52 
flies during the successive generation periods. If there are recog¬ 
nizable and considerable variations in abundance, or definite breaks 
in the insect series, correlated in a uniform way with waves of 
pellagra increase presently* following; and if exceptions to this cor¬ 
relation are to be clearly explained by exceptional circumstances, 
we shall have strong reason to believe that one or more of the 
species of Simulinm then and there present are causally related to 
the conveyance of pellagra from one person to another. Other and 
more direct lines of operation on this problem belong to the pathol¬ 
ogist rather than to the entomologist. 
