- Email: [email protected]
The viability of nearctic freshwater turtles submerged in anoxia and normoxia at 3 and 10°C
Camp. Biochem. Physiol.Vol. SlA, No. 3, pp. 607-611, 1985
030&9629/85$3.00+ 0.00 0 1985Pergamon Press Ltd
Printed in Great Britain
THE VIABILITY OF NEARCTIC FRESHWATER TURTLES SUBMERGED IN ANOXIA AND NORMOXIA AT 3 AND 10°C CORDON
R. ULTSCH*
Department of Biology, University of Alabama, Tuscaloosa, AL 35486, U.S.A. Abstract- 1. Survival times of temperature-acclimated freshwater turtles submerged in normoxic and anoxic water were determined. 2. Juvenile Chrysemys scripta from Alabama and adult Chrysemyspicta bellii from Wisconsin exhibited the maximal survival times of the 10 species and subspecies investigated, both being able to survive at least a half-year of submergence at 3°C in normoxic water. 3. Minimal viability was exhibited by Sternotherus odoratus and Trionyx spinzferus submerged in anoxia at 10°C (5.2 and 2.6 days, respectively). 4. All species lived longer in normoxic water than in anoxic water, attesting to the importance of extrapulmonary gas exchange during submergence at low temperature. 5. While freshwater turtles are often reputed to overwinter buried in mud, which is an anoxic microenvironment, this behaviour seems unlikely for southern Trionyx and Sternotherus, although it may be possible for northern Chrysemys. INTRODUCTION In 1933, Johlin and Moreland documented the excep-
tional tolerance of turtles to anoxia by exposing them to a nitrogen atmosphere. Similar studies (Belkin, 1963, 1968a; Bellamy and Petersen, 1968) have confirmed this finding and further shown that turtles are not only very tolerant of anoxia and hypoxia, but are very much more so than other reptiles, provided that the circulation remains functional (Belkin, 1968a). While nitrogen breathing produces anoxia, it does not have an ecological analog, and controlling of the behavior (and resultant rate of energy expenditure) of the turtles in such conditions is impractical. Of more interest is the hypoxia produced by prolonged dives. At 20°C and above, this hypoxia approaches anoxia internally, regardless of the state of oxygenation of the water (up to air-saturation levels), for turtles that do not have efficient extrapulmonary gas exchange. Thus kinosternid (mud and musk) and trionychid (softshelled) turtles, well known for such extrapulmonary gas exchange, fare much better in normoxic water than in anoxic, but the difference in survival time for emydids (pond and river turtles, the largest family) under the two conditions is not significant (Belkin, 1968b). Survival times of freshwater turtles during diving at 20°C and above, even in normoxic water, are measured in hours or days. In contrast, turtles are reputed to hibernate underwater for months at low temperatures, often buried in anoxic mud (Ultsch and Lee, 1984). Sucn increases in submergence tolerance, particularly in anoxia, considerably exceed what one would predict based on a simple Q,,relationship for metabolic rate. This either means that turtles undergo some extraordinary physiological adjustments while hibernating underwater buried in mud, or that they periodically surface to breathe, or that they seek (at least occasionally) a normoxic area where they can utilize extrapuhnonary gas exchange. *Please address correspondence to the author at: Department of Zoology, University of Florida, Gainesville, FL 32611, U.S.A.
Until recently, it has been suggested that extrapulmonary gas exchange might be an important factor in the survival of hibernating emydid turtles (Mussachia, 1959; Smith and Nickon, 1961). Recent reports have proven this to be the case for the emydid Chrysemys picta and have detailed the physiological implications of such aquatic respiration (Jackson and Heisler, 1982 and 1983; Jackson and Ultsch, 1982; Ultsch and Jackson, 1982a,b; Ultsch, Hanley and Bauman, in press). Here I report on the submergence viability at low temperatures of several species and families of freshwater turtles. MATERIALSAND METHODS Experimental protocol
Turtles were acclimated to 10°C by placing them in tanks in a cold room and allowing their water to cool overnight; they were maintained at this temperature for at least a week before submergence. Submergences at 10°C were begun in May-Jan. Turtles acclimated to 3°C were cooled from room temperature to 3°C at l”C/day and maintained at least an additional 2 days at 3°C prior to submergence. All experiments at 3°C were initiated in late Ott-Dec. Submergence was maintained by placing a plastic screen in the water column that prevented the turtles from surfacing. Normoxia was ensured by constant aeration of the water above the screen. Anoxia (here defined as an 0, tension of < 2 mmHg) was produced by covering the tanks with a plastic lid seated on closed-cell foam insulation and continuously bubbling N, through the water; in some cases a layer of mineral oil was also floated on the water surface. The water was kept clear by periodic flushing with appropriately prepared clean tap water. Animals were deemed dead when they no longer responded to vigorous prodding and made no effort to right themselves when inverted. Occasionally an animal would recover upon removal and slow warming to room temperature, but in over 95% of the cases there was no recovery. Animals
Turtles were maintained indoors at 18-22°C in either tap water or well water. They were fed chopped fish, dog food, liver, catfish chow, and/or lettuce, depending upon the species. The source and treatment of the turtles were as follows (sample sizes are given in Fig. 1):
608
R.
GORDON
ULTSCH
e/vAANOXIA
NORMOX IA
II
~HRYSEMYS
SCRIPTA
(JUVENILES)
N=7
l///mwrw/~
..‘/IN=8
.
t////4ev
.
. N
.
I
4
N=4 N”. 4
.
+N=7
q
.
(ADULTS)
.
)
N=7
CHRYSEMY$
8
CONCINNA
KINOSTERNON
N=3
CHELYDRA
SUBRUBRUM -___
SUBRUBRUM
SERPENTINA
)
1
. . A=6,0=3,
c_. ScRiPTA
9:
/////:t’/t//:
&
*
IN=7
.
