on Flower-Haunting Dipfera. 125 



remarkable, and similar differences occur in almost all 

 orders and more than one genus. Miiller tabulates 

 several orders with their visitors, and his example has 

 been followed in this country by Mr. J. C. Willis aud 

 others. 



However, as a rough classification, in order to obtaiu 

 some insight into the intelhgence of these Diptera, I 

 arranged all the flowers which I have examined into six 

 divisions : — 



1. Uanuncidm group, including all open flowers rich 

 in pollen, into which any insect may enter bodily. 



2. Crucifer group, including such forms as Umhelli- 

 ferx, Galium, Alisma, as well as all Compositse with 

 extremely short florets, such as, e.g., Daisy. 



3. Veronica, Mijosotlx, etc., i.e., flowers with a distinct 

 though short tube, and which involve higher intelligence 

 in their clients than the preceding. 



4. The long-tubed Compositse such as Cardu us, and also 

 Lychnis, etc. 



5. The smaller Legnminosfe, Medicago, Ornithopu>>, and 

 Trifolium procumhens. 



6. The larger Legmninosse, 



To these six classes I have added Ajiiga [or Euphrasia) , 

 Scilla, and Orchis;. 



The result is expressed on the next page :— 



