288 Mr. R. I. Pocock on the 



separated from Polyxenus the presence of these copulatory 

 feet. 



The occurrence of these organs in the Glomeridaa and 

 Inlidse is due to the existence of similar physiological require- 

 ments, but that the existence of similar physiological require- 

 ments in two groups is not a sign of affinity between them 

 need now-a-days hardly be urged. It would be as justifiable 

 to consider the branched trachese of Glomeris and Scolopendra 

 to be a bond of union between the two genera as to think 

 that the presence of the copulatory feet is a sign of affinity 

 between Glomeris and lulus. 



The possession by the Glomeridse of the branched tracheae, 

 referred to above, shows, as Mr. Bourne has pointed out, that 

 great specialization has taken place ; and great specialization 

 signifies in this case great differentiation from the ancestral 

 form, for it is very probable that the ancestor of the Chilo- 

 gnatha resembled Peripatus and the lulus-like Myriopods in 

 the possession of tufted trachese. 



Another important particular in which the Glomeridse and 

 lulidas differ is the position of the foramina repugnatoria. 

 Whether these glands be or be not homologous in the two 

 groups it is difficult to say ; but it seems that the suggestions 

 made by Prof. Moseley (Encycl. Brit.) with regard to the 

 stigmata of Scutigera are equally applicable to the apertures 

 in question. However that may be, it is, by the way, an 

 exceedingly remarkable thing that in the most highly special- 

 ized member of each of the two divisions of the Myriopoda 

 {Glomeris in the one case and Scutigera in the other) a series 

 of apertures, which in allied forms is found to be situated on 

 each side of the body, exists as a single row in the dorsal 

 middle line. Whether this single median dorsal series in 

 Glomeris represents in reality the paired lateral series in lulus 

 must for the present be left an open question. 



The straightness of the digestive tract in lulus and the 

 absence of distinct pleuree in the body -rings, though characters 

 of significance, are of less significance than the characters 

 mentioned above, and the freedom of the anal valves in 

 Glomeris is but a consequent of the incompleteness of the 

 skeleton of the posterior somite. 



Having now seen that the Diplopoda are divisible into two 

 groups, the Pselaphognatha and the Chilognatha, and that 

 the Chilognatha are in turn divisible into two groups, the 

 first to contain the Glomeridae, for which the name Onisco- 

 morpha is proposed, and the second lulus and allied genera 

 and the closely-related but in some respects aberrant Poly- 

 zonium, it remains but to consider the structure of the 



