Insects. 



5897 



in very many cases the two antennae differ in this respect in the same 

 individual. The number of the segments of the abdomen remains as 

 before, and the forceps has undergone no change except an increase 

 in size. 



In the next stage, when the length varies from to 8 lines, the 

 first faint indications of sex appear. The antennae have now reached 

 twenty-two joints ; indeed, one specimen (an incipient male) had 

 twenty-three joints in one and twenty-two in the other antenna, of 

 both of which the terminal joint was certainly perfect. The number 

 of abdominal segments in all the specimens still continues to be nine, 

 but the forceps, in such as I presume are to be developed as males, 

 has become much thicker at the expense of its length, though the 

 basal tooth possessed by the mature insect is yet wanting. The 

 whole of the specimens have during the last two changes so com- 

 pletely assumed the colour and appearance, and in many examples 

 even the size, of the perfect state that at first sight they might readily 

 be mistaken for small mature females. 



On the final change there is no longer any doubt of their maturity. 

 The antennae of two or three specimens, out of probably a hundred, I 

 have ascertained to consist now of twenty -three joints, but I think this 

 is not the maximum when really perfect, for the apical joint in one or 

 two instances bears certain marks of not being the terminal one when 

 submitted to a powerful compound microscope. A considerable num- 

 ber of my specimens possess twenty-two joints in one or other of their 

 antennae, but the bulk of them range from twenty-one to nineteen, and 

 many even fewer. These have all undoubted traces of mutilation. As 

 before remarked, they are not seldom dissimilar in the same insect. 

 Fieber gives seventeen as the number; Serville says, " more than 

 twenty." The antennae may be described as filiform, or tapering 

 very gradually towards the apex. The first joint oblong-ovate, nar- 

 rowed at the base, as long as the second, third and fourth taken 

 together, rather smooth, with a few scattered hairs ; the second very 

 short, almost quadrate or even transverse, slightly pubescent ; the 

 third elongate, not quite so stout as the second and three times 

 its length, and, like the twelve following ones, slightly contracted at 

 the base and pubescent ; the fourth short, but rather longer than the 

 second ; the fifth short, a little longer than the fourth ; the sixth and 

 following six or seven gradually increasing in length, after which the 

 remainder are linear and as long as the second and third united, and 

 the terminal ones very closely and very densely pubescent. The ab- 

 domen of the male still retains the nine segments, and the forceps has 

 XVI. G 



