The large-sized species tended to occur in lower densities but, contra
ry to the expectation, their ranges tended to be less extensive than t
hose of the smaller-sized species. Correlation between the density and
the size of the range of the species was not significant. Disturbed p
opulations and marginal ranges of many species seemed to explain the d
eviations from expectations. Environmental factors indicating producti
vity seemed to be important in explaining density variation in many, b
ut not all of the species considered. The number of species and divers
ity in local assemblages increased, while the total density decreased
with increasing latitude. Tree species composition of forests in combi
nation with latitude explained best the geographical variation in the
number of species. The total density of birds of prey was best explain
ed by the combination of those variables that indicated productivity,
the contribution of the proportion of fields being the most pronounced
. Environmental variables indicating gradients in productivity also ex
plained best the variation in the structure of the local communities a
s a whole. The availability of forests seemed to explain the large sca
le difference in the community structure between the central and perip
heral parts of the country.