302 



ciably so. In no case is the larval leg of the last instar of full 

 size, but it usually has a small claw. 



There are eighteen in which the basal farts are more or less 

 implicated in the amputation (figs. 27, 29). Of these five 

 have only four tarsal joints, and all of these have the leg very 

 rudimentary in the last larval instar, less than a fourth of the 

 normal femur in size. Thirteen have the legs complete with 

 five- jointed tarsi, three or four of practically normal size, the 

 rest somewhat reduced but none very small ; the leg in the 

 last larval instar is usually complete, but generally rather 

 smaller than the normal femurs. In the two intermediate 

 instars it often seems absent or represented by a not definitely 

 namable chitinous plate. 



First Instar (five skins preserved) (fig. 33). — There are 

 two such specimens in which the whole leg and some of the 

 base appears to have been removed. In one of these the third 

 instar presents the basal plates of small size, and the leg as a 

 chitinous shield; in the fourth instar, the leg has three joints, 

 but is only as long as the opposite femur. In the last instar 

 the leg is about half the normal length, and a little deformed ; the 

 imago appears normal. In the other specimen the leg appears 

 as a small shield in the fourth instar, one later than in the other 

 specimen, and is not so well developed in the last instar ; the 

 imaginai leg is visibly shorter than the other femur, being 4'5 mm. 

 against 5, tibia nearly normal, tarsus 3 mm. instead of 5 mm. 



There are some specimens that do not quite admit of being 

 classified with those so far noticed. For example, fig. 20 

 shows a specimen in which amputation in the third instar was 

 intended, but resulted only in some basal injury to the leg. 

 The third instar preparation does not show the injured leg ; 

 the fourth instar shows it present, but curiously deformed ; the 

 fifth shows the deformity to persist, but exaggerated in the 

 preparation by twisting having occurred at the weak zone — 

 the imaginai result is a nearly normal tarsus, but very deformed 

 femur and tibia, with an attempt to produce two supernumerarj' 

 legs at the femoro-tibial articulation, both very minute, but 

 one with several joints. 



Fig. 15 illustrates a very similar specimen in which the third 

 instar skin is not photographed. 