.
N=l
TRIONYX
m
1 N=II*TRlONYX
A=4.
SPINIFERUS ASPERUS
STERNOTHERUS
31
‘N=’
MlNOR
N= IO* .
1 STERNOTHERUS ___-
IN.8 0=4.
ODORATUS
N=ll
.
. l
PELTIFER
l
.a0
.
////////////////,r//////////LLd
3 I/////*///i////////A r
I
*
CHRYSEMYS *N;,OPICTA
0
N=IO
l
)~.6
N = 24*
I 50
1
SUBMERGENCE
1 100
C _.
_PICTA
1
VIABILITY
BELI
11
~DORSALIS
1 150
I
7 200
(DAYS)
Fig. 1. Survival times of freshwater turtles submerged at 3 and 10°C in aerated (normoxia) or N,-bubbled (anoxia) water. A star after the sample size indicates that not all of the individuals died, some being removed for blood sampling or allowed to recover (see the species accounts for details). Individual deaths are indicated by blackened circles; other symbols are used to indicate days when relatively large numbers of animals died.
1. Chrysemys (= Pseuakmys) scripta (scripta elegans x scripta scripta) adults (985-1541 g) were collected from Tuscaloosa and Sumpter counties in Alabama. Submergence was initiated on 9 Nov 1982. Juveniles (47-130 g) were from Tuscaloosa and Jefferson counties in Alabama; submergence commenced on 6 and 9 Nov 1982 for anoxia and normoxia, respectively. Chrysemys (= Pseudemys) concinna (6klSlS g) were from Jefferson County, Alabamaa%: mergence was started on 9 Nov 1982. 3. Kinosternon subrubrum subrubrum were from Tuscaloosa County in Alabama. Adults (151-230 g) were submerged on 10 Nov 1982. 4. Chelydra serpentina were collected in Tuscaloosa County and included juveniles and adults (68-5211 g). Submergence was initiated between 27 May 1982 and 14 Jan 1983. 5. One adult male Trionyx ferox (949g) from Alachua County, Florida, was submerged on 10 Nov 1982. 6. Trionyx spiniferus asperus juveniles and adults (g-3021 g) were collected in Tuscaloosa and Jefferson coun-
ties, Alabama. Individuals were submerged in anoxic water between 4 June and 26 Ott 1982, and in normoxic water between 27 May and 15 Ott 1982. 7. One adult (91 g) Sternotherus minor pelt& from Tuscaloosa County, Alabama, was submerged on 10 Nov 1982. 8. Sternotherus odoratus submerged at 3°C were from Jefferson County, Alabama. Submergence experiments at 10°C used adults from Tuscaloosa and Jefferson counties in Alabama. 9. Chrysemys picta bellii adults were purchased from Kons Scientific Company in Wisconsin, in which state the turtles were collected. Animals used at 3°C were submerged on 31 Ott 1979; those used at 10°C were submerged on 11 Dee 1982. 10. Chrysemys picta dorsalis that died during anoxic submergence at 3°C were from Lake County, Tennessee. Although 24 adults were submerged, it was not the purpose of the study utilizing these turtles to determine maximum viability of the entire group (Ultsch, Hanley and Bauman, in press). However, of the 3 that died during submergence, the last was a large (506g), healthy, female, that was left
Submergence viability of freshwater turtles submerged specifically to gain an estimate of viability. Where statements concerning the significance of results between 2 groups are made, they are derived from a t-test at the 95% level of significance.
RESULTS AND DISCUSSION
Critique of methods Several previous studies of turtles have used nitrogen atmospheres to induce anoxia (Belkin, 1963; Bellamy and Petersen, 1968; Clark and Miller, 1973; Johlin and Moreland, 1933). Turtles, however, are essentially aquatic organisms, and their behaviour out of water in an N,atmosphere could influence their survival time. In addition, diving emydid turtles at temperatures of 20°C and above undergo a metabolic depression of large proportions (Jackson, 1968; Jackson and Schmidt-Nielsen, 1966), which is likely to lengthen their survival time if anoxia is induced in an aquatic environment as opposed to an aerial one. Therefore survival times reported here are presumed to be more representative of maximal viability than would be the case in experiments where anoxia was induced otherwise. For these reasons, previous results with submerged turtles are probably the most informative with regard to the limits of viability in anoxia, and certainly so in the case of normoxic submergence. However, of such studies, only Belkin (1968b) gives the geographic origin of the turtles, their mass, and the time of the year that the experiments were conducted; furthermore, his was the only study where the 0, tension was regulated and monitored. Other reports (Mussachia, 1959; Smith and Nickon, 1961), while instructive, lack information on one or more of these points. It appears that turtles submerged at low temperature in the winter survive longer than those submerged at low temperature in the summer (Belkin, 1963), indicating that there may be some circannual rhythms involved in preparation for hibernation (Gregory, 1983). Therefore all turtles studied here at 3°C were submerged during the fall or early winter. However, this precaution was not always taken for turtles studied at 10°C which was considered a cool, but not cold, temperature. It is possible that survival at 10°C could have been enhanced if all of these turtles were submerged during the same period as the animals studied at 3°C. Viability results for emydid turtles in normoxic water were frequently confounded by fungal infections. Such infections occurred only in normoxia, indicating a dependence of the fungus upon oxygen. The infections were unpredictable and generally resistant to treatment. Therefore, when survival times are being considered for animals that were so infected (see species accounts), they should be treated as minimal estimates of viability. Not all of the data reported here come from investigations that were designed only to test viability, and in some of those that were so designed, not every turtle was allowed to die. Deviations from a protocol requiring the death of all submerged animals are indicated on Fig. 1 by a star after the sample size, and are discussed under the accounts of results for each species.
