by Eli B. Greenbaum
Department of Biology
Northeast Louisiana University
Monroe, LA 71209
Reprinted from The Bulletin of the Chicago Herpetological Society, Vol. 34, No. 4 April 1999.
In August of 1996, I was presented with the task of describing gonad formation in the spiny softshell turtle, (Apalone spinifera). Ostensibly, this seemed like a job that would have little interest, but I soon learned that it would have great significance to scientists interested in sex determination. The reason? Softshell turtles are one of the few turtles currently known to have genetic sex determination (Table 1).
|Table 1. Sex determination in reptiles.|
|Genetic Sex Determination (GSD)|
|Some turtles, including:|
|Subfamily Staurotypinae (Kinosternidae)|
|Siebenrockiella crassicollis (Batiguridae)|
|Chelodina longicollis (Chelidae)|
|Emydura signata (Chelidae)|
|Clemmys insculpta (Emydidae)|
|Apalone mutica (Trionychidae)|
|Apalone spinifera* (Trionychidae)|
|Temperature-dependent Sex Determination (TSD)|
|Kinosternidae, except Staurotypinae|
|*This species is the focus of this study.|
Most turtles actually have what is known as temperature-dependent sex determination (TSD), a condition in which the sexual outcome of the hatchlings is determined by the ambient incubation temperature. Among reptiles, three types of TSD are known: designated Pattern Ia, Pattern Ib and Pattern II. In Pattern Ia, high temperatures yield a high frequency of females. Pattern Ib reptiles yield males at high temperatures. And Pattern II species yield females at high and low temperatures, with males emerging from intermediate temperatures. It is probable that the latter pattern of TSD is the most primitive. Note that the Kinosternidae, Carettochelyidae and Trionychidae are three families of turtles that are closely related, yet display widely varying forms of sex determination (Gaffney, 1984). The ineluctable question is why?
Charnov and Bull (1977) proposed an explanation for the existence of TSD. If a differential advantage exists between males and females in an environment for resource use, predator avoidance, or mate acquisition, then it pays for TSD to exist. In this way, a cue from the environment (i.e., temperature) can be employed to determine the sex of the developing embryos to their best advantage. Because it pays for these embryos to get as much environmental information as possible before irreversibly advancing towards the formation of an ovary or testis, it is to their best advantage to delay sexual differentiation as much as possible before hatching. This is precisely what happens in species with TSD.
In discussing embryological development in turtles, scientists have developed a system in which 26 stages of development are utilized (Yntema, 1968). Stage 1 is about the time of egg laying and stage 26 is hatching. Gonads start to develop into an ovary or testis between stages 18 and 20 in TSD turtle species. However, changes in temperature can reverse the path of sexual differentiation after it has begun (Table 2). Ultimately, temperature can delay sexual differentiation in TSD turtle species until stage 22 in some cases.
|Table 2. Sexual differentiation.|
|* Changing the temperature at these stages can reverse
the path of sexual differentiation from male to female or
Although many aspects of the phenomenon are not understood, the theory above gives a basic framework for one to consider TSD. But what about the few known cases in which genetic sex determination (GSD) exists? Charnov and Bull (1977) hypothesized that GSD would arise from species with TSD if there was not an advantage to one sex in the environment. GSD appears to have arisen independently four to six times in turtles (Ewert and Nelson, 1991). But if it does not pay for these turtles to delay sexual differentiation, would it actually occur earlier in embryonic development? To my surprise, none of the scientists studying this phenomenon in turtles had addressed this question in the 20-year history of the research.
Since the spiny softshell turtle (Apalone spinifera) has GSD and is relatively common in Louisiana, I decided to investigate this species to determine at what stage sexual differentiation occurs. I hypothesized that sexual differentiation would occur earlier than in TSD species since Apalone has no adaptive advantage to delaying it. In order to do this study, I would have to obtain a couple hundred eggs, sacrifice embryos at various stages of development, make slides of the gonads to look for signs of sexual differentiation, and compare what I found to a recent study of sexual differentiation in the emydid Trachemys scripta (Wibbels et al., 1991).
The first task of the experiment, involving egg collection, was horribly unsuccessful. For months in the spring and summer of 1997, I searched for softshell turtle nests at Black Bayou Lake National Wildlife Reserve near Monroe, Louisiana. It was easy to spot the large disturbed areas of gravel or sand where females had dug nests, but in almost every case predators such as raccoons, crows, or fire ants destroyed the eggs before I could find them. I turned to Jesse Evans, a turtle farmer in Wildsville, Louisiana, to obtain more eggs. He graciously donated 180 eggs to the research.
The eggs were randomly divided and put into incubators, one set for 31°C and the other for 26°C. This was to ensure that Apalone spinifera did indeed have GSD, and that temperature would have no effect on sex determination of the embryos. This fact was confirmed by the conclusion of the experiment.
