6 
204 (June 16, 
Packard.] 
embryo Geophilus correspond to the antenne, two pairs of mouth-parts 
and three pairs of legs of the larval Julus. 
The phenomenon of two pairs of limbs to a segment, so unique in Tra- 
cheata, may be explained by reference to the Phyllopoda among the 
Branchiata. The parallel is quite exact. The larvee in both groups have 
but a single pair of appendages to a segment; the acquisition of a second 
pair in the diplopods is clearly enough a secondary character, and perhaps 
necessary in locomotion in a cylindrical body with no sterna.* 
The larval Julusand the ancestral Chilognaths were hexapod Tracheata, 
but sufficiently different to indicate plainly that the Myriopods branched 
off from a much more primitive form than the Scolopendrella-like hexapod 
ancestor, and which form somewhat agrees with our hypothetical lepti- 
form ancestor of all Tracheata, 
The Myriopods also differ from Hexapoda in that the genital armature 
of the male (the females have nothing corresponding to the ovipositor of 
Hexapoda) is not homologous with that of true insects ; moreover, the 
armature is not homologous with the limbs or jointed appendages of the 
myriopodous body. On the contrary, the apparatus of hooks arises from 
the sternum of the sixth segment, between, but a little in advance of the 
origin of the eighth pair of legs. It should be observed that the legs in 
Myriopods are outgrowths between the tergites and sternites, there being 
no pleurites differentiated, and in this important point also, the myriopods 
are quite unlike the Hexapodous Tracheates. 
Affinity and systematic position of the Pauropoda. The nearest living 
forms which approaches the larval Diplopod are Pauropus and Eury- 
pauropus. These organisms are practically primitive diplopods. Looking 
at the lowest Chilognath, Polyvenus, and comparing Pauropus with it, it 
will be seen that the latter scarcely differs from it ordinally. Pauropus 
has a head with a pair of antenne and two pair of mouth-appendages, 
The antenne are quite unlike any other myriopods, being 5-jointed and 
bifurcate, somewhat as in certain Coleopterous larve ; the peculiar sense 
filaments may be the homologues of the flattened sense-sete at the end of 
the antenne of Diplopod Myriopods. 
The ‘‘mandibles’’ are rudimentary, very simple, and are scarcely more 
like Chilopod than diplopod protomale ; there is a second pair of append- 
ages which, as Lubbock states, are ‘‘minute and conical ;’’ they bear a 
closer resemblance in position and general appearance to the ‘‘under lip ”’ 
of Chilognaths, especially the under lip of Siphonophora ; in fact, the 
*Tt is plain that, as Balfour suggests, Comparative Embryology p. 824, the 
double segments have not originated from a fusion of two primitively distinet 
segments. There is, however, a misconception as to the nature of the double 
segments.” They are not so in fact. The scutes are single, undivided, but the 
ventral region is alone imperfectly double, bearing two pairs of append- 
ages, just as single segments of Apodidse may bear from 2-6 appendages; the 
differentiation is confined to the ventral limb-bearing region and limbs alone; 
the dorsal part of the segment does not share in the process, 
