WEATHER AS AN INDIRECT INFLUENCE 303 



necticut Hill area has varied 12 days during the study. Apparently, therefore, other forces also 

 are involved. 



Analysis of the data, taking into acc(junt hours of sunshine, temperature, precipitation and 

 humidity, revealed but one valid relation.ship. There has been a decided tendency for the aver- 

 age nesting date on Connecticut Hill to advance a few days in seasons when the average mini- 

 mum temperature during the first 20 days of April was above normal. Conversely, later nesting 

 has accompanied lower temperatures. 



Physiologists have long been aware that temperature is a major factor controlling the rate 

 of physiological reactions. When temperature is low. radiation of heat from the body takes 

 place at a greater rate than when it is high and the increased energy required to compensate 

 for this leaves less for the various other body functions. This is undoubtedly the case with 

 grouse. Increasing daylight is probably the fundamental force motivating the reproductive 

 cycle, but the rate of advance seems, in some measure, to be controlled by the average mini- 

 mum temperature just previous to the egg-laying period. Yearly differences in this have ap- 

 parently resulted in the minor variations in nesting date that have been observed. 



During the Brood Period 



Losses among grouse broods each year are normally high and, as discussed elsewhere*, 

 represent one of the major unsolved problems in the ecology of the species. For several reasons 

 the number of dead chicks found by the survey has been too few to shed much light on the 

 question. Thus it has been necessary to depend, to a considerable extent, on circumstantial 

 evidence. 



In studying the possible influences of weather, certain variations have been found to be 

 regularly associated with increases or decreases in brood mortality. These correlations are pre- 

 sented here, not as conclusions, but as informalidn mi highly interesting fields for further 

 research. 



Both temperature and precipitation during the early brood period appear to be important. 

 Yet no connection was observed between weather and early season mortality. Rather it was 

 that of the last half of the sunnner (July 1()-August 31) which seemed to be affected. Losses al 

 this time have been constantly higher in those years in which June temperatures'^ have been 

 below average and, conversely, they have been lower when temperatures have been above aver- 

 age. When the number of days on which .01 inch or more of rain fell during the first three 

 weeks after the average hatching date each year was added to the analysis, the correlation 

 became more significant. 



With respect to precipitation, the number of days on which rain fell seemed to be more 

 important than the total amount'. This is logical in view of the fact that light rains soak the 

 vegetation to much the same extent as heavy ones and the vegetation may remain wet as long 

 after a fall of .02 inch as after a 1-inch storm. 



Although the variations in brood mortality have not been large, it is of interest, in view of 

 the degree of relationship, to consider the mechanism which might be involved. Laboratory 

 experiments with grouse and related species, such as poultry, indicate that physiological ad- 

 justments to changes in temperature are reflected by variations in the amount of energy needed 

 for heat production, activity and growth. As the temperature drops, more food is utilized for 



* See Chapter XII. p. 523. 



A The temperatute value usetl was what tlie Weather Biiieau terms "accumulated temperature". It represents the net deviation 



of the mean temperature recorded daily for the month from the standard mean for the same period. 

 t Flood conditions would, of course, be an exception, but none were experienced by the Investigation before mid-summer. 