I started sacrificing the embryos at stage 11 of development, and all went smoothly for a while. However, the older the embryos became, the harder it was to compare them to Yntema’s (1968) staging criteria for the common snapping turtle, (Chelydra serpentina). In addition, a more recent staging scheme for painted turtles (Chrysemys picta) confused things by using only 23 stages of development (Mahmoud et al., 1973). By stage 24 (see Figure 1), none of the snapping turtle criteria could be used with my softshell embryos, and it became apparent that I would need to make my own staging criteria for Apalone spinifera if I were to proceed with the experiment. I spent months meticulously developing a staging methodology and discovered that some tiny details could distinguish softshell and snapping turtle embryos as early as stage 13!
When the staging problem was finally resolved in early 1998, I was able to extract gonads from the embryos and prepare histological slides. Germ cells, changes in the cortex structure, and Müllerian duct development showed whether the gonad was developing into an ovary or testis (Figure 2). The majority of the gonads had sexually differentiated by stage 19, exactly when such differentiation was occurring in TSD species. Superficially, there appeared to be no difference in the timing of gonadal differentiation between Apalone spinifera and TSD turtle species.
However, I did discover something intriguing about the Müllerian ducts. These ducts eventually form the oviducts in sexually mature females, and in males they degenerate as soon as the testes begin to develop. Apparently, some chemical produced by functioning testes breaks down the duct in males. In males of A. spinifera, the duct is first seen at stage 16, degenerates at stage 20, and then becomes undetectable by stage 21.
Although other turtles do experience degeneration of Müllerian ducts in males, very few of them degenerate to the point of being undetectable (Fox, 1977; Risley, 1933). However in a study of Carettochelys insculpta, a turtle species closely related to the trionychids, it was found that some males may have no recognizable Müllerian duct by hatching (Webb et al., 1986). The complete disappearance of this duct might be a useful tool for future systematics work.
But what were the veritable consequences of the finding that Apalone spinifera differentiates sexually by stage 19? Although this chronology is synchronous with TSD turtle species, the fact that sexual differentiation in softshells is irreversible changes the interpretation of the finding. Since TSD species can reverse their path of sexual differentiation up to three stages later, it appears that A. spinifera has a different agenda. This study suggests that A. spinifera is irreversibly sexually differentiating earlier than TSD species.
What adaptive significance does this have? I propose that GSD species of turtles are trying to reach sexual maturity as soon as possible in order to maximize their reproductive fitness over their lifetime. Although further study is needed to confirm this theory, anecdotal observations of A. spinifera suggest that males are reaching sexual maturity much earlier than TSD turtle species (Dr. J. L. Carr, pers. com.; Webb, 1962; Graham, 1991). Scientists are still far from understanding the phenomenon of sex determination in reptiles, since less than 10% of known reptiles have been studied with regard to their mode of sex determination (Ewert and Nelson, 1991). However, this study has brought us one step closer to solving the puzzle and shows just how intriguing this aspect of reptilian biology really is.
Dr. John L. Carr guided me in every step of this project from theoretical conceptualization to refinement of the thesis transcript, and I owe him a tremendous debt of gratitude. Drs. John Knesel, Neil Douglas, Frank Pezold, and Mike Ewert assisted me in several aspects of this study. I am also grateful for the efforts of Dennis Bell, Martha Ann Messinger, George Patton, and Paul Smith. This work was supported by a Grant-in-Aid-of-Research from Sigma Xi and a Grant-in-Herpetology from the Chicago Herpetological Society.
Charnov, E. L., and J. J. Bull. 1977. When is sex environmentally determined? Nature 266:828-830.
Ewert, M. A., and C. E. Nelson. 1991. Sex determination in turtles: Diverse patterns and some possible adaptive values. Copeia 1991:50-69.
Fox, H. 1977. The urinogenital system of reptiles. In: C. Gans, editor, Biology of the Reptilia. Volume 6. New York: Academic Press Inc. Gaffney, E. S. 1984. Historical analysis of theories of chelonian relationship. Syst. Zool. 33:283-301.
Graham, T. E. 1991. Life history notes: Apalone spinifera spinifera (Eastern Spiny Softshell). Pattern dimorphism. Herp. Rev. 22:97.
Mahmoud, I.Y., G. L. Hess, and J. Klicka. 1973. Normal embryonic stages of the western painted turtle, Chrysemys picta belli. J. Morph. 141:269-280.
Risley, P. L. 1933. Contributions on the development of the reproductive system in the musk turtle, Sternotherus odoralus (Latreille). II. Gonadogenesis and sex differentiation. Z. Zellforsch. 18:493-541.
Webb, G. J.W., D. Choquenot and P. J. Whitehead. 1986. Nests, eggs and embryonic development of Carettochelys insculpta (Chelonia: Carettochelidae) from northern Australia. J. Zool., Lond. (B)1:521-550.
Webb, R.G. 1962. North American recent soft-shelled turtles (Family Trionychidae). Univ. Kan. Pub., Mus. Nat. Hist. 13:429-611.
Wibbels, T., J. J. Bull and D. Crews. 1991. Chronology and morphology of temperature-dependent sex determination. J. Exper. Zool. 260:371-381.
Yntema, C. L. 1968. A series of stages in the embryonic development of Chelydra serpentina. J. Morph. 125:219-251.