with Observations on the General Arrangement of the Articulata. 293 



The seventh and eighth segments, which together form the suh-hasilar seg- 

 ment (fig. 15, c) in Arthronomalus, unite at a very early period, even before any 

 appendages are developed. But a trace of the seventh subsegment exists in all 

 the true Geophili (fig. 10.), and even in Arthronomalus longicornis. Although 

 the sub-basilar segment itself remains distinct, and supports the first pair of 

 legs in the true Geophili, it becomes united to the basilar in the higher genus 

 of this family, Mecistocephalus (fig. 1/.), and forms with it one large segment 

 (6,7,8.), that bears a pair of atrophied legs posterior to the proper organs of 

 nutrition. These atrophied legs indicate the original distinctness of the sub- 

 segments in all the Geophilidue and Scolopendridce ; but these also are lost in 

 the Lithohiidce, in which the whole basilar region of the head is reduced on the 

 dorsal surface to a narrow ring (figs. 27 and 29, b, c), and the cephalic region 

 (a) is enormously developed. 



To trace the manner in which these changes in the structure of the head 

 are effected, we must return again to the young Geophilus. We have seen that 

 shortly after the animal has left the ovum, the cephalic segments are nearly 

 all of the same size, but that the fourth is soon enlarged, so as to equal in 

 extent the whole of the others ; while the first is more and more retarded in 

 its development. The enlargement of the cephalic region of the head in Chi- 

 lopoda thus takes place in a backward, as we shall presently find the basilar 

 region does in di forward direction ; the two regions thus tending to one com- 

 mon centre. This fact is proved by a comparison of the head in the different 

 families of Chilopoda. In Cryptops (fig. 20.) the cephalic region has acquired 

 a large extent, as compared with the Geophili, and this increased size is main- 

 tained in most of the Hcolopendridce (fig. 4.), until, in Lithobius (fig. 29.), we 

 find the cephalic region constituting nearly the whole head. It has been de- 

 veloped backwards, not only so as to cover the greater portion of the united 

 basilar and sub-basilar segments, but also to occasion them to become atro- 

 phied to a simple short segment (b, c). In Cermatia (fig. 36, a) this change 

 has been carried to its maximum extent. The cephalic region now forms the 

 whole upper surface of the head, and entirely covers the basilar region, which, 

 in consequence, is so completely atrophied as to leave but a trace of its exist- 

 ence on the upper surface, covered by the cephalic portion ; while the formative 

 powers have been expended in the development forwards of the remaining 



2q2 



