Doctoral Dissertations

Date of Award


Degree Type


Degree Name

Doctor of Philosophy

Major Professor

J.T. Tanner


A study was conducted in the Great Smoky Mountains National Park to determine the validity of the mimicry hypothesis which has been proposed to explain the resemblance between red-cheeked color variants of Desmognathus ochrophaeus carolinensis and Plethodon jordani jordani, the red-cheeked salamander. Five basic rules of mimicry were tested in both field and laboratory approaches to the problem.

The color patterns of carolinensis were recorded in 587 individuals observed. Of this number, 25.7 per cent displayed some cheek coloration which included color variations of red to orange, yellow, and white. Red to orange cheek colors were found in 92 per cent of all carolinensis having cheek coloration. These supposed mimics of jordani differed markedly in appearance from the normal carolinensis.

An attempt was made to delimit the range of jordani within the Park. Observations were also made of the presence or absence or red-cheeked carolinensis within and outside this range of jordani. D.O. carolinensis was found with jordani in seven general areas investigated. Red-cheeked carolinensis were found in the vicinity of Blanket Mountain, but no jordani were found in this general area. The presence of mimics outside the range of models in not what one would expect if mimicry exists between the two species. Also, no red-cheeked carolinensis were found within a large sample of carolinenesis collected on Hyatt's Ridge, even though jordani was present in the area. Records of red-cheeked carolinensis occurring outside the Park were cited from the literature. The frequency of mimics at Wayah Bald, North Carolina, may be higher than one would expect under the rules of mimicry. No evidence was available to indicate that the frequency of red-legged carolinensis was any higher within the range of P.j. shermani than in any other area.

Eight transects were placed at elevations ranging from 2,925 feet to 5,390 feet within the Great Smokies to determine the relative abundance of jordani and red-cheeked carolinensis. It was found that a positive correlation seemed to exist between the number of jordani and the frequency of red-cheeked carolinensis in the areas sampled. The percentage of supposed mimics in the total jordani red-cheeked carolinensis population varied in the areas sampled from 25 per cent to 60 per cent. It is questionable whether the supposed mimicry is effective enough to support this high frequency of mimics.

P.j. jordani and red-cheeked and normal carolinensis were exposed to four different types of predators in an attempt to detect differences in the survival rates between the two species of salamanders. The predators used in these laboratory feeding experiments included the eastern garter snake (Thamnophis sirtalis sirtalis), the large short-tailed shrew (Blarina brevicauda), the mountain spring salamander (Gyrinophilus danielsi danielsi), and the black-bellied salamander (Desmognathus quadra-maculatus).

All of these predators readily fed upon all salamanders presented to them. No differences were noted between the survival rates of carolinensis and jordani when exposed to these predators under laboratory conditions. Preliminary feeding experiments were also conducted with the long-tailed deer mouse (Peromysccus maniculatus) and the red-backed mouse (Clethrionomys gapperi), but additional feeding experiments were not conducted with these mice since they did not seem to be natural predators. An attempt was also made to detect bird predation upon salamanders under natural conditions. No predation was found to exist under the conditions of the feeding experiment.

No evidence supporting either Batesian or Mullerian mimicry could be found under the methods and conditions of this study. Alternate hypotheses for the explanation of the resemblance between red-cheeked carolinensis and jordani were discussed. These alternate hypotheses include the following: (1) hybridization has taken place or is occurring, (2) genetic drift has occurred, (3) the red cheek coloration attracts prey, and (4) the cheek coloration is a pleiotropic characteristic which is linked to a physiological mutation which has selective value. Of these four possible explanations, the fourth was considered to be the most plausible explanation of the resemblance between the red-cheeked carolinensis and jordani.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Included in

Zoology Commons