negative correlations were found throughout the season. Thus, abundance of Pedicia hannai is not 

 directly associated with abundance of its principal prey. It is possible, though, that the negative 

 correlations result from increased predation on CoUembola where Pedicia hanni are abundant. 



Nematocera larvae occurred in greatest numbers on the dry polygon top plots, and showed a 

 strong negative correlation with Pedicia hannai larvae, which may prey upon them. This agrees with 

 results obtained from the artificial manipulation site, which differed from the intensive site 2 in 

 having fewer Pedicia hannai but more Nematocera larvae. On the community gradient site Nematocera 

 larvae resembled Pedicia hannai larvae in showing a,net decline over the season. A similar decline, 

 although of less magnitude, was also seen on the intensive site. 



Ground beetles (Carabidae) occurred with equal frequency on the mesic and dry plots, but quite 

 rarely on the wet plots. Abundance of larvae varied from 18/m^ on 2 and 12 July to 60/m' cxi 11 

 August. Adults were found in all three summer months; however, they were never common. The 

 overall ratio of larvae to adults (108:8) could be interpreted as indicating life cycles of several years, 

 short adult life spans, or high mortality of larvae and pupae. All of these factors may be involved. 



Rove beetles (Staphylinidae) were found most frequently on the polygon plots and least frequently 

 on the wetter plots. Their density was about one-third that of the Carabidae. Larvae of leaf beetles 

 (Chrysomelidae) were found infrequently in all habitats. 



The emergence of adult crane fUes was followed using two emergence traps, each covering 1 m% 

 on each of the six study plots. Tipula carinifrons was captured in 10 of the 12 traps; the empty 

 traps for this species were on a polygon top and in the polygon trough. The greatest number of 

 captures in one trap was 11; the other traps captured 1 (4 traps), 2 (4 traps), or 3 (1 trap) individuals. 

 Thus, Tipula carinifrons is able to complete its life cycle in a wide range of habitats but is not 

 exceedingly abundant in any of these. With the wide range of habitats involved, the emergence was 

 not highly synchronous. For instance, 17 males emerged between 3 and 27 July, with the median 

 emergence on 18 July and the central 75% of the emergence falling in a 13-day period between 12 and 

 25 July, Sticky-board traps on the same plots gave somewhat different results. The median capture 

 of 125 male Tipula carinifrons occurred on 11 July, with the central 75% captured between 7 and 17 

 July. The difference is probably due to the small number of captures in the emergence sample. 



Pedicia hannai emerged into 7 of the twelve traps. None emerged into the polygon top traps, 

 and no larvae were encountered in these habitats. The other traps captured from 1 (3 traps) to 

 33 individuals. The two traps in the wet meadow produced a mean of 26 adult Pedicia hannai/m^. 

 In both crane fly species the emergence of adults was not sufficient to account for the large decline 

 in larval populations which occurred from early to mid-season. 



The emergence of Pedicia hannai was more synchronous than that of Tipula carinifrons. The 

 median emergence of 56 males occurred on 13 July, and the median of 31 females on 14 July. In 

 both sexes 75% of the adults emerged in a 4-day period between 12 and 16 July. Similar results 

 were obtained by sticky-board trapping. 



The median dates of emergence on the community gradient site 4 were several days later than 

 the medians for site 2. This correlates with a later snowmelt on the community gradient site, 

 and supports the hypothesis that timing and synchrony of emergence are controlled by a system of 

 time or temperature summation rather than by some seasonally varying environmental stimulus. 



Avian populations and production 



Dave Norton* University of Alaska 



Steve MacDonald University of Alaska 



George West University of Alaska 



Populations of breeding birds and the production of these populations were estimated by 

 sampling within a 700 x 500-m rectangular plot encompassing sites 1 and 2 at Barrow, Alaska. In 



♦Principal author. 



48 



