POINTS OP THE ANATOMY OP POLYXENUS LAGURUS. 33 



General Conclusions. 



The principal interest of Polyxenus lies in the likeness of 

 some of the features of its anatomy to the anatomy of the 

 Chilopods. While it agrees with the Chilognatha in the 

 position of its generative organs and the duplication of some 

 of its segments (the first four segments are provided with only 

 one pair of appendages^ the next four have two pairs, and the 

 last one pair — Latzel and Bode), and in the fact that it is a 

 vegetable-feeding animal, in connection with which fact its 

 salivary glands are long and tubular, like those of other Chi- 

 lognaths ; it nevertheless resembles the Chilopods in the form 

 of its spermatozoa, which are long and filiform, and are con- 

 tained in spermatophores ; in the general structure of the 

 segments, having the legs wide apart, with a ventral region 

 between them ; and in the differentiation of the ventral nerve- 

 cord. The single artery given off in each segment seems at 

 first sight to afford a resemblance to the Chilopod circulatory 

 organ, but I believe this resemblance to be superficial. In its 

 essential characters the heart resembles that of the Chilognaths, 

 and remembering that the Chilognath heart is formed by the 

 confluence of spaces in the tissues of the body, the formation 

 of the arteries is not a deep-seated character. 



Characters peculiar to itself are the peculiar form of the 

 second pair of mouth appendages, and the absence of stink- 

 glands and the substitution for them of numerous spines as a 

 means of defence. 



In a former paper (5) I advanced certain views about the 

 phylogeny of Myriapods, and came to these conclusions : — 

 First, that Myriapods were descended from some Peripatus- 

 like ancestor ; secondly, that the Chilopods and Chilognaths 

 branched off from some common ancestor, not differing much 

 from the fossil Archipolipoda. Now the characters in which 

 Polyxenus resembles the Chilopods are characters common to 

 the larval Julus, and to the Archipolipoda (5). With regard 

 to the absence of stink-glands, and the substitution of spines 

 arranged in tufts over the body, I found in Julus that the 



