natrix cyclopion

and Pit:-opbis catenifer..............................

Minute volume as a function of oxygen consumption

in Natrix fasciata pictiventris and Th'mnophis

sauritus................................ .... ..

Minute volume as a function of oxygen consumption

in Natrix fasciata compressicacde ....................

The fraction of oxygen in the expired air as

a function of oxygen consu-mption in Natrix

fasciata pictiventris...............................

The fraction of oxygen in the expired air as

a function of oxygen consumption in Natrix

fasciata compressicauda.................................

The fraction of oxygen in the expired air as a

function of oxygen consumption in Thamnophis

sauritu:c........................ .. .....................











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

THE ECOLOGICAL AND PHYSIOLOGICAL CONTROL OF WATER LOSS
IN SNAKES

By

Thomas Henry 1rakauer

August, 1970

Chairman: Brian K. McNab
Major Deparbnent: Zoology

The cutaneous and respiratory water loss of Natrix fasciata

pictiventris, N. f. cmpressicauda and Thianophis sauritus were

measured using a flowing air system and a dew-point hygrometer.

Cutaneous water loss, when expressed as a function of the vapor

pressure deficit, did not change with temperature. Cutaneous

water loss was almost three times greater in 1. f. jfetiJventris

(fresh iater) tha' in N. f. co-pressicauda (salt water) and

amounted to 92 and 80 per cent (25C) of the total water loss

respectively.

Deterimin.tiens of skin penneability in aqueous media by in vivo

rimeasurcfients of tritiated water flux indicate thM-t external salinity

has no effect on the effliu from the animal. The skin penneability

of N. f. jDitiLvent.ris in water is greater than that of N. f.

ceompyessic3uda.


vii





No differences, between subspecies, in respiratory water locs

were found at 15, 25 and 320C. Respiratory water loss of Thamnoahis

was low-er, correlating with a lower level of oxygen consumption. A

decrease in the amount of water lost per ml of oxygen consumed tended

to keep the respiratory water loss constant with increased temperature

(corrected for saturation deficit). Respiratory wcter loss increases

less rapidly wiith activity than does oxygen consumption.

The rate of respiratory water loss is not modified, except by

changes in oxygen con;uniption, while the cutaneous permeability is

responsive to ecological demands.


viii










INTRODUCTION

Reptiles are able to remain in water balance under a wide

range of environmental conditions. Marine and terrestrial species

live in dehydrating media, while fresh-water reptiles are exposed

to .nn e-,vironl;-n!e.n th favors the entry of water into th. animal.

Water loss takes place at three sites: cloaca, respiratory tract,

and skin. This study deals with the cutaneous and respiratory

components.

Classically, reptiles were viewed to have an impermeable

integunent, with the respiratory tract stated to be the route of

most of the waterr loss (Chew, 1961; Pettus, 1950). Reccnt :tudi es,

hoc-ver, indicate that rates of water loss of reptiles vary fro-c

those typical of "waterproof" insects to those of am.phibinne

(see Schmidt-Niclsen, 1969; Krakauer et al., 1968), .iand thc the

major component of water loss is cutaneous (Ecntlcy and Sclmid--Nielsenl

1966; Schmidt-Nielsen and Bentley, 1966; Clausen, 1967; Dawson c_ l1.,

1966; Prange and Scfrnidt-Nielsen; 1969). The rates of water loss

correlate with habitat aridity (Bogert and Ccwles, 1947; see also

Krakauer et rl., 1968; Gans et al., 1968).

In this study I will exrmine the water loss of three closely

related, but ecologically distinct, snakes: a brackish-water snake,

the mangrove water snake (Natri'xL f-cnciata co_ ressiecu1d)* ; a frich-'iater

snake, the Florida banded water snake (N. f. plctivYltris); and a ore









terrestrial rsn-:ke, the ribbon snake (ThE.nophis snturitsC The

Natrix and Thajnophis live nez: nater but differ in the amount of

time vpent in water. The dercrt gopher snake (Pituophis cntenifr)

and the green water snake (Natrix cycjoion) were also studied. The

variability of water loss is examined to determine hotj the rate of

water loss can be modified. Data are presented on the changes in

cutaneous and respiratory iyater loss with changes in ecology, body

weight, temperature, activity, shedding cycle, and external medium-

(dry air, distilled r-ater, and sea water).










MATERIALS AND METHODS

Natrix fasciata pictiiventris, f. cyclonion, and T. sauritus

were collected at several localities in Alachua County, Florida.

The specimen of P. catenifer was obtained without data from an

animal dealer in southern California. N. f. co^.pressi.cu.da wer

collected at Matheson H raock, Dade County, Florida and in Monroe

County, 35 milcs southwest of 1Homestead, Dade County, Florida.

These sites were pioneer and secondary mangrove association with

salinity that varied with rainfall (14-41 parts per thousand;

Tabb et al., 1962).

The Nntrix and Thamnophis were maintained in the laboratory

on a diet of fish in dry cages, and held at 25 + 10C with 12 hours

of light daily. Tap water was provided. The Pituophs was kept

under similar conditions except at fluctuating room temperature

(ca.260C). All water loss measurements were made on animals after

they had fasted for at least one week and at known times during the

shedding cycle.

Skin Per ca-bility in Air

A flowing air desiccation system tws used to determine water

loss to dry air (Fig. 1). The respiratory and cutaneous water

losses were measured separately. A partition between a head and

body chaMnbcr was created by sliding the punctured tip of a tapered

latex balloon (Trojar-enz INo. 175) around the animal's neck, and

the open end over a collar on the ple:iglas hWad cha-iber (320 ml).

The snake's body was tied to a coarse moeh wire screen by a pipe

3

























Fig. 1 Diagram of equipment. The dark
triangles indicate the points where
the rate of flow was regulated.























I
II
I -







OXYGEN

'ANALYZER





I I
I I









cleaner around the neck and inserted into one of two body ch.mablers

(2.5 or 3.0 liter glass cylinders; 7 cm diameter). The screen

permitted air circulation around the animal. The head and body

chambers were then clamped together and placed in a 240 liter

Forma water bath.

Measurements were made at 15, 25 and 350C over 3-10-hour

periods during daylight. If defecation or urination occurred,

the measurFments were discarded. The animals were weighed to the

nearest 0.1 g before and after each experiment. A diaphragm pump

maintained flow: rates between 40 and 1,200 cc/min to keep the

relative humidity in the chamber approximately between 10 and

30 per ceuL. Flow control acs maintained upstream of the chamber

to prevent a large pressure buildup in the ch-iaber.

The airstream from the body chamber passed through a flowameter

and either to room air, or through a dcw point hygrometer (Cambridge

Systems Model 922-cl). The airstream from the head chamber passed

through a flo.meter and either directly through silica gel and

ascarite tubes (to remove water vapor and carbon dioxide) to a

Beckalnn F3 Parcmagnetic o;:ygen analyzer, or through the hygrometer

and then through the tubes to the oxygen analyzer. Reducing the

oxygen concentration 1 per cent belo- ambient resulted in full-

scale deflection of the chart recorder. Chart recorder output from

both machines provided continuous measurements of 02 consumption

and alternate m eacsurcnxent, of water loss.

At the end of each experiment the animal was removed, the

chamber resealed, and the background dew points detennined at the

experimental flio rates. 'acigro-'.nd dc. points averaged -10.6 and









-20.10C for the head and body chambers respectively. The higher

water content of the head chamber resulted from the lower flow rates

required to obtain sufficient scale deflection from the oxygen

analyzer. Water loss was calculated by subtracting the background

water loss (flow rate x water content) from the experimental water

loss (flow rate x water content). Rates of water losn calculated

from the hygrometer were within 10 per cent of the rates calculated

from weight change of a container of water placed in the syLtem.

The water loss data are presented as a function of the water vapor

pressure deficit (mmHiig) to allow comparisons between values obtained

at different temperatures and relative humidity. The oxygen con-

sumption values were converted to values STPD.

Skin Permeability in Aqueous Media

To measure the unidirectional flow of water from the snakes to

the aqueous media, snakes were injected intraperitoneally with

sufficient tritiated water (THO) in saline (specific activity

100 uc/ml) to obtain an internal concentration of 1-2 uc/ml body

water. The injected volume was less than 1 per cent of the estimated

body water. The snakes were placed in the chamber described above,

and left for one hour to allow the TH'O to become uniformly distributed

in the animal's water space (see Gans et al., 1968). Tneu the body

chamber was filled with 2.0 to 2.5 liters of either distilled water

or filtered sea water (28-29 parts per thousand) and immersed in a

water bath at 250C. A vibrator pump recirculated the fluid through

the body chamber at a rate of 200 ml/min (40.6 cm/min). The snakes

were exposed to each solution for at least four hours. Periodically

1 ml aliquots were removed from the medium and counted in a liquid









scintillation counter using Br-y's solution as a scintillator. After

at least 11 samples had been taken, the snake was removed from the

chamber and 0.1 nl of placma obtained by cardiac puncture was prepared

for counting. The plasma counts vcre corrected for quenching. Room

air was pumped through the head ch;abcr (600 ml/min) to glass tubes

iimerced in a l-methoxy-2-propanol-dry ice bath to freeze dry the

airstream.

The water efflusr: was calculated from the increae of T O0 in

the external mediLta and from the plasma TEO:

Flux = (V*O CR)/(P -t)
X
where the flux is in milliliters per minute, V is the volune in

milliliters of the external comparLtmsnt, AC;i is the iicrease in

THO in the medium during the sampling period, P is the mean T'H
X
count per ml of the body water space during the scr pling period,

and t is the duration in minutes of the sampliin period. The plana

counts corresponding to each snmpling time were calculated by adding

V -* CI/0. 66W

to the final plasma counts; where W equals body weight, and Ci is

the mean TO0 activity during the se.apling period (0.66 is an estimate

of the per cent body water; Bentley, 1959). An alternative value of

0.73 for the per cent body water of freh--wvater reptiles is presented

by Thorson (1960). Flux values usinS this figure are less than

2 per cent greater then when 0.66 is used. The specify c activity

of the bnth rc:ziined a small fraction of that in the animal (103),

so backflu;x was neglected. Sharp breaks in the cu-rvs of CPi against

time were taken to indicate urination and wer;e e:;cluded frcm the

calculations of mean water loss.











RESULTS

Figures 2-10 present resting values for cutaneous water loss,

oxygen consumption and respiratory water loss. Cutaneous water

loss is not affected by the shedding stage until the outer epidermal

generation splits (see Gans et al., 1968). Logarithmic plots are

employed because water loss and oxygen consumption are most simply

expressed as power functions of body weight. Tables 1, 2 and 3

present the sample means and the regression equations calculated

from the logarithmically transformed data by the method of least

squares. The mean and standard error of the cutaneous fraction of

total water loss are presented in Table 4 (only those pairs of water

loss values with similar oxygen consumption were used). Regression

analysis of these data indicates that this fraction remains constant

with changing weight. There is no statistically significant difference

between the TII efflux to fresh and salt water (Fig. 11), although the

mean flux through the skin of both subspecies was higher to fresh

water than to sea water.

Analysis of these data is left to the ensuing section.





















Fig. 2. Cuteneous water loss as a function of wzigh-
in N trix fasciata pictiventris.























Fig. 3. Cutaneous water loss as a function of eight
in N1tri:: f.scinta comprer.sicada.












0 A t t r t-
200

2.2 r
r a

ccl
S100= 2 0



I.B -"x o
0 e
50 *
W -
e 1.6"--".
BODY WATER LOSS
Q13 N. F. PICTIVENTRIS
c F
1.4 250 C
o 15C C
20 V mncrt


1.6 1.8 2.0 2.2 2.4 26 2.6
LOG WEIGHT (G)

50 100 2uU 500
WEIGHT (G)



r -t'---'-r^:t'zc-cLtMatin t-.t[tt___rr-3'.,~rrzrr... -t-t:.,-;r.s' --
2 .2



2.0 x



1.8 X
te x
E o
I,




o 1.4 X X
xx e
BODY WATER LOSS S
1.2 h. COMPRS!CtAUDA 0 D

X 32 C

1.0 25 ce
1 15' '



1.0 1.2 1. 16 1.86 2. 2.0 2.2 2.4 2.6 2.

LOG CODY WEIGHT (C.)

50 1I0 20C' 50C
CODY WEIGHT (G)



































Fig. 4. Cutaneous water loss as a function of weight at
250C. The solid square represents the rate of water
loss of a ThcmLnophis sauritus after a heavy feeding.













2.6


Ic








2.2


t 0 I






0

a1.4 60

I .


BOOY WATER LOSS 25e C
SN. F. COM!PRESSICUO
X H. F. FICIIVENlTRIS
o0 J.. S;URilius C









S1.4 1. 1. 2.0 2.2 2.4 2 2.

LOG BODY V!EIGHT (G)

50 100 200
BODY \WEIIG[IT (G)





















Fig. 5. Oxygen consurlp-iof -4s r, function Of Weight
in Nat.Ycix- fn cicI, pictXivepr!Lri's.























Fig. 6. Oxygeti coiiy~t:npt:ion- -Is a function of x-cight:
in N.trix foscinta cornpr1e.c nuda.












