228 
Willis and Bur kill. — Flowers and 
identify, thus rendering this paper far more accurate and 
complete than it could otherwise have been made. 
Dr. D. Sharp, F.R.S. (Coleoptera and many others). 
G. H. Verrall, Esq., F.E.S. (Diptera). 
R. C. L. Perkins, Esq., M.A. (Hymenoptera Aculeata). 
In addition to naming our Coleoptera, &c., Dr. Sharp has 
given us several valuable suggestions and criticisms, for 
which we desire to express our thanks. 
i. Introductory. 
The first person to emphasize the need of studying the 
insects that visit flowers, as well as the flowers themselves, 
was H. Muller. In his classical books he gives descriptions 
of the mechanisms of many flowers, and also full lists of 
their insect-visitors. From these it is easily seen that there 
are degrees of specialization in the flowers — more effectual 
shelter of pollen and honey, more conspicuousness, scent, &c. 
— and corresponding degrees of specialization in their insect- 
visitors — greater adaptation of mouth-parts to procuring 
pollen and honey, greater cleverness in finding concealed 
honey, preference for certain colours or scents, &c. Muller 
himself was the first to put these facts into shape, as the 
‘ Theory of Flowers ’ which is enunciated in his ‘ Alpen- 
blumen ’ (1881 ; it was partly, however, set forth in earlier 
papers). Entomophilous flowers are there divided into 
several classes ; beginning with the lowest, these are : — 
1. Pollen-flowers (Po) of simple type, offering pollen only 
to their visitors ; examples are Helianthemum vulgare , Spiraea 
Ulmaria . 
2. Flowers with freely exposed honey (A); Umbelliferae, 
Parnassia. 
3. Flowers with partially concealed honey (AB) ; Cruciferae, 
Potentilla . 
4. Flowers with fully concealed honey (B) ; Mentha , Calluna . 
5. Capitula with fully concealed honey (B') ; Compositae, 
Phyteuma. 
