american marten ( martes americana turton) are endangered in nova scotia and the population on cape breton island is critically low. a marten recovery strategy is in place but there are large gaps in information regarding future prey abundance and forest structure in managed stands across the cape breton highland plateau. the crowdis mountain study area was established in 2002 by storaenso, now new page, port hawkesbury limited, to study the effects of variable-retention harvesting techniques on habitat requirements of marten. the goal of this study was to determine the response of small mammals, standing and downed dead wood (sddw), and understory ground vegetation to these alternative harvesting techniques. sampling occurred from may to september pre-treatment in 2002 and post-treatment 2003 and 2005. it was concluded that treatments did not have a significant effect on small mammals or density of standing and volume of downed dead wood. small mammals displayed an increasing trend over the entire study area and these increases were correlated with percent cover of fern, slash/fine debris, and cwd volume. increases in sddw were found when snag density and cwd volume data were combined for all treatment units but treatment effect was found to be non-significant. this study confirmed that because of past silvicultural practices, stands in the crowdis mountain study area had low small-mammal abundances, understory cover, and sddw. experimental harvesting treatments implemented were economical and maintained minimum coarse stand-type requirements of marten and did not negatively affect small-mammal abundances. however, at 50 years old, stands were showing signs of wind damage and increasing trends in small-mammal abundance and sddw recruitment, independent of silvicultural intervention.