OXYGEN CONSUMPTIqO
o N.F. PICTIVWMTnT
o0 5v c

"0 o 2 C uinr-e
'o STRAIl'E3D


xi Li s
c c

S'0
SLO-







x

LOG BODY WEIGHT (G)


20 50 100 200 500
BODY WEIGHT (G)





OXYGIM CONSUMPTION
Fs c cc-r .CA'
i T. 0 32 C '
251 C
X 150 C






0
1.2 i.4 .0 .4 2.o 2. b
















LOG BCOI WEIGT(G)
S0 0U 20 00 o















BODY ,EIGHT (G)
to 3C"'o' ""o oo> 0


1.0.

x ac





x x
1.6 2.0 2.2 2.4 2.6 2.
LOG OD/ WEIGHT(G)

BODY WEIGHT(G)

























Fig. 7. Oxygen cc'iption as a function of
eightt at 230C.











! OOr C2.0 .. O ox


n 0, I0 io XX e1
::0 2.o




ST. AUTU
0










C- !* -
cC

O OXYGEN CONS:MP11 ON 25"C i


'. o I .' CO .... ":.:.olCAU."I/
L;,C2 --_ 'i------


-" 0"r1 C I. C;A U D:;]A



i i ii
', I n p 7 _' i




.0 .2 1.4 G 2.0 2.2 2.r4 2.6 2.
LOG I3ODY WEIGHT (G)
rTn_ -rr''S~T~~r~C1FCP -C





















Fig. 8. Reaspi ti-ory wi(-er Injs os n f,.,nction of
weight in Na.trix ftscinta pictiventris.

























Fig. 9. Respiratory water loss as a function of
weight in Natrix fascianta conpcressicauda.







19




















l" oI 4
*







1.0- HAD V TER LOSS ,




Sn I 32' C








& 25 oc
K 30"-' s













o 31" C
S 1S C 2





o0 o or
LOG BODY HEIGHT (G)


--arr-tsnaF.---, I0l
BODY MtIGHT IG)






10 I 00 *i


























.O C W(IG rT KO
S S x 0

0 3 O C oo









S
x





1.8 HEAD VPTER LOSS

0 o 32'% I
15'C




1. 6 2 0 2 2 2.4 2.& 281
IOG BODY WE IGT (G)

60 M W IG CX.0 .

























Fig. 10. Respiratory water loss as a function of
weight at 250C. The solid square represents
the rate of water loss of a Th~.nophis sauritus
after a heavy feeding.















, i ,i






',

rY ij

I!


S :;




0 i
n,
c i'
, i
<*




i '
I 'l
', j
'11

I 1
i.j


!A
i.2 o < il
1.2 U=,




1.0 X
c 2
oo
D; 0






,,i nX
0.6




c L 0 X
xX

0.4 HEAD WATER LOSS 25 OC o

,C DC0 N.F. COMPRESSICAUDA .
X N.F. PICTIVENTRIS

0.2 O T. SAURITUS
C N. CYCLOPION

p P. CATENIFER
O0.0W 7_. n n
1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6

LOG BODY W;:GHT (G)
;' --- p



































Fig. 11. Isotope flu:: rates as a function of weight
at 2500. The mean rate of cutaneous water loss
in dry Pir is included.








1.,



1.












0



0.61
i'
| J \



0.8






\ o



1 0
0.2 0




0

0.0
1.80 2.-2 2.
2.0 ~2T 2.4


t N.F. P. S V
X N.F. P. 250C DRY AIR
O N.F. CO:2PRESSICAUDA FYW
* N.F.r.. s
N.F.C. 25 C DRY AI

I




o L 1


i


1'




0




2-3









2.6 2.8 3.'


LOG BODY 'WEIGHT (G)


B100 20 -E0 -
BODY WEIGfIT (G)
















SPECIES

N. f. nictiventris








N. f. ccmDressicauda








T. sauritus


TT'P.

15


25

32

Isotope

15

25

32

Isotope h

25


Sb

151.6

182.7

159.9

278.3

192.1

134.6

136.5

173.3

27.6


TL d

0.36

1.77

2.43

7.19

0.43

0.75

1.60

2.02


WL/mmrNg c

37.56

84.87

63.62



30.81

30.58

41.79



53.99


r e

0.80

0.76

.91i

0.86

0.16

0.57

0.59

0.87

0.68


byx f Sb g

-0.37 0.16

-1.22 0.28

-0.60 0.14

-2.48 0.66

-0.21 0.49

-0.52 0.33

-0.68 0.37

-3.27 0.63

-1.63 1.00


M


--

















SPECIES


N. f. pictiventris








N. f. comprosaicauda






T. ca.-ritus


N


TEMP.


15

25

25 c

32

15

25

32

25


Q1O


172.3

202.1

54.3

159.9

166.9

198.2

159.7

23.4


24.46

62.48

58.56

98.53

33.44

60.82

121.80

80.67


byx


0.05

-0.32

-0.22

-0.45

-0.19

-0.42

0.18

-0.30


0.15

0.43

0.59

0.75

0.45

0.79

0.43

0.30


0.12

0.20

0.80

0.19

0.13

0.12

0.11

0.43


2.49



2.00



1.81

2.68


__


















SPECIES

-f. f. nictiventris





N. f. ccmorecicauda





T. sauritus


TEMP.

15

25

32

15

25

32

25


N


146.1

175.2

167.9

178.7

185.7

146.4

25.2


WL/m-mKf

5.080

6.439

7.324

1.342

6.165

5.255

9.051


WTL

0.072

0.101i

0.233

0.048

0.141

0.178

0.165


r

0.87

0.77

0.72

0.62

0.57

0.38

0.58


byx

-1.27

-0.54

-0.57

-0.99

-0.39

-0.35

-0.70


Sb

0.36

0.18

0.32

0.51

0.27

0.28

0.40


















15iC


N. f. ctiventris'


N. f. coenTresicaud


86.5 : 1.52 a
N = 4

91.4 2.27
N -10


T. snuritus


N. cyclonion


P. catenifer


250C


92.2 0.90
N = 13

30.4 1 3.50
N =6

81.3 + 4.29
N=4

96.8
N=2

72.2
N= 2


90.3 1 ..18
N =5

90.2 1 0.94
N =5


S7EIS .0


SF CIE.T r











DISCUSSION


Cutaneous Water Loss

In snakes, the relationship of total water loss to body weight

has been shown to be similar to that of surface area to body weight

(Cans et el., 1968). It was hoped in this study to determine if

this similarity reflected the importance of cutaneous water loss,

or whether it resulted from a fortuitous combination of independent

relationships of cutaneous and respiratory water loss to body weight.

Table 4 indicates that in these snakes cutaneous water loss does

predominate, but the variability of the data, probably enhanced by

restraining the animals, prevents comparisons of water loss/body

weight and surface nrea/body eight curves. The steepness of the

water loss/body weight slopes probably reflects an increase in

stockiness with an increase in weight, and a change in the relative

amount of the evaporative surface area (see below).

Any variation in the rates of water loss at a common weight

cannot be due to differences in the drying po.er of the air, since

water losts is expressed as a function of the vapor pressure deficit.

Body temperature does not appear to have any clearcut effect on

cutaneous afterr loss (Figs. 2., 3; Table 1). This indicates that

an increase in cmtab:olis~ n nd, presumably, a change in peripheral

perfusion have no effect on the rate of water loss. One snake

even continued to lose uater at its normal resting rate after

death.








Other data, when recalculated to factor out saturation deficit,

also indicate that no saturation deficit-independent increase in

cutaneous water loss occurs with increasing temperature (Bentley and

Schmidt-Nielsen, 1966; ScL-idt-Nielsen and Bentley, 1966). The data

of Dawson et al. (1965) indicate that the relationship between

cutaneous water loss (not corrected for saturation deficit changes)

and ambienL temperature depends upon the species studied. With an

increase in air temperature from 20 to 300C, the saturation deficit

has a Q10 of 1.8; yet one species of lizard showed no increase in

the rate of water loss (Q10 = 1.0; Amphiiolurus ornatus), one had

a Q10 of 1.2, and the rate of water loss of the third species

increased at the same rate as the saturation deficit.

In contrast, Roberts (1968) states that air temperature is

almost as important a determinant of water loss as is saturation

deficit for the lizard Uta stansburiana. Gens et al. (1968) report

that flux of THO to water increases with temperature (Elapbe

climacophora; 27-34OC; Q10 = 2.2). Flux through excised skin in

an osmometer increases with temperature (Q10 = 1.8; Tercafs and

Schoffeniels, 19C5). Thus, the unidirectional flux through the

skin can increase in the absence of a change in the vapor pressure

deficit.

No single factor explains all these data, but habitat aridity

is important. The lizards that show the temperature-independent

water loss have a xeric distribution. However, U. st,_-sburiann

has an increase in cutaneous water loss that is independent of

saturation deficit (Roberts, 1968; Clauson, 1967) yet has the sane

rate of water loss per unit: of surface area as A. ornn-tu (D:n.son ct al.,

1966) and also occurs in arid habitats.








The level of cutancous water loss strongly reflects the species'

ecology (Fig. 4). At 250C cutaneous water loss varies from 5.9 per

cent of the initial body weight per day in the completely aquatic N.

cyclopion, which is 40 per cent of that of an equal-sized amphibian

(calculated from Schmid, 1965), to 0.3 per cent per day in the diuri:al,

desert P. catenifer. Thamnophis sauritus lost water at the rate of

2.1 per cent per day, reflecting its small size and large weight-

specific surface area. The fresh-water N. f. pictiventris loses

water at a mean rate of 4.25 per cent per day, while the salt-water

N. f coma essicauda has a rate equal to a 1.8 per cent per day.

The levels of cutaneous water loss ere similar to those previously

reported for snakes (Prange and Sclhmidt-Nielsen, 1969; Chew and

Darmmann, 1961).

Although water stress is the primary long term determinant of

cutaneous permeability, the snake's body posture, feeding state and

activity provide short term modifications. However, to urdersnand

their effects, some features of reptilian skin must be examined.

Squamate skin has an outer and an inner scalar surface, and a

hinge region. The most superficial layers of all three parts are

keratinized, with B-type (Beta) keratin predominant on the outer

scalar surface and A-type (alpha) keratin predominant on the inner

scalar surfaces and in the hinge region (Mercer, 1961; from Maderson,

1964, 1965). The A-type keratin is flexible and permits expansion

of the skin during breathing, during activity, and after swallowing

large prey. Norailly the' exposed skin surface is the inflexible

B-type keratin of the outer scalar surface. If water diffuses

through the A-type more rapidly than through B-type keratin, then

considerations of water conservation would work against skin distensibility.









I have two types of evidence indicating that the interscalar

surfaces are more permeable than the scalar surfaces. Figure 4

includes data for post-absorptive T. sauritus, and for the same

snake after being fed sufficient fish to spread the scales so that

they no longer overlapped and the interscalar surfaces were exposed.

A five-fold increase in cutaneous water loss resulted, which was not

due to increased activity (oxygen consumption was 1.4 times the

post-absorptivc rate).

MaximTum scale overlap occurs when the snake is straight; overlap

is reduced on the outside of curves. Therefore body position should

modify cutaneous water loss. Figure 12 illustrates a selected segment

of a water loss record with the values plotted as percentage deviations

from an arbitrarily chosen baseline (inactive and curved). Water loss

decreased when the animal switched from a curved sinusoidall as

distinct from coiled) to a straight position. Tight coiling in an

area of low convection would, hoYwever, create a shell of high

humidity air around the snake and reduce the water loss.

Since cutaneous water loss does not increase with increasing

metabolism, the increase in water loss with activity seems to be

due to a flushing out of high humidity pockets under the scales

(see Gans et al., 1968). In this study, cutaneous water loss

increased as much as 3.3 times when the snakes were active. Although

in most cases the increase was less than 2 fold.

If overlapping scales are important in reducing water loss for

most snakes, then a study of the skin structure of achrochordid snakes

would be interesting. These are heterosaline snakes with nonoverlapping

scales (Carpden-.lMaii, 1970).























Fig. 12. The effects of body position and activity
on cutaneous water loss. The data are plotted
as percentage deviations frcm an arbitrarily
chosen baseline equal to the rate of water
loss of the snake when inactive and curved.













"- 01 ( : COPRESS UDA
25"C
20 !13.8 g
SItNACTIVE
SX ACTIVE "z o




CURVED STRAIGHT CURVED rgr 3 r


S-10



-! r M

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 4-1 4 48 .3
TIME (MIN.)








The balance between the need for skin distension and for a reduced

cutaneous water exchange determines the lower limit of cutaneous water

loss. High skin perm cability may even be advantageous by permitting

cutaneous gas exchange (see Crawford and Schultetus, 1970). It

appears that the evolutionary determinacion of this balance point

is sensitive to conditions in the environment,

.Oxygen Consumption

Since respiratory water loss depends upon the ven.tilatory voluY1c,

the o:tygen consumption was examined (Fig. 5, 6; Table 2). Rates of

oxygen consumption are similar to those reported by Galvao et al.

(1965) for snakes, and have temperature sensitivities (Q10) similar

to those previously reported for lizards (Dawson, 1967).

Oxygen consumption of the restrained N. f. pictiventris was 1.46

times that of equal-sized unrestrained specimens. The maximl.'m increase

in oxygen consumption with voluntary activity in the restrained snakes

was 7.6. Although in most cases the increase was between 2 and 6 fold.

N. f. comnress-cauda and N. f. pictiventris have similar rates of o::ygen

consumption at 250C. At 15 and 320C variability of the data and

uncertainty about the regression lines prevent statistical separation

of the two subspecies (Fig. 7; Table 2).

