A Blog by Ed Yong

Turtle embryos bask against the warmest side of their own eggs

Many animals, including the majority of reptiles, cannot produce their own body heat. To control their temperature, they have to use their environment. They bask in the sun to heat up and lounge in the shade to cool down. And some of them start before they’ve even hatched.

Wei-Guo Du from the Chinese Academy of Sciences has found that the embryos of soft-shelled turtles “bask” inside their eggs. “People usually think reptilian embryos are immobile,” says Du. After all, their limbs are tiny stumps and they have few places to move to. But that doesn’t stop them. Du found that the embryos can not only move, but they can snuggle up to the warmest side of their eggs.

He collected 260 eggs from a local turtle farm, placed them in individual jars and warmed them with heat lamps. The eggs were about one degree Celsius warmer on the sides closest to the lamps, and the turtles could sense this. After a few days, they had pressed up against the warmer side. When Du moved the heat lamps around, the embryos followed.

Du also buried 540 eggs in more natural nests, either in a flat field or a sloping river bank. He arranged all the eggs so the embryos were sitting in the uppermost half (he could tell where they were by shining a candle behind them and looking at the silhouette). After 20 days, the embryos in the field, with the sun warming them from directly overhead, stayed in the same position. However, the embryos buried on the slope, with the sun shining down on them at an angle, shifted towards the bank.

Do these results have any relevance to natural conditions? Du thinks so. “Most reptiles lay eggs in underground nests, which are heated by sunlight falling on the ground above,” he writes. These can create uneven temperatures within the nest – some eggs will be warmer than others and some parts of each egg will be warmer than other parts.

Soft-shelled turtle embryos can sense these differences and respond to them, basking in the warmth before they’re even hatched. “To our knowledge, no previous study has looked for this ability, presumably because embryos were thought incapable of such behaviour,” says Du.

A small difference in temperature can mean the world to a developing turtle. It can affect how quickly it grows, how big it gets, and when it hatches. Heat a turtle embryo up by one degree – the temperature difference that Du found in his experiment – and it will hatch around 4.5 days earlier. That’s 4.5 days when it won’t be a sitting duck for egg predators, and when it can escape from extremes of temperature or drought.

Mother turtles may decide where to bury the eggs, but the embryos aren’t passive. By moving about in their cramped homes, they have a small say in their own fates. Du says, “The embryo is not simply a work in progress, but is a functioning organism with surprisingly sophisticated and effective means of affecting its own destiny.”

This discovery raises a tantalising possibility. For many reptiles, the temperature of the egg determines the sex of the babies. In some turtles, if eggs are incubated at 22.5 to 27 degrees, the babies will be all male. If they’re incubated at 30 degrees, they’ll all be female. Temperatures in the middle will yield a mixed-sex clutch. For the moment, Du doesn’t know if the soft-shelled turtle relies on temperature in this way, or if other reptiles that do rely on temperature can move about in their own eggs. If either statement is true, then they may be able to control their own sex.

Reference: Du, Zhao, Chen & Shine. 2011. Behavioral thermoregulation by turtle embryos. PNAS http://dx.doi.org/10.1073/pnas.1102965108

More on reptiles and temperature:

10 thoughts on “Turtle embryos bask against the warmest side of their own eggs

  1. “For many reptiles, the temperature of the egg determines the sex of the babies. ”

    I remember crocodiles have the same Temperature-dependent sex determination (TSD). I always wonder wether such reproductive pattern is vulnerable to global climate change. If temps go one direction permantly, they all end up with one gender and extinction. Did dinosaurs have similar TSD?

  2. Word on the street is that Chinese softshells are genetically determined (jury’s still out, but see http://www.ncbi.nlm.nih.gov/pubmed/19729916), but that doesn’t preclude a temperature effect on determination, as has been seen in some other reptilians (e.g., Bassiana duperreyi). Even with genetic determination, temperature can still influence a vast array of fitness-related parameters, as you noted. I’ve always considered temperature a too-overlooked environmental input (The Journal of Thermal Biology notwithstanding), but that’s probably just my bias. Even spiders have shown influences of temperature, and there’s probably so much more to find out about what temperature does across the animal kingdom during development. I’m thrilled to see this post. Thanks.

