In winter grouse (Canachites canadensis) in Alaska feed on only the needles of black (Picea mariana) and white (P. glauca) trees. The birds prefer to feed in P. glauca trees rather than P. mariana, partly because the densely growing needles of P. mariana make more difficult. P. mariana contains more crude fat than P. glauca, less ash, and less protein in some habitats, any of which also may result in discrimination against P. mariana. Within P. glauca, the needles of trees that are selectively browsed are less dense and contain less crude fiber than the needles of unbrowsed trees. The foliage of browsed and unbrowsed P. glauca trees does not differ in crude protein (N X 6.25), soluble carbohydrates, P, K, Ca, Na, Fe, Mg, Mn, Cu, Zn, Sr, Ba, Al, and B. J. WILDL. MANAGE. 40(2):205-213 Tetraonids often feed selectively on certain species of plants among those available, and within a species they may prefer certain individual plants. The basis of this selection is not always evident, but nutrient content of the vegetation has been reported to influence food discrimination among the following species of tetraonids: grouse (Gurchinoff and Robinson 1972); red grouse (Lagopus 1. scoticus) (Moss 1967, 1972); willow ptarmigan (L. lagopus) (Pulliainen and Salo 1973); rock ptarmigan (L. mutus) (Gardarsson and Moss 1970, Pulliainen 1970a); capercaillie (Tetrao urogallus (Pulliainen 1970b); and ruffed grouse (Bonasa umbellus) (Svoboda and Gullion 1972). Investigators of selective have given less attention to physical attributes of plants that may influence selection, though Seiskari (1962:53) thought that the selection of feeding pines by capercaillie might be determined by morphology of the trees, as well as by their nutrient content. The selection of heather (Calluna vulgaris) by red grouse may be determined by particle size (Moss 1972) and plant height (Moss et al. 1972) in conjunct on with chemical factors. The pu pose of this study was to examine winter food selection by Alaskan gro se, which feed exclusively on the ne dles of P. glauca and P. mariana from November to April (Ellison 1966, 1972:57). The birds show a marked preference for P. glauca over P. mariana, and within P. glauca certain individual trees are heavily browsed by groups of birds for weeks or months while nearby trees are untouched. Furthermore, some of these same feeding spruce are known to be browsed selectively for up to 4-5 years. In an attempt to explain these preferences, I measured several physical and chemical characteristics among P. mariana trees and browsed and unbrowsed P. glauca trees. I sincerely appreciate the aid and encouragement of A. S. Leopold, whose interest rendered this study possible. M. A. Williams helped interpret nutritional aspects of the study. P. Heilman, Forest Science Laboratory at Fairbanks, Alaska, analyzed vegetation for nitrogen content during a preliminary study. The U.S. Fish and Wildlife Service provided field accommodations on the Kenai National Moose Range. I wish also to thank two anonymous reviewers, whose stimulating comments altered my interpretation of some data. 'Study funded by the Union Wildlife Foundation Fund, University of California Patent Fund, National Wildlife Federation, and U.S. Fish and Wildlife Service. 2Present address: D6partement de biologie, Universit6 Laval, Quebec 10, Canada G1K 7P4. J. Wildl. Manage. 40 (2):1976 205 This content downloaded from 207.46.13.58 on Wed, 12 Oct 2016 04:40:07 UTC All use subject to http://about.jstor.org/terms 206 WINTER FOODS OF SPRUCE GROUSE * Ellison STUDY AREA AND METHODS The study was conducted during the winters of 1969-70 and 1970-71 on the Kenai Peninsula, south-central Alaska, in conjunction with studies of movements and social organization (Ellison 1973) and population dynamics (Ellison 1974) of