Hatching plasticity is associated with a more advanced stage at hatching in an Ambystoma with terrestrial eggs

Abstract

Hatching plasticity allows animals to initiate hatching in response to environmental cues including predation, flooding, and hypoxia. In species with terrestrial eggs but aquatic larvae, hatching plasticity often manifests as extended development of embryos when water is not available. Although these effects are taxonomically widespread, little attention has focused on differences in plasticity across closely related species with terrestrial and aquatic embryos. We propose that the terrestrial embryonic environment favors slower and prolonged development and, consequently, that we should see differences in development between closely related species that differ in where they lay their eggs. We test this hypothesis by comparing embryonic development between two mole salamanders, Ambystoma opacum and A. annulatum. Most Ambystoma lay eggs submerged in ponds but A. opacum lays its eggs on land, where hatching is triggered when eggs are submerged by rising pond levels. Embryos of both species were reared under common laboratory conditions simulating both aquatic and terrestrial nest sites. Consistent with our hypothesis, we found that A. opacum embryos exhibited slower development and took longer to hatch than A. annulatum embryos in both rearing environments. Furthermore, we observed in A. opacum a plasticity in hatching stage that was absent in A. annulatum. Our results indicate that the terrestrial-laying A. opacum has evolved slower and prolonged development relative to its aquatic-laying congener and suggest that embryonic survival in the unpredictable terrestrial environment may be facilitated by developmental plasticity.

Keywords: amphibian; embryo; extended development; hatching plasticity; larva.

FIGURE 1

(a) Female marbled salamander (Ambystoma opacum) with embryos. Nest was under a log on the margins of a wildlife pond in Ozark‐St. Francis National Forest, Pope County, AR. (b) Ringed salamander (A. annulatum) egg masses attached to sticks in a wildlife pond in Ozark‐St. Francis National Forest, Franklin County, AR. Photos by R. E. Hale.

FIGURE 2

Oviposition bins like those used for marbled salamander (Ambystoma opacum) at the Pulaski County (Little Rock) site in 2019. Bins were 0.6 m diameter × 0.5 m height and contained 10–15 cm depth of leaf litter, had holes in the bottom to prevent water retention, and were covered with netting attached by an elastic cord. One or two gravid females and one to two males were placed inside, under leaf letter. We returned 3–7 days after placement to inspect for deposited eggs. Bins in photograph were used in North Carolina in 2014 by R.E. Hale.

FIGURE 3

Number of embryos per jar, out of 20, that hatched in air and water rearing environments. Horizontal spread is added to increase visibility of individuals points. The air treatment simulates the terrestrial environment in which Ambystoma opacum typically deposits embryos. The water treatment simulates the aquatic environment in which A. annulatum typically deposits embryos. There were no significant differences in number that hatched between species or treatments.

FIGURE 4

(a) Age, (b) Harrison (1969) stage, and (c) mass at hatching for embryos of ringed (Ambystoma annulatum) and marbled (A. opacum) from six sites in Arkansas. Ringed salamander embryos were collected from Baxter County (Bax), Franklin County (Fra), Madison County (Mad), and Washington County (Was). Marbled salamander embryos were collected from Pulaski County (Pul15 and Pul19), Pope County (Pop), and Phillips County (StF). The two boxplots for Pulaski County represent two days of collection. Embryos from air and water treatments are combined within a site.

FIGURE 5

(A) Age, (B) Harrison (1969) stage, and (C) mass at hatching for embryos of ringed (Ambystoma annulatum) and marbled (A. opacum) salamanders reared in either moist air or water, at 15°C. The air treatment simulates the terrestrial environment in which A. opacum typically deposits embryos. The water treatment simulates the aquatic environment in which A. annulatum typically deposits embryos. Horizontal spread is added to increase visibility of individuals points. Boxplots with the same letter were not significantly different, as evaluated using the Tukey method and the eemeans function in R. See text for a summary of differences between treatments.

FIGURE 6

Relationships between (a) age and stage at hatching and (b) age and mass at hatching, and (c) stage and mass at hatching for each species and treatment combination. Dashed lines indicate regression for water treatment, solid lines for air treatment. Stage at hatching increased with log(age) at hatching in both species, with a greater positive slope for marbled than ringed salamanders. There was no relationship between stage and mass in marbled salamander (Ambystoma opacum). There was a weak negative relationship (slope = −0.09, effect size η2 = 0.05) between stage and mass for ringed salamander (A. annulatum). The relationship between age and mass at hatching are visualized but not statistically evaluated, as we decided a priori to test for relationships only between age and stage at hatching and between mass and stage at hatching.