Respiratory Uater Loss

Just as with oxygen consumrpticn, the rates of respiratory water

loss of thie two subspecies of !J. fasci- t were equal at 250C

(Fig. 10; Table 3). Th1nriohis sauritus, w ich had a lower rate

of oxygen consumption than N'ntrix, also had a lo-er rate of respiratory

water loss. At 15 and 320C there is considerable overlap of respiratory

water loss valuet- of the two N.ntrix subspecies. Statistical separation

of the two fon. s is again not possible.








The ecologically raticnalizable trends of respiratory water loss

cited by Bentley and Schmidt-Nielsen (1966) for lizards, turtles, and

crocodilians do not emerge from the data on snakes. The respiratory

water loss of N. taxispilotq equals that of P. catenifer (Prange and

Schmidt-Nielsen, 1969). However, elevated rates of oxygen consumption

in these two species resulted in rates of respiratory water loss that

are greater than those reported in Table 3. In the snakes that have

been studied the respiratory water loss is mainly a fnnrtion of

oxygen consumption.

The relationship between oxygen consumption and respiratory water

loss is not straightforward. Figures 8 and 9 and Table 3 reveal

little increase in saturation deficit-independent water loss between

15 and 32C, while oxygen consumption showed a Q10 in excess of 2.0

throughout the temperature range.

At a common oxygen consumption, the amount of water lost per

milliliter of oxygen consumed decreases only slightly with increasing

temperature, but at any one temperature except for N. f. compressi cud

at 150C, respiratory water loss increases as a fractional exponential

function of oxygen consumption that is independent of activity and body

weight (Figs. 13, 14 include both resting and active values. One con-

sequence of having water loss per milliliter of oxygen consumed decrease

with increasing c:ygen consumptirn is that respiratory water loss

increases with body size and activity loss quickly than oxygen

consumption (Tabl].e 2, 3). Therefore, in terms of water loss, the

cost of activity is reduced relative to the change in oxygen con-

sumption. Water loss continues to decrease relative to oxygen con-

sumnption and saturation deficit until evaporative cooling is actively

increased by panting (Templeton; 1960, Dawson and Tcmpleton, 1963).




















Fig. 13. U!ter loss per milliliter of oxygen consu.r-ed eas
a function'nn nof oxygen cnn-',rtio"nn in N.trrix frci tan
pictiventxis and Tihamnophin, saurituso Thit graph
includes wvlues for inctive and active animals.























Fig. 14. Uater loss per milliliter of oxygen consumed as
a function of o::ygce consumption in INatrix fascia-a
compiressi. a, N.. cyjclopion and Fituophis cateni for.
This graph includes values for inactive and active
animals.














28 z
N.F. PICTIVENTXIS

aS x 15, C
S 25 C

21 0 32 C
2X V T SAU1'ITUS 25 C

X

V0

0




0o 00 *o





LI

C X XPI



















*1 25C C
0





1 2
V



























o -.
-j-
00 0.2 04 0.6 0 8 10 1.2 14 .6 1.8 2.0 2
LOG OXYGEN CONSU'PTIO (MIL/HR)

2.0 .O o0o 2 o0 00
OXYGEN CONSUMPTION (ML/VHR)













t c
2.6 25 C "t










x



0 00 3 C








00
I' c







0 0 02 0.4 0.6 0.6 1.0 2 14 I.G 1. 2.:

LOG OXYGEN CCOjSU',P;TION (;,LL/HIR)
2 5 50 2050 100
OXYGEN CCOISUV"PnTION (ML/HR)








Minute volume and oxygen extraction were examined to uncover the

respiratory mechanisms of the decrease of water loss per milliliter

of oxygen consumed at higher rates of metabolism, NIinute volume was

calculated from the water locs data by dividing the respiratory water

loss (g/inin) by the density of water in saturated air at ambient

temperature (m!ng/l; Figs. 15, 16), It w\as assum-aed that the exhaled

air was at ambient temperature. Oxygen extraction was calculated

from the r.inute volume and oxygen consumption (Dejours, 1966;

assuming RQ = 0.7; Figs. 17, 18, 19).

The savings in respiratory iater loss are achieved primarily

through the relative reduction in minute volume with increasing

oxygen constumption (Figs. 15, 16). However, at resting metabolic

rates, the lack of temperature dependence of ventilation rate (4-8

breaths/min at 25 and 320C) indicates that tidal volume is increasing

with temperature, and the relative dead space ventilation is decreasing

with temperature. This results in further vtcer savings at high

temaperatures. Unfortunrtely I have insufficient data on breathing

rates to calculate tidal, volumes corresponding to each wai:er loss

point.

Partitiow-n? of Ulatter LOS

Within a Ospcies, the partitioning of rater loss reflects the

level of respiratory after loss since cutaneou5 water loss is

temperature independent (Table 4) The respiratory component does

not increase greatly with temperature and the parti.tioninig remains

within r.nrrow bounds, 'JThe changes it; partitioning are more complex

in lizards since there re a species differences; in the temperature

sensitivity of cutc.neous water loss rs well ,s active aug-,mentnation

of respiratory water lons at hb.ihr tcrmpe'ratures.





















Fig. 15. Minute volume as a function of o::ygc.n consuption
in Natrix frsciata pictiventris and Thanr-,ophis sauritts.
This graph includes values for inactive and active
animals.
























Fig. 16. Minute volume as a function of oxygen constmnption in
NatJrix fascinta comprescicau0n. This graph includes
values for inactive and active animals.














Fig. 17. The fraction of oxygen in the expired nir as a
oxygen constmption in N .trix fa3csct:a .pticventris.
includes values for inactive and active animals.


function of!
Th!is graph


Fig. 18. The fraction of oxygen in the expired air as a function of
o:4gen consumption in Nt.rix fasc-ita comresnicaudca. This graph
includes values for inactive and active animals.



























0


x x


MItiUFE VOLUME.

N.E PICTIVENTRIS
X 150L

25 C

0 32 C


T. SAURITUS
D ,0


r* V 25




o0 20 30 40 50 L E 0 80 90 100 110 12

OXYGEN CONSUMPTION (ML/HR)








9 9











K


MINUTE VOLUME

N.F. COMPtRSSICAUD;1
X 15S C

25 C

0 2 'C


so x


1u 2 1j Y4 5 2. It 8) 3) 1. Ili )2

OXYiEN CONSUMVPTIO\ (MIL /HR)


2









0

S



xO
*
X M
Q


0 0
x0 0
x 0
K
I


~"P~--- ~V~"~"iT~T~J"-~~r"~"~-r;rr-Y~-I~-r5~I~L~ I


































Fig. 19. The fraction of o:qgen in the expired air as a
function of oxygen consumption in Thnmnophis snuritus.
This graph include values for inactive and active
animLals.











S. B CXYrSN EUiRV nI -

ii "
x* u'c
x x 25 aC
x 1 x is *c

7. x .
tO * x x t




rX
I- a


rx'aGn CONsu 10tron (MPip.)






,+L. -^ .5


F e .


a


r* o





15






113
1i
!i I

1 a





.I', .3"P i !



































Fig. 19. The fraction of or.ygen in the expired air as a
function of oxygen consumption in Thsmnophis sauritus.
This graph include; values for inactive and active
animals.















I'o OXYCLE EXTRACTION c
I T. SAURILUS
O 0
0o o 25 C









0
I 0 I




*-











___ ,- '1 rl -- ..
10 420 30 40 5
OXYGEN CONSUMPTION
(M L/HR)








In these snakes, the ecology of the species dictates the

differences in partitioning since the magnitude of the cutaneous

component is ecologically labile, while the respiratory fraction

is dependent upon the organism's demand for oxygen. The level of

the basal rate of metabolism is probably also in part a function

of the animal's ecology, but is more directly a function of size.

The cutaneous fraction of water loss found in this study

(72.9-96.8 per cent) is similar to values from the literature for

reptiles of comparable ecology (for review, see Schmidt-Nielsen,

1969). The cutaneous fraction of N. cyclopion is higher than any

previously reported.

Water Exchnge With Aqueous Media

Since one finds high skin permeability in air occurring in

animals that normally live in or near fresh water and thus are

subjected to hydration stress, data on the relative rates of water

influx and efflux should be reviewed. The literature is chaotic,

with reports of equal permeability in both directions, efflux but

no influx, and no flux in either direction.

Tercafs and Schoffeniels (1965) and George (1947) present data

on excised reptile skin that indicate equal permeability in both

directions. Tercafs and Schoffeniels (1965) also report that

certain lizards (but no snakes) gain weight when the animal is in

contact with free water. Fluorescein dye placed in the water did

not appear in the digestive tract. This was interpreted as indicating

influx through the C-lin.

Capillary flow along the scales to the mouth followed by drinking

was shcw~ to be the mechanism of allegedly cutaneous water uptake of

the desert lizard Moloch horridus (Bentley and Blu-er, 1962). Th'ey








speculate that a similar nechauism is responsible for the other records

of apparent water uptake through the skin. Krakcuer et al. (1968)

report that several burrowing reptiles (nnakes and amphisbaenians)

that lose water at the same rate as equl-siized anphibians were

unable to regain water efter de hydration, unless the mouth ;as able

to contact the substrate (moist sand or wet sponge). Cloudsley-

Thompson (1966) m~'ae similar observations on Nile crocodiles.

Pe-ti',a (1958) repo-rs that nVo wi~tcr -asses through the excised

skin of ~. f. co-fluens (fresh-"weter) and N. f. cla ki (a brackish-

watert subspecies). An o.smrotic gradient ;es created across the ckin

from salt ;ater to distilled water, with either solution in contact

with the external skin surface. Presumed death of the skin after

24 hours in the osmomueter did not increase the water flux. Pettus'

data were the basis of the sta-tcient that reptilian water loss is

primarily r-spirttory (Che:, 1961).

It is obvious that Pettus' conclusion is erroneous; it may have

been due to a high rcsistatce 1ma,:.ineter fluid coupled with skin

distension that charged the volume in his oszmomte: ccampartments.

The results that indicate little wite influx; may reflect natural

conditions more closely than those indicating syianmetrical skin

pcrmenability because they were performed on intact animals,

The isotope experir.cnts in this study (Fig. 11) measure only

one-1way gross per:eability, and do not indicate .c t flow7 of licqid.

l!;rever, since tle isotope measurements were made in the presence

of an osmotic gradic:nt (Lody-fresh water, body-salt water), bulk

flow along the gradient should interact with the diffusional flow;

(see lhotaif, et rl,, 1969). The isotope fliu r-:ins't the ocmotic









gradient to distilled water is the same as the flux with the gradient

to sea water, thus indirectly supporting an hypothesis of a reduced

or asymmetrical membrane permeability. Unfortunately no reliable

measurements of direction and magnitude of net flux were obtained.

Pettus (1963) reports weight loss of around 0.6 per cent per

day for N. f. clarki immersed in both distilled and salt water. As

is N. f. compressicauda, this is a brackish-water snake that

rnori.aliy relies on the preformed water in its diet to keep in water

balance. Therefore, since the snakes were kept without food, and

since they do not drink salt water, the rates of weight loss represent

the rate of water loss plus the weight loss due to metabolism. The

weight loss due to metabolism is probably small. The rate of weight

loss in water is lower than the rate of water loss of N. f. comr.ressicaiida

in air (1.8 per cent per day) and much lower than the apparent rates

from the isotope flux rates (9.6 per cent per day; value normalized

to the mean weight of the animals used in the dry air runs at 250C).

Since the weight of the animals used in his study was not given, the

rates of water loss cannot be rigorously compared.

If the osmotic flow of water is small when the animal is in

water, then an explanation must be found for the high diffusion

rates through wetted skin observed by Tcrcafs and Schoffeniels

(1965), Gans et al. (1968) and in this study. Tercafs and Schoffeniels

(1965) suggest that snakes can regulate permeability depending upon

its surroundings, and support this suggestion by pharmacologically

moclify.ng water loss rates (atropine and curare increase flux rates

across wetted skin). Gans et a]_. (1968) suggest that hydration of

the skin proteins, analogous to the mechanism suggested for insects







(Becment, 1961), may be responsible for the increase in the diffusion

rate across wetted skin. A third possible explanation that is

consistent with the data of Cans et al. (1968) and the data presented

here concentrates on the difference in magnitude of the scouring

forces exerted on the skin under different conditions.

The scouring force is exerted parallel to the skin and determines

the width of the inmobile shell around the animal. This force may be

caic.' ate; by the following equation:

T= Cf p U

where T is the scouring force in dynes/cm2, Cf is the coefficient

of friction, 0 is the density of the medium, and U equals the

velocity of the medium. Cf depends upon the roughness of the

surface, and the Reynolds number (RL; Sutton, 1953; Knudsen and

Katz, 1958).

The Reynolds number is dimensionless and depends upon the

velocity of the medium, the length of the flopath, and the

kinematic viscosity (Newtonian viscosity/density)o The Reynolds

numbers, calculated using the mean water and air flow rates, were

in that dubious zone between numbers that indicate laminar or

turbulent flow. (air 49.1, water 574.2). Laminarity waas assumed,

allowing Cf to be calculated from the formula:

Cf = 1.328/(RL)0.5

The scouring force with water as the external medium will be

21.08 times larger than the force with air as the external medium.

The rate of diffusion will depend upon the steepness of the gradient.

