CLASSIFICATION OF THE ASCALAPIIIDiE. 225 



the female with the appendices of the last segment ; and both fall 

 to the ground, and rest on some plant. Their position is then 

 similar to that of the Noctuidce and other Lepidoptera. A few 

 days after pairing the female lays her eggs. These are arranged 

 in two parallel rows, to the number of forty or fifty, on some 

 plant, generally grass. When at large I have observed them 

 to hunt chiefly Lepidoptera and small beetles." The young 

 larva3 scatter themselves little from the position in which they 

 are bred, and grow very slowly until the end of winter (though 

 born in August) ; they are then very difficult to^ find, and appear 

 to feed chiefly on Aphides, hiding mostly amongst moss and small 

 stones. In the spring they begin to grow more rapidly, and 

 take to larger food ; and in June they spin cocoons amongst low 

 herbage, in which they change to pupa). The larvae have a 

 process on the sides of each thoracic and abdominal segment, 

 though far less developed than in Ulula ; and the possession of 

 these processes seems to bo one of the best characters whereby 

 to separate the larvae of the Ascalapliidce from those oi Ifymeleo- 

 nidcB, which latter have no processes. In the same Journal for 

 1867, p. 966, Brauer briefly alludes to a larva of this family from 

 Eockhampton, which, I think, is probably that of a Suphalasca. 

 He describes it as having only one long tooth to the mandibles. 



I possess the eggs of a species of the family from Saugor, Cen- 

 trallndia, given to me by Mr. F. Moore, of the India Museum. 

 They are arranged in two or three rows on a dead twig of mulberry, 

 to the number of nearly sixty. These eggs produced larvae thir- 

 teen days after they were discovered. The larva is about 3'" in 

 length, the head rather broader than long, with two produced 

 eye-bearing tubercles in front, and very deeply concave on its 

 hitidor margin, extremely rough ; the mandibles with three large 

 teeth and many smaller ones. Each thoracic and abdominal seg- 

 ment has a subcylindrical process on each side furnished with 

 long and strong spines. Neither with these eggs nor with those 

 of AsealapJiiis macaronius is there any vestige of the repagula 

 mentioned by Guilding. 



A larva given to me by Mr. Bates, captured by him in tlie 

 Amazon region, evidently belongs to the family, and may possibly 

 be that of a Ulula. It is 6|"' long, without the mandibles (or 8|"' 

 including those members), and nearly 5'" broad at its broadest part. 

 The mandibles have three equidistant long teeth, between which are 

 very short tubercular teeth. The head is nearly quadrate and sea- 



