THE CANADIAN ENTOMOLOGIST. 379 



the four regions is 28.8 per cent., while for the other ten regions, all but 

 one of which contain at least one genus of this class, average 6.5 per cent. 

 These numbers are easily compared with those for all plants. It will be 

 observed that while the average per cent, of endemic apically dehiscent 

 genera is 16.9 for the four regions, and .9 for the ten others, for all 

 endemic genera the average is 10.8 and 2.6 per cent, respectively. Thus, 

 in regions 3, 4, 6 and 7 the relative abundance of the Solanum-Cassia type 

 may be expressed as +5.6 and for the other ten as — 1.7. Comparing 

 the relative abundance of all genera occurring, we find that for the four 

 regions it may be represented by +4.5, while for the other ten regions it 

 is — 3.3. While the Solanum-Cassia type is abundantly represented in 

 the Indian region, the per cent, of endemic forms and all forms of this 

 type occurring there is something more than one less than the per cents 

 of all the genera of flowering plants which are found in the flora. Next 

 to the tropical American region the flora of the Indian region is the 

 richest of the fourteen regions recognized, and the abundance of the 

 Solanum-Cassia type there seems to be due rather to the richness of the 

 whole flora than to any special conditions favouring its development. 

 Considering only the three regions, 4, 6 and 7, we find that the average 

 per cent, of genera of the Solanum-Cassia type endemic is 18.6, while for 

 the other eleven regions it is 1.9. For all genera of the Solanum-Cassia 

 type occurring, the three regions average 29.3 per cent., while the other 

 eleven regions average 8.4 per cent. Comparing these figures with those 

 obtained for all genera of plants, we find that in the tropical American, 

 Australian and extra-tropical South-American regions the per cent, of 

 apically dehiscent genera endemic in the several regions is S.;^ more 

 than that for all genera, while in the other eleven regions it is 1.7 less, and 

 for all apically dehiscent genera of the Solanum-Cassia type occurring the 

 per cent, for the three regions averages 10 more than that of all forms, 

 while for the other regions it averages 3. i less. 



The present work is essentially a comparison of the distribution of 

 floral structures, but these floral structures are thought to be adaptations 

 to a factor in the environment, which so differs in potency in the several 

 regions under consideration as to bring about a difference in the frequency 

 of occurrence of these floral types. 



In a problem of biogeography which involves taxonomic, morpholog- 

 ical and ecological considerations, it is difficult to decide just what shall be 

 the basis for comparison. The characteristics of genera probably furnish 

 most satisfactorily the morphological units which we seek, but ecologically 

 the importance of the genus in the flora may be vastly increased by 

 specific differentiation. 



Without attempting any comparison with the number of species of 

 the whole flora, we may examine the distribution of the species of the 

 Solanum-Cassia type. The differentiation of Solanum and Cassia in 

 tropical South America first called attention to the distributional phase of 



