546 W. L. TOWER, 



trunk are derived from the hypotrichal membram by its evaginatioo^ 

 This I believe to be true of all insects. In L. decemlineaia the cells 

 of the hypotrichal membrane begin to divide, forming a conical pro- 

 jection (PI. 19, Fig. 63) which grows rapidly in length (PI. 20, Fig. 69) 

 and has a delicate lumen as seen in transverse section (PI. 20, Fig. 70). 

 This lumen begins first at the basal end of the trachea and abuts 

 directly against the intima of the tracheal trunk. In early stages it 

 is distinguished with difficulty (PI. 20, Fig. 69), but it soon becomes 

 well developed (PI. 20, Figs. 71 and 72) and here shows the structure 

 usually found in larval wings. At about the stage shown in Fig. 71 

 (PI. 20) the intima begins to be formed and soon the taenidia make 

 their appearance (PI. 20, Fig. 76). At the distal end of the growing 

 trachea the cells become much flattened and cover relatively large 

 areas (PI. 20, Fig. 74). The lumen of the trachea is, as has been so 

 frequently observed, intercellular even in the smaller parts, Ijut in the 

 finer ramifications the tubules are intracellular, like the tracheoles 

 which develop for the temporary tracheal system. 



In the pupal and adult wings there are numerous tracheoles, 

 which are intracellular tubes developed from the tracheae of the 

 wings, and these ramify from the main trunks all through the tissue 

 of the wing. Although several authors (Meinert, Krüger and others) 

 have failed to find tracheae in the elytra of beetles they are present 

 in all of the twenty odd species examined by myself and by Comstock 

 & NeedhAxM (1899). Even in dried material the tracheae are easily 

 discovered, and I cannot account for the failure of several authors to 

 find them. These tracheae are always lodged in well defined veins which 

 are established before either haemolymph or tracheae enter the wing. 



The number of tracheae is always six, the costal, subcostal, 

 ramous, médius, cubital and anal, some of which may be greatly 

 reduced or entirely absent (?), a fact well shown by Comstock & 

 Needham (1899) in some Cerambycidae. Moreover, when any branch 

 is reduced it is reduced in both pairs of wings to about the same 

 degree. This is strong positive evidence tending to proove that the 

 elytra are true wings, for as Comstock & Needham (1899) remark: 

 "If the elytra and hind wings are not homodynamous organs, it is 

 not probable that the modifications of the two [tracheal systems] would 

 be so closely correlated. We conclude, therefore, that the elytra are 

 modified wings." I would add that if the elytra and hind wings are 

 not homodynamous organs it is also improbable that they would have 

 the same type of tracheation and the same number of tracheal branches. 