The more rapidly the scouring of molecules away from the surface of

the skin, the steeper the gradient that will be maintained across

the skin and thus the greater the flux.









A ratio of flux rates to water and to air of up to 21 times

can be rationalized. The observed ratios were 5.3 for I. f.

coprlressicauda and 11.3 for N. f. pictiventris. The increase in

diffusion of water to water may just reflect the fluid dynamics of

a flowing aquatic medium.

The quantitative meaning of the unidirectional flux measurements

is not clear, although these determinations indicate that the brackish-

water tr.>: has a lower skin permeability in water 'th-"! d0oes the

fresh-z;ater fomn. Apparently dehydration stress exerts greater

selective pressure than does hydration stress. Unfortunately the

relationship between water loss and water gain is still not known.








CONCLUSIONS

Cutaneous water loss in snakes can be modified by structural

means and by behavioral changes in posture and habitat selection.

However, the structural means m!ust be compatible with feeding

and locc'otino, .lhicli deim.nid;, free spreading of the scales

(Maderson, 1964), and the behavioral changes must remain consistent

with the animal's energetic needs, e.g. activity rhyth.ns are ret by

the times of prey availability.

Physiologically and ecologically the conclusions appear clear.

The brackish-water snake, N. f. compressiccaudd, has reduced skin

permeability both in %water and on land. This is appropriate for a

snake that inhabits saline areca, and lacks accessory salt glan-d

(SchmidL"tielsen and Fange, 1958). Its prey are vertebrates, which

provide a good source of free water, so it is subject to dehydration

stress no more severe than is a desert reptile. The r. ora moder ..te

temperatures of the water sn::ke's habitat decrease the problem of

maintaining a water balance. Fresh-water snakes are seldom under

dehydration stress, and can benefit from the non-pulmonary gas

exchange normally associated v.ith high water permea-bility. If there

is an increz sed influx in fresh water associated wlth the higher skin

permeability of aquatic snakes, then the kidneys iust get rid of the

added water load.

Respiratory water loss can be changed by behavioral or by

anatomical and physiological modifications. l:hspiratory water loss

depends upon the total pulmonary ventilation, the humidity of the

50









inspired air, and the temperature of the expired air (since it is

saturated with water vapor). It can be reduced by reducing the

oxygen consumption to decrease the ventilation volume. But the

level of the basal rate of metabolism is determined by the

homeostatic demands of the organism. Apparently N. f. pictiventris

and N. f. compressicac d are similar enough that in spite of differences

in the water streC;s of the environment, vater conservation cannot be

achieved by reducing the rate of Lietabolis-.l. T. sauritus has a

reduced respiratory water loss that corresponds to its lowered

rate of metabolism.

Since respiratory water loss ai: rest is only a small fraction of

the total water locs, it appears that the most significant adaptations

are not those that decrease the resting rate, but are those that

decrease the rate of increase with activity. Ventilation volume

is reduced by an increase in oxygen extraction at higher rates

of metaboliFm. This results in water savings, but requires changes

in the acid-base balance and respiratory characteristics of the blood.

Reduction of the relative dead-space ventilation by increasing tidal

volume also serves to decrease the ventilation volume corresponding

to a given rate of metabolism.








SUMMARY

The ecological and physiological control of water loss was studied

in closely related but ecologically distinct snakes. The main con-

clusions are:

1. Cutaneous water loss measured in dry moving air does not increase

with increasing temperature when the water loss is expressed as a

function of the vapor pressure deficit.


2. Cutaneous water loss correlates with the water availability in

the natural environment of each of the species and varies from 5.9

per cent of the initial body weight per day for the wholly aquatic

Natrix .cjclooion to 0.3 per cent per day for the desert snake

Pituophis catenifer. The fresh-water N. fasciata pictiventrjs had

a iiti'gheI -,kin permeability (4.25 per cent pir udy) Laiii ie 1~I ckib

water N. f. compressicauda (1.8 per cent per day).


3. In the interscalar regions of the skin permeability is higher

than in the scalar regions, resulting in changes in cutaneous water

loss with changes in posture, activity and feeding state.


4. Rates of respiratory water loss are not ecologically rationalizable,

but rather depend upon the level of oxygen consumption.


5. Respiratory water loss when expressed as a function of oxygen

consumption and saturation deficit increases as a fractional po.-er

of oxygen consumption. Respiratory water" loss therefore increases

less rapidly with activity than does oxygen coniSmpt:'.on,

52






53


6. Cutaneous water loss contributes between 72 and 97 per cent of the

total water loss. With the rate of respiratory water loss fiy.xd by the

level of oxygen consumption, the differences in water loss partitioning

are due to the ecologically rationalizable differences in cutaneous

water loss.

7. Skin permeability measured in aqueous media is greater in N. f.

Ecitiventris than in N. f. compressicauda.












REFERENCES


Beament J. W. L. (1961) The water relations of insect cuticle.
Biol. Rev. 36, 281-320.

Bentlev P. J. (IO59) S-,,dipi on 'tli wnar .nd nrl PlPct--, l retn oli
of the lizard Trachysaurus :ruosus. J. Physiol. 145, 37-47.

Bentley P. J. and Bluimer W. F. C. (1962) Uptake of water by the
lizard Moloch horridus. Nature 194, 699-700.

Bentley P. J. nr:d Schmidt-Nielsen K. (1966) Cutaneous water loss
in reptiles. Science 151, 1547-1549.

Bogert C and Cowles R. (1947) Results of the Archbold expedition,
No. 58. Moisture loss in relation to habitat selection in so.,e
Floridian reptiles. Amer. Mus. Novitates 1358, 1-34.

Campden-Main S. M. (1970) A field guide to the snakes of South Vietnam,
Division of Reptiles and Amphibians. U. S. Natl. Mus. Smithsonian
Inst., Washington. 1-114.

Chew R. M. (1961) Water metabolism of desert inhabiting vertebrates.
Biol. Rev. 36, 1-31.

Chew P.. M. Eani DatiIIanii A. E. ( ..I) Evapuz-tLive :c2aLT-r Uoss of small
vertebrates, as measured with an infrared noalyser. Science 133,
384-385.

Clarsen D. L. (1967) Studies of water loss in two species of lizards.
Comp. Biochem. Physiol. 20, 115-130.

Cloudsley-Thompson J. L. (1968) Water relations of crocodiles.
Nature 220 (5168), 708.

Crawford E. C., Jr. and Schultetus R. R. (1970) Cutaneous gas exchange
in the lizard Sauromalus obesus. Copoia 1970, 179-180.

Dawson 17. R. (1967) Interspecific variation in physiological response
of lizards to temperature. 230-257. Jn W. U. Milstead (ed)
Lizard Ecology: A Symposium." Univ. Missouri Press, Columbia.

Dawson W. R., Shoemaker V. H. and Licht P. (1966) Evaporative water
losses of so.e small Australi.en lizards. Ecology 47, 589-594.

Dawson W. R. nod Templeton J. R. (1963) Physiologic:: responses to
temperature in the lizard Cro: '_tus collrris. Physiol. Zool.
36, 219-236.







Dejours P. (1966) Respiration. Oxford Univ. Press, New York.

Galvao P. E., Tarasautchi J. and Guertzenstein P. (1965) Heat pro-
duction of tropical snakes in relationship to body v:cight and
body surface. Am. J. Physiol. 209, 501-506.

Gans C., Krakauer T. and Paganolli C, V. (1968) Water loss in snakes:
Interspecific and intraspecific variability. Comp. Biochem.
Physiol. 27, 747-761.

George J. C. (1947) A comparative study of permeability to water of
the sk i in nomne reprrenent;'itive verrt:ebhrar tes- ... Univ. BonM y ..16
28-32.

Knudsen J. c and Kat? D. L. (1958) Fluid Dynamics and Heat T.ransfer.
McGrarw-lill Look Co., New York.

Krakauer T., GaCrs C. and Paga;nlli C, V. (196;) Ecological correlation
of water loss in burrowing reptiles. Nature 218 (5142), 659-660.

Naderson P. F. A. (1964) The skin of lizards and snake. Brit. J.
Herpetology 3, 151-154.

Naderson P. F. A. (1965) Histological changes in the cpidemnis of
snakes during the sloughing cycle. J. Zool. 146, 98-113.

Nercer E. H1. (1961) Keratin and Keratinization. Perg.non, .ondon,

Motais R., Isaia J., Rankin J. C. and iaetz J. (1969) Adaptive changes
of the water permeability of the teleostean gill epithelium in re-
lation to external salinity. J. Exp. Biol. 51, 529-546.

Pettus D. (1958) Water relations in Natrix sipjedon. Copeia 1958,
207-211.

Pettus D. (1963) Salinity and subspeciation in IlatrJx si2edon.
Copeia 1963, 499-504.

Prange II, D. and Schmidt-Nic!sen K. (1969) Evaporative water loss in
snckes. Cop. Biochlm. Physiol. 28, 973-975.

Roberts L. A. (1968) Water loss in the desert lizard Uta rtansburiana.
Com.p. Biochen. Physiol. 27, 583-589.

Schmid W. D. (1965) Some aspects of the water economics of nin'
species of eniphibians. Ecology 46, 261-269.

Schimidt-Nielsern :.. (1969) The neglected interface: The biology of
water as a liquid-gas system. Quart. Rev. Biophys. 2, 283-304.

Sclmidt-Niielsen K. and Bentley P. J. (1966) Desert tortoise CGoherus
.r.aSazii: Cutrneous water loss. Science 154, 911.

Scluhidt-Njolron K. and Fnge R. (1958) Salt glandc in marin reptiles.
Nature 182, 783-785.





56


Sutton 0. G. (1953) Micrometerology. McGrt...-Hill Book Co., New York.

Tabb D. C., Dubrow D. L. and Manning R. B. (1962) The ecology of
northern Florida Bay and adjacent estuaries. State of Florida,
Board of Conservation Tech. Ser. No. 39.

Templeton J. R. (1960) Respiration and water loss at higher
temperatures in the desert igauno, Dipsosaurus dorsalis.
Physiol. Zool. 33, 135-145.

Tercafs P. R. and Schoffeniels E. (1965) Phcnom:ene's de permrabilitc
u nuve2u de aI pc-u dcs reptiles. A.,. Roy. Soc. Zool. Il.lii.
96, 9-22.

Thor~on T. .. (1968) Body fluid partitioning in reptilia. Copeij
1968, 592-501.










BIOGRAPHICAL SKETCH

Thomas Henry Krakauer was born September 6, 1.942, at Buffalo,

New York. In June, 1959, he was graduated from The Park School

of Buffalo. In June, 1964, he received the degree of Bachelor of

Art, ai'h a major in Biology from the University of Rochester. In

June, 1966, he received the degree of Master of Science with a major

in Biology from the University of Miami. In 1966 he enrolled in the

Graduate School of the University of Florida. He worked as an Inter.im

Instructor in the Departnent of Zoology until June, 1967. His work

toward the degree of Doctor of Philosophy was further supported by a

National Defense Education Act, Title IV Fello-w.ship and a University

of Florida Graduate School Fellowship.

Thomas Henry Krakauer is married to the former Janet MacColl.











This dissertation was prepared under the direction of the

cha:irl-,a-i o1 the candidate's supervisory comr.ittee and has been

approved by all members of that committee. It was submitted to the

Dcai of th, College of Arts and Sciences and to the Graduate

Cou;nci, and was approved as partial fulfillment of the requirements

for the degree of Doctor of Philosophy.

August, 1970.





Dean, College f Arts -d Sciences




Dean, GrCduate School
Supervlisory Cormii ttec:



1CiL/dk iL fi?
Cli ;nirin



A l^





<=E^ yv^J _(2. l d fo^




The ecological and physiological control of water loss in snakes
CITATION SEARCH THUMBNAILS DOWNLOADS PDF VIEWER PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00097724/00001
 Material Information
Title: The ecological and physiological control of water loss in snakes
Added title page title: Water loss in snakes
Physical Description: viii, 57 leaves : illus. ; 28 cm.
Language: English
Creator: Krakauer, Thomas Henry, 1942-
Publisher: University of Florida
Place of Publication: Gainesville
Gainesville
Publication Date: 1970
Copyright Date: 1970
 Subjects
Subjects / Keywords: Snakes -- Physiology   ( lcsh )
Zoology thesis Ph. D
Dissertations, Academic -- Zoology -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Bibliography: Bibliography: leaves 54-56.
Additional Physical Form: Also available on World Wide Web
General Note: Manuscript copy.
General Note: Thesis - University of Florida.
General Note: Vita.
 Record Information
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000549683
oclc - 13272884
notis - ACX3979
System ID: UF00097724:00001

Downloads

This item has the following downloads:

( PDF )

( INSTR )


Full Text






THE ECOLOGICAL AND PHYSIOLOGICAL CONTROL

OF WATER LOSS IN SNAKES













By
THOMAS HENRY KRAKAUER













A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY










UNIVERSITY OF FLORIDA
1970















ACKNOWLEDGEMENTS


I would like to thank Drs. Brian K. McNab, Frank G. Nordlie

and Daniel A. Belkin for help during the course of the study and

during the preparation of this manuscript. Dr. John F. Anderson

and Mr. Paul E. Moler helped in many ways including serving as

valuable sounding boards for some of my ideas. My wife, Janet,

helped in all the ways that wives usually do, and helped in the

preparation of the illustrations.