609
Species accounts The survival times of individual turtles submerged under various conditions of temperature and oxygenation are presented in Fig. 1. Appropriate notes follow: 1. Chrysemys scripta adults in normoxia became infected with fungus about the head and limbs, while the juveniles did not. Therefore, while it is tempting to invoke surface to mass ratio arguments concerning the longer survival of the smaller animals, it would not be appropriate to do so with these data. Only two of the juveniles in normoxia died while submerged; the others were removed and allowed to warm to room temperature in 1 cm of water. One removed on day 179 died 6 days later; of 4 removed on day 182 (chosen to be equivalent to half a year of submergence), one died 8 days later and 3 recovered completely and were released. All animals removed alive appeared bloated, indicating at least a partial failure of water balance mechanisms, and all had an unknown eye infection that responded to treatment with an ophthalmic cream. There was no significant difference in survival time between adults and juveniles in anoxic water. 2. Chrysemys concinna in normoxic water also had fungal infections, but they did not appear as early, or seem as severe, as with C. scripta. 3. Kinosternon subrubrum has been shown to be capable of hibernation on land, and this is one of very few aquatic turtles that may do so routinely (Bennett, 1972; Bennett et al., 1970). Terrestrial hibernation may be related to the low survival time of this species in normoxic submergence. However, the sample size here is small, therefore these results should be considered as preliminary. 4. The difference in survival times of Chelydra serpentina in normoxia and anoxia at 10°C was not significant, but I suspect that this was probably more due to sample size than to any other factor, and that there would be a significant difference at 3°C. There was no problem with fungus in the normoxic situation. 5. Trionyx spiniferus asperus was the most tolerant of all species of normoxic submergence at 10°C and the least tolerant of anoxic submergence. After 100 days of normoxic submergence, 7 of 11 were still alive; the 4 that had died were small individuals (l&8 1 g). Small animals seemed especially susceptible to a fungus that attacked the shell. In anoxia, there was a trend toward longer survival of larger individuals, suggestive of an importance of decreasing weight-specific metabolic rate (lactate formation, in this case) with increasing size and its interaction with whole-body buffering capacity. This was the only trend related to body size in this study where the sample size (15), size range (9.2-3021 g), and the lack of fungal infections, suggested that body size might be a parameter influencing submergence viability. 6. Sternotherus odoratus is also capable of efficient extrapulmonary gas exchange (Root, 1949) as was evidenced by the survival of 5 of 8 turtles for 100 days in normoxic water at 10°C. These animals were in good health when removed for blood sampling (Ultsch et al., 1984). As with Trionyx, survival at 10°C in anoxic water was of short duration. At 3°C the results are qualitatively similar to those at 10°C. The
610
GORDONR. ULTSCH
first death in anoxia occurred between days 16 and 2 1, and all died by day 31. In normoxia, the first death did not occur until between days 135 and 150. After 150 days, 4 turtles were removed for physiological studies, and 5 were placed in shallow water at room temperature. Of the 5, 3 died within 2 days and 2 recovered completely and were released. 7. Belkin (1968b) found that Sternotherus minor minor from Florida lived only 2.5 days when submerged in normoxic water at 22°C. I found a single Sternotherus minor pelt@ from Alabama to survive 126-128 days of normoxic submergence at 3°C. This result emphasizes the importance of temperature in affecting survival. 8. The painted turtle (Chrysemys picta) is particularly interesting because it is the only nearctic species with a range that is continuous from the east to west coasts and is also one of only a few species that range from the Gulf Coast to southern Canada. Chrysemys picta bellii is the most northerly subspecies, and its survival at 3°C has been summarized previously (Ultsch and Jackson, 1982a). Of 6 removed from normoxia after 189 days, 5 appeared to be in moderately good health, but they were not allowed to attempt recovery. Turtles removed from anoxia on days 155-168 were in very poor health. At 10°C the shortened survival of animals in normoxia was partially due to the higher temperature and possibly partially due to fungal infections. It is not possible to segregate the relative importance of the two effects, but it appears likely that temperature was the more influential, since the ratio of survival at 10°C to that at 3°C is about the same for both normoxia and anoxia. Southern painted turtles (C. p. dorsalis) were submerged in anoxia at 3°C in a study designed to compare their physiological responses to those of their northern counterparts (Ultsch et al., in press). This necessitated periodic removals of animals, and only one was kept submerged to give an indication of maximal viability. That animal, chosen as one of the apparently healthiest of the group, lived 86 days, only about half as long as the northern form. Physiological and ecological trends and implications. It is apparent that all of the species studied can benefit from extrapulmonary gas exchange while submerged in normoxic water at low temperatures. This was to be expected for Sternotherus and Trionyx, which are thin-skinned turtles with favorable surface to volume ratios (due to size in the case of Sternotherus and shape in the case of Trionyx), and which also utilize buccopharyngeal respiration. It was, however, only recently that the importance of extrapulmonary gas exchange was demonstrated for an emydid species (Ultsch and Jackson, 1982a and b; Jackson and Ultsch, 1982). I suggest that probably all freshwater turtles can efficaciously utilize extrapulmonary gas exchange when submerged in normoxic water at low temperature. The relative inability of certain species (Sternotherus and Trionyx) to tolerate submergence anoxia suggests that they may be excluded from hibernation in mud, reputed to be a typical microenvironment for overwintering freshwater turtles. This hypothesis, however, is subject to several reservations. Most of the turtles used here were from the southern United
