The June 16 runs of Circle routes A and B were found to have a similarity of 

 0.96, indicating that these two routes sample essentially the same avifauna. The 

 two June runs of route A (June 16 and June 22 had a similarity of 0.83. Thus, we 

 may consider any routes having a similarity greaterthan approximately 0.80 to be 

 essentially identical. Differences resulting from a change from route A to route 

 B were minor. This change should not interfere with comparisons of future runs 

 of route B with 1968-1977 data obtained for route A. 



The Species-abundance relation and sampling adequacy . Species-abundance 

 plots (the species curves of Preston 1948) for pooled breeding season (May-July) 

 runs are shown in figure 49 for all five roadside wildlife survey routes, and 

 the four standard routes which do not include the Dreyer Ranch route. Also shown 

 in this figure are the closest-fitting lognormal curves for each case. An excel- 

 lent fit is obtained for the five routes {l^= 6.94 10 d.f.) where S=10.99 and 

 a=0.2075. The corresponding value of S^ (that is, the theoretical total number 

 of species in the universe defined by the technique and the sample areas) was 

 found to be 91.6. Since the total number of species actually observed on these 

 runs was 36, sampling appeared to be 91.6% complete. This value of S^ (91.6) is 

 reasonably close to the actual number of summer resident, permanent resident, and 

 summer visitor bird species observed in the field during 1977, that is, 105. Thus, 

 the technique appears to be quite effective in estimating total species number. 

 Indeed, various autnors (Edden 1971, Longuet-Higgins 1971) have suggested that S^ 

 may be one of the most meaningful community parameters in describing community 

 diversity, as it is largely independent of sample size. Year-to-year changes in 

 S^ obtained during monitoring should provide a reliable description of changes 

 in community structure during the monitoring period. 



For the four standard routes (figure 49), a fairly good fit (X^=9.57, 10 d.f.) 

 IS obtained where S =10.29 and a = 0.2075. S. may then be calculated for this 

 smaller universe as 87.9. This compares to 81 species actually observed on these 

 runs. Sampling in this case appears to be 92.2% complete. The values of S^ for 

 individual routes (followed in parentheses by the indicated precentage of sampling 

 effectiveness) were found to be as follows: Circle, 71.7 (68.3%); Flowing Well, 

 49.2 (85.4%); Prairie Elk, 63.5 (89.8%); Missouri River, 70.0 (90.0%); and 

 Dreyer Ranch, 53.0 (98.1%). The estimate of S^for the Circle Route is believed 

 to be high due to the relatively large number of singletons (species providing 

 only one observation). We can thus conclude that the roadside survey technique 

 is effective in sampling roughly 90% of the total number of species in the sample 

 area which are capable of being detected by the technique. 



Habitat Relations . Pearson product-movement correlation coefficients re- 

 vealed significant correlations between habitat variables and sample abundances 

 of certain species sampled by the four standard roadside wildlife survey routes 

 (Species Narratives). In nearly all cases, these correlations agreed with the 

 subjective impression of the observer regarding habitat selection by these species. 

 An attempt to further elucidate habitat requirements of individual species by 

 stepwise multiple regression using the same data was not as successful as had 

 been hoped; the twelve habitat variables used in the regression analysis accounted 

 for only 5% to 43% of the variation in sample abundances for those species analyzed. 

 Multiple regression equations for those species showing the clearest habitat rela- 

 tions are given below. The coefficient of multiple determination (r2), which 

 equals the amount of variation in the dependent variables, is given as a percen- 

 tage in parentheses after each equation, and significance is indicated by aster- 

 isks (*0.01<p<0.05, **p<0.01, DF=187): 



169 



