112 DAVID JENKINS AND ADAM WATSON 



ever 38-58 per cent of the tabbed birds were lost each December (Table VIII), 

 and so the total number of individuals missing from the population may- 

 have been much greater. Clearly many more grouse were lost, presumably 

 through emigration as well as death, than would appear from the counts 

 alone. 



The constancy of the population density at these times was evidently a 

 result of an active process regulated by the grouse themselves. This main- 

 tained total numbers steady despite alteration of the individual membership 

 of the population. Despite the apparent stability of the population, individual 

 grouse must have constantly moved into and through the area from outside, 

 replacing the casualties and emigrants. 



A similar process may have operated during the late winter when count 

 totals each year were again similar and also during the summer-time. In 

 both 1958 and 1959 old birds that died during the breeding season were 

 apparently replaced. In 1958 there were seventy-eight known casualties 

 (mostly unmarked) on the study area between April and July (Table VII). 

 Thus with a population of 370 in spring, the calculated adult population 

 would be about 290 in August. In fact the estimate was about 410. Similarly 

 in 1959 sixty-one deaths were recorded between April and July; yet the 

 respective figures for spring and August populations of old birds were 225 

 and 255 (Table II). The differences between 370 and 410, and between 225 

 and 255, are slight (plus 11 per cent and plus 13 per cent), but possibly the 

 spring level was actively maintained throughout the breeding season. Further 

 observations are needed. 



DISCUSSION 



This is a preliminary account of our fmdings and it might be premature to 

 form any defmite conclusions. Nevertheless it appears as though the grouse 

 population density on our study area has to a certain extent been regulated 

 by the grouse themselves. External factors, such as predators and disease, are 

 certainly important but apparently this population has been to some extent 

 self-regulatory. We do not know how the mechanism works nor why the 

 grouse themselves permit their population density to stabihze at different 

 levels. Possibly there are annual behaviour differences, maybe correlated 

 with changes in resources. 



Observations suggest that numbers change when different types of 

 territorial behaviour begin. During the summer there is little aggression 

 and numbers build up to a maximum through breeding. Then in August 

 and September the cocks take up territories and some aggressive display is 

 seen. This coincides with the first fall in numbers. During the autumn, 

 temporary pair associations are observed and one function of these may be 



