The live-trap plot near the intensive site was run between 16 and 21 August; no lemmings were 

 captured here, either. Brown lemmings were occasionally seen in this area during the summer; how- 

 ever, this area was repeatedly visited by a large number of alert observers and the observations may 

 have been of transient lemmings, or of individuals in a population so sparse as to make detection by 

 trapping unlikely. Thus, while the primary objective in trapping was not accomplished, the results 

 emphasized that, at low points in the cycle, lemming abundance approached zero. 



While few data were acquired directly, this was, perhaps, an ideal season to begin an intensive 

 study of the effects of lemming grazing on other ecosystem components. The tundra vegetation was 

 last subjected to heavy grazing in 1965; since then considerable recovery has occurred. This is 

 particularly evident in the large amount of standing dead material in the vegetative canopy. A study 

 of the feeding ecology of Lemmus uimucronatus, emphasizing food selection by lemmings and its 

 influence on the vegetation, was begun by Melchior. 



In a pilot study, the food preferences of six hand-captured brown lemmings were tested in 

 individual experimental enclosures. Each lemming was permitted to graze a Vz-m^ area containing 

 four plant species. Each plant species was presented as a homogeneous 625-cm^ block of sod with 

 its natural standing crop. At the end of a 4-hour period, the lemmings were removed from the test 

 enclosures. The wet and oven-dry biomass of standing crop remaining was determined and compared 

 with the wet and oven-dry biomass of standing crop on ungrazed (control) blocks of sod. The food 

 preference of each lemming was ranked from 1 (greatest biomass consumed) to 4 (least biomass 

 consumed). The mean rank of each of the four plant species was: 1 for Eriophorum scbeucbzeri, 

 2.3 for Dupontia (iscberi, 2.7 for Carex aquatilis, and 3.6 for Arctopbila (ulva. As indicated by 

 these data, all of the lemmings showed a preference for Eriophorum scbeucbzeri; however, lemmings 

 must be tested on many additional combinations of grasses and sedges and the preferences among 

 these then related to lemming densities and the distribution of plant species. In anticipation of 

 further work on the feeding ecology of Lemmus trimucronatus, a large lemming enclosure was con- 

 structed and a point frame (see section on primary production) was used to characterize the pretreat- 

 ment (ungrazed) condition of the vegetation within the enclosure. Several areas within the enclosures 

 were protected from grazing; these served as controls. A live-trap grid was established near the 

 enclosure to monitor the unconfined lemming populations. 



An additional study, which although not conducted primarily at Barrow is critical to an under- 

 standing of the role of lemmings in the ecosystem, is being made by Coady and West on the energetics 

 of the brown lemming. This study is investigating the relationship of lemming metabolism to ambient 

 temperature, time of season, phase of cycle, and other relevant factors and will provide estimates of 

 digestive efficiency of lemmings consuming their natural diet. The results will allow us to convert 

 estimates of lemming density to estimates of energy utilization. When this is combined with data on 

 feeding behavior, the intensity of grazing on the vegetation throughout the population cycle can be 

 estimated. 



Metabolic rates of lemmings have been measured in each season of the year. A sample result, 

 which shows the oxygen consumption of spring-acclimatized lemmings measured at constant tempera- 

 ture over a 24-hour period, is given in Figure 19. Animals used for these measurements were 

 supplied with food and water and had sufficient space to move around freely. Average daily metabolic 

 rates can be calculated from such records and will indicate at least minimum values for animals in 

 their natural habitat at the same time of year. 



When a series of measurements is made of oxygen consumption over a range of temperatures the 

 relationship between ambient temperature and metabolic rate may be calculated. For example, using 

 animals captured at Barrow during the summer season, the following relationship was found: 



©2 consumption (cc/g hr) = 2.89 - (0.074 x temp in °C). 



43 



