363 



The second appendages meanwhile have fused, and now 

 enclose a cavity. Since the inner half of each of the second 

 appendages has become invaginated into the outer half, it 

 follows that the tube formed by their fusion must be lined 

 internally by the ectoderm. 



Shrinking of the appendages continues; in the tliirty-six 

 hour pupa the second appendage is no longer recognizable as 

 a compound structure ; it appears simply as a tube, lined bv a 

 single layer of ectodermal cells, the cavity containing 

 mesoderm. 



During the third day chitinisation commences. The outer 

 portions of the first appendages chitinise strongly; their inner 

 parts, however, remain as flexible membranes, similar to those 

 of the leg joints, neck joints, etc. (fig. 42). The compound 

 second appendage chitinises in a very remarkable manner. 

 The cells at its tip have previously arranged themselves so as 

 to present a serrated tip to the ovipositor; chitinisation of this 

 results in the characteristic sawing extremity. The remainder 

 of the compound appendage becomes semilunar in section ; the 

 two outer thirds chitinise heavily, and the chitinous prisms 

 so formed are connected by the median portion, whose walls 

 develop into tough membranes, and enclose a quantity of 

 mesodermal tissue. In side view the chitinised ovipositor 

 shows a very pronounced ''spiral" pattern, not unlike that 

 of large tracheae. 



The hard chitinous sheaths and the tough membranes of 

 the first and second appendages thus enclose^ between them a 

 firm, yet pliable passage, down which the eggs pass during 

 oviposition. 



An important part, of the female egg-laying apparatus 

 is the short appendages of the last abdominal segments; the 

 general nature of these appendages has been described in con- 

 nection with the general features of the insect; it remains to 

 point out the nature of the tactile organs with which the 

 appendage is so well supplied. 



The general features of the development of the appendage 

 are identical with those of the body integument. The bristle- 

 secreting cells, however, do not entirely chitinise; on the 

 contrary, they seem to grow in siz-e, and growing backwards 

 from the bristle, pass well into the cavity of the appendage, 

 remaining in connection with the cell only by a long delicate 

 process (cf. fig. 50). A large nerve enters each appendage, 

 then breaks up into nerve fibres, which communicate each with 

 a bristle cell. I could find no trace of an intervening meso- 

 dermal cell, similar to that described in the tactile bristles of 

 the leg or antenna. The whole appendage, however, is filled 



