Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Life Sciences

Major Professor

Gordon M. Burghardt

Committee Members

Arthur C. Echternacht, David A. Etnier, Neil Greenberg


Most of what is known about the control and expression of appetite comes from studies of laboratory rodents. Such animals, which take small, frequent meals, are located at one end of a spectrum of feeding frequencies. The theme of the present work was to explore appetite in an animal that takes relatively large meals at long intervals. I studied some of the factors that control food intake and influence the expression of hunger in Northern water snakes, Nerodia sipedon; Serpentes: Colubridae). Subjects were collected in streams in eastern Tennessee, especially the Little River in Blount County.

In order to establish criteria for satiety, I examined feeding in single meal sessions. After ingesting a large prey item, Northern water snakes exhibit a behavioral satiety sequence that closely resembles that in mammals: feeding is followed by grooming, then locomotory activity, and finally, by resting. However, satiety sequences were qualitatively similar after meals of 10 and 30% of snake mass. Because of ambiguity in interpreting this sequence, a temporal criterion derived from satiation curves was used to establish satiety.

Neonate water snakes eat much smaller meals when offered an excess of small prey than when offered a single large prey. At prey sizes from 5- 20% of body mass, meal size of neonates was 23.5% (average of 6 litter means). In contrast, neonates can readily ingest single prey equal to 40% of their mass, and sometimes as large as 55%. Thus, the largest meal that these snakes can take is a single, maximal item. Since snakes exhibited this response in their first feeding, the response is unlearned.

Because snakes are competent forager over a large range of body sizes, the are ideal models for investigating ingestive allometry. When prey size (in %) is held constant, the meal size of hungry water snakes scales as an isometric function of body mass (log-log slope=0.97). Over a body mass range of 2.5-449.0 g, meal size averaged 22.4% of body mass.

An ad libitum feeding study was conducted with four Northern water snakes (mean mass=70 g) offered live minnows (mean mass=2 g). All feeding events were recorded with time lapse video. On this ration, snakes converted 24% of ingested mass into body tissue. Log survivorship analysis indicated that these snakes fed prandially, i.e. in temporally discreet meals. Median meal sizes where 2.7-6.0% of initial snake mass. Snakes typically took much smaller meals (44-71% of meals consisting of one fish) than their maxima (4-17 fish). Median intermeal intervals were 22.7-32.8 h, and the shedding cycle did not appear to affect the timing of meal taking. Intermeal intervals of approximately one and two days were prevalent, with the likely cause being a strong deil cyclicity. Feeding was significantly nocturnal, and more specifically, crepuscular (19-78% of meals initiated during first hour of darkness). Most meal pattern correlations were not significant.

Using the method of Diana (1979), I estimated the feeding frequency of adult female Northern water snakes in the field. The time needed at 26 °C to complete gastric digestion of a prey of typical size was 49 h. The proportion of adult females with good was 18/34, or 52.9%. I estimated that adult female water snakes feed, on average, once every 3.9 d (95% discrimination interval=2.9-5.0 d).

Given the importance of timing factors, I investigated the influence of the lunar cycle on snake foraging activity in the field. I conducted field censuses along a standardized transect in the Little River. There was a significant positive correlation between number of snakes seen and water temperature. After accounting for thermal differences, I found a significant effect of lunar phase on snake foraging activity. More snakes were seen during phases when the moon was dark in the early evening. Predation both by and upon the snakes may be relevant to understanding this phenomenon. Changes in the timing of lunar ascent/descent may be an important part of lunar rhythmicity in addition to changes in illumination.

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

Included in

Life Sciences Commons