Variation in the long-tailed field-mouse (Apodemus sylvaticus (L.)) in north-west Scotland I. Comparisons of individual characters

Abstract

Earlier workers have suggested that a distinct species of long-tailed field-mouse ( Apodemus hebridensis (de Winton)) occurs in the Inner and Outer Hebrides of north-west Scotland. They have further claimed that ten subspecies of A. hebridensis can be recognized although no one island has been recorded as supporting more than one subspecies. The present paper attempts to examine by comparison of individual characters the extent of differentiation between various island and mainland populations. A total of 470 specimens of Apodemus were collected between July and September in 1957 to 1960. The mainland localities were Applecross in Wester Ross and Laga, Ardnamurchan, Argyllshire. Collections were made from the Islands of Raasay, Rhum, Mull, Colonsay, Lewis, North Uist, South Uist and Barra. In the analyses of differences between the various populations the following measurements were used : occipito-nasal length, tail length, hind foot length, weight, length of tooth row, tooth wear, incisor width, length of pectoral stripe, values and chroma. The last two are measures of colour using Munsell notation. The pectoral stripe measurement was analysed using a logarithmic transformation. The means have been calculated of each character for every locality. Tooth wear is a measure of age and a regression adjustment was used to bring all other characters to a uniform age class. The variance ratio (between localities)/(within localities), and the standard errors of the differences between each pair of localities have been calculated from the adjusted values for each character. Regression coefficients of weight, tail length, hind foot length, pectoral stripe length, tooth row length and incisor width on occipito-nasal length are given for each locality and the differences between localities have been tested statistically. Finally, means adjusted to a common occipito-nasal length have been computed together with their standard errors. The results suggest that size differences account for a large amount of the variation between island stocks. There are also other differences between localities. The mice from certain groups of localities show close affinity to each other although the groups are not sharply defined. There is no clear overall island pattern that readily separates these mice from those from the mainland.