Alter incubation temperatures to achieve optimal/desired sex ratio: Tortoises, terrapins, side-necked & softshell turtles

Supporting evidence from individual studies

1. A replicated, randomized study in 1992 in a laboratory in the USA (Burke et al. 1996) found that altering the incubation temperature of gopher tortoise Gopherus polyphemus eggs resulted in different sex ratios of hatchlings. Cooler incubation temperatures produced more males, whereas warmer temperatures produced more females (26°C: 4 male, 0 female; 29°C: 3 male, 1 female; 32°C: 0 male, 4 female). Overall hatching success was 77% (20 of 26 eggs), and incubation period was longer at lower temperatures (26°C: 115 days; 29°C: 97 days; 32°C: 86 days). In June 1992, four wild-caught tortoises were induced with oxytocin and the 26 eggs produced were randomly assigned to the 26, 29 or 32°C treatment. Eggs were placed in containers of moist vermiculite, and every two weeks containers were rotated, and water levels topped up. After 10 months in captivity 17 hatchlings had died and 13 of these were sexed successfully.

   Study and other actions tested

   Referenced paper

   Burke R.L., Ewert M.A., McLemore J.B. & Jackson D.R. (1996) Temperature-dependent sex determination and hatching success in the gopher tortoise (Gopherus polyphemus). Chelonian Conservation and Biology, 2, 86-88.

2. A replicated study in 2002–2004 in a laboratory in Oklahoma, USA (Ligon & Lovern 2009) found that incubating alligator snapping turtle Macrochelys temminckii eggs at higher temperatures resulted in strongly female-biased sex ratios, though the coldest and warmest temperatures resulted in very low hatching success. Warmer incubation temperatures produced almost all female hatchlings (28.5°C or 30.5°C resulting in 97% and 100% of female hatchlings), whereas incubating at 26.5°C resulted in 81% male hatchlings. The coldest (23.0 and 24.5°C) and warmest (31.0°C) incubation temperatures resulted in 0% hatching success, and hatching success was higher at 26.5°C (85%) compared to cooler (26.0°C: 33%) or warmer (30.5°C: 40%) temperatures. Eggs were obtained in 2002 (3 clutches of 15–37 eggs, 88 total) and 2004 (6 clutches of 17–42 eggs, 186 in total) and split evenly between six incubation temperatures in 2002 (23.0, 24.5, 26.0, 26.5, 28.5 and 31.0°C; 12–13 eggs/temperature) and three temperatures in 2004 (53 eggs at 26.5°C; 51 at 28.5°C; 47 eggs at 30.5°C). Eggs were incubated in damp vermiculite (1:1 ratio with water by mass). Hatching success was assessed and hatchlings were sexed 267–278 after hatching by observing gonads via a non-lethal surgical procedure.

   Study and other actions tested

   Referenced paper

   Ligon D.B. & Lovern M.B. (2009) Temperature effects during early life stages of the alligator snapping turtle (Macrochelys temminckii). Chelonian Conservation and Biology, 8, 74-83.

3. A study in 2005–2007 in laboratory conditions in Colombia (Paez et al. 2009) found that higher incubation temperatures increased Magdalena river turtle Podocnemis lewyana hatching success, and that females were only produced above a temperature threshold. In the first year, eggs incubated at 33.0°C had a higher hatching success rate (28 of 29, 97% eggs hatched) than eggs incubated at 28.0°C (14 of 38, 37% eggs hatched) and all hatchlings at both temperatures were male. In the second year, five of 10 eggs (50%) incubated at 33.4°C and seven of seven eggs (100%) incubated at 34.7°C produced female hatchlings. The authors reported that nests monitored in the field showed a similar pattern, with the coldest nests having lower hatching rates (see original paper for details). In 2006, river turtle eggs were obtained from 28 nests and incubated in a laboratory at: 28.0°C (38 eggs), 29.5°C (43 eggs), 32.0°C (39 eggs) and 33.0°C (29 eggs). In 2007, river turtle eggs were obtained by inducing four female river turtles using an injection of oxytocin and incubated at: 33.4°C (14 eggs) and 34.7°C (13 eggs). Eggs were monitored through to hatching. Only hatching success data from 2006 laboratory eggs are included as the authors reported that the use of oxytocin to obtain eggs in 2007 may have affected hatching rates. Natural nests were monitored in the field in 2005 (8 nests) and 2006 (11 nests) for incubation temperature and hatching success.

   Study and other actions tested

   Referenced paper

   Paez V.P., Correa J.C., Cano A.M. & Bock B.C. (2009) A comparison of maternal and temperature effects on sex, size, and growth of hatchlings of the Magdalena river turtle (Podocnemis lewyana) incubated under field and controlled laboratory conditions. Copeia, 2009, 698-704.

4. A before-and-after study in 2006–2009 at a captive breeding facility in southern California, USA (Kuchling et al. 2013) found that radiated tortoise Astrochelys radiata most eggs incubated at 28.9°C or higher produced female hatchlings. Results were not statistically tested. At 28.9°C, twenty-three of 25 hatchlings (92%) were female, and at 30°C, all 29 hatchlings were female. Eggs from captive tortoises were collected and incubated in modified wine coolers at 28.9°C in 2006–2007, and at 30°C in 2008–2009. Eggs were incubated in vermiculite and water at a 2:1 ratio.

   Study and other actions tested

   Referenced paper

   Kuchling G., Goode E. & Praschag P. (2013) Endoscopic imaging of gonads, sex ration, and temperature-dependent sex determination in juvenile captive-bred radiated tortoises, Astrochelys radiata. Chelonian Research Monographs, 6, 113-118.

