directed booklets. Antennae short, small, one or two 

 jointed and fleshy. Body smooth or provided with soft 

 conical projections and bristles; below usually with seven 

 pairs of abdominal feet. At the posterior end the body 

 terminates in a more or less elongate tube, single or 

 double, the stigmata. This tube sometimes forms a short, 

 almost chitinized, tubercular projection on the dorsal 

 part of the last segment; at other times it is very long, 

 longer than the body, slender and composed of two joints, 

 the one sliding within the other, like the joints of a tele- 

 scope. In changing to the pupal condition, the larval 

 skin contracts to form the pupal envelope, and the body 

 becomes shorter, more oval and of a darker color, the 

 elongated" respiratory tube, in the 'rat-tailed' species, 

 being curved over the back. Unlike most other Cyclor- 

 rhapha, the frontal lunule is not used in pushing off the 

 cap to the puparium. 



The habits of the larvae are more variable than those 

 of the adult flies. A large number live in decaying wood, 

 or other vegetation, or in ordure, or decomposing animal 

 remains. Some live in the stems of various plants, some 

 in fungi. The larvae of many species of Syrphus and al- 

 lied forms are aphidophagous, crawling about on the 

 stems of plants frequented by plant-lice, and destroying 

 them. Some live in ants' nests and may be parasitic; 

 others in the nests of humble bees. 



Because of the large number of genera in the family, 

 and the consequent length of the table, I give at the close 

 an auxiliary grouping of the larger part of those genera 

 which present decisive characters. By examining any 

 specimen for the eleven characters given, and noting the 

 agreement in numbers, the student may, in many cases, 

 be more assured of his determinations. This family is a 

 peculiarly difficult one to define clearly all the genera in 

 a dichotomic table. I have introduced a number of cross- 

 24 



