532 



I'RODI'CTIVITY OF GROVSF. POPULATIONS 



member that the effectiveness of all these decimating agents is affected in varying degrees 

 hy a number of factors such as cover quality, weather and buffer abundance. The basic data 

 of llic Imcstigation are not influenced by the factor of hunting since this has been prohibited 

 on the jHJinarv study areas. Neither has there been any evidence of starvation observed. 



The following table summarizes the analyses of the dead grouse picked up. Some of the 

 specimens charged to predalion very likely died of other causes but the available evidence 

 gives no reason to believe that the proportion of such instances has ever exceeded the dif- 

 ference between the figures listed below and the 15-odd per cent estimated to cover such 

 losses*. 



TABLK 83. PROPORTION OF DEAD GROUSE FOUND ON VARIOUS STUDY AREAS 



ATTRIBUTED TO VARIOUS CAUSES— 1930-1942 



On the various study areas, causes other than predation were identified only on Connecti- 

 cut Hill. The instances of accidental death are all attributable, so far as is known, to col- 

 lisions with trees, fences and sometimes buildings. Also they have occurred chiefly during the 

 fall and, in most cases, appear to have been associated with the so-called "crazy flight". Losses 

 resulting from crippling by hunters have not been a factor on the areas from which this ma- 

 terial was collected, although on heavily shot tracts they undoubtedly enter in'^. Incidental 

 records from elsewhere over the State have also involved a number of other types of accident. 

 Along improved highways birds are often struck by automobiles. Females have been found 

 burned to death on their nests following forest fires. And. now and then, a grouse has been 

 caught in a fox trap. The disease relationships observed are discussed in Chapter X. 



The number of adults surviving from year to year has been only slightly correlated with 

 differences in the level of the preceding fall po|)ulation (figure 45). 



On the other hand, the number of adults lost each winter has borne a marked relationship 

 to fall abundance. As discussed elsewhere''', other conditions being similar, the more numerous 

 grouse are the more frequcntb they will be encountered and killed by predators. The data 

 for each year on the two main study areas are compared in figure IC)*. In both cases, as 

 the fall population has risen, so also has the mortality. 



Yet the proportion lost has iiccn less related (figure 17)'. Although the trend has been 

 similar, the degree of change has b<-en less and exceptions more frequent. The correlation 

 has been stronger on the Adirdiidack area than on Coiuiccticut Hill. 



At the same time, when iIk- data were analyzed on the basis of average values after the 



• Sec Cho|.l.r Vn, [.. 319. 

 A Sec Chaplrt IX. !>. .176. 

 t Src Chopler VII. p. .121. 



t BertUBC ol the wide tlifTrrence in the numerical popuUtiona involved (»n lhi>«r two arean the tinia covered 

 been rediirrit to percentage deviation from the mean. 



thia graph have 



