SOME POINTS IN THE ANATOMY OF THE PLATYDESMIDA. 511 
evidence to support it, but, so far as 1 know, the condition of 
matters in Platydesmide and Craspedosoma is one that 
is only known in Polyxenus, if Bode is correct in his view (I 
think heis). So that in favour of the first we have the fact that 
Polyxenus shows a connecting-link between the almost 
vanishing tergum and the big pleure of Platydesmus on 
the one hand, and the common condition of a large tergum 
and small pleurz on the other, while I can think of no evi- 
dence in favour of the latter. 
Tt will be observed that to those who hold to the continuity 
of species evolution, the first view will appear most probable, 
while from the standpoint of discontinuous evolution the 
formation of anew structure like the narrow intercalated plate 
will not offer the same difficulty. 
There is, however, another point of view which is worth 
considering. Ina paper by Ray Lankester (“ Structure and 
Classification of the Arthropoda,” ‘Quart. Journ. Micr. 8ci.,’ 
1904), his thirteenth law or generalisation of metamerism 
asserts that homologous meromes of adjacent somites tend to 
fuse with one another, as in the case of the double somites of 
Diplopoda. In Heymon’s investigation on the embryology of 
the Scolopendride (‘ Die Entwickelungsgeschichte der Scolo- 
pendrer,’ Stuttgardt, 1901) he has made out that in the 
development of the exoskeleton of the dorsal region two 
parallel longitudinal furrows