166 ME. a. C. BOURNE ON THE 



The vascular system, alimentarj tract, gonads, and genital ducts 

 were so badly preserved in my specimens as not to admit of accu- 

 rate investigation ; but so far as I was able to investigate them, 

 tbey do not depart in any important particulars from the struc- 

 ture characteristic of the Julidse. The alimentary tract is bent 

 once upon itself, as in Glomeris. 



The tracheal system, however, is quite unlike that of the 

 majority of the Diplopoda. In them, as is well known, the 

 stigmata open into as many dilated vesicles, from each o£ which 

 a tuft of short unbranched tracheal stems, having a very feebly 

 developed spiral filament, are given off. In SpJicBrotherium the 

 tracheal stems are stout, have a well-developed spiral filament, 

 and are much ramified, in this respect resembling those of the 

 Chilopoda and Insecta ; but they differ from these in taking their 

 origin, not directly from the stigmata, but from peculiar sac- like 

 cavities, with firm chitinous walls, into which the stigmata open. 

 These internal chitinous structures, which I shall call tracheal 

 sacs, occur in connexion with each pair of walking-legs. A single 

 tracheal sac, with its tracheal plate attached, is represented in 

 fig. 8. It is triangular in outline, somewhat compressed from 

 side to side ; and its walls are composed of firm but translucent 

 chitin. At one of its angles it is attached to the tracheal plate, 

 and opens through it to the exterior by an oblique slit-like stigma. 

 I'rom its other two angles stout tracheae are given off. Each 

 tracheal sac lies with its longest axis at right angles to the long 

 axis of the body of the animal. The tracheae springing from its 

 innermost angle, near to the median line, are distributed to the 

 viscera ; those springing from its outermost or distal angle supply 

 the powerful lateral and dorsal muscles of the body. Three stout 

 tracheae, springing from the first pair of tracheal sacs, are dis- 

 tributed to each side of the head. 



The tracheal sacs and their stigmata have precisely the same 

 relation to the segmentation of the body as have the walking- 

 legs. Thus each of the first three body-segments has one pair of 

 tracheal sacs, the remaining nine segments two pairs each, mating 

 twenty-one pairs in all. A diagram of the tracheal sacs and their 

 relations is given in fig. 8. 



Glomeris, which is in all its characters closely allied to Splice- 

 rotherium, is also distinguished by the presence of branched 

 tracheae (Gregenbaur's Comp. Anat., English edit, p, 288); and 

 the same is doubtless the case in the allied genus Zephronia, 



