168 MR. G. C. BOURNE ON THE 



As is shown in my diagram (fig. 1-1), there is a complete corre- 

 spondence between the walkiiig-legs, the stigmata, and the ganglia 

 of the ventral nerve-cord ; and these structures occur tv^ice in 

 every segment after the third. I have not yet seen a satisfactory 

 explanation of this double segmentation in the hinder region of 

 the bodies of the Diplopoda ; nor am I able to offer one myself, 

 though I have given much thought to the subject. Until we arrive 

 at a satisfactory solution of the origin of metameric segmentation 

 itself, it will scarcely be possible to explain this additional com- 

 plication which has been grafted on it. 



The nervous system is quite similar to that of Glomeris, 

 which is described by Leydig (Tafeln zar vergleicbenden Ana- 

 tomie: erstes Heft, Taf. vii. figs. 3& 5). As stated above, there 

 is a small ganglion on the ventral nerve-cord corresponding with 

 each pair of legs. From each ganglion a pair of nerves is given 

 olF to the legs, and a pair of nerves runs outwards dorsally to the 

 tracheal sacs. The accessory appendages of the male derive their 

 nerve-supply from the last ganglion of the ventral chain. 



The eyes are aggregate, each being formed of a group of closely 

 apposed simple eyes. Those of my specimens were unfit for 

 microscopical examination. 



The antennary sense-organ is especially well developed in 

 Splicer otheiHum. Leydig was the first to call attention to the 

 existence of terminal sense-organs on the antennge o^Julus; and 

 in his figures of the nervous system of Glomeris (loc. cit.) he 

 draws the nerve-endings on the antennge of that genus. 



Althougli these antennary sense-organs are conspicuous, their 

 histological structure was not described until Biitschli gave a 

 short account of them in the ' Biologisches Centralblatt,' No. TV. 

 1884, p. 113. He gives a woodcut of the structure of the 

 sense-organ of Glomeris ; but his account and his figure are 

 mostly from memory, and he does not pretend to very great 

 exactness. 



In SjihcErotheriiim the terminal joint of the antenna is trun- 

 cated, and the truncated surface is covered with a thin flat 

 plate of chitin. On the surface of this pkte stand up a number 

 of conical or spike-like chitinous projections, each being sur- 

 rounded by a white nng with a dark pigmented border. In the 

 antenna which I have figured (figs. 9 & 10) there are eighteen such 

 projections ; but the number is not constant, varying from fifteen 

 to twenty. Their number is much less in the Julidse : in a 



