Australasian Science: Australia's authority on science since 1938

Mother Knows Best

An adult female green turtle returning to the sea after nesting.

An adult female green turtle returning to the sea after nesting. Photo: T. Franciscus Scheelings

By Anthony Rafferty

Why do turtles lay eggs when their close relatives evolved live birth? A study of their reproductive physiology reveals how egg-laying improves the survival prospects of hatchlings.

Turtle embryos stop developing inside their eggs at just a few days old – when they are still inside the mother’s reproductive tract. They only start growing again after the eggs are laid.

By careful studies in both freshwater and marine turtles, we have found that this strategy gives turtle mothers the flexibility to choose when and where to lay their eggs for the best chances of survival of their young. What’s more, we have been able to work out how they do it.

The Evolution of Live Birth

There are two main ways to reproduce: by laying eggs or giving birth to live young. Egg-laying is considered an ancestral reproductive mode from which live birth evolved. The transition from egg-laying to live birth requires periods of extended egg retention during which embryos develop to increasingly advanced stages inside the mother.

Live birth is generally associated with mammals, but has evolved in nearly 140 reptile lineages and been documented in various prehistoric aquatic reptiles. In these latter instances, live birth presumably allowed an exclusively aquatic existence, free from the need to lay eggs on land.

Surprisingly, live birth of reptiles has only evolved in lizards and snakes but never in turtles or crocodiles, which is fascinating considering these latter groups live a primarily aquatic lifestyle. So why have these reptiles remained egg-layers dependant on land for reproduction when their aquatic predecessors and close relatives evolved live birth?

To answer this question you must first consider the developmental processes that occur during embryonic development in turtles and crocodiles.

In turtles, embryos fail to advance because they stop developing during early stages of cellular division while they are still inside the mother. This extraordinary process is termed preovipositional developmental arrest, and is universal among all turtle species.

In contrast, crocodile embryos are incapable of arresting development, and either become deformed or die when eggs are retained inside the mother for longer than normal.

Considering that embryos of both groups either become arrested or deformed inside the mother if retained for extended periods, it is plausible to think that something necessary for advanced growth inside the reproductive tracts is missing or in limited supply. But what could this be?

The Limiting Factor

Although this question seemed quite simple to my colleagues and I when it was first posed, it became more and more complex the further we waded through the literature looking for answers. Luckily, previous research ruled out temperature, limited water availability or an increase in carbon dioxide concentration as likely candidates for consideration.

Several researchers also hypothesised that limited oxygen may play a role, yet none dared, or knew exactly how, to go about delving into the depths of the reptile reproductive tracts to find out exactly what environment embryos were being exposed to inside mothers. Armed with small endoscopic cameras and tiny oxygen probes, this is exactly what we set out to do, focusing specifically on the process of arrested embryonic development in turtles for our investigation.

Initially, many a day was spent waist-high in the swampy waters of Victoria and Western Australia gathering our freshwater participants, which included the western oblong turtle, the eastern longneck turtle and the Murray River turtle. We also included the green sea turtle in the study, which meant visiting Heron Island on several occasions – the life of a researcher is hard at times!

Palpation of pregnant females identified the presence of calcified eggs, which were later confirmed by X-ray. Full calcification of the eggshell coincides with the onset of developmental arrest in turtles, so we were confident that all eggs were at the appropriate stage of development for the study. Similarly, by only selecting marine turtle females that were preparing to nest on the beach, we knew that all eggs in their tracts were arrested and ready to be laid.

In the lab, pregnant freshwater females were anaesthetised and marine mothers were restrained to reduce movement. An endoscope with a tiny oxygen probe attached to the tip was then directed through the reproductive opening and into the reproductive tracts where eggs were being stored. The entire procedure was visualised on a monitor so that correct placement of the probe inside the tracts could be observed.

With complete astonishment, the entire team watched the oxygen tension measurements plummet in the oviducts of the first turtle. The near absence of any trace of oxygen inside the female mystified all of us and we wondered how this was possible. With each consecutive turtle the results remained the same, stimulating eager conversation that quickly identified the next line of investigation.

It was unanimously concluded that we needed to identify the role that this extremely low oxygen level was having on the process of developmental arrest of the embryos. We hypothesised that the low oxygen was maintaining developmental arrest inside the mothers and that when eggs were laid, the increased oxygen level in the air triggered embryos to start developing again.

To find out if we were right, we artificially induced the pregnant freshwater mums to lay their eggs in the lab, which had become a make-shift labour ward for this aspect of the study. Immediately upon laying, eggs were quickly placed in airtight egg incubation boxes, which were flooded with nitrogen gas to create a low oxygen environment that mimicked the reproductive tract.

The same protocol was also strictly adhered to for the marine turtles, requiring a team member to wait patiently behind a female turtle for her to lay her eggs while shoulder-deep in a nest on the beach. A wayward hind flipper to the face was not uncommon.