  3. AG…some people have certainly speculated that this pattern is vulnerable. Some have even gone so far as to speculate that the age of the dinosaurs ended because of an interaction of TSD and climate change (e.g., http://rsbl.royalsocietypublishing.org/content/7/2/295.abstract). Crocodilians do all exhibit TSD, as do many turtles, some lizards, and the tuatara. It’s been hypothesized that TSD is a lingering life history strategy from the deep evolutionary history of reptilians. Given the age of the tuatara, that may certainly be the case. But…birds (as dinosaurian descendents) are genetically determined, so the TSD family tree is a confusing one, at best. As I noted above, some species even exhibit both. I’ve also published a paper on the interaction of temperature and environmental contaminants that mimic steroid hormones (the endpoint of temperature’s influence in TSD species), and they do interact. There’s certainly evidence that temperature changes and other factors, such as these contaminants, might influence sex ratios.

  4. That’s the coolest thing I’ve seen in a long time, also love the pictures! Just wow.
    Regarding the possible controlling of sex – how would the embryo choose which sex to become? Or rather why? It seems damn hard to prove that choice of sex is a factor influencing movement in the egg…

  5. Sam…one speculative response to your question is that the temperature trigger draws that “taxis” of the embryo, something that perhaps is selected for so that embryos that would be better fit at a given sex with an earlier emergence do, in fact, emerge earlier and with the “right” sex because of the warmer incubation temp they receive. The embryo doesn’t actively make the choice, of course, but perhaps there is selection for earlier emergence at, say, a peak temp of the season for a specific sex for whatever reason: more food, smaller organism (because of shorter incubation) can evade predators better, blah blah blah. Depends on the species, too. Some turtles do high/boys–low/girls, others reverse that, and some animals show low-high temps=one sex, mid-range=the other sex…so the “fitness” factors and interactions with sex are likely very species specific.

    As a general example, a turtle that develops as male at low temps but as female at high temps might be male early in the season when being male is beneficial for whatever reasons, but female if incubated later in the season when female might be beneficial for various reasons. The mothers play a role in terms of their choice of nesting site, i.e., in shade, in sun, etc., too. It’s all delightfully fascinating and complex.

  6. Pretty cool!
    I’m very skeptical, however of any connection to TSD. First, afaik the entire family of softshells has genetic SD, or at least all those tested so far do. Second, even in a species with TSD, a thermal gradient of one degree C (the maximum obtained experimentally here) will only affect sex for those few eggs whose temperature would otherwise have been less than 1 degree lower than the pivotal incubation temperature.
    The behavior is far more likely aimed at speeding up the rate of growth and development and thereby getting an earlier start as a larger hatchling.

  7. IMHO it is somewhat doubtful that climate change will have serious effects via TSD (allthough it will have effects through other mechanisms) since many turtle species have rather wide latitudinal ranges with temperature differences along that gradient that will dwarf any that will occur because of AGW. Painted turtles (Chrysemys picta), for example range from northernmost Mexico to the southern edge of the boreal forest in Canada. The temperature differences between nests even at the same site are also very large relative to climactic effects. That is not to say that climate is not important to the developing embryos; it is. Cold climate seems to limit the distribution of northern turtles by limiting the time that embryos have to develop (rather than by constraints on adult hibernation, for example). Also, the pivotal temp at which embryos shift from one sex to another can shift in response to selection (it varies latitudinaly).

    Emily, I would be interested to hear which species of turtles have a cold female/hot male TSD system. The ones that I can think of all have either type Ia (hot female/cold male) or type II (cold female/intermediate male/hot female). It sounds like you know more about this than I do so let me know.

  8. Matt…sorry, that was misleading. I shouldn’t have said “turtles.” After “depends on the species,” I should have said “reptilians,” not “turtles.” I have turtles on the brain. Agama agama follows that high-males/low-females. There are probably others. But you’re right about the turtles. Also, I’ve been out of the TSD biz for about 8 years (although I follow the literature), and things may have changed with more info.

    Regarding temp shifts and climate change…there is also the phenomenon of hatchling overwintering, which some species, like T. scripta show in northern latitudes.

Leave a Reply

Your email address will not be published. Required fields are marked *