States, and southern forms of species with wide latitudinal ranges may not be as tolerant of prolonged submergence as northern forms; this has been shown to be the case for Chrysemys picta (Fig. 1; Ultsch et al., in press). Secondly, I report no data here for Trionyx at 3X, although it is difficult to conceive of physiological adjustments that would allow this species to tolerate months of anoxia in the northern part of its range when southern animals only survived 0.8-6.8 days at 10°C. Finally, Ultsch and Lee (1984), found that 3 of 5 Chelydra serpentina in Rhode Island survived at least 3.5 months buried in mud during the winter, although air-breathing could not be ruled out for two of them, or buccopharyngeal respiration from the water column for the third. There is a virtual complete lack of solid information on the behaviour of freshwater turtles in the winter, other than occasional notes on turtles swimming under the ice or caught by muskrat trappers in winter (R. Meeks, personal communication), and until individual turtles are tracked in their home ranges throughout a winter, we cannot be certain whether they are hibernating in anoxia or have at least partial access to either dissolved or atmospheric oxygen. While the southern turtles in this study generally fared poorer than the one northern form (C. p. bellii), all southern species survived normoxic submergence longer than they would be likely to have to do so in their natural environment during the winter. In contrast, the survival in anoxia of southern turtles, including the southern form of C. picta dorsalis, while much less than that of C. p. bellii, is not necessarily mismatched with the time that they might have to spend in hibernation during a typical cold period during the winter. If turtles do in fact spend most of their hibernation time in anoxic mud, then the viability of all the forms studied here (save perhaps Trionyx and Sternotherus) is reasonably well matched to their expected hibernation period. Thus the northern limit of each race may be dependent upon its tolerance to anoxic submergence, and this tolerance may be rather plastic in an evolutionary sense, within a species. However, such reasoning is speculative at this time. There is a critical need for definitive field studies of hibernation in freshwater turtles (and other freshwater ectotherms), and until these are available, physiological studies will continue to be only suggestive. Acknowledgements-I thank Dennis Lee and Kenneth O’Brien for technical aid. This study was partially supported by NSF Grants PCM 79-l 1609 to the author and PCM 78-
22333 to Donald C. Jackson of Brown University and by a Research Grants Committee grant from the University of Alabama. Robert Guthrie was of invaluable aid in collecting the turtles. REFERENCES Belkin D. A. (1963) Anoxia:
tolerance
139,492-493. Belkin D. A. (1968a) Anaerobic
in reptiles.
Science
brain function: effects of stagnant and anoxic anoxia on persistance of breathing in reptiles. Science 162, 1017-1818. Belkin D. A. (1968b) Aquatic respiration and underwater survival of two freshwater turtle species. Respir. Physiol. 4, 1-14. Bellamy D. and Petersen J. A. (1968) Anaerobiosis and the
Submergence viability of freshwater turtles toxicity of cyanide in turtles. Comp. Biochem. Physiol. 24, 543-548.
Bennett D. H. (1972) Notes on the terrestrial wintering of mud turtles (Kinosternon subrubrum). Herpetologica 28, 2455247.
picture of the turtle after complete anoxia. J. biol. Chem. 103, 107-l 14. Musacchia X. J. (1959) The viability of Chrysemys picta submerged at various temperatures. Physiol. Zoo. 32, 47-50.
Bennett D. H., Gibbons J. W. and Franson J. C. (1970) Terrestrial activity in aquatic turtles. Ecology 51,738-740. Clark V. and Miller A. (1973) Studies on anaerobic metabolism in the freshwater turtle (Pseudemys scripta eleguns). Comp. Biochem.
611
Physiol. A, 55-62.
_
Greeorv P. T. (1983) Rentilian hibernation. In Biology of the Reptilia. (Edited by Gans C. and Pough F. H.), Vol. 13, pp. 53-154. Academic Press, New York. Jackson D. C. (1968) Metabolic depression and oxygen depletion in the diving turtle. J. appl. Physiol. 24,503-509. Jackson D. C. and Heisler N. (1982) Plasma ion balance of submerged anoxic turtles at 3°C: the role of calcium lactate formation. Respir. Physiol. 49, 1599174. Jackson D. C. and Heisler N. (1983) Intracellular and extracellular acid-base and electrolyte status of submerged anoxic turtles at 3’C. Respir. Physiol. 53,187-201. Jackson D. C. and Schmidt-Nielsen K. (1966) Heat production during diving in the fresh water turtle, Pseudemys scripta. J. cell. Physiol. 61, 225-232.
Jackson D. C. and Ultsch G. R. (1982) Long-term submergence at 3°C of the turtle, Chrysemys pictu bellii, in normoxic and severely hypoxic water-II. Extracellular ionic responses to extreme lactic acidosis. J. exp. Biol. 96, 29-43.
Johlin J. M. and Moreland F. B. (1933) Studies of the blood
Root R. W. (1949) Aquatic respiration in the musk turtle. Physiol. Zool. 22, 172-178.
Smith H. M. and Nickon D. C. (1961) Preliminary experiments on the role of the cloaca1 bursae in hibernating turtles. Natural History Miscellanea of the Chicago Academy of Sciences No. 178. 8 pp. Ultsch G. R., Hanley R. W. and Bauman T. R. (1985) Responses to anoxia during simulated hibernation in northern and southern painted turtles. Ecology. In press. Ultsch G. R., Herbert C. V. and Jackson D. C. 1984) The comparative physiology of diving in North American freshwater turtles-I. Submergence tolerance, gas exchange and acid-base status. Physiol. Zool. 57, 62M31.
Ultsch G. R. and Jackson D. C. (1982a) Long-term submergence at 3°C of the turtle Chrysemys picta bellii, in normoxic and severely hypoxic water-I. Survival, gas exchange and acid-base status. J. exp. Biol. 96, 1l-28. Ultsch G. R. and Jackson D. C. (1982b) Long-term submergence at 3°C of the turtle, Chrysemys picta bellii, in normoxic and severely hypoxic water-III. Effects of changes in ambient PO, and subsequent air breathing. J. exp. Biol. 97, 87-99.
