— 1 5 8— 



Hymenoptera (Cresson's estimate) Phytophaga 573 



Entomophaga 2166 

 Praedores 1078 



Anthophila 633 4,450 



Lepidoptera; Butterflies (Edward's List, 1884), ... 614 



Moths (Grote Check List, 1882), '. 3,184 



Tineidae (not included by Grote, 



Chamber's List, 1 878), 779 



Diptera (Osten-Sacken's estimate in 1878), 2,500 



Coleoptera (Henshaw List of 1885), 9, 507 



Hemiptera-Homoptera (Uhler estimate), 1,200 



Heteroptera (Uhler Check List, 1886), 1,448 



Orthoptera (Scudder estimate), 450 



Neuroptera, not estimated — perhaps 1,000 



Giving a total of 25,132 



Upon former occasions when it has been my privilege to address 

 this Club officially, I have urged the importance of the study of the 

 earlier stages of our insects upon its members and upon all students in 

 entomology. I beg leave to repeat the recommendation in view of the 

 value of such study in a scientific classification, and the necessity of it in 

 economic investigation, for of but comparatively a small number of our 

 insects are the earlier stages known, and of only a few do we possess full 

 life-histories. 



Although the Coleoptera have been the most thoroughly worked of 

 any of the orders, yet it will surprise some of you to learn, as it did me, 

 when I recently received the information from an eminent Coleopterist, 

 of how small a proportion are the earlier stages known. 



By permission of the writer, Mr. E. A. Schwarz, 1 give the com- 

 munication addressed to me in reply to an inquiry made, in which the 

 information above referred to is embodied. It is of so much interest 

 and value, that it deserves publication: 



"I have endeavored to keep track of the descriptions of Coleopterous 

 larva; from all countries. I herewith select at random a few of the larger 

 families, and do not believe that the proportion of the known larvae to 

 the total number of species would be materially changed if I would go 

 through all families. For the Chrysomelidce, unfortunately, I cannot get 

 at the figures without much trouble. 



In Carabidace, including the Cicindelidq, there are described, in 

 round numbers, the larvae of 120 species representing 40 genera, out of 

 a total of 9300 species. 



