Jeffrey Howard and Norman Reichenbach

Biology and Chemistry Department

Liberty University

Lynchburg, Virginia 24506

Worldwide, many amphibian populations have been declining and undergoing range reductions (Blaustein et al., 1998; Houlahan et al., 2000; Alford et al., 2001). Causes for the declines are often difficult to determine but fungal infections, habitat destruction, changes in local climate, pollution and increased ultraviolet (UV-B) radiation have been listed as potential causes (Blaustein and Wake, 1995; Houlahan et al., 2000). Most studies for assessing effects of UV-B radiation on amphibians occurred in Australia, Europe or northwest United States (Blaustein et al., 1998) and only one has been conducted in the eastern United States (Starnes et al., 2000). Several of these studies have included ambystomatid salamanders (Ambystoma macrodactylum and A. gracile) and an adverse UV-B effect on embryonic survival has been demonstrated (Blaustein et al., 1995, 1998). In contrast, Starnes et al. (2000) did not show a statistically significant UV-B effect on A. maculatum embryonic survival although decreased incidence of deformities was noted for embryos shielded from UV-B. Our goal was to determine if the results from Starnes et al. (2000) apply to A. maculatum populations in Virginia.

Materials and Methods

Ten egg masses of the clear jelly type (Hardy and Lucas, 1991) were collected from a ditch in a dirt road near Riverville, Amherst County, Virginia on 24 February 2001. The egg masses were returned to the laboratory and kept at 4 C for 3 days until placement in the experimental units on 27 February. Counts were made of the number of embryos per egg mass and the embryonic developmental stage was determined using the scale by Harrison (1969). Egg masses were randomly assigned to numbered plastic cases that were either shielded (n = 5) with Luminar/Plexiglas shields (shields completely block wavelengths below 380 nm and therefore UV-B radiation [280-320 nm], Lizana and Pedraza, 1998) or left exposed to sunlight using plastic mesh covers (n = 5). Each case had mesh-covered holes to allow for free flow of water around
the egg masses and a styrofoam 'skirt' to keep the case afloat. A completely randomized design was used to place the cases in a linear array in an exposed location in a lake near Liberty University, Lynchburg, Virginia.

During the field experiment, one egg mass was lost when a storm blew off the Luminar/Plexiglas shield and two other cases lost their styrofoam skirts at the end of the experiment; thus, some larvae were able to escape from these cases. After hatching, all egg masses and live larvae were removed to the laboratory. Mortality was assessed as eggs that did not hatch. Live larvae were anesthetized, preserved in formalin and then examined for deformities.

Percent survival and deformities for shielded and unshielded egg masses were analyzed using t-tests. Because of the small sample size, when the null hypothesis was not rejected a power analysis was done to see how large a sample size would be needed to conclude the effect was significant at µ = 0.05 and power = 0.8. If the sample size needed was large, we concluded the effect was not significant. If the sample size was small, then the failure to reject the null hypothesis (H₀: average for unshielded egg masses = average for shielded egg masses) would likely be attributable to small sample size (Gregory, 2001). Methods used for calculating power and sample size were from Zar (1996).

Results and Discussion

The number of embryos per egg mass ranged from 17 to 106 with an average of 51 embryos. The embryos were at Harrison stages 6-8 (cleavage stages to early blastula or approximately 20-30 hours old) when the egg masses were placed in the field (27 February). Hatching was completed by 17 April 2001. Percent survival averaged 96.5 and 97.8 for shielded and unshielded egg masses, respectively, and these were not significantly different (t = 0.4, d.f.= 7, P = 0.7, Table 1). All unshielded egg masses produced deformities, whereas only one of the shielded egg masses did so. The average percent deformities of 0.4 and 1.2 for shielded and unshielded egg masses, respectively, were not significantly different (t = 1.4, d.f.= 5, P = 0.2, Table 1).

Table 1. Percent survival and deformity for spotted salamander (Ambystoma maculatum) embryos exposed to and shielded from UV-B radiation. Means are followed by one standard deviation.

Case Types	% Survival	% Deformities
Shielded	100	0
	100	1.7
	100	0
	86.1	0
	----^A	----^A
	96.5 ± 7.0	0.4 ± 0.9
Unshielded	100	0.9
	94.3	0.9
	94.6	1.8
	100	----^B
	100	----^B
	97.8 ± 3.0	1.2 ± 0.5

A) Egg mass lost during storm.

B) Percent deformities could not be assessed due to missing larvae since cases which lost styrofoam skirts allowed for newly hatched larvae to escape.

