R. T. LEWIS ON THE LARVA OF MANTISPA. 215 



pecies after two further moults diflfered only in being larger and 

 somewhat lighter in colour. 



As regards Mantispa the ease is entirely different. It has 

 already been mentioned that the eggs are laid separately, each 

 being elevated on the top of a slender stalk. Of those received 

 from my friend in Natal, about a dozen hatched out after their 

 arrival here; the larvae were very small and bore not the 

 slightest resemblance in any respect to the perfect insect. In 

 total length they were slightly less than 1 mm. ; the head, about 

 0*25 mm. long by 0*2 mm. wide, bore two simple eyes and was 

 armed with what appeared to correspond to mandibles, projecting 

 forward, and tapering to a point so as to form a piercing rostrum 

 when closed together. The antennae were three-jointed, 0*16 mm. 

 in length exclusive of the terminal hair ; the basal joint short ; the 

 imiddle joint about four times as long, slender at its point of 

 attachment to the basal joint but expanding towards the other 

 extremity; the third joint being about half the length of the 

 second, and terminating with a long straight hair of the same 

 length as the three joints together. 



The palpi were seven-jointed, their whole length being 0*125 mm.; 

 the basal joint being slightly larger than that of the antennae, 

 the second joint about the same length but much more slender, 

 the next four quite short and of equal width, and the last slightly 

 longer and tapering to a point. Of the three thoracic segments 

 the prothorax is much the shortest, the meso- and metathorax 

 being about twice its length. The nine abdominal segments 

 gradually decrease in diameter towards the posterior extremity of 

 the body, which terminates in a scoop-shaped plate extending 

 from the anal orifice. The six legs are of equal size and shape, 

 0*3 mm. in length, the tarsi being very remarkable in that they do 

 not terminate with the usual claw or claws, but with a hollow 

 trumpet-shaped appendage the use of which is not very apparent. 



The subsequent history of this larva is of extreme interest, since 

 it affords a conspicuous example of what Fabre has termed hyper- 



