Stomach contents were examined from 90 individuals of Sorex longirostris and 167 individuals of S. cinereus from hardwood floodplain habitats in Vigo Co., Indiana. The five most important foods of S. longirostris were spiders, Lepidoptera larvae, Gryllidae, Coleoptera adults and harvestmen. The five most important foods of S. cinereus were the same except that Coleoptera larvae replaced Gryllidae. Maintenance of a contiguously allopatric pattern of distribution within the relatively small geographic area of range overlap is probably important in minimizing competition between these two morphologically similar species. INTRODUCTION The southeastern shrew (Sorex longirostris) occurs throughout the southeastern U.S. ranging as far W as Louisiana and Arkansas, and N to Missouri, Illinois and Indiana (French, 1980a). The masked shrew (S. cinereus) is the most widespread North American Sorex, with a range extending across North America (Hall, 1981). Ranges of the two species overlap in portions of Illinois and Indiana, and in Indiana both species occur in Knox, Spencer, Tippecanoe and Vigo counties (Mumford and Rippy, 1962; French, 1980b; Rose, 1980). In Vigo County, these shrews are allopatric, however (French, 1980b). Sorex cinereus is found within the boundaries of the glacial floodplain of the Wabash River. Sorex longirostris is found in similar habitats and on some of the same soil series as S. cinereus but only outside the boundaries of the glacial floodplain of the Wabash River. The purpose of this study was to determine the degree of similarity in diets of Sorex cinereus and S. longirostris that occupy the same type of habitat but do not overlap in distribution in Vigo Co., Indiana. It was hypothesized that morphological similarities between these two shrews reflect similar diets that contribute to allopatric distribution patterns in Vigo County. MATERIALS AND METHODS Sorex longirostris and S. cinereus were trapped with no. 10 cans set as pitfalls from 1976 to 1979 in Vigo Co., Indiana. Each can contained ca. 10 cm of water and shrews drowned quickly after entering the cans. Most specimens of both species were trapped in floodplain hardwoods along the Wabash River and its tributaries. Specimens of S. longirostris and S. cinereus were collected during all seasons as follows: January (1,0); February (1,0); March (9,2); April (11,72); May (9,12); June (13,3); July (6,23); August (14,3); September (3,1); October (18,21); November (3,21), and December (3,10). Stomach contents were preserved in 70 % isopropyl alcohol and examined in shallow petri dishes using a binocular dissecting microscope. The percent of total volume for each food item was estimated by visual inspection (Whitaker and Mumford, 1972). Items were identified by comparison with materials collected in the field. For aid in identification, microscope slides were made of all questionable food items; these were studied in detail at a later time. Shrews with empty stomachs were not included in the analysis. A similarity coefficient (Whittaker, 1970) was used to compare the results of this 1 Present address: Eastern Regional Heritage Program, The Nature Conservancy, 294 Washington Street, Boston, Mass. 02108.
Read full abstract