Ultsch G. R. and Lee D. (1984) Radiotelemetric observations of wintering snapping turtles (Chelydra serpentina) in Rhode Island. J. Alabama Acad. Sci. 54,20@-206.
030&9629/85$3.00+ 0.00 0 1985Pergamon Press Ltd
Printed in Great Britain
THE VIABILITY OF NEARCTIC FRESHWATER TURTLES SUBMERGED IN ANOXIA AND NORMOXIA AT 3 AND 10°C CORDON
R. ULTSCH*
Department of Biology, University of Alabama, Tuscaloosa, AL 35486, U.S.A. Abstract- 1. Survival times of temperature-acclimated freshwater turtles submerged in normoxic and anoxic water were determined. 2. Juvenile Chrysemys scripta from Alabama and adult Chrysemyspicta bellii from Wisconsin exhibited the maximal survival times of the 10 species and subspecies investigated, both being able to survive at least a half-year of submergence at 3°C in normoxic water. 3. Minimal viability was exhibited by Sternotherus odoratus and Trionyx spinzferus submerged in anoxia at 10°C (5.2 and 2.6 days, respectively). 4. All species lived longer in normoxic water than in anoxic water, attesting to the importance of extrapulmonary gas exchange during submergence at low temperature. 5. While freshwater turtles are often reputed to overwinter buried in mud, which is an anoxic microenvironment, this behaviour seems unlikely for southern Trionyx and Sternotherus, although it may be possible for northern Chrysemys. INTRODUCTION In 1933, Johlin and Moreland documented the excep-
tional tolerance of turtles to anoxia by exposing them to a nitrogen atmosphere. Similar studies (Belkin, 1963, 1968a; Bellamy and Petersen, 1968) have confirmed this finding and further shown that turtles are not only very tolerant of anoxia and hypoxia, but are very much more so than other reptiles, provided that the circulation remains functional (Belkin, 1968a). While nitrogen breathing produces anoxia, it does not have an ecological analog, and controlling of the behavior (and resultant rate of energy expenditure) of the turtles in such conditions is impractical. Of more interest is the hypoxia produced by prolonged dives. At 20°C and above, this hypoxia approaches anoxia internally, regardless of the state of oxygenation of the water (up to air-saturation levels), for turtles that do not have efficient extrapulmonary gas exchange. Thus kinosternid (mud and musk) and trionychid (softshelled) turtles, well known for such extrapulmonary gas exchange, fare much better in normoxic water than in anoxic, but the difference in survival time for emydids (pond and river turtles, the largest family) under the two conditions is not significant (Belkin, 1968b). Survival times of freshwater turtles during diving at 20°C and above, even in normoxic water, are measured in hours or days. In contrast, turtles are reputed to hibernate underwater for months at low temperatures, often buried in anoxic mud (Ultsch and Lee, 1984). Sucn increases in submergence tolerance, particularly in anoxia, considerably exceed what one would predict based on a simple Q,,relationship for metabolic rate. This either means that turtles undergo some extraordinary physiological adjustments while hibernating underwater buried in mud, or that they periodically surface to breathe, or that they seek (at least occasionally) a normoxic area where they can utilize extrapuhnonary gas exchange. *Please address correspondence to the author at: Department of Zoology, University of Florida, Gainesville, FL 32611, U.S.A.
Until recently, it has been suggested that extrapulmonary gas exchange might be an important factor in the survival of hibernating emydid turtles (Mussachia, 1959; Smith and Nickon, 1961). Recent reports have proven this to be the case for the emydid Chrysemys picta and have detailed the physiological implications of such aquatic respiration (Jackson and Heisler, 1982 and 1983; Jackson and Ultsch, 1982; Ultsch and Jackson, 1982a,b; Ultsch, Hanley and Bauman, in press). Here I report on the submergence viability at low temperatures of several species and families of freshwater turtles. MATERIALSAND METHODS Experimental protocol
Turtles were acclimated to 10°C by placing them in tanks in a cold room and allowing their water to cool overnight; they were maintained at this temperature for at least a week before submergence. Submergences at 10°C were begun in May-Jan. Turtles acclimated to 3°C were cooled from room temperature to 3°C at l”C/day and maintained at least an additional 2 days at 3°C prior to submergence. All experiments at 3°C were initiated in late Ott-Dec. Submergence was maintained by placing a plastic screen in the water column that prevented the turtles from surfacing. Normoxia was ensured by constant aeration of the water above the screen. Anoxia (here defined as an 0, tension of < 2 mmHg) was produced by covering the tanks with a plastic lid seated on closed-cell foam insulation and continuously bubbling N, through the water; in some cases a layer of mineral oil was also floated on the water surface. The water was kept clear by periodic flushing with appropriately prepared clean tap water. Animals were deemed dead when they no longer responded to vigorous prodding and made no effort to right themselves when inverted. Occasionally an animal would recover upon removal and slow warming to room temperature, but in over 95% of the cases there was no recovery. Animals
Turtles were maintained indoors at 18-22°C in either tap water or well water. They were fed chopped fish, dog food, liver, catfish chow, and/or lettuce, depending upon the species. The source and treatment of the turtles were as follows (sample sizes are given in Fig. 1):
608
R.
GORDON
ULTSCH
e/vAANOXIA
NORMOX IA
II
~HRYSEMYS
SCRIPTA
(JUVENILES)
N=7
l///mwrw/~
..‘/IN=8
.
t////4ev
.
. N
.
I
4
N=4 N”. 4
.
+N=7
q
.
(ADULTS)
.
)
N=7
CHRYSEMY$
8
CONCINNA
KINOSTERNON
N=3
CHELYDRA
SUBRUBRUM -___
SUBRUBRUM
SERPENTINA
)
1
. . A=6,0=3,
c_. ScRiPTA
9:
/////:t’/t//:
&
*
IN=7
.
.
N=l
TRIONYX
m
1 N=II*TRlONYX
A=4.