TABLE OF CONTENTS

Acknowledgements............................ .......****......... ii

List of Tables.................................................. iv

List of Figures................................................. v

Abstract........................................................ vii

Introduction................... .......... ..... 00* .......********** **.. 1

Materials and Methods........................................... 3

Skin Permeability in Air................................... 3

Skin Permeability in Aqueous Media......................... 7

Results............................................*...*........ 9

Discussion...................................................... 28

Cutaneous Water Loss ....................................... 28

Oxygen Consumption........................................ 34

Respiratory Water Loss.................................... 34

Partitioning of Water Loss................................. 38

Water Exchange with Aqueous Media.......................... 45

Conclusions............ ..................... ..... .......... 50

Summary ..... ... .............................. ............. ..... 52

References...................................................... 54

Biographical Sketch ........... ......................... ..... 57












LIST OF TABLES

Table 1 Cutaneous water loss................................. 24

Table 2 Oxygen consumption............................... 25

Table 3 Respiratory water loss............................... 26

Table 4 Percent of total water loss that is cutaneous......... 27













LIST OF FIGURES

Figure 1 Diagram of equipment................................. 5

Figure 2 Cutaneous water loss as a function of weight

in Natrix fasciata pictiventris ...................... 11

Figure 3 Cutaneous water loss as a function of weight

in Natrix fasciata compressicauda.................... 11

Figure 4 Cutaneous water loss as a function of weight

at 25C.............................................. 13

Figure 5 Oxygen consumption as a function of weight

in Natrix fasciata pictiventris...................... 15

Figure 6 Oxygen consumption as a function of weight

in Natrix fasciata compressicauda.................... 15

Figure 7 Oxygen consumption as a function of weight at

25oc ................................................. 17

Figure 8 Respiratory water loss as a function of weight

in Natrix fasciata pictiventris...................... 19

Figure 9 Respiratory water loss as a function of weight

in Natrix fasciata compressicauda.................... 19

Figure 10 Respiratory water loss as a function of weight

at 25C..................... ......... .............. ... 21

Figure 11 Isotope flux rates as a function of weight

at 25C...................... ............ ........... 23

Figure 12 The effects of body position and activity on

cutaneous water loss.............. .... .............. 33







Figure 13







Figure 14








Figure 15





Figure 16



Figure 17





Figure 18





Figure 19


Respiratory water loss per milliliter of oxygen

consumed as a function of oxygen consumption in

Natrix fasciata pictiventris and Thamnophis

sauritus............................ *...............

Respiratory water loss per milliliter of oxygen

consumed as a function of oxygen consumption in

Natrix fasciata compressicauda, Natrix cyclopion

and Pituophis catenifer ...............................

Minute volume as a function of oxygen consumption

in Natrix fasciata pictiventris and Thamnophis

sauritus.......... ....... .........................

Minute volume as a function of oxygen consumption

in Natrix fasciata compressicauda....................

The fraction of oxygen in the expired air as

a function of oxygen consumption in Natrix

fasciata pictiventris......... ..................

The fraction of oxygen in the expired air as

a function of oxygen consumption in Natrix

fasciata compressicauda..............................

The fraction of oxygen in the expired air as a

function of oxygen consumption in Thamnophis

sauritus............................................













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

THE ECOLOGICAL AND PHYSIOLOGICAL CONTROL OF WATER LOSS
IN SNAKES

By

Thomas Henry Krakauer

August, 1970

Chairman: Brian K. McNab
Major Department: Zoology

The cutaneous and respiratory water loss of Natrix fasciata

pictiventris, N. f. compressicauda and Thamnophis sauritus were

measured using a flowing air system and a dew-point hygrometer.

Cutaneous water loss, when expressed as a function of the vapor

pressure deficit, did not change with temperature. Cutaneous

water loss was almost three times greater in N. f. pictiventris

(fresh water) than in N. f. compressicauda (salt water) and

amounted to 92 and 80 per cent (250C) of the total water loss

respectively.

Determinations of skin permeability in aqueous media by in vivo

measurements of tritiated water flux indicate that external salinity

has no effect on the efflux from the animal. The skin permeability

of N. f. pictiventris in water is greater than that of N. f.

compressicauda.


vii






No differences, between subspecies, in respiratory water loss

were found at 15, 25 and 320C. Respiratory water loss of Thamnophis

was lower, correlating with a lower level of oxygen consumption. A

decrease in the amount of water lost per ml of oxygen consumed tended

to keep the respiratory water loss constant with increased temperature

(corrected for saturation deficit). Respiratory water loss increases

less rapidly with activity than does oxygen consumption.

The rate of respiratory water loss is not modified, except by

changes in oxygen consumption, while the cutaneous permeability is

responsive to ecological demands.


viii













INTRODUCTION

Reptiles are able to remain in water balance under a wide

range of environmental conditions. Marine and terrestrial species

live in dehydrating media, while fresh-water reptiles are exposed

to an environment that favors the entry of water into the animal.

Water loss takes place at three sites: cloaca, respiratory tract,

and skin. This study deals with the cutaneous and respiratory

components.

Classically, reptiles were viewed to have an impermeable

integument, with the respiratory tract stated to be the route of

most of the water loss (Chew, 1961; Pettus, 1958). Recent studies,

however, indicate that rates of water loss of reptiles vary from

those typical of "waterproof" insects to those of amphibians

(see Schmidt-Nielsen, 1969; Krakauer et al., 1968), and that the

major component of water loss is cutaneous (Bentley and Schmidt-Nielsen,

1966; Schmidt-Nielsen and Bentley, 1966; Clausen, 1967; Dawson et al.,

1966; Prange and Schmidt-Nielsen; 1969). The rates of water loss

correlate with habitat aridity (Bogert and Cowles, 1947; see also

Krakauer et al., 1968; Gans et al., 1968).

In this study I will examine the water loss of three closely

related, but ecologically distinct, snakes: a brackish-water snake,

the mangrove water snake (Natrix fasciata compressicauda); a fresh-water

snake, the Florida banded water snake (N. f. pictiventris); and a more




2


terrestrial snake, the ribbon snake (Thamnophis sauritus). The

Natrix and Thamnophis live near water but differ in the amount of

time spent in water. The desert gopher snake (Pituophis catenifer)

and the green water snake (Natrix cyclopion) were also studied. The

variability of water loss is examined to determine how the rate of

water loss can be modified. Data are presented on the changes in

cutaneous and respiratory water loss with changes in ecology, body

weight, temperature, activity, shedding cycle, and external medium

(dry air, distilled water, and sea water).













MATERIALS AND METHODS

Natrix fasciata pictiventris, N. cyclopion, and T. sauritus

were collected at several localities in Alachua County, Florida.

The specimen of P. catenifer was obtained without data from an

animal dealer in southern California. N. f. compressicauda were

collected at Matheson Hammock, Dade County, Florida and in Monroe

County, 35 miles southwest of Homestead, Dade County, Florida.

These sites were pioneer and secondary mangrove association with

salinity that varied with rainfall (14-41 parts per thousand;

Tabb et al., 1962).

The Natrix and Thamnophis were maintained in the laboratory

on a diet of fish in dry cages, and held at 25 + 10C with 12 hours

of light daily. Tap water was provided. The Pituophis was kept

under similar conditions except at fluctuating room temperature

(ca.260C). All water loss measurements were made on animals after

they had fasted for at least one week and at known times during the

shedding cycle.

Skin Permeability in Air

A flowing air desiccation system was used to determine water

loss to dry air (Fig. 1). The respiratory and cutaneous water

losses were measured separately. A partition between a head and

body chamber was created by sliding the punctured tip of a tapered

latex balloon (Trojan-enz No. 175) around the animal's neck, and

the open end over a collar on the plexiglas head chamber (320 ml).

The snake's body was tied to a coarse mesh wire screen by a pipe

3






































0 :3
0.
4) u
0H W






) (0 0






OS
6 <4





w .0
bOl44J
aot
r4 m5
0




.1.41











k- X -


----1 --- ---
1 \ I i







cleaner around the neck and inserted into one of two body chambers

(2.5 or 3.0 liter glass cylinders; 7 cm diameter). The screen

permitted air circulation around the animal. The head and body

chambers were then clamped together and placed in a 240 liter

Forma water bath.

Measurements were made at 15, 25 and 350C over 3-10-hour

periods during daylight. If defecation or urination occurred,

the measurements were discarded. The animals were weighed to the

nearest 0.1 g before and after each experiment. A diaphragm pump

maintained flow rates between 40 and 1,200 cc/min to keep the

relative humidity in the chamber approximately between 10 and

30 per cent. Flow control was maintained upstream of the chamber

to prevent a large pressure buildup in the chamber.

The airstream from the body chamber passed through a flowmeter

and either to room air, or through a dew point hygrometer (Cambridge

Systems Model 922-cl). The airstream from the head chamber passed

through a flowmeter and either directly through silica gel and

ascarite tubes (to remove water vapor and carbon dioxide) to a

Beckman F3 Paramagnetic oxygen analyzer, or through the hygrometer

and then through the tubes to the oxygen analyzer. Reducing the

oxygen concentration 1 per cent below ambient resulted in full-

scale deflection of the chart recorder. Chart recorder output from

both machines provided continuous measurements of 02 consumption

and alternate measurements of water loss.

At the end of each experiment the animal was removed, the

chamber resealed, and the background dew points determined at the

experimental flow rates. Background dew points averaged -10.6 and







-20.1C for the head and body chambers respectively. The higher

water content of the head chamber resulted from the lower flow rates

required to obtain sufficient scale deflection from the oxygen

analyzer. Water loss was calculated by subtracting the background

water loss (flow rate x water content) from the experimental water

loss (flow rate x water content). Rates of water loss calculated

from the hygrometer were within 10 per cent of the rates calculated

from weight change of a container of water placed in the system.

The water loss data are presented as a function of the water vapor

pressure deficit (mmHg) to allow comparisons between values obtained

at different temperatures and relative humidity. The oxygen con-

sumption values were converted to values STPD.

Skin Permeability in Aqueous Media

To measure the unidirectional flow of water from the snakes to

the aqueous media, snakes were injected intraperitoneally with

sufficient tritiated water (THO) in saline (specific activity

100 uc/ml) to obtain an internal concentration of 1-2 uc/ml body

water. The injected volume was less than 1 per cent of the estimated

body water. The snakes were placed in the chamber described above,

and left for one hour to allow the THO to become uniformly distributed

in the animal's water space (see Gans et al., 1968). Then the body

chamber was filled with 2.0 to 2.5 liters of either distilled water

or filtered sea water (28-29 parts per thousand) and immersed in a

water bath at 250C. A vibrator pump recirculated the fluid through

the body chamber at a rate of 200 ml/min (40.6 cm/min). The snakes

were exposed to each solution for at least four hours. Periodically

1 ml aliquots were removed from the medium and counted in a liquid







scintillation counter using Bray's solution as a scintillator. After

at least 11 samples had been taken, the snake was removed from the

chamber and 0.1 ml of plasma obtained by cardiac puncture was prepared

for counting. The plasma counts were corrected for quenching. Room

air was pumped through the head chamber (600 ml/min) to glass tubes

immersed in a 1-methoxy-2-propanol-dry ice bath to freeze dry the

airstream.

The water efflux was calculated from the increase of THO in

the external medium and from the plasma THO:

Flux = (V-A CPM)/(P .t)
X
where the flux is in milliliters per minute, V is the volume in

milliliters of the external compartment, &CPM is the increase in

THO in the medium during the sampling period, P_ is the mean THO
X
count per ml of the body water space during the sampling period,

and t is the duration in minutes of the sampling period. The plasma

counts corresponding to each sampling time were calculated by adding

V.*P'/0.66W

to the final plasma counts; where W equals body weight, and CPM is

the mean THO activity during the sampling period (0.66 is an estimate

of the per cent body water; Bentley, 1959). An alternative value of

0.73 for the per cent body water of fresh-water reptiles is presented

by Thorson (1968). Flux values using this figure are less than

2 per cent greater than when 0.66 is used. The specific activity

of the bath remained a small fraction of that in the animal (103,

so backflux was neglected. Sharp breaks in the curves of CPM against

time were taken to indicate urination and were excluded from the

calculations of mean water loss.













RESULTS

Figures 2-10 present resting values for cutaneous water loss,

oxygen consumption and respiratory water loss. Cutaneous water

loss is not affected by the shedding stage until the outer epidermal

generation splits (see Gans et al., 1968). Logarithmic plots are

employed because water loss and oxygen consumption are most simply

expressed as power functions of body weight. Tables 1, 2 and 3

present the sample means and the regression equations calculated

from the logarithmically transformed data by the method of least

squares. The mean and standard error of the cutaneous fraction of

total water loss are presented in Table 4 (only those pairs of water

loss values with similar oxygen consumption were used). Regression

analysis of these data indicates that this fraction remains constant

with changing weight. There is no statistically significant difference

between the THO efflux to fresh and salt water (Fig. 11), although the

mean flux through the skin of both subspecies was higher to fresh

water than to sea water.

Analysis of these data is left to the ensuing section.



















Fig. 2. Cutaneous water loss as a function of weight
in Natrix fasciata pictiventris.





















Fig. 3. Cutaneous water loss as a function of weight
in Natrix fasciata compressicauda.










- *- -- -,
h Il


200








80
C,.








S50 ,








20 m=


i.61 *0
BODY WATER LOSS
--r-f .1..