To test our hypothesis, we created three treatments that involved exposing eggs to low oxygen levels for 3, 6 and 9 days. Additional eggs were also incubated under normal atmospheric oxygen conditions from the time of laying to act as a control.

Several eggs were examined before and after each treatment so that the stage of embryonic development could be determined and compared between treatments. As predicted, eggs remained in a state of developmental arrest while they were in the low oxygen environment, but development recommenced soon after they were exposed to normal oxygen levels.

After recovering from the initial euphoria of our findings, we all began to question how the oxygen tension inside the reproductive tracts was so low considering that the walls lining the tracts were filled with vessels carrying oxygen-rich blood to the area. Surely there had to be a barrier blocking the diffusion of oxygen from the lining of the tracts to the eggs.

Immediately, we all thought back to the experience that one of the team had encountered while collecting green sea turtle eggs for the incubation study. Head-down in a nest rummaging for eggs, the trusty egg collector copped a mouthful of a sticky secretion as it poured from the female’s reproductive opening while she expelled eggs during laying.

At the time we considered the gooey substance a great source of amusement rather than the key to fully understanding the process of developmental arrest. However, contemplating all of our findings since then, it was reasonable to assume that the thick secretion could possibly be forming a barrier around the eggs. Heading back to Heron Island for the final time to collect secretion samples, the prospect of ultimately solving the puzzle was at the forefront of our minds.

Again, we found ourselves smothered in sand while collecting secretions from the reproductive opening of nesting females. However, this time the sand was not so easy to remove as the goo produced by the females coated our hands and arms like glue.

In the lab, we tested the rate of oxygen diffusion through the secretions and compared it to diffusion through saline. To our excitement, the results confirmed that the secretion retarded oxygen diffusion and was likely the barrier preventing the transport of oxygen from the tract walls to the eggs.

Finally, our question had been answered. Extremely low oxygen levels maintain a state of preovipositional developmental arrest of embryos while eggs are inside the reproductive tracts of female turtles. Recommencement of development then occurs after eggs are exposed to an increase in oxygen tension when they are laid. Diffusion of oxygen from the tract walls to the eggs is retarded by viscous secretions that coat the eggs inside the female.

But why stop asking questions there? In true scientific fashion, the next question was already beginning to formulate despite reaching our intriguing conclusion. We began wondering why egg-laying persisted in turtles when so many ancestors and close relatives had abandoned this reproductive tactic in favour of live birth. Surely developmental arrest was providing this group of egg-layers with some sort of advantage?

Unfortunately, the answer to this question did not require another epic adventure to Heron Island and could be found in the existing literature.

Continued Success

Turtle embryos enter a state of arrested development approximately 1 week after eggs are fertilised, and remain in this condition while they are in the reproductive tracts, regardless of the duration that mothers retain their eggs. This period of egg retention varies largely among turtle groups, but typically extends for just several days in the majority of marine species, with the exception of the olive ridley sea turtle and Kemp’s ridley sea turtle.

Both species exhibit a unique mass nesting behaviour called “arribada”, during which hundreds and sometimes thousands of females come onto the beach to lay their eggs during a single nesting event. This strategy is thought to overwhelm predators waiting to eat the eggs because there are simply too many eggs for them to eat.

In order to achieve this simultaneous nesting event, the majority of females put embryonic development on hold until all females are ready to nest together. In the case of the olive ridley sea turtle, females carry arrested eggs for up to 63 days before laying, yet the eggs laid are at the same developmental stage as those laid by females carrying eggs for shorter periods.

By laying all their eggs at the same stage, embryonic development and hatching of different clutches tends to occur within the same timeframe. This results in mass hatchling emergence from the nest, which again overwhelms predators and ensures greater offspring survival. This is a great example of how turtles benefit from arrested development, and gives us a glimpse of one of the reasons why turtles might have retained this developmental strategy throughout evolutionary time.

Freshwater species also typically exhibit protracted periods of egg retention during which eggs remain arrested for weeks or months. In all instances, arrested embryonic development gives females the freedom to choose the most appropriate time to nest, which generally coincides with climatic changes in temperature, air pressure or rainfall.

Arguably the most interesting example of how the climate influences the timing of nesting and the duration of developmental arrest in freshwater turtles is observed in the northern long-necked turtle found in the Northern Territory. This species nests during the wet season, laying arrested eggs into underwater nests at the peak of the season.

Eggs of this species remain in a prolonged state of pre­ovipositional arrest while underwater, and only recommence development when exposed to normal atmospheric oxygen levels after the water recedes and the nests dry out during the dry season. Development and hatching coincides with the onset of the following wet season, when food is thought to be more readily available to the hatchlings.

What can be deduced from these examples is that temporarily halting embryonic development allows turtle mothers the flexibility to choose when to lay their eggs. In doing so, mothers can ensure a greater chance of offspring survival by outsmarting predators or by laying eggs during periods when the weather favours successful development or coincides with abundant food availability.

In other words, mother knows best!

Anthony Rafferty is a reproductive and developmental biologist in the School of Biological Sciences at Monash University.