SPINIFERUS ASPERUS
STERNOTHERUS
31
‘N=’
MlNOR
N= IO* .
1 STERNOTHERUS ___-
IN.8 0=4.
ODORATUS
N=ll
.
. l
PELTIFER
l
.a0
.
////////////////,r//////////LLd
3 I/////*///i////////A r
I
*
CHRYSEMYS *N;,OPICTA
0
N=IO
l
)~.6
N = 24*
I 50
1
SUBMERGENCE
1 100
C _.
_PICTA
1
VIABILITY
BELI
11
~DORSALIS
1 150
I
7 200
(DAYS)
Fig. 1. Survival times of freshwater turtles submerged at 3 and 10°C in aerated (normoxia) or N,-bubbled (anoxia) water. A star after the sample size indicates that not all of the individuals died, some being removed for blood sampling or allowed to recover (see the species accounts for details). Individual deaths are indicated by blackened circles; other symbols are used to indicate days when relatively large numbers of animals died.
1. Chrysemys (= Pseuakmys) scripta (scripta elegans x scripta scripta) adults (985-1541 g) were collected from Tuscaloosa and Sumpter counties in Alabama. Submergence was initiated on 9 Nov 1982. Juveniles (47-130 g) were from Tuscaloosa and Jefferson counties in Alabama; submergence commenced on 6 and 9 Nov 1982 for anoxia and normoxia, respectively. Chrysemys (= Pseudemys) concinna (6klSlS g) were from Jefferson County, Alabamaa%: mergence was started on 9 Nov 1982. 3. Kinosternon subrubrum subrubrum were from Tuscaloosa County in Alabama. Adults (151-230 g) were submerged on 10 Nov 1982. 4. Chelydra serpentina were collected in Tuscaloosa County and included juveniles and adults (68-5211 g). Submergence was initiated between 27 May 1982 and 14 Jan 1983. 5. One adult male Trionyx ferox (949g) from Alachua County, Florida, was submerged on 10 Nov 1982. 6. Trionyx spiniferus asperus juveniles and adults (g-3021 g) were collected in Tuscaloosa and Jefferson coun-
ties, Alabama. Individuals were submerged in anoxic water between 4 June and 26 Ott 1982, and in normoxic water between 27 May and 15 Ott 1982. 7. One adult (91 g) Sternotherus minor pelt& from Tuscaloosa County, Alabama, was submerged on 10 Nov 1982. 8. Sternotherus odoratus submerged at 3°C were from Jefferson County, Alabama. Submergence experiments at 10°C used adults from Tuscaloosa and Jefferson counties in Alabama. 9. Chrysemys picta bellii adults were purchased from Kons Scientific Company in Wisconsin, in which state the turtles were collected. Animals used at 3°C were submerged on 31 Ott 1979; those used at 10°C were submerged on 11 Dee 1982. 10. Chrysemys picta dorsalis that died during anoxic submergence at 3°C were from Lake County, Tennessee. Although 24 adults were submerged, it was not the purpose of the study utilizing these turtles to determine maximum viability of the entire group (Ultsch, Hanley and Bauman, in press). However, of the 3 that died during submergence, the last was a large (506g), healthy, female, that was left
Submergence viability of freshwater turtles submerged specifically to gain an estimate of viability. Where statements concerning the significance of results between 2 groups are made, they are derived from a t-test at the 95% level of significance.
RESULTS AND DISCUSSION
Critique of methods Several previous studies of turtles have used nitrogen atmospheres to induce anoxia (Belkin, 1963; Bellamy and Petersen, 1968; Clark and Miller, 1973; Johlin and Moreland, 1933). Turtles, however, are essentially aquatic organisms, and their behaviour out of water in an N,atmosphere could influence their survival time. In addition, diving emydid turtles at temperatures of 20°C and above undergo a metabolic depression of large proportions (Jackson, 1968; Jackson and Schmidt-Nielsen, 1966), which is likely to lengthen their survival time if anoxia is induced in an aquatic environment as opposed to an aerial one. Therefore survival times reported here are presumed to be more representative of maximal viability than would be the case in experiments where anoxia was induced otherwise. For these reasons, previous results with submerged turtles are probably the most informative with regard to the limits of viability in anoxia, and certainly so in the case of normoxic submergence. However, of such studies, only Belkin (1968b) gives the geographic origin of the turtles, their mass, and the time of the year that the experiments were conducted; furthermore, his was the only study where the 0, tension was regulated and monitored. Other reports (Mussachia, 1959; Smith and Nickon, 1961), while instructive, lack information on one or more of these points. It appears that turtles submerged at low temperature in the winter survive longer than those submerged at low temperature in the summer (Belkin, 1963), indicating that there may be some circannual rhythms involved in preparation for hibernation (Gregory, 1983). Therefore all turtles studied here at 3°C were submerged during the fall or early winter. However, this precaution was not always taken for turtles studied at 10°C which was considered a cool, but not cold, temperature. It is possible that survival at 10°C could have been enhanced if all of these turtles were submerged during the same period as the animals studied at 3°C. Viability results for emydid turtles in normoxic water were frequently confounded by fungal infections. Such infections occurred only in normoxia, indicating a dependence of the fungus upon oxygen. The infections were unpredictable and generally resistant to treatment. Therefore, when survival times are being considered for animals that were so infected (see species accounts), they should be treated as minimal estimates of viability. Not all of the data reported here come from investigations that were designed only to test viability, and in some of those that were so designed, not every turtle was allowed to die. Deviations from a protocol requiring the death of all submerged animals are indicated on Fig. 1 by a star after the sample size, and are discussed under the accounts of results for each species.