Power analysis of the survival data indicated that a sample size of 235 for

each treatment would have been required to reject the null hypothesis. Taking into consideration the large sample size that would have been needed in addition to the higher survivorship being with the unshielded treatment, we would conclude that the UV-B effect on survivorship was really not significant. In contrast, for percent deformities, the sample size needed would have only been 14 egg masses per treatment. It is likely that our lack of rejecting the null hypothesis was due to small sample size.

Our results follow those of Starnes et al. (2000) indicating there is no statistically significant evidence that UV-B increases either mortality or deformities in developing spotted salamander embryos. For percent deformities, Starnes et al. (2000, Figure 1) showed a decline from approximately 6% to 3% for clear egg masses (n=10/treatment) unshielded and shielded (98% UV-B blocking treatment), respectively. Perhaps if they had used a larger sample size, the null hypothesis for percent deformities would have been rejected. The question of biological significance would, though, still need to be addressed since the overall deformity rates were below 10% for all treatments both in this study and in Starnes et al. (2000).

The high survivorship and low rates of deformity may be due to various factors which reduce the risk of developing embryos to UV-B such as DNA repair mechanisms, reduced transmission of UV-B due to the jelly surrounding the embryo, cloud cover, and water turbidity (Hatch and Burton, 1998; Starnes et al., 2000). Even though the risk of UV-B to A. maculatum embryos appears minimal at this time, we consider it prudent that Virginia amphibian species that lay eggs in exposed sites be monitored for UV-B effects. There may be species-specific and/or locational differences in risk to UV-B. For example, UV-B increased Hyla chrysoscelis and Pseudacris triseriata embryonic deformity rates (Starnes et al., 2000). In addition, pollutants in some locations might work in a synergistic fashion with UV-B similar to that noted by Hatch and Burton (1998). They found photoinduced toxicity of fluoranthene (a polycyclic aromatic hydrocarbon) on spotted salamander larvae. Therefore, research on the effects of UV-B on amphibian reproductive success in a variety of locations in Virginia is encouraged.

Acknowledgments

We would like to thank the students of the 1999 Environmental Biology class at Liberty University for the construction of the cases and the 2001 class for their aid in the setup of the cases at the site on Lake Hydaway. We would also like to thank the staff at Thomas Road Baptist Church for the use of their property for the performance of this study.

Literature Cited

Alford, R., P. Dixon, and J. Pechmann. 2001. Global amphibian population declines. Nature 412: 499-500.

Blaustein, A., B. Joseph, and J. Keisecker. 1995. Ambient ultraviolet radiation causes mortality in salamander eggs. Ecological Applications 5: 740-743.

Blaustein, A., J. Kiesecker, D. Chivers, D. Hokit, A. Marco, L. Belden, and A. Hatch. 1998. Effects of ultraviolet radiation on amphibians: Field experiments. American Zoologist 38: 799-812.

Blaustein, A. and D. Wake. 1995. The puzzle of declining amphibian populations. Scientific American 272: 52-57.

Gregory, P. 2001. Feeding, thermoregulation, and offspring viability in gravid garter snakes (Thamnophis sirtalis): What makes laboratory results believable. Copeia 2001: 365-371.

Hardy, L. and Lucas, M. 1991. A crystalline protein is responsible for dimorphic egg jellies in the spotted salamander, Ambystoma maculatum (Shaw)(Caudata: Ambystomatidae). Comparative Biochemistry and Physiology 100: 653-660.

Harrison, R. 1969. Harrison stages and description of the normal development of the spotted salamanders, Ambystoma punctatum (Linn.). Pp 44-66 in Harrison, R. ed. Organization and Development of the Embryo. Yale University Press, New Haven, CT.

Hatch, A. and G. Burton. 1998. Effects of photoinduced toxicity of fluoranthene on amphibian embryos and larvae. Environmental Toxicology and Chemistry 17: 1777-1785.

Houlahan, J., C. Findlay, B. Schmidt, A Meyer and S. Kuzmin. 2000. Quantitative evidence for global amphibian declines. Nature 404: 752-755.

Lizana, M. and E. Pedraza. 1998. The effects of UV-B radiation on toad mortality in mountainous areas of Central Spain. Conservation Biology 12: 703-707.

Starnes, S., C. Kennedy, and J. Petranka. 2000. Sensitivity of embryos of

southern Appalachian amphibians to ambient solar UV-B radiation. Conservation Biology 14: 277-282.

Zar, J. 1996. Biostatistical Analysis. 3^rd ed. Prentice Hall, Upper Saddle River, NJ. 662pp.