Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)


Natural Resources

Department name when degree awarded

Fisheries and Wildlife

Committee Chair(s)

John A. Bissonette


John A. Bissonette


Tom Edwards


Barrie Gilbert


Jim Long


Doug Ramsey


Although the American marten (Martes americana) generally is recognized as an obligate late-seral species, the factors dictating this association are poorly understood. Martens were studied in Newfoundland, Yellowstone National Park, and in a captive setting. As expected, use of habitat types was not proportional to availability P < 0.001). Defoliated and late-seral conifer stands were used more than expected, while all other types indicated expected or less than expected use. Habitat selection by martens was detectable at spatial scales greater than 80 m (P < 0.001).

Newfoundland martens were radio-collared and monitored for diet activity during the winters of 1990 and 1991. A regression of the percent active fixes on temperature had a negative slope (P = -4.45, P= 0.084, n = 12), indicating that martens did not minimize their exposure to low temperatures. A log-linear model suggested that the presence or absence of light was the only factor associated with marten activity patterns (P < 0.001).

Martens in Western Newfoundland and a population in Yellowstone National Park were tested for their response to predation risk using bait stations in various habitat types. Visitation rates of martens at bait-boxes were not different between study sites (P = 0 .190). However, martens visitation by habitat was different (P = 0.001). Martens use of bait-boxes was similar in old-growth and defoliated habitats, suggesting that foliar cover may not have a strong influence on the risk of predation for martens during winter. Martens did use bait-boxes in defoliated stands to a greater extent than those in open habitats (P < 0.001), suggesting that they perceived stem structure as decreasing predation risk.

In captive experiments, martens selected areas with both overhead cover and woody stem structure (P = 0.012). I detected no difference between the use of areas with only overhead cover and those having only stem structure (P = 0.671). However, martens decreased foraging activity in response to a predatory cue (P = 0.004). The inability of martens to use food resources in areas lacking cover during the summer suggests a perception of increased threat associated with these areas.