17G Ofi the (icorjraphical Distribution of Insects. 



composed of sporiidical species. Sueli a genus cannot be of 

 the latter character, because its species have a limited geogra- 

 phical extension ; but there is nothing to prevent it being rich 

 in species, which in such an event are accumulated in a more 

 or less restricted district. This is what takes place with many 

 intertropical genera especially, which thus constitute so many 

 exceptions to the preceding rule, that genera are so much the 

 more sporadical as their species ai-e more numex-ous. 



Being unable, in the present state of entomology, to calcu- 

 late the area of the different groups, as rigorously as is done 

 by botanists m regard to the objects of their study, M. La- 

 cordaire lias attempted to give a view of the distribution of the 

 genera of each family of the Coleoptera, considering them 

 under different aspects, which enables us to perceive their 

 greater or less degree of sporadicity, and the progression they 

 follow from the poles to the equator. For this purpose De 

 Jeans' C'atalogue has been again employed, which contains 

 2211 genera, sufhcient materials to operate upon with some 

 decree of exactness. It is of little importance to the inquiry, 

 whether these genera be all subsequently adopted or not ; it is 

 enough if they indicate differences of form, however minute, 

 and that these differences be proper to such and such countries. 

 The rapidity with which entomological discoveries succeed each 

 other, gives us reason to suppose, that, in future, their num- 

 ber will be increased rather than diminished. 



Bv a preliminary proceeding analogous to one of those em- 

 ployed by botanists, an effort has been made to determine the 

 number of regions which the genera of each family inhabit. 

 Only, instead of taking as a basis the rather numerous regions 

 established on a former occasion (winch would have been im- 

 possible, since we do not know the habitation of the different 

 groups with sufficient exactness), the number has been limited 

 to eight, which has already served for some previous calcula- 

 tion. In this way the following table has been obtained. 



