We developed a method of quantifying levels of fluorescence in the whiskers of wild stoats (Mustela erminea) using fluorescence microscopy and Axiovision 3.0.6.1 software. The method allows for discrimination between natural fluorescence present in or on a whisker, and the fluorescence resulting from the ingestion of the systemic marker Rhodamine B (RB), although some visual judgement is still required. We also developed a new high performance liquid chromatography (HPLC) protocol for detecting the systemic marker iophenoxic acid (IPA) in the blood of laboratory rats (Rattus norvegicus) and wild stoats. With this method, the blood of an animal that has consumed IPA can be tested for the presence of the foreign IPA compound itself. This is a more reliable test than the previous method, which measured the raised level of natural blood protein-bound iodine correlated with IPA absorption. The quantity of blood required from animal subjects is very small (10 ml), so the testing is less intrusive and the method can be extended to smaller species. The extraction technique uses Z02034; Online publication date

Stoats (Mustela erminea), introduced to New Zealand in the late nineteenth century, are common in New Zealand beech (Nothofagus sp.) forests, where populations of feral house mice (Mus musculus) fluctuate between years much as voles do in the northern hemisphere. We present new field evidence and two models demonstrating (i) a strong correlation between density indices for young stoats in summer and for mice in the previous spring, and (ii) a significant linear relationship between productivity per female and spring density of mice up to 25 mice captures per 100 trap-nights. These models confirm that short-lived small mustelid predators dependent on fluctuating populations of prey have evolved means of matching their productivity to the prospects of success across a wide range, from total failure in rodent crash years to >12 independent young per female in rodent peak years. We suggest that the enhanced reproductive success of female stoats when rodents are abundant is due to a combination of critical improvements in both the reproductive physiology and the foraging behaviour of female stoats in rodent peak years. Conversely, a drastic shortage of rodents increases the mortality of embryos and nestlings, while the adult females are able to survive, and even remain relatively fat, on other foods.

The dispersion, age structure and diet of stoats (Mustela erminea) in beech forest in the Borland and Grebe Valleys, Fiordland National Park, were examined during December and January 2000/01, 20 months after a heavy seed-fall in 1999. Thirty trap stations were set along a 38-km transect through almost continuous beech forest, at least 1 km apart. Mice were very scarce (<1 capture per 100 trap nights, C/100TN) along two standard index lines placed at either end of the transect, compared with November 1999 (>60/100TN), but mice were detected (from footprints in stoat tunnels) along an 8 km central section of the transect (stations 14-22). Live trapping with one trap per station (total 317.5 trap nights) in December 2000 caught 2 female and 23 male stoats, of which 10 (including both females) were radio collared. The minimum range lengths of the two females along the transect represented by the trap line were 2.2 and 6.0 km; those of eight radiotracked males averaged 2.9 ±1.7 km. Stations 14-22 tended to be visited more often, by more marked individual stoats, than the other 21 stations. Another two marked males were recovered dead. The stoat population showed no sign of chronic nutritional stress (average fat reserve index = 2.8 on a scale of 1-4 where 4 = highest fat content); and only one of 63 guts analysed was empty. Nevertheless, all 76 stoats handled were adults with 1-3 cementum annuli in their teeth, showing that reproduction had failed that season. Prey categories recorded in descending frequency of occurrence were birds, carabid beetle (ground beetle), weta, possum, rat, and mouse. The frequencies of occurrence of mice and birds in the diet of these stoats (10% and 48%, respectively) were quite different from those in stoats collected in Pig Creek, a tributary of the Borland River (87%, 5%), 12 months previously when mice were still abundant. Five of the six stoat guts containing mice were collected within 1 km of stations 14-22.

Exclusion fencing is being increasingly used to protect threatened wildlife. This project provides managers with information to help make decisions about appropriate fence design. Summary Exclusion fencing is being increasingly used to protect areas of high conservation value or to create 'islands' of protected habitat for native fauna. The objective of this study was to test a range of fence designs to assess the optimum physical and/or electrical barrier required to exclude feral Cats ( Felis catus ) and Red Foxes ( Vulpes vulpes ). We tested against six fence designs by placing individual animals (18 Cats and 18 Foxes) in 20 × 20 m pens and recording their responses to various fence components. Fence design 1 was 1.8-m high with one electric wire at 1200 mm and another at 1500 mm offset by 80 mm and a curved overhang 600 mm in diameter. Fence design 2 and 3 had single electric wires placed at 400 mm and 1700 mm, respectively. Fence design 4 had a single electric wire at the end of the overhang. Fence design 5 had no electric wire and fence design 6 was 1.2-m high with an overhang and no electric wire. Neither feral Cats nor Foxes were able to scale the 1.8-m fence regardless of the position or absence of electric wires. One Fox and one feral Cat successfully breached the 1.2-m high fence by jumping onto the overhang and climbing over. Four Foxes excavated holes underneath the fence and one chewed through the mesh. Results from this work indicate that fences designed to exclude feral Cats and Foxes should be 1.8-m high, have an overhang that is at least 600 mm in circumference that is curved or shaped in such a way that prevents animals climbing over from underneath, and have an apron with a mesh hard enough to prevent Foxes chewing through. Exclusion fences do not require electric wires. The omission of electric wires will provide significant savings in building and maintenance costs for exclusion fencing.