5. A replicated, randomized study in 2008–2009 in laboratory conditions in California, USA (Geist et al. 2015) found that altering incubation temperatures of western pond turtle Actinemys marmorata eggs resulted in variable hatching success, and that no female hatchlings were produced at lower temperatures. Hatching success varied with temperature, with the highest reported success rate at 29°C (26 of 28, 93%) and the lowest at 31°C (3 of 7, 43%). Hatching success at other temperatures (26, 27, 28 or 30°C) ranged from 68–82%. Eggs at 26–27°C produced all male hatchlings and those at 28–29°C were highly skewed towards males (28°C: 92% males; 29°C: 85% males), while those at 30°C produced all females. In 2008–2009, eggs were obtained from 44 wild turtle nests. Eggs were distributed evenly between five plastic containers that were partially filled with moist vermiculite (5:1 ratio with water by volume) and incubated at constant temperatures. Five temperatures were chosen in 2008, and these were all decreased by 1°C in 2009, resulting in the following number of eggs/treatment: 15 eggs at 26°C; 28 at each of 27, 28 and 29°C; 25 at 30°C; seven at 31°C. In 2009, the sex of fifty-nine turtles was determined through a non-lethal surgical procedure that allowed gonads to be observed (see paper for details).

   Study and other actions tested

   Referenced paper

   Geist N.R., Dallara Z. & Gordon R. (2015) The role of incubation temperature and clutch effects in development and phenotype of head-started western pond turtles (Emys marmorata). Herpetological Conservation and Biology, 10, 489-503.

6. A replicated, randomized study in 2012 in laboratory conditions in Columbia (Gómez-Saldarriaga et al. 2016) found that incubating Magdalena River turtle Podocnemis lewyana eggs at lower temperatures produced more male hatchlings, and higher temperatures produced more females. Hatching success ranged from 57–100%. Lower incubation temperatures resulted in fewer female hatchlings (29°C: 8% female; 31°C: 18%; 34.7°C: 86%). When incubating at 29°C, a 10-day high temperature pulse resulted in more female hatchlings compared to the constant temperature if it came during day 21–30 (37% female), but a similar number if it came at day 31–50 (4–20% female). At 31°C, a pulse during day 21–50 resulted in more females than the constant temperature (32–67% female). In 2012, a total of 227 eggs were collected from 10 nests (14 beaches searched). Eggs were incubated at either a constant temperature (29, 31 or 34.7°C; 30–31 eggs/temperature), or at 29 or 31°C with a 10-day period at a high temperature (35.1–35.5°C) during day 21–30, 31–40 or 41–50 (19–25 eggs/treatment). Sex of 20 individuals was determined by assessing gonadal histology. Detailed morphology of these 20 individuals was used to estimate the sex of all hatchlings (see paper for details).

   Study and other actions tested

   Referenced paper

   Gómez-Saldarriaga C., Valenzuela N. & Ceballos C.P. (2016) Effects of incubation temperature on sex determination in the endangered Magdalena river turtle, Podocnemis lewyana. Chelonian Conservation and Biology, 15, 43-53.

7. A controlled study in 2003 on a sandy beach in Amazonas, Brazil (Eisemberg et al. 2017) found that covering six-tubercled Amazon River turtle Podocnemis sextuberculata nests with black plastic sheeting increased the proportion of female hatchlings, but decreased hatching success.  River turtle nests covered with black plastic sheeting produced half the number of male hatchlings (1.5 of 5, 30% hatchlings/nest were male) compared to uncovered nests (3.3 of 5, 66% hatchlings/nest were male) and covered nests had lower hatching success (80%) than uncovered nests (92%). In September-November 2003, thirty turtle nests laid on a river-side beach (2 km long, 600 m wide) in a reserve were monitored from within 12 hours of being laid through to hatchling emergence. Fifteen of 30 nests were covered with a sheet of black plastic (0.1 mm thick covering a 2 m² area) in order to influence hatchling sex ratios. The remaining fifteen nests were monitored but not covered. Black plastic was removed after 50 days of incubation and nests were covered with nets to capture hatchlings as they emerged. Sex ratios were determined by sacrificing five hatchlings/nest and carrying out an examination.

   Study and other actions tested

   Referenced paper

   Eisemberg C.C., Drummond G.M. & Vogt R.C. (2017) Boosting female hatchling production in endangered, male-biased turtle populations. Wildlife Society Bulletin, 41, 810-815.

8. A study in 1986 in a captive rearing facility in Galápagos, Ecuador (Sancho et al. 2017) found that less than half of artificially incubated Española giant tortoise Chelonoidis hoodensis eggs hatched in captivity and that the sex ratio was temperature dependent. Results were not statistically tested. Española giant tortoise eggs artificially incubated at 25.5°C had a hatching success of 16% and 10 of 11 (91%) sexed hatchlings were male. Eggs incubated at 29.5°C had a hatching success of 40% and five of 15 (33%) sexed hatchlings were male. No eggs artificially incubated at 33.5°C hatched successfully. In 1986, giant tortoise eggs laid in captivity as part of a head-starting programme were incubated at three different temperatures: 25.5, 29.5 and 33.5°C (67 eggs/temperature, 189 total eggs). Eggs were placed in plastic boxes with damp vermiculite, covered and put in incubation chambers. Hatchlings were sexed by direct observation, examination of dead hatchlings’ gonads (35 individuals) or key-hole surgery (15 individuals). Data from six hatchlings that hatched earlier in the season in the same facility were included in the results.

   Study and other actions tested

   Referenced paper

   Sancho A., Gutzke W.H.N., Snell H.L., Rea S., Wilson M. & Burke R.L. (2017) Temperature sex determination, incubation duration, and hatchling sexual dimorphism in the Espanola giant tortoise (Chelonoidis hoodensis) of the Galapagos Islands. Amphibian & Reptile Conservation, 11, 44-50.