observed on these plots; however, both within-plot and between-plot heterogeneity on the control 

 plots was very high, and obscured any possible treatment effect. 



The three manipulated plots, when they were first sampled, had fewer mites and Collembola 

 than the control plots. This is attributed to cliance variation in the plots selected (randomly) for 

 manipulation. This difference was maintained in the plot (clip and clear) from which ori;anlc 

 matter was removed. However, in the two plots in which organic matter was made available to 

 soil-dwelling animals this difference disappeared and by August the density of Collembola was 

 greater than or equal to the density in the control plots. This finding should be confirmed by the 

 use of replicated manipulation plots. If this is really a phenomenon, it could be due to the 

 nutritive value of the litter added or to extension of the physical habitat, providing more living 

 space. The physical and nutritive effects could be separated by the addition of non-nutritive 

 litter to tundra plots. This possible increase in the number of Collembola was the only effect of 

 natural manipulations on arthropod populations detected in the first season following manipulation. 

 Additional effects might be expected in subsequent seasons, as saprovores and decomposers utilize 

 the added litter, and are then fed on by carnivores. 



Decomposers - microbiology and soil nutrient 



The effect of manipulation on decomposer populations was examined for the fertilized (20-10-10), 

 oil-treated, and track plots only. In the oO-treated plots at sites 1 and 3 there was a 10-fold 

 increase in the bacterial plate count, and on the fertilized plot there was a twofold increase of the 

 bacterial flora. However, at the site 3 track site there was a 15-fold decrease in the bacterial plate 

 count as the result of compaction. It may be concluded that some of the manipulations used in this 

 study altered the microbial biomass and may have had a greater effect on the activity of organisms 

 in the soil at the start of the manipulation. No activity measurements were made of any plot at 

 site 2, but the activity measurements at sites 1 and 3 are in agreement with this idea. 



Complete soil nutrient analyses were performed on the manipulated plots at several sample 

 intervals. The most conspicuous results were seen on the two fertilizer plots. The added nutrients 

 were detectable in soil solution soon after fertilizer application. The added nitrogen (NH3) was no 

 longer evident at peak season, although the PO, level remained high. Other soil nutrients, however, 

 were less concentrated in the soil solution of the fertilized plots than in the control plots. Thus, 

 it appeared that fertilization stimulated biological activity to the point that nutrients (other than 

 those added) in the soil solution were depleted relative to those in the control plots. Since fertili- 

 zation did not lead to a demonstrable increase in above-ground production, the difference is 

 attributed to nutrient content of the vegetation, storage by root systems, or nutrients bound by some 

 groups other than primary producers. Samples for plant nutrient have been analyzed but data are not 

 completely available at this writing. 



In general, high soil moisture is related to the predominance of a moss layer. The difference 

 between sites 1 and 2 is evident in the soil moisture values. All of the site 1 plots had lower soil 

 moisture than the site 2 plots. The effects of treatment on soil moisture were not obvious at either 

 site 1 or site 2. 



The heated air manipulation appeared to have considerable effect on the soil and soil solution. 

 Soil moisture was similar to that in the control plot; however, redox potential was lower, and soil 

 temperature and pH were higher in the treated plot than in the control plot. Concentrations of Fe*~^, 

 NHj, Si, Ca and Mg in the soil solutions were lower in the heated plot than in the control plot, and 

 PO4 concentration was much greater in the heated plot. The effects of physical disturbance were 

 similar to those seen on the heated air plot. In all cases, the mechanisms relating the manipulation 

 to observed effects on soil properties are not clear; and at this time in the analyses it is difficult 

 to separate the treatment effects from the heterogeneity in soil properties encountered at both sites. 



65 