1.4 25 C
o 15C



1.6 1.8


.-- -- --- 2.2
2.0 2.2


LOG WEIGHT (G)

50 lu,
WEIGHT (G)


D



2.4 2.6


2.8

51.l1l


x



e

0f


. .3
- D ;. .


x X


BODY WATER LOSS *
1.2 N. F. COMPRESS!CAUDA 0 x

X 32 *c

1.0 *a 25 c
0 15c



1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.1

LOG BODY WEIGHT (G)


50 100 200 500
BODY WEIGHT (G)


* B


1!~

I00
80 c




.3:
50



a- SO *
en
<


p *


-%






























Fig. 4. Cutaneous water loss as a function of weight at
250C. The solid square represents the rate of water
loss of a Thamnophis sauritus after a heavy feeding.

















C
r L.A
*C x









X
X


0 0





0
-

-

0 .9

BODY WATER LOSS 25C
N N.F COMPRESSICAUDA
S N. F. PICTIVENTRiS
3 J. SAURI IraS


P P, CATENIFER

P
p




.4 1.6 1. 2.0 2.2 2.4 26 2.8- -
1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8


LOG BODY WEIGHT (G)


- ..
so


100 200
Y WEIGHT (G)


C




C


C


1.2


- oo



















Fig. 5. Oxygen consumption as a function of weight
in Natrix fasciata pictiventris.






















Fig. 6. Oxygen consumption as a function of weight
in Natrix fasciata compressicauda.




















020



2.0' B. *"-' ^ ,.,


300]
o200 -



0



o
80 Z



z


x -J
0


300=



-200







20I
0














20c.


OXYGEN CONSUMPTION
N.F IC.lTIV lT:" "
0 32.
25 C

&TRAINED
X 1I C


^ ^ e ^ ~-~-^ a ^, C^ ^
0

63 0 *.L
a2 B O e~

U U B


0.05 15 C X -
X1


1.0 1.2 1.4 1.6 2i. 2.0 2.2 2.4 2.fA,
LOG BODY WEIGHT (G)


20 50 100 200 500
BODY WEIGHT (0)


x x
- _- _--7 -- - ..- .X.-.. .-.
16 1.8 20 :, 24 2, 2.8
LO ; & ll", '.EilI I("
p. .._ p-- __ .--._=_- .....
It.L ,, v J
L s:0., BO [O,;H (,)


1.6



1.4



































44
0


*r4









0
0C










*rt
0
"4




o m











ri






























<



o 0 ;
ua :-

si al6 -1 KL
~, ~' < I


X o -


^ -^ .--^-.-^ -^^ ^ ^ ^^ __. ,.,,------ ^
o 0




0- N3 A- 0

a a aL+ II. ++,+ ,


Esflg
AC)








k;0
rO







0~9
L!J





0

Be



















Fig. 8. Respiratory water loss as a function of
weight in Natrix fasciata pictiventris.
























Fig. 9. Respiratory water loss as a function of
weight in Natrix fasciata compressicauda.




















0 0








X


x


HEAD WATER LOSS
N.F.PICTIVENTRIS
o 32' C
25' C


7 .. 15



LOG BODY WEIGHT (G)
,reL-.-.,, -- ':--. -_ . 8 ,I-.,- -
a', i, ...-., ,. -J, .
B[,',' '. ".H' I I


Tg a HEAD WATER LOSS
N.F. COMP'E SSIT'UT
Q 32% x

T.6 1. 15 'C

x


LOG BODY WEIGHT (G)


50 WIT(
BODY WEIGHT (G)


bo-
| 5.0


300


on'.


0.8 E-
































44 f


4U .1
4. S .w4
$ a


0




to
e W 94
4 0 0
U 0. *rI




cO


04

0 C3
W 0

0-4 4




4 0 0
J3 -4.v4





i 4* *
Co u

6 0
044
a*r 00
600a
0 :440

600
i 0 ..
o 0^ 4















xc


(J x.


oo
0
e,.


D 1


- o D c


((9!.-dH t>-9/9Tr)SS01 yWIIvA DO]


(0 CO ) C .-

01^-2H 1^ -O/H1^ SS 01 831VIA


r0


:r)
F*-

1C1




:) CO











JM
is


c-i



































Fig. 11. Isotope flux rates as a function of weight
at 25oC. The mean rate of cutaneous water loss
in dry air is included.




23







0 N.F. PICTIVENTRIS F W
S\ N.F. P. SW
S\ X N.F. P. 250C DRY AIR
1.4 U \o N.F. COMPRESSICAUDA FW c
S\ 0 N.F.C. o SW
0 \ N.F.C. 25 C DRYAIR




I II
10 1.0









0.00
CD
a 0.8










2 0





1 0.06 1



2.0 2.2 2.4 2.6 2.8 3.0

LOG BODY WEIGHT (G)


100 200 500
BODY WEIGHT (G)
















%0 Go
-4 C4
* N



0 0



Io cM
*4




o '0
o o
oo
O



'0 0





\0 1
(n I.-





* *
o *-4








'0 IF-

r- 4.




*
M- N
(v oo









1-4


-4
o *




0 0








0 0
o I







S4 -4










(N I
* I





on r-.
0 C


Ij, I.-
- ('s1


o o
* *





-4 C,.
N '.0
o o
* *





-4 U,

0 0



I4 I-
o o


I- W 0










I I I






0 0 0



o N4

N- (N


1-4 co I 0i
* S *

On On Nt Ln

o o fn4 S 0




- 4 -4 4


%.o Ch ON 0 0 In km


ull tn ( 0 Ln LM o 0
0 0
o 0
I- -I





*r4


4 1

4 %4


0





0





4J








Ca
0 w

2to
1- 1
0-4

Ii










.0




0
"4 4 0











0 a a0
0 0 "4 44
^. ^ 0 0.-4










SW4044
0 :i 0 14 UV
0 -1 0- 0v4 0 0 0
0 C 0 1t



14 U r.. (
fa g $.1 0 o






im 44 af
0)0 U' 04 M 0


0
0
0


14
0


44

0
3
0
0







i-4

(-4





25









% 0 00
0 -t 0 0 N o
1-4 C.







.0 * . . *
M 0 0 0 0 0 0 0 0



(4 C14 Ln 0% c(' 00 0
X 0 rn 4 cn



a * * * *
.0 0 0 0 0 0 0 0 0




un cn aO O O ON cn o
l-i ~ L 4- ^ /t i r-^ 4l cn
~~rt ~C C; 0% IC f % ;




S0 0 0 0 0 0 0


.0
14 NT 4 oo m 0 * * * *
4 M 0 -14 0



('4 %O N 0% Cs '0 C* COC

(3 (q m %D 00 01 4n
4 4 0 CM 0% C

















OO CM &OO O O (
N 0 Ul 1 i l %D a% ( C4
--4 -4 -4


00 N' %0 %0 0 m% (n N-
z*
-4~~- $4 -4-
T-4



0 w
E- 000V
0 6
m L un 1f C'4 Un LI C14 in
~ ~ 0 m/ u"> m S i ui( iu>J
0 r-4 C'. ('4 ('^ ,-4 ('4 CV ('4 O C
0 E-w 0
4.1 *Cl4 4 Q
& q4.4 & 4.4

0 01 2a
co
0 (344 0 0l

9 4J -A0w

w0 r. w r-4
6w w W 0 t>0 -r
S> w0 >1 2
0 00 Ai


I- 0 'l 0-' 02 B <
E-4 CO ZI U iI 0).0 0-
^-1 0041 n4

co
pQ 0La E41* *





26







S00 N -4 r*% 00 0
(.a -4 c n ts 1M 4
A0 * . . .
to 0 0 0 0 o 0 0



t.. r-. 0% 0% t 0
C4 N Ir >n m M M N
* . *
A0 -4 0 0 0 0 0 0
I I I I I I I


r00 N N o c00
C4 00 r n I .0 -4 co 'A
* S S S
o 0 0 0 0 0 0


t%. 0 C' *4 4f r. '0
0 1- N 0 D
: * L 1 *
0 0 0 0 0 0 0







1 O O^ 4 N 'A ur 4f

E a n %0 co %a Go




t o. o r 0 4 N



I'A 4l a0 00 '0 0 00



w a 1-1 tn C- n C4 L
**
0-4 m-44N m ~
-4
-4





00
*



DrID
**




0 w a
0 *M <- (M C>IC4






u0
0 4
a c

14 a0
o 1.4





040



W i .
01 *1




-0 C- I 0
a >o a Bj



















00 N
$4 0!
co .t.I I
-0 0 o
a C4 +1 11 +1 11
> en a
cva MZ C4
o 0 o

*0

0

14
0

t


0
) 0 c+ + +



0 C' 00 I I0 a"I
4.



-4





0 L* C4

0 0 i,
(' 0 0 0 N













,u1- UD Z M










40 0 0:
0 4M
4a 44 4 I






D3 0

0o r4 0

0 0
o .



I44
0 o $4




0 "M o 0
0 -4 *
S.0
* 04
4:1 w
to P
H W.












DISCUSSION


Cutaneous Water Loss

In snakes, the relationship of total water loss to body weight

has been shown to be similar to that of surface area to body weight

(Cans et al., 1968). It was hoped in this study to determine if

this similarity reflected the importance of cutaneous water loss,

or whether it resulted from a fortuitous combination of independent

relationships of cutaneous and respiratory water loss to body weight.

Table 4 indicates that in these snakes cutaneous water loss does

predominate, but the variability of the data, probably enhanced by

restraining the animals, prevents comparisons of water loss/body

weight and surface area/body weight curves. The steepness of the

water loss/body weight slopes probably reflects an increase in

stockiness with an increase in weight, and a change in the relative

amount of the evaporative surface area (see below).

Any variation in the rates of water loss at a common weight

cannot be due to differences in the drying power of the air, since

water loss is expressed as a function of the vapor pressure deficit.

Body temperature does not appear to have any clearcut effect on

cutaneous water loss (Figs. 2, 3; Table 1). This indicates that

an increase in metabolism and, presumably, a change in peripheral

perfusion have no effect on the rate of water loss. One snake

even continued to lose water at its normal resting rate after

death.







Other data, when recalculated to factor out saturation deficit,

also indicate that no saturation deficit-independent increase in

cutaneous water loss occurs with increasing temperature (Bentley and

Schmidt-Nielsen, 1966; Schmidt-Nielsen and Bentley, 1966). The data

of Dawson et al. (1966) indicate that the relationship between

cutaneous water loss (not corrected for saturation deficit changes)

and ambient temperature depends upon the species studied. With an

increase in air temperature from 20 to 300C, the saturation deficit

has a Q10 of 1.8; yet one species of lizard showed no increase in

the rate of water loss (Q10 = 1.0; Amphibolurus ornatus), one had

a Q10 of 1.2, and the rate of water loss of the third species

increased at the same rate as the saturation deficit.

In contrast, Roberts (1968) states that air temperature is

almost as important a determinant of water loss as is saturation

deficit for the lizard Uta stansburiana. Gans et al. (1968) report

that flux of THO to water increases with temperature (Elaphe

climacophora; 27-34oC; Q10 = 2.2). Flux through excised skin in

an osmometer increases with temperature (Q10 = 1.8; Tercafs and

Schoffeniels, 1965). Thus, the unidirectional flux through the

skin can increase in the absence of a change in the vapor pressure

deficit.

No single factor explains all these data, but habitat aridity

is important. The lizards that show the temperature-independent

water loss have a xeric distribution. However, U. stansburiana

has an increase in cutaneous water loss that is independent of

saturation deficit (Roberts, 1968; Clausen, 1967) yet has the same

rate of water loss per unit of surface area as A. ornatus (Dawson et al.,

1966) and also occurs in arid habitats.








The level of cutaneous water loss strongly reflects the species'

ecology (Fig. 4). At 250C cutaneous water loss varies from 5.9 per

cent of the initial body weight per day in the completely aquatic N.

cyclopion, which is 40 per cent of that of an equal-sized amphibian

(calculated from Schnid, 1965), to 0.3 per cent per day in the diurnal,

desert P. catenifer. Thamnophis sauritus lost water at the rate of

2.1 per cent per day, reflecting its small size and large weight-

specific surface area. The fresh-water N. f. pictiventris loses

water at a mean rate of 4.25 per cent per day, while the salt-water

.N. f, compressicauda has a rate equal to a 1.8 per cent per day.

The levels of cutaneous water loss are similar to those previously

reported for snakes (Prange and Schmidt-Nielsen, 1969; Chew and

Dammann, 1961).

Although water stress is the primary long term determinant of

cutaneous permeability, the snake's body posture, feeding state and

activity provide short term modifications. However, to understand

their effects, some features of reptilian skin must be examined.

Squamate skin has an outer and an inner scalar surface, and a

hinge region. The most superficial layers of all three parts are

keratinized, with B-type (Beta) keratin predominant on the outer

scalar surface and A-type (alpha) keratin predominant on the inner

scalar surfaces and in the hinge region (Mercer, 1961; from Maderson,

1964, 1965). The A-type keratin is flexible and permits expansion

of the skin during breathing, during activity, and after swallowing

large prey. Normally the exposed skin surface is the inflexible

B-type keratin of the outer scalar surface. If water diffuses

through the A-type more rapidly than through B-type keratin, then

considerations of water conservation would work against skin distensibility.








I have two types of evidence indicating that the interscalar

surfaces are more permeable than the scalar surfaces. Figure 4

includes data for post-absorptive T. sauritus, and for the same

snake after being fed sufficient fish to spread the scales so that

they no longer overlapped and the interscalar surfaces were exposed.

