Diet of stoats (Mustela erminea) in an Alpine habitat: The importance of fruit consumption in summer Adriano Martinoli*, Damiano G. Preatoni, Barbara Chiarenzi, Luc A. Wauters, Guido Tosi Dipartimento di Biologia Strutturale e Funzionale, Università degli Studi dell’Insubria, Via J.H. Dunant, 3, 21100 Varese, Italy Received 9 February 2000; revised 1 December 2000; accepted 5 January 2001 Abstract – The diet of stoat, Mustela erminea, in the Italian Alps was studied from May to October 1996 and 1997 through faecal analysis. Scats were collected along twenty transects and near dens within a 10-km 2 study area. In the same area, seasonal variation in the available biomass of different prey types was estimated using capture-mark-recapture (small rodents), pitfall trap grids (insects), and fruit counts. About 60 % of scats (n = 734) contained small rodents, indicating that they were the main prey for stoat. However, the frequency of occurrence of fruits in the diet increased significantly in August, after rodent biomass had dropped by more than 50 % in July, but increased again to previous August levels. Fruit consumption decreased in September and October, although available biomass of fruits remained constant. Thus stoat included a large amount of alternative food in their diet when fruits were mature and their availability, and probably their profitability, increased.We were, however, unable to measure absolute consumption of prey. We conclude that primary prey, rodents, is always harvested, suggesting that the costs of harvesting rodents, abundant throughout spring–autumn, are never high. The consumption of alternative prey is independent of its availability, and increases only when relative benefits of harvesting alternative prey (fruits) increases, which is consistent with optimal foraging theory. © 2001 Éditions scientifiques et médicales Elsevier SAS alternative feeding strategy / diet / food specialist / Mustela erminea / trophic ecology 1. INTRODUCTION Optimal foraging theory states that a predator should choose prey types based on a trade-off between costs and benefits that will give the maximum net benefit to the individual [27]. The profitability of certain prey types might change in time. As long as the most profitable prey is very abundant, the predator should specialise on that prey alone, independent of the availability of the secondary prey type. Only when the energy intake rate of foraging solely on primary prey falls below the profitability of harvesting alterna- tive prey will the diet be expanded (e.g. [13, 28, 33]). Since handling time will have great effects on prey profitability, a decrease in prey numbers (biomass) will increase foraging costs. Other factors involved can be: (i) nutrient quality of different prey types; (ii) risk of the predator being attacked by larger predators; and (iii) increased energy demands, e.g. breeding (lactation in females, increased mobility in males searching for a partner), winter fattening, increased costs of ther- moregulation, or migration. Finally, a breeding female might change the spatial pattern of hunting behaviour in order to always stay close to its den, covering its territory in a patchy manner. In this case profitability of primary prey, as a result of local depletion and increased handling time, may decrease rapidly in a given patch close to the den. The dietary choices of small carnivores, such as the stoat (Mustela erminea L., 1758), will depend prima- rily on the temporal variation of foraging costs [13, 33, 41], which are mainly affected by availability of the primary prey species. Stoats primarily feed on small mammals throughout the year [7, 9, 13]. Even though some authors use the term ‘semi-generalist’ [26, 36], the results of theirs and other similar studies indicate that the stoat is a ‘specialist’ predator, feeding mostly on small rodents or on other food resources of animal origin [9, 8, 13]. Moreover, small rodent availability not only determines the presence of small mustelids, but also their activity and spacing behaviour, and *Correspondence and reprints: fax +39 0332 421554. E-mail address: adriano.martinoli@uninsubria.it (A. Martinoli). Acta Oecologica 22 (2001) 45-53 © 2001 Éditions scientifiques et médicales Elsevier SAS. All rights reserved S1146609X0101102X/FLA