214 



liairs of its tail are in contact with the pilose portion of the style and 

 become pollinated, if the flower has recently come into liloom and the 

 style has not yet shed its hairy coating. But this leaf-cutter is not wholly 

 dependent on its position while collecting honey for a supply of pollen. 

 On several occasions it was seen clinging to the style and transferring 

 pollen to its alxlomen with its hind-legs, a maneuver that no other bee 

 seems capable of performing. With the hairs of its tail charged with 

 pollen it is easy to understand how cross-fertilizatiou is effected, as it 

 passes from one flower to another: and so systematic are the movements 

 that they appear to be evolved for the purpose they fulfll. So far as 

 the writer has been able to discover, no other insect than .1/. hrevis is of 

 use in fertilizing the tall liellflower. Another leaf-cutter. M([!<ichHv iii- 

 fni(/i!is Cresson, was often seen collecting honey from Inipatinis imrca 



Fig-ures. Campanula Americana L. 



a. Triaugiilar bases of stamens. 



b. Pilose end of style covered with pollen and bee collecting honey. 



c. Style denuded of hairs, bee about to brush against lobes of stigma. 



d. Style denuded of hairs and bee in position on style while collecting honey. 



-Muhl. and pollen on HdhnifJuis anninis L. growing nearby, but was never 

 seen on C. Aiiicricaiid. 



The tall bellflower. on which the observations descrilied were made, 

 grew in the back yard of a city residence, and was in bloom from .Tulv 



