Opportunistic breeding in the Cape spiny mouse ( Acomys subspinosus )

T eight species of Acomys have an extremely wide distribution, occurring in arid regions from southern Pakistan through the Middle East and most of Africa to the Cape (Nowak 1991). Two species are recognized within southern Africa, the spiny mouse A. spinosissimus, and the Cape spiny mouse A. subspinosus (Dippenaar & Rautenbach 1986). A. subspinosus has a restricted range (de Graaff 1997) and is described as a true fynbos endemic (Breytenbach 1982). Little is known of the life history, diet or habits of A. subspinosus, and this species was included in the 1976/1977 South African Red Data Book as rare, owing to low numbers and a lack of information on its basic biology (Meester 1976). The species is no longer listed since ‘its rocky mountainous terrain is unlikely to be degraded’ (Smithers 1986). We carried out a study of the mammal-pollinated Protea humiflora (Proteaceae) in the Riviersonderend Mountains of the western Cape, South Africa (Fleming & Nicolson, in press). A. subspinosus is a flower visitor and potentially important pollinator of P. humiflora. Here we record information on the density and breeding of A. subspinosus at Jonaskop (33°55’S, 19°31’E, 630 m) and compare this with published data for the species. The slope at Jonaskop is north-facing, with exposed rock and heterogeneous vegetation densities, ranging from near-open ground, with little cover, to areas of dense growth of P. humiflora bushes, some 1–1.5 m high. The vegetation is typical of the fynbos/karoo ecotone, and is dominated by Asteraceae and Thymelaeaceae/Rhamnaceae; P. humiflora was usually the only species of Proteaceae present. We trapped over three nights every month, on the new moon, from June to November 2000, thus comparing winter months, when P. humiflora was flowering, with the following spring. Two adjacent grids (50 m apart) of 64 Sherman traps each (10 m apart, baited with peanut butter and oats), yielded a total trapping area of 1.28 ha. Although P. humiflora inflorescences were removed over one of these grids, for the purposes of the present study, the grids are not considered separately. Each captured animal was identified, weighed, measured and individually marked by toe–clipping. Breeding status was recorded: females were recorded as perforate or non-perforate and nipple development was noted. Juveniles were considered post hoc as those individuals <17 g on the basis that the smallest sexually active individual