609
Species accounts The survival times of individual turtles submerged under various conditions of temperature and oxygenation are presented in Fig. 1. Appropriate notes follow: 1. Chrysemys scripta adults in normoxia became infected with fungus about the head and limbs, while the juveniles did not. Therefore, while it is tempting to invoke surface to mass ratio arguments concerning the longer survival of the smaller animals, it would not be appropriate to do so with these data. Only two of the juveniles in normoxia died while submerged; the others were removed and allowed to warm to room temperature in 1 cm of water. One removed on day 179 died 6 days later; of 4 removed on day 182 (chosen to be equivalent to half a year of submergence), one died 8 days later and 3 recovered completely and were released. All animals removed alive appeared bloated, indicating at least a partial failure of water balance mechanisms, and all had an unknown eye infection that responded to treatment with an ophthalmic cream. There was no significant difference in survival time between adults and juveniles in anoxic water. 2. Chrysemys concinna in normoxic water also had fungal infections, but they did not appear as early, or seem as severe, as with C. scripta. 3. Kinosternon subrubrum has been shown to be capable of hibernation on land, and this is one of very few aquatic turtles that may do so routinely (Bennett, 1972; Bennett et al., 1970). Terrestrial hibernation may be related to the low survival time of this species in normoxic submergence. However, the sample size here is small, therefore these results should be considered as preliminary. 4. The difference in survival times of Chelydra serpentina in normoxia and anoxia at 10°C was not significant, but I suspect that this was probably more due to sample size than to any other factor, and that there would be a significant difference at 3°C. There was no problem with fungus in the normoxic situation. 5. Trionyx spiniferus asperus was the most tolerant of all species of normoxic submergence at 10°C and the least tolerant of anoxic submergence. After 100 days of normoxic submergence, 7 of 11 were still alive; the 4 that had died were small individuals (l&8 1 g). Small animals seemed especially susceptible to a fungus that attacked the shell. In anoxia, there was a trend toward longer survival of larger individuals, suggestive of an importance of decreasing weight-specific metabolic rate (lactate formation, in this case) with increasing size and its interaction with whole-body buffering capacity. This was the only trend related to body size in this study where the sample size (15), size range (9.2-3021 g), and the lack of fungal infections, suggested that body size might be a parameter influencing submergence viability. 6. Sternotherus odoratus is also capable of efficient extrapulmonary gas exchange (Root, 1949) as was evidenced by the survival of 5 of 8 turtles for 100 days in normoxic water at 10°C. These animals were in good health when removed for blood sampling (Ultsch et al., 1984). As with Trionyx, survival at 10°C in anoxic water was of short duration. At 3°C the results are qualitatively similar to those at 10°C. The
610
GORDONR. ULTSCH
first death in anoxia occurred between days 16 and 2 1, and all died by day 31. In normoxia, the first death did not occur until between days 135 and 150. After 150 days, 4 turtles were removed for physiological studies, and 5 were placed in shallow water at room temperature. Of the 5, 3 died within 2 days and 2 recovered completely and were released. 7. Belkin (1968b) found that Sternotherus minor minor from Florida lived only 2.5 days when submerged in normoxic water at 22°C. I found a single Sternotherus minor pelt@ from Alabama to survive 126-128 days of normoxic submergence at 3°C. This result emphasizes the importance of temperature in affecting survival. 8. The painted turtle (Chrysemys picta) is particularly interesting because it is the only nearctic species with a range that is continuous from the east to west coasts and is also one of only a few species that range from the Gulf Coast to southern Canada. Chrysemys picta bellii is the most northerly subspecies, and its survival at 3°C has been summarized previously (Ultsch and Jackson, 1982a). Of 6 removed from normoxia after 189 days, 5 appeared to be in moderately good health, but they were not allowed to attempt recovery. Turtles removed from anoxia on days 155-168 were in very poor health. At 10°C the shortened survival of animals in normoxia was partially due to the higher temperature and possibly partially due to fungal infections. It is not possible to segregate the relative importance of the two effects, but it appears likely that temperature was the more influential, since the ratio of survival at 10°C to that at 3°C is about the same for both normoxia and anoxia. Southern painted turtles (C. p. dorsalis) were submerged in anoxia at 3°C in a study designed to compare their physiological responses to those of their northern counterparts (Ultsch et al., in press). This necessitated periodic removals of animals, and only one was kept submerged to give an indication of maximal viability. That animal, chosen as one of the apparently healthiest of the group, lived 86 days, only about half as long as the northern form. Physiological and ecological trends and implications. It is apparent that all of the species studied can benefit from extrapulmonary gas exchange while submerged in normoxic water at low temperatures. This was to be expected for Sternotherus and Trionyx, which are thin-skinned turtles with favorable surface to volume ratios (due to size in the case of Sternotherus and shape in the case of Trionyx), and which also utilize buccopharyngeal respiration. It was, however, only recently that the importance of extrapulmonary gas exchange was demonstrated for an emydid species (Ultsch and Jackson, 1982a and b; Jackson and Ultsch, 1982). I suggest that probably all freshwater turtles can efficaciously utilize extrapulmonary gas exchange when submerged in normoxic water at low temperature. The relative inability of certain species (Sternotherus and Trionyx) to tolerate submergence anoxia suggests that they may be excluded from hibernation in mud, reputed to be a typical microenvironment for overwintering freshwater turtles. This hypothesis, however, is subject to several reservations. Most of the turtles used here were from the southern United
