348 G. O. Batzli et al. 



Salix and herbaceous dicotyledons. Only Eriophorum ranked highly as a 

 preferred item for both lemmings. 



Comparison of dietary composition in the field with the vegetational 

 composition of the habitat shows that Lemmus selects Dupontia fished. 

 The percentage in the diet is about twice the percentage in available for- 

 age during midsummer (Batzh 1981). Consumption of the most abun- 

 dant and widely distributed graminoid, the sedge Carex aquatilis, is 

 highly correlated with availability (/• = 0.91 , p < 0.01), and it is taken at a 

 level of about half of its availability. Surprisingly, Eriophorum (E. 

 angustifolium, E. russeolum and E. scheuchzeri, taken together) shows a 

 lower preference rating despite its higher ranking in the laboratory trials. 



In winter, graminoid shoots die back to about 1 cm above the 

 ground, and tissue below this point freezes while green and nutritious. 

 Food preferences are less apparent then, and lemmings consume the ma- 

 jor graminoids approximately in proportion to their availabihty. Because 

 polygon troughs, the favored winter habitat, contain much Dupontia, 

 this species is also an important winter food of lemmings. 



Energy Requirements 



The total assimilated energy required by a lemming can be estimated 

 by summing the energy demands for maintenance, growth and reproduc- 

 tion and the amount of energy lost in urine. Urinary loss of energy for 

 lemmings on natural diets equals approximately 5% of respiratory 

 energy (Batzli and Cole 1979). Maintenance energy requirement, or res- 

 piration, can be expressed as average daily metabolic rate, which is a 

 function of body size, ambient temperature and activity. Collier et al. 

 (1975) have developed a model of lemming energetics. The current ver- 

 sion of that model (Peterson et al. 1976) is 



ADMR = (1. 28 Pr "-0.457+6.40)4.19 



where admr is average daily metabolic rate in kJ day"'. Wis live body 

 weight in grams and Tis ambient temperature in °C. Measurements of 

 the metabolic rate of free-living Lemmus using doubly labeled water, 

 D2'*0, indicated that extrapolations from laboratory results probably 

 underestimate the metabolic rates of lemmings in the field by 40% 

 (Peterson et al. 1976). Therefore, the following calculations based upon 

 the curve of Collier et al. (1975) give conservative estimates of energy 

 requirements. 



The above equation was based upon animals with stable weights; 

 additional energy is required for growing animals. The amount needed is 

 determined by the growth rate of the individual and the efficiency with 



