Science is—and how else can I say it?—most fun when it plays with interesting ideas, examines their implications, and recognizes that old information might be explained in surprisingly new ways.

—Stephen Jay Gould

The emergence of hatchlings from the nest is the culmination of a process that begins in loggerhead turtles (Caretta caretta), and presumably other species of marine turtles as well, some 4 d prior to the event (Godfrey and Mrosovsky 1997). It starts when the hatchling breaks through the shell and escapes from the confines of its egg, buried some 40–60 cm below the sand surface.

After this escape, the turtles complete their final stages of preemergence preparation. The turtles gradually unfold from the “curled” position assumed inside the egg to a dorsoventrally flattened position that expedites more-efficient locomotion. At the same time, the yolk sac is pulled inside the body (Miller et al. 2003). A space is created in the nest by compaction of the eggshells below the mass of hatchlings and by fluid drainage from their eggs. The hatchlings dig upward within that space, displacing sand from above them to below. About a day or two before they emerge, the hatchlings can often be found resting at the bottom of an air-filled cavity beneath the surface sand. Just before emergence, the final phase of digging activity by the turtles typically causes the surface sand to sag into the cavity, forming a depression at the surface. The depression collapses when the hatchlings finally emerge.

In this article, we reexamine the factors that control the temporal pattern (rhythm) of hatchling emergence. Emergence is primarily a nocturnal activity because high surface sand temperatures during the day inhibit hatchling digging activity (Hendrickson 1958; Bustard 1967; Mrosovsky 1968; Moran et al. 1999). Thus, if the hatchlings digging toward the surface encounter hot sand, they stop. Those temperatures decline rapidly after sunset, release that inhibition, and trigger emergence activity which at any nesting beach is primarily confined to time periods between dusk and dawn (Fig. 1). However, to fully understand the rhythm, it is essential to know not only how it is triggered but also how it is expressed through the dark period and what mechanisms bring the cycle to an end. There have been many studies dealing with correlations between decreasing temperatures and the initiation of the emergence rhythm. Less understood are the reasons why both populations and species differ in how the rhythm is expressed and what causes emergence activity to end at dawn. Here, we summarize the evidence on those points and present a hypothesis that might explain how the rhythm ends each morning.

View Figure

• Download as Powerpoint
• Download Figure

Emergence rhythms are usually quantified at the population level by recording how often they occur at half- or full-hour intervals throughout the night (Witherington et al. 1990; Gyuris 1993). Counts are usually made for many nests over many evenings. Another way of estimating emergences is to count the number of crawling hatchlings that fall into traps each hour, set between the nest and the ocean. The number of nests from which they came can then be estimated by dividing the counts by the proportion of the clutch that on average exits the nest (Glen et al. 2006). The resulting distribution pattern (Fig. 1) varies from one that can show normality (e.g., loggerheads in Melbourne Beach, Florida, USA) to a pattern that is strongly asymmetrical (hawksbills [Eretmochelys imbricata] at Antigua, West Indies; green turtles [Chelonia mydas] at Heron Island, Australia). A more-typical pattern is skewed toward the early evening but shows less asymmetry (e.g., loggerheads in North Carolina, USA; green turtles at Ascension Island). In all of the asymmetrical distributions, however, more hatchlings emerge from nests during the first half than during the second half of the dark period. Why that should be so is a matter of speculation.

A possible explanation for the skewed pattern has been suggested by Glen, Hays, and coworkers (see citations below). Nests in different locations on the same beach can expose eggs and hatchlings to environments that differ in thermal, moisture, and other conditions. When those conditions are relatively uniform within a nest, the hatchlings develop in synchrony and emerge at the same time. Conversely, when conditions vary within a nest the hatchlings will develop at different rates and emerge over several evenings (“multiple emergences”), as several workers have documented (Witherington et al. 1990; Hays et al. 1992; Houghton and Hays 2001; Glen et al. 2005). Similarly, variation in conditions between nests results in hatchlings that dig to the surface at different times of the day and night. Those that reach hot sand during the day will stop, wait, and accumulate throughout the day. It is those nests, containing hatchlings that for hours have been competent to emerge, that are responsible for the surge in emergences that occur earlier in the evening (Glen et al. 2006). Additionally, sand temperatures cool most rapidly during the early evening (Fig. 2). Hatchlings emerge sooner when sands at shallow depths (15 cm) below the surface cool rapidly (Hays et al. 1992). The emergences that occur later in the evening (forming the tail of the asymmetrical distribution) are likely from nests where the hatchlings dug to the surface hours later, when sands were cooling more slowly, if at all (Fig. 2). However, this explanation does not account for why distributions vary in degree of asymmetry or (like those of loggerheads in Florida) show normality.

View Figure

• Download as Powerpoint
• Download Figure

There still remains the problem of what ends the rhythm in the early morning (Fig. 1). This cessation of emergence activity is unlikely to be caused by a temperature-related inhibition, as emergences stop before solar radiation reheats the surface sand (Fig. 2). Put another way, the emergence rhythm is suppressed twice each day: by hot beach sands (up to 50°C; Witherington et al. 1990) during the midday and in the early morning by a factor or factors unknown. It is also unlikely that the cycle ends because all of the turtles at that beach, competent to leave their nest, have in fact left. Moran et al. (1999) hypothesize that they have; they suggest that the hatchlings begin their ascent in time to arrive at the beach surface at night. We think this is unlikely for the reasons given above; embryos can develop asynchronously, and so hatchlings probably initiate a dig toward the surface (and arrive near there) at any time of the day or night.

But Moran et al. (1999) may be correct when they imply that hatchlings might have a sense of time. Perhaps the turtles that fail to emerge in the morning are those that sense that they have arrived at the surface too late to emerge and so must dig no farther. Those turtles (at least in Florida) rarely break through the surface sand, and so exposure to light is not a likely cue that stops them. What does? We suggest that they might have a “time sense,” perhaps triggered by the rapid decline in sand temperatures following sunset, although any part of a diel cycle of temperature variation could provide a cue and start such temporal “countdown.” Those changes would be most apparent to hatchlings that have dug their way to < 20 cm below the beach surface (Fig. 2). The timing mechanism they possess could be analogous to an “hourglass” rhythm (first described by Enright and Hamner 1967) that, once triggered, runs its temporal course; it must be triggered again to repeat the cycle and influence the behavior of the hatchlings that dig near to the surface the next morning.

In summary, an additional element (a time sense) is required to account for how emergence rhythms are controlled from start to finish each day. Emergence rhythms (as documented by others) are 1) suppressed by hot sand temperatures during the day, 2) initiated by the temperature decline after sunset, 3) persist throughout the night while sand temperatures remain low, and (as we propose here) 4) are stopped in the morning by some type of time sense. How long that effect lasts remains to be determined; perhaps it ends once beach temperatures start to rapidly rise once again (Fig. 2).

Field observations consistent with this hypothesis were made some years ago by Gyuris (1993) at Heron Island. During several days of overcast and rain, she reported that “. . . hatchling emergence was greatly reduced for the first day or so . . . then returned to normal numbers although these were spread throughout the 24 hours of the day . . .” (italics ours). Thus, hatchlings that reached the surface when sands were cool and which should have emerged at any time—if only sand temperature inhibited their activity—failed to do so for about 24 hrs. Apparently, the turtles waited in the nest only so long before emerging arhythmically (during the day and at night) probably because, in the absence of a signal to emerge, they must eventually do so to begin an offshore migration. Waiting indefinitely sacrifices energy reserves needed to power that migration (Hays et al. 1992).

Clearly, some interesting experiments remain to be done!