Academic journal article The American Midland Naturalist

Evidence of Introgression between Masked Shrews (Sorex Cinereus), and Prairie Shrews (S. Haydeni), in Minnesota

Academic journal article The American Midland Naturalist

Evidence of Introgression between Masked Shrews (Sorex Cinereus), and Prairie Shrews (S. Haydeni), in Minnesota

Article excerpt

ABSTRACT.-Several studies support specific status for the prairie form of Sorex cinereus, designated as S. haydeni. Evidence for introgression between these species has been found in Alberta, despite significant sequence differences (>15%) in mitochondrial DNA. Using mtDNA and morphological criteria, we identified to species 94 masked shrews to assess the distribution of these two species in Minnesota and to examine the extent of introgression in zones of sympatry. Only four specimens scored genetically and morphologically as S. haydeni, indicating a restricted distribution. Four specimens had incongruent genetic and morphological identifications, suggesting introgression between the two species.


Field studies of masked shrews (Sorex cinereus) in Minnesota have yielded specimens from essentially every county (Hazard, 1982) in both forested and grassland habitats (Tester and Marshall, 1961; Iverson et al., 1967). The species was considered to be a widely distributed ecological generalist with the subspecies S. c. haydeni restricted to prairies. However, van Zyll de Jong's (1976, 1980) studies of geographic variation in morphological characters suggested that the prairie form of the masked shrew merited specific status as S. haydeni.

Most subsequent research has supported specific status for Sorex haydeni. Studies of allozyme variation suggested to George (1988) that S. cinereus and S. haydeni were related very closely, probably last sharing a common ancestor in the Wisconsinan which ended approximately 10,000 y ago. Stewart et al. (1993), however, were unable to discriminate between S. cinereus and S. haydeni using only allozyme data. Volobouev and van Zyll de Jong (1994) found the two to be distinct chromosomally, differing in both 2n (64 in S. haydeni and 66 in S. cinereus) and FN numbers (66 in S. haydeni and 70 in S. cinereus). Comparisons of mitochondrial DNA (mtDNA) sequences (Stewart and Baker, 1997) indicated that S. haydeni and S. cinereus are quite distinct on the basis of a large sequence difference (>15%). In addition, S. haydeni has an insertion that is absent in S. cinereus. Patterns of mtDNA variation suggested limited introgression in Alberta, resulting in some individuals morphologically characteristic of S. haydeni possessing haplotypes of S. cinereus (Stewart and Baker, 1997). Van Zyll de Jong and Kirkland (1989) conducted morphological studies of the two taxa in eastern and central U.S. Based on significant divergence of cranial measurements, they concluded there was no evidence of intergradation in a large area of geographic overlap encompassing most of the original grassland portion of western Minnesota, Iowa and some prairie portions of Canada. However, the number of specimens from several Minnesota counties included in the study was low (approximately half of the counties were represented by a single specimen), and lack of mtDNA data obscured putative hybrids.

We trapped specimens in western Minnesota to re-examine the distribution of Sorex cinereus and S. haydeni and to test for the possibility of intergradation. We analyzed mtDNA sequences for 94 shrews and independently identified specimens to species using morphological criteria (van Zyll de Jong, 1980). We then merged the data to reveal the distribution of S. cinereus and S. haydeni mitochondrial genomes and the concordance between mtDNA and morphological species-level identification. The latter allowed us indirectly to measure the extent of introgression between S. cinereus and S. haydeni and to assess the use of morphological characteristics to distinguish between the two species.


Field collection.-Forty-two specimens were trapped in 1999 by ECB, AKB and KMK at 23 localities covering 10 western counties. Three traplines of approximately 30 traps each were set for one night at each location. At each trap locality we recorded data on habitat type. All specimens were prepared in the field and tissue samples preserved in DMSO buffer. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.