States, and southern forms of species with wide latitudinal ranges may not be as tolerant of prolonged submergence as northern forms; this has been shown to be the case for Chrysemys picta (Fig. 1; Ultsch et al., in press). Secondly, I report no data here for Trionyx at 3X, although it is difficult to conceive of physiological adjustments that would allow this species to tolerate months of anoxia in the northern part of its range when southern animals only survived 0.8-6.8 days at 10°C. Finally, Ultsch and Lee (1984), found that 3 of 5 Chelydra serpentina in Rhode Island survived at least 3.5 months buried in mud during the winter, although air-breathing could not be ruled out for two of them, or buccopharyngeal respiration from the water column for the third. There is a virtual complete lack of solid information on the behaviour of freshwater turtles in the winter, other than occasional notes on turtles swimming under the ice or caught by muskrat trappers in winter (R. Meeks, personal communication), and until individual turtles are tracked in their home ranges throughout a winter, we cannot be certain whether they are hibernating in anoxia or have at least partial access to either dissolved or atmospheric oxygen. While the southern turtles in this study generally fared poorer than the one northern form (C. p. bellii), all southern species survived normoxic submergence longer than they would be likely to have to do so in their natural environment during the winter. In contrast, the survival in anoxia of southern turtles, including the southern form of C. picta dorsalis, while much less than that of C. p. bellii, is not necessarily mismatched with the time that they might have to spend in hibernation during a typical cold period during the winter. If turtles do in fact spend most of their hibernation time in anoxic mud, then the viability of all the forms studied here (save perhaps Trionyx and Sternotherus) is reasonably well matched to their expected hibernation period. Thus the northern limit of each race may be dependent upon its tolerance to anoxic submergence, and this tolerance may be rather plastic in an evolutionary sense, within a species. However, such reasoning is speculative at this time. There is a critical need for definitive field studies of hibernation in freshwater turtles (and other freshwater ectotherms), and until these are available, physiological studies will continue to be only suggestive. Acknowledgements-I thank Dennis Lee and Kenneth O’Brien for technical aid. This study was partially supported by NSF Grants PCM 79-l 1609 to the author and PCM 78-
22333 to Donald C. Jackson of Brown University and by a Research Grants Committee grant from the University of Alabama. Robert Guthrie was of invaluable aid in collecting the turtles. REFERENCES Belkin D. A. (1963) Anoxia:
tolerance
139,492-493. Belkin D. A. (1968a) Anaerobic
in reptiles.
Science
brain function: effects of stagnant and anoxic anoxia on persistance of breathing in reptiles. Science 162, 1017-1818. Belkin D. A. (1968b) Aquatic respiration and underwater survival of two freshwater turtle species. Respir. Physiol. 4, 1-14. Bellamy D. and Petersen J. A. (1968) Anaerobiosis and the
Submergence viability of freshwater turtles toxicity of cyanide in turtles. Comp. Biochem. Physiol. 24, 543-548.
Bennett D. H. (1972) Notes on the terrestrial wintering of mud turtles (Kinosternon subrubrum). Herpetologica 28, 2455247.
picture of the turtle after complete anoxia. J. biol. Chem. 103, 107-l 14. Musacchia X. J. (1959) The viability of Chrysemys picta submerged at various temperatures. Physiol. Zoo. 32, 47-50.
Bennett D. H., Gibbons J. W. and Franson J. C. (1970) Terrestrial activity in aquatic turtles. Ecology 51,738-740. Clark V. and Miller A. (1973) Studies on anaerobic metabolism in the freshwater turtle (Pseudemys scripta eleguns). Comp. Biochem.
611
Physiol. A, 55-62.
_
Greeorv P. T. (1983) Rentilian hibernation. In Biology of the Reptilia. (Edited by Gans C. and Pough F. H.), Vol. 13, pp. 53-154. Academic Press, New York. Jackson D. C. (1968) Metabolic depression and oxygen depletion in the diving turtle. J. appl. Physiol. 24,503-509. Jackson D. C. and Heisler N. (1982) Plasma ion balance of submerged anoxic turtles at 3°C: the role of calcium lactate formation. Respir. Physiol. 49, 1599174. Jackson D. C. and Heisler N. (1983) Intracellular and extracellular acid-base and electrolyte status of submerged anoxic turtles at 3’C. Respir. Physiol. 53,187-201. Jackson D. C. and Schmidt-Nielsen K. (1966) Heat production during diving in the fresh water turtle, Pseudemys scripta. J. cell. Physiol. 61, 225-232.
Jackson D. C. and Ultsch G. R. (1982) Long-term submergence at 3°C of the turtle, Chrysemys pictu bellii, in normoxic and severely hypoxic water-II. Extracellular ionic responses to extreme lactic acidosis. J. exp. Biol. 96, 29-43.
Johlin J. M. and Moreland F. B. (1933) Studies of the blood
Root R. W. (1949) Aquatic respiration in the musk turtle. Physiol. Zool. 22, 172-178.
Smith H. M. and Nickon D. C. (1961) Preliminary experiments on the role of the cloaca1 bursae in hibernating turtles. Natural History Miscellanea of the Chicago Academy of Sciences No. 178. 8 pp. Ultsch G. R., Hanley R. W. and Bauman T. R. (1985) Responses to anoxia during simulated hibernation in northern and southern painted turtles. Ecology. In press. Ultsch G. R., Herbert C. V. and Jackson D. C. 1984) The comparative physiology of diving in North American freshwater turtles-I. Submergence tolerance, gas exchange and acid-base status. Physiol. Zool. 57, 62M31.
Ultsch G. R. and Jackson D. C. (1982a) Long-term submergence at 3°C of the turtle Chrysemys picta bellii, in normoxic and severely hypoxic water-I. Survival, gas exchange and acid-base status. J. exp. Biol. 96, 1l-28. Ultsch G. R. and Jackson D. C. (1982b) Long-term submergence at 3°C of the turtle, Chrysemys picta bellii, in normoxic and severely hypoxic water-III. Effects of changes in ambient PO, and subsequent air breathing. J. exp. Biol. 97, 87-99.
Ultsch G. R. and Lee D. (1984) Radiotelemetric observations of wintering snapping turtles (Chelydra serpentina) in Rhode Island. J. Alabama Acad. Sci. 54,20@-206.