A five-fold increase in cutaneous water loss resulted, which was not

due to increased activity (oxygen consumption was 1.4 times the

post-absorptive rate).

Maximum scale overlap occurs when the snake is straight; overlap

is reduced on the outside of curves. Therefore body position should

modify cutaneous water loss. Figure 12 illustrates a selected segment

of a water loss record with the values plotted as percentage deviations

from an arbitrarily chosen baseline (inactive and curved). Water loss

decreased when the animal switched from a curved sinusoidall as

distinct from coiled) to a straight position. Tight coiling in an

area of low convection would, however, create a shell of high

humidity air around the snake and reduce the water loss.

Since cutaneous water loss does not increase with increasing

metabolism, the increase in water loss with activity seems to be

due to a flushing out of high humidity pockets under the scales

(see Gans et al., 1968). In this study, cutaneous water loss

increased as much as 3.3 times when the snakes were active. Although

in most cases the increase was less than 2 fold.

If overlapping scales are important in reducing water loss for

most snakes, then a study of the skin structure of achrochordid snakes

would be interesting. These are heterosaline snakes with nonoverlapping

scales (Campden-Main, 1970).






























^ON
>% 10

"4 04
>U 4
"4 0 "4t 1 *0
*u 1 r1 4 u
0 14 0
S4J 4
go 14 4 U 1
g oao'a.w
0 E-.S0'

o c a
41 >4




0" w 'u
o>##
1 0 8 "4
0a 0 r4J
0M 00
.0 0 r-4 0




M Q< &
0 '-44 0,-
"-4 U a
co u4r u0'

48 0a 610-

44 9 0 141
40 w4.8 QA


4) o > a 0
c41 4 04.
AOCU '44
46J


E44J 14 0 0
0i 041 %

3 1 a W
.o ca
00
E- 0A 0

v-4 0 0 U 9-4

60
'"4




















































Y3







C)C


F E













.(.0
e I :1







L |















rr-
9




















SI



















SSO7 U31VM 3'.,.H) %







The balance between the need for skin distension and for a reduced

cutaneous water exchange determines the lower limit of cutaneous water

loss. High skin permeability may even be advantageous by permitting

cutaneous gas exchange (see Crawford and Schultetus, 1970). It

appears that the evolutionary determination of this balance point

is sensitive to conditions in the environment.

Oxygen Consumption

Since respiratory water loss depends upon the ventilatory volume,

the oxygen consumption was examined (Fig. 5, 6; Table 2). Rates of

oxygen consumption are similar to those reported by Galvao et al.

(1965) for snakes, and have temperature sensitivities (Q10) similar

to those previously reported for lizards (Dawson, 1967).

Oxygen consumption of the restrained N. f. pictiventris was 1.46

times that of equal-sized unrestrained specimens. The maximum increase

in oxygen consumption with voluntary activity in the restrained snakes

was 7.6. Although in most cases the increase was between 2 and 6 fold.

N. f. compressicauda and N. f. pictiventris have similar rates of oxygen

consumption at 250C. At 15 and 32C variability of the data and

uncertainty about the regression lines prevent statistical separation'

of the two subspecies (Fig. 7; Table 2).

Respiratory Water Loss

Just as with oxygen consumption, the rates of respiratory water

loss of the two subspecies of N. fasciata were equal at 250C

(Fig. 10; Table 3). Thamnophis sauritus, which had a lower rate

of oxygen consumption than Natrix, also had a lower rate of respiratory

water loss. At 15 and 320C there is considerable overlap of respiratory

water loss values of the two Natrix subspecies. Statistical separation

of the two forms is again not possible.







The ecologically rationalizable trends of respiratory water loss

cited by Bentley and Schmidt-Nielsen (1966) for lizards, turtles, and

crocodilians do not emerge from the data on snakes. The respiratory

water loss of N. taxispilota equals that of P. catenifer (Prange and

Schmidt-Nielsen, 1969). However, elevated rates of oxygen consumption

in these two species resulted in rates of respiratory water loss that

are greater than those reported in Table 3. In the snakes that have

been studied the respiratory water loss is mainly a function of

oxygen consumption.

The relationship between oxygen consumption and respiratory water

loss is not straightforward. Figures 8 and 9 and Table 3 reveal

little increase in saturation deficit-independent water loss between

15 and 320C, while oxygen consumption showed a Q10 in excess of 2.0

throughout the temperature range.

At a common oxygen consumption, the amount of water lost per

milliliter of oxygen consumed decreases only slightly with increasing

temperature, but at any one temperature except for N. f. compressicauda

at 150C, respiratory water loss increases as a fractional exponential

function of oxygen consumption that is independent of activity and body

weight (Figs. 13, 14 include both resting and active values. One con-

sequence of having water loss per milliliter of oxygen consumed decrease

with increasing oxygen consumption is that respiratory water loss

increases with body size and activity less quickly than oxygen

consumption (Tables 2, 3). Therefore, in terms of water loss, the

cost of activity is reduced relative to the change in oxygen con-

sumption. Water loss continues to decrease relative to oxygen con-

sumption and saturation deficit until evaporative cooling is actively

increased by panting (Templeton, 1960; Dawson and Templeton, 1963).

















Fig. 13. Water loss per milliliter of oxygen consumed as
a function of oxygen consumption in Natrix fasciata
pictiventris and Thamnophis sauritus. This graph
includes values for inactive and active animals.





















Fig. 14. Water loss per milliliter of oxygen consumed as
a function of oxygen consumption in Natrix fasciata
compressicauda, N. cyclopion and Pituophis catenifer.
This graph includes values for inactive and active
animals.
























































































SIC1


0.0 0.2 04 0.6 0.8 1.0 1.2 14 1.6
LOG OXYGEN CONSUMPTION (ML/HR)


OXYGEN CONSUMPTION (ML/HR)




x N.F. COMPRESCA
X 15 C
25 C
0 32 C

C N.CYCLOPION

P P.CATE NFER
SC


x




0 0


00 0 2 0 4 0 6 0 10 I 1 1.6 .6 2.0 2.2
LOG OXYGEN CONSUMPTION (ML/HR)

2 5 10 20 50 100
OXYGEN CONSUMPTION (ML/HR)


28 X

N.f PICTIVENTRIS

aZs x 15" C
X 25"C


2.1 0 32 C

X V T SAURITUS 25 C
X
2 *v V a X


0 **

0


0 0





0 V X x

1.4


300'









I f 0



20.




lo.


IO-








10-


300.










I-




60J







10]





10"


1.G 2.0 2.2


I.

a





a








Minute volume and oxygen extraction were examined to uncover the

respiratory mechanisms of the decrease of water loss per milliliter

of oxygen consumed at higher rates of metabolism. Minute volume was

calculated from the water loss data by dividing the respiratory water

loss (mg/min) by the density of water in saturated air at ambient

temperature (mg/ml; Figs. 15, 16). It was assumed that the exhaled

air was at ambient temperature. Oxygen extraction was calculated

from the minute volume and oxygen consumption (Dejours, 1966;

assuming RQ = 0.7; Figs. 17, 18, 19).

The savings in respiratory water loss are achieved primarily

through the relative reduction in minute volume with increasing

oxygen consumption (Figs. 15, 16). However, at resting metabolic

rates, the lack of temperature dependence of ventilation rate (4-8

breabhs/min at 25 and 320C) indicates that tidal volume is increasing

with temperature, and the relative dead space ventilation is decreasing

with temperature. This results in further water savings at high

temperatures. Unfortunately I have insufficient data on breathing

rates to calculate tidal volumes corresponding to each water loss

point.

Partitioning of Water Loss

Within a species, the partitioning of water loss reflects the

level of respiratory water loss since cutaneous water loss is

temperature independent (Table 4). The respiratory component does

not increase greatly with temperature and the partitioning remains

within narrow bounds. The changes in partitioning are more complex

in lizards since there are species differences in the temperature

sensitivity of cutaneous water loss as well as active augmentation

of respiratory water loss at higher temperatures.

















Fig. 15. Minute volume as a function of oxygen consumption
in Natrix fasciata pictiventris and Thamnophis sauritus.
This graph includes values for inactive and active
animals.























Fig. 16. Minute volume as a function of oxygen consumption in
Natrix fasciata compressicauda. This graph includes
values for inactive and active animals.


























0


* a
0

xO S



o 0 "


MINUTE VOLUME

N.E. PICTIVENTRIS
X 1]SC

?s C

0 32 C


T. SAURITUS
V Is"


4 .u Wu 4* Lu u e ) Ilk ij i01

OXYGEN CONSUMPTION (ML /HR)




42 0

40








30 0

28



-. 24


MINUTE VOLUME

N.F. COMPRESSICAUDA
X i50 C


x 0 25.s

S0 32' C

x x
X

.. CO S PT I (0 IR
OXYGEN CONSUMPTION (ML /HR)


48"


* 0
0

xo


0
* o

























44 .0
0

0 60
ObO


*U C
'a
0a T
g
44






0.
4 I
Cc
go "41
0 93










UCO
w "M4-
105.
4.11



*O 4
o 0



w< V
K044



C: 0





0 *rl
60
. U4







5410

60 141-9


0 B4
00h
B. 0.

B4 0 <
0 r
Tl00 C



p09


.0
CL
440c

00

0

93
0 w
44-A F


0 .





u E-4

4 0
0 `
*Hco w





ce
M te
K u
-A4 (d


J 0
*00



0
04464
-p4



0 $4
a049 49
M 0
0.4
4. 04
0 ".4





00
4 1


44 09 >



*
oM
E-4 0 *0

60;0


F44


























S-




F**

, .


KJ
a k


* *



% *
* *
xc

** *



- - a. -
V. -
0^


A I"








_ 1










- < a
4< o



VI

*














H4 0
cie
aS







a I a
0 .a
- -


































Fig. 19. The fraction of oxygen in the expired air as a
function of oxygen consumption in Thamnophis sauritus.
This graph includes values for inactive and active
animals.















20 ,''o OXYGEN EXTRACTION
20. o
I T. SAURITLUS
o o0 250 C
0 V 150 C
19


I
18L
S0
17 o








15



14




0 10 20 30 40 50
OXYGEN CONSUMPTION
(ML/HR)








In these snakes, the ecology of the species dictates the

differences in partitioning since the magnitude of the cutaneous

component is ecologically labile, while the respiratory fraction

is dependent upon the organism's demand for oxygen. The level of

the basal rate of metabolism is probably also in part a function

of the animal's ecology, but is more directly a function of size.

The cutaneous fraction of water loss found in this study

(72.9-96.8 per cent) is similar to values from the literature for

reptiles of comparable ecology (for review, see Schmidt-Nielsen,

1969). The cutaneous fraction of N. cyclopion is higher than any

previously reported.

Water Exchange With Aqueous Media

Since one finds high skin permeability in air occurring in

animals that normally live in or near fresh water and thus are

subjected to hydration stress, data on the relative rates of water

influx and efflux should be reviewed. The literature is chaotic,

with reports of equal permeability in both directions, efflux but

no influx, and no flux in either direction.

Tercafs and Schoffeniels (1965) and George (1947) present data

on excised reptile skin that indicate equal permeability in both

directions. Tercafs and Schoffeniels (1965) also report that

certain lizards (but no snakes) gain weight when the animal is in

contact with free water. Fluorescein dye placed in the water did

not appear in the digestive tract. This was interpreted as indicating

influx through the skin.

Capillary flow along the scales to the mouth followed by drinking

was shown to be the mechanism of allegedly cutaneous water uptake of

the desert lizard Moloch horridus (Bentley and Blumer, 1962). They







speculate that a similar mechanism is responsible for the other records

of apparent water uptake through the skin. Krakauer et al. (1968)

report that several burrowing reptiles (snakes and amphisbaenians)

that lose water at the same rate as equal-sized amphibians were

unable to regain water after dehydration, unless the mouth was able

to contact the substrate (moist sand or wet sponge). Cloudsley-

Thompson (1968) made similar observations on Nile crocodiles.

Pettus (1958) reports that no water passes through the excised

skin of N. f. confluens (fresh-water) and N. f. clarki (a brackish-

water subspecies). An osmotic gradient was created across the skin

from salt water to distilled water, with either solution in contact

with the external skin surface. Presumed death of the skin after

24 hours in the osmometer did not increase the water flux. Pettus'

data were the basis of the statement that reptilian water loss is

primarily respiratory (Chew, 1961).

It is obvious that Pettus' conclusion is erroneous; it may have

been due to a high resistance manometer fluid coupled with skin

distension that changed the volume in his osmometer compartments.

The results that indicate little water influx may reflect natural

conditions more closely than those indicating symmetrical skin

permeability because they were performed on intact animals.

The isotope experiments in this study (Fig. 11) measure only

one-way gross permeability, and do not indicate net flow of liquid.

However, since the isotope measurements were made in the presence

of an osmotic gradient (body-fresh water, body-salt water), bulk

flow along the gradient should interact with the diffusional flow

(see Motais et al,, 1969). The isotope flux against the osmotic







gradient to distilled water is the same as the flux with the gradient

to sea water, thus indirectly supporting an hypothesis of a reduced

or asymmetrical membrane permeability. Unfortunately no reliable

measurements of direction and magnitude of net flux were obtained.

