436 R. T. POCOOK. 



major or two minor somites coming together. And snrely 

 we should expect the alteration to be effected, as indeed it 

 is, by the juxtaposition of two major instead of two minor 

 metameres if, as has been suggested, all the somites belonged 

 originally to the major category. 



That the change sets in in Lithobius close to the middle 

 of the body, and in Scolopendra in the anterior third, may 

 perhaps be explained by some such consideration as the 

 following. The juxtaposition of two major somites is no 

 doubt unfavourable to free flexibility ; and since it is advan- 

 tageous for the greatest possible freedom of movement to lie 

 in the anterior and posterior ends of the body, which bear 

 the principal external organs set apart for the performance 

 of the vegetative and generative functions, the reason becomes 

 clear for the juxtaposition of two major somites to take place 

 as near as can be to the middle of the body in Lithobius, 

 where the body is relatively short, few-jointed, and less 

 flexible. In Scolopendra, on the other hand, with its longer, 

 more jointed, and therefore more flexible body, it is probably 

 immaterial where the juxtaposition of the major somites 

 exactly occurs, provided it is sufficiently distant from both ends 

 not to interfere with the full flexibility of the head and tail. 



There appears, then, to be no insuperable or even serious 

 difficulties in the way of believing that the similarity in 

 segmentation between the Lithobiidse and typical members 

 of the Scolopendridae is assignable to homoplasy. If, on the 

 other hand, it is attributable to the direct homology of the 

 somites concerned, the view that Crater ostigm us exhibits 

 a stage of segmentation intermediate between those of the 

 other two must be wrong. In that case no genetic signifi- 

 cance can be attached to the presence of twenty-one tergal 

 plates in Craterostigmus and in Scolopendra. The 

 numei'ical likeness must be regarded as purel}' coincidental. 

 This view of the. matter I find it impossible to adopt when 

 the other many and deep-seated structural details in which 

 the two forms approximate to one another are taken into 

 consideration. 



