540 



PRODUCTIVITY OF GROUSE POPULATIONS 



area reveals a distinct tendency for greater relative increases to be associated with lower breed- 

 ing populations*. 



To examine further the effect of abundance the records for each area were grouped accord- 

 ing to the density of the breeding population (table 88). This method too shows an inverse 

 relationship" between the number of birds in the spring and productivity by the following 

 fall. But just what interaction of factors may be involved is not clear. Certain features of 

 the data mav. h(jwever. be mentioned. 



TABLE 88. HEL\TI0NSH1P OF DENSITY OF BREEDING POPUIATIONS TO 

 PRODUCTION RATIO ON CONNECTICUT HILL AND 

 ADIRONDACK STUDY AREAS— 1930-1912 



* Does not cover years after 1938 when spring censuses were not made. 



A According to same groups as used in table 83, 



In the first place, the proportion of the loss in potential productivity as of September 1 

 which occurred during the breeding and nesting season has averaged 69.5 per cent on Con- 

 necticut Hill and 50.6 per cent on the Adirondack area as against 30.5 per cent and 49.4 per 

 cent respectively for the brood period. On the former area loss of productivity up to this 

 point, therefore, has been due more to breeding and nesting failure than it has to brood mor- 

 lalitv. while on the northern tract the two have had nearly equal weight. 



To analyze this relationship still further these ratios were broken <lowii Italtie 89) accord- 

 ing to the same groupings of breeder density considered in table 83. 



TMU.E 89. LOSS i)V POTENTIAL PRODUCTIVITY DURING BRi:i:niNG AND NESTING 

 SEASON CO.MPARED WITH TH.VT DURING BROOD PERIOD ON 

 CONNECTICUT HILL AND ADIRONDACK STUDY A RE.\S— 19.30-1912 



According to density i>f brcrders a» in table 8.1. 



It is nolt'worlln that on Cunncctic iit lliil as tlic density of the spring; |)i)[)ulali(.)n increased 

 so also did the proportion of the loss in productivity which look place during the breeding 



* On Cnnnprlicut Hill thi* rorrrlatinn in tlalittically signifiraDt wtiilr on tlir othrr aret it it ffomewhat Icm drBnjte. 

 A Analysii of these by clii-squarc ■huwt that on Coniircticut Hill Croup H baa been very close to ibo total average, but the 

 drviations of Groups 1 and HI arr hiichly significant. 