Pettus (1963) reports weight loss of around 0.6 per cent per

day for N. f. clarki immersed in both distilled and salt water. As

is N. f. compressicauda, this is a brackish-water snake that

normally relies on the preformed water in its diet to keep in water

balance. Therefore, since the snakes were kept without food, and

since they do not drink salt water, the rates of weight loss represent

the rate of water loss plus the weight loss due to metabolism. The

weight loss due to metabolism is probably small. The rate of weight

loss in water is lower than the rate of water loss of N. f. compressicauda

in air (1.8 per cent per day) and muchilower than the apparent rates

from the isotope flux rates (9.6 per cent per day; value normalized

to the mean weight of the animals used in the dry air runs at 250C).

Since the weight of the animals used in his study was not given, the

rates of water loss cannot be rigorously compared.

If the osmotic flow of water is small when the animal is in

water, then an explanation must be found for the high diffusion

rates through wetted skin observed by Tercafs and Schoffeniels

(1965), Gans et al. (1968) and in this study. Tercafs and Schoffeniels

(1965) suggest that snakes can regulate permeability depending upon

its surroundings, and support this suggestion by pharmacologically

modifying water loss rates (atropine and curare increase flux rates

across wetted skin). Gans et al. (1968) suggest that hydration of

the skin proteins, analogous to the mechanism suggested for insects







(Beament, 1961), may be responsible for the increase in the diffusion

rate across wetted skin. A third possible explanation that is

consistent with the data of Gans et al. (1968) and the data presented

here concentrates on the difference in magnitude of the scouring

forces exerted on the skin under different conditions.

The scouring force is exerted parallel to the skin and determines

the width of the immobile shell around the animal. This force may be

calculated by the following equation:

T=Cfp U2

where T is the scouring force in dynes/cm2, Cf is the coefficient

of friction, p is the density of the medium, and U equals the

velocity of the medium. Cf depends upon the roughness of the

surface, and the Reynolds number (RL; Sutton, 1953; Knudsen and

Katz, 1958).

The Reynolds number is dimensionless and depends upon the

velocity of the medium, the length of the flowpath, and the

kinematic viscosity (Newtonian viscosity/density). The Reynolds

numbers, calculated using the mean water and air flow rates, were

in that dubious zone between numbers that indicate laminar or

turbulent flow (air 49.1, water 574.2). Laminarity was assumed,

allowing Cf to be calculated from the formula:

Cf = 1.328/(RL)05.

The scouring force with water as the external medium will be

21.08 times larger than the force with air as the external medium.

The rate of diffusion will depend upon the steepness of the gradient.

The more rapidly the scouring of molecules away from the surface of

the skin, the steeper the gradient that will be maintained across

the skin and thus the greater the flux.







A ratio of flux rates to water and to air of up to 21 times

can be rationalized. The observed ratios were 5.3 for N. f.

compressicauda and 11.3 for N. f. pictiventris. The increase in

diffusion of water to water may just reflect the fluid dynamics of

a flowing aquatic medium.

The quantitative meaning of the unidirectional flux measurements

is not clear, although these determinations indicate that the brackish-

water Natrix has a lower skin permeability in water than does the

fresh-water form. Apparently dehydration stress exerts greater

selective pressure than does hydration stress. Unfortunately the

relationship between water loss and water gain is still not known.














CONCLUSIONS

Cutaneous water loss in snakes can be modified by structural

means and by behavioral changes in posture and habitat selection.

However, the structural means must be compatible with feeding

and locomotion, which demands free spreading of the scales

(Maderson, 1964), and the behavioral changes must remain consistent

with the animal's energetic needs, e.&. activity rhythms are set by

the times of prey availability.

Physiologically and ecologically the conclusions appear clear.

The brackish-water snake, N. f. compressicauda, has reduced skin

permeability both in water and on land. This is appropriate for a

snake that inhabits saline areas, and lacks accessory salt glands

(Schmidt-Nielsen and Fange, 1958).. Its prey are vertebrates, which

provide a good source of free water, so it is subject to dehydration

stress no more severe than is a desert reptile. The more moderate

temperatures of the water snake's habitat decrease the problem of

maintaining a water balance. Fresh-water snakes are seldom under

dehydration stress, and can benefit from the non-pulmonary gas

exchange normally associated with high water permeability. If there

is an increased influx in fresh water associated with the higher skin

permeability of aquatic snakes, then the kidneys must get rid of the

added water load.

Respiratory water loss can be changed by behavioral or by

anatomical and physiological modifications. Respiratory water loss

depends upon the total pulmonary ventilation, the humidity of the

50







inspired air, and the temperature of the expired air (since it is

saturated with water vapor). It can be reduced by reducing the

oxygen consumption to decrease the ventilation volume. But the

level of the basal rate of metabolism is determined by the

homeostatic demands of the organism. Apparently N. f. pictiventris

and N. f. compressicauda are similar enough that in spite of differences

in the water stress of the environment, water conservation cannot be

achieved by reducing the rate of metabolism. T. sauritus has a

reduced respiratory water loss that corresponds to its lowered

rate of metabolism.

Since respiratory water loss at rest is only a small fraction of

the total water loss, it appears that the most significant adaptations

are not those that decrease the resting rate, but are those that

decrease the rate of increase with activity. Ventilation volume

is reduced by an increase in oxygen extraction at higher rates

of metabolism. This results in water savings, but requires changes

in the acid-base balance and respiratory characteristics of the blood.

Reduction of the relative dead-space ventilation by increasing tidal

volume also serves to decrease the ventilation volume corresponding

to a given rate of metabolism.













SUMMARY

The ecological and physiological control of water loss was studied

in closely related but ecologically distinct snakes. The main con-

clusions are:

1. Cutaneous water loss measured in dry moving air does not increase

with increasing temperature when the water loss is expressed as a

function of the vapor pressure deficit.


2. Cutaneous water loss correlates with the water availability in

the natural environment of each of the species and varies from 5.9

per cent of the initial body weight per day for the wholly aquatic

Natrix cyclopion to 0.3 per cent per day for the desert snake

Pituophis catenifer. The fresh-water N. fasciata pictiventris had

a higher skin permeability (4.25 per cent per day) than the brackish

water N. f. compressicauda (1.8 per cent per day).


3. In the interscalar regions of the skin permeability is higher

than in the scalar regions, resulting in changes in cutaneous water

loss with changes in posture, activity and feeding state.


4. Rates of respiratory water loss are not ecologically rationalizable,

but rather depend upon the level of oxygen consumption.


5. Respiratory water loss when expressed as a function of oxygen

consumption and saturation deficit increases as a fractional power

of oxygen consumption. Respiratory water loss therefore increases

less rapidly with activity than does oxygen consumption.

52




53


6. Cutaneous water loss contributes between 72 and 97 per cent of the

total water loss. With the rate of respiratory water loss fixed by the

level of oxygen consumption, the differences in water loss partitioning

are due to the ecologically rationalizable differences in cutaneous

water loss.

7. Skin permeability measured in aqueous media is greater in N. f.

pictiventris than in N. f. compressicauda.













REFERENCES

Beament J. W. L. (1961) The water relations of insect cuticle.
Biol. Rev. 36, 281-320.

Bentley P. J. (1959) Studies on the water and electrolyte metabolism
of the lizard Trachysaurus rugosus. J. Physiol. 145, 37-47.

Bentley P. J. and Blumer W. F. C. (1962) Uptake of water by the
lizard Moloch horridus. Nature 194, 699-700.

Bentley P. J. and Schmidt-Nielsen K. (1966) Cutaneous water loss
in reptiles. Science 151, 1547-1549.

Bogert C. and Cowles R. (1947) Results of the Archbold expedition,
No. 58. Moisture loss in relation to habitat selection in some
Floridian reptiles. Amer. Mus. Novitates 1358, 1-34.

Campden-Main S. M. (1970) A field guide to the snakes of South Vietnam,
Division of Reptiles and Amphibians. U. S. Natl. Mus. Smithsonian
Inst., Washington. 1-114.

Chew R. M. (1961) Water metabolism of desert inhabiting vertebrates.
Biol. Rev. 36, 1-31.

Chew R. M. and Dammann A. E. (1961) Evaporative water loss of small
vertebrates, as measured with an infrared analyser. Science 133,
384-385.

Clausen D. L. '(1967) Studies of water loss in two species of lizards.
Comp. Biochem. Physiol. 20, 115-130.

Cloudsley-Thompson J. L. (1968) Water relations of crocodiles.
Nature 220 (5168), 708.

Crawford E. C., Jr. and Schultetus R. R. (1970) Cutaneous gas exchange
in the lizard Sauromalus obesus. Copeia 1970, 179-180.

Dawson W. R. (1967) Interspecific variation in physiological response
of lizards to temperature. 230-257. In W. W. Milstead (ed)
Lizard Ecology: A Symposium. Univ. Missouri Press, Columbia.

Dawson W. R., Shoemaker V. H. and Licht P. (1966) Evaporative water
losses of some small Australian lizards. Ecology 47, 589-594.

Dawson W. R. and Templeton J. R. (1963) Physiological responses to
temperature in the lizard Crotaphytus collaris. Physiol. Zool.
36, 219-236.







Dejours P. (1966) Respiration. Oxford Univ. Press, New York.

Galvao P. E., Tarasantchi J. and Guertzenstein P. (1965) Heat pro-
duction of tropical snakes in relationship to body weight and
body surface. Am. J. Physiol. 209, 501-506.

Gans C., Krakauer T. and Paganelli C. V. (1968) Water loss in snakes:
Interspecific and intraspecific variability. Comp. Biochem.
Physiol. 27, 747-761.

George J. C. (1947) A comparative study of permeability to water of
the skin in some representative vertebrates. J. Univ. Bombay 16,
28-32.

Knudsen J. G. and Katz D. L. (1958) Fluid dynamics and Heat Transfer.
McGraw-Hill Book Co., New York,

Krakauer T., Gans C. and Paganelli C. V. (1968) Ecological correlation
of water loss in burrowing reptiles. Nature 218 (5142), 659-660.

Maderson P. F. A. (1964) The skin of lizards and snakes. Brit. J.
Herpetology 151-154.

Maderson P. F. A. (1965) Histological changes in the epidermis of
snakes during the sloughing cycle. J. Zool. 146, 98-113.

Mercer E. H. (1961) Keratin and Keratinization. Pergamon, London.

Motais R., Isaia J., Rankin J. C. and Maetz J. (1969) Adaptive changes
of the water permeability of the teleostean gill epithelium in re-
lation to external salinity. J. Exp. Biol. 51, 529-546.

Pettus D. (1958) Water relations in Natrix sipedon. Copeia 1958,
207-211.

Pettus D. (1963) Salinity and subspeciation in Natrix sipedon.
Copeia 1963, 499-504.

Prange H. D. and Schmidt-Nielsen K. (1969) Evaporative water loss in
snakes. Comp. Biochem. Physiol. 28, 973-975.

Roberts L. A. (1968) Water loss in the desert lizard Uta stansburiana.
Comp. Biochem. Physiol. 27, 583-589.

Schmid W. D. (1965) Some aspects of the water economies of nine
species of amphibians. Ecology 46, 261-269.

Schmidt-Nielsen K. (1969) The neglected interface: The biology of
water as a liquid-gas system. Quart. Rev. Biophys. 2, 283-304.

Schmidt-Nielsen K. and Bentley P. J. (1966) Desert tortoise Gopherus
agassazii: Cutaneous water loss. Science 154, 911.

Schmidt-Nielsen K. and Fange R. (1958) Salt glands in marine reptiles.
Nature 182, 783-785.





56


Sutton 0. G. (1953) Micrometerology. McGraw-Hill Book Co., New York.

Tabb D. C., Dubrow D. L. and Manning R. B. (1962) The ecology of
northern Florida Bay and adjacent estuaries. State of Florida,
Board of Conservation Tech. Ser. No. 39.

Templeton J. R. (1960) Respiration and water loss at higher
temperatures in the desert igauna, Dipsosaurus dorsalis.
Physiol. Zool. 33, 136-145.

Tercafs P. R. and Schoffeniels E. (1965) Phenomenes de permiabilite
au niveau de la peau des reptiles. Ann. Roy. Soc. Zool. Belgium
96, 9-22.

Thorson T. B. (1968) Body fluid partitioning in reptilia. Copeia
1968, 592-601.













BIOGRAPHICAL SKETCH

Thomas Henry Krakauer was born September 6, 1942, at Buffalo,

New York. In June, 1959, he was graduated from The Park School

of Buffalo. In June, 1964, he received the degree of Bachelor of

Arts with a major in Biology from the University of Rochester. In

June, 1966, he received the degree of Master of Science with a major

in Biology from the University of Miami. In 1966 he enrolled in the

Graduate School of the University of Florida. He worked as an Interim

Instructor in the Department of Zoology until June, 1967. His work

toward the degree of Doctor of Philosophy was further supported by a

National Defense Education Act, Title IV Fellowship and a University

of Florida Graduate School Fellowship.

Thomas Henry Krakauer is married to the former Janet MacColl.













This dissertation was prepared under the direction of the

chairman of the candidate's supervisory committee and has been

approved by all members of that committee. It was submitted to the

Dean of the College of Arts and Sciences and to the Graduate

Council, and was approved as partial fulfillment of the requirements

for the degree of Doctor of Philosophy.

August, 1970.





Dean, College of Ats and Sciences




tiean, Graduate School
Supervisory Committee:




Chairman







i^ Pe JL






"'0/
/~~ P4,/

Y













































/00


~$


0 6.s