471 



jn which the hair lies. At first the hair is without any chitin and has 

 only a thin external layer of homogeneous substance. The hair may 

 be much curved as it lies in the cavity (Fig. 7), which is filled with a 

 fluid secretion of the cell. After the hair is completely formed the 

 secretion of this fluid is very rapid, the cavity becomes large and the 

 nucleus and protoplasm of the cell are crowded down and to one side. 

 The fluid, however, is not stored in the cell, but in the cup-shaped 

 invagination in which the hair lies. Small globules of this material 

 can be seen in the cytoplasm, but they are evidently extruded before 

 any great amount is collected in the cell. This secretion is largely 

 water, but a small precipitate forms when the animal is killed. This 

 precipitate is finely granular, occasionally stringy, and stains precisely 

 like the exuvial fluid which is found a few hours later, separating the 

 hypodermis and cuticula. It does not resemble the haemolymph: it 

 stains difl'erently with all analins; and it is more finely granular and 

 often shows a peculiar stellate form of coagulation which the haemo- 

 lymph never does. This latter stellate form of coagulation appears 

 after killing with sublimate acetic, Hermann, Fleming (strong), or 

 vom Rath, and is probably not due to the peculiar action of any one 

 killing fluid. 



The withdrawal of the hair, and the secretion of the fluid 

 require from thirty to eighty hours for their completion. When the 

 time for ecdysis arrives, the hair and the liquid contents of the cavity 

 about it, are, by the contraction of the cell, extruded between the old 

 cuticula and the hypodermis. The liquid serves as a cushion, protecting 

 the delicate hypodermis from injury by being brought into too violent 

 contact with the hard cuticula, and allowing room for the new cuti- 

 cula to form. This now develops as a thin, continuous layer covering 

 the whole body surface as well as the hair. In the later stages of the 

 larva the hairs are lost and these cells remain as glandular looking 

 cells lying beneath the cuticula. It was in this stage that I first noticed 

 them. In the younger larval stages the cells do not change much in 

 character during ecdysis, excepting that there is a decrease of size due 

 to the extrusion of material to form the exuvial fluid. In the last two 

 larval stages almost all the hairs have been lost and these cells exist 

 as large unicellular glands, which now take on the peculiar and charac- 

 teristic structure shown in Fig. 3. 



At pupation these cells are turgescent with the exuvial fluid 

 (Fig. I), which is extruded between the hypodermis and cuticula, 

 separating them. During the expansion of the wings, mouth parts and 

 legs, and the formation of the pupa, this fluid surronnds them, and 

 its solid part is gradually precipitated upon the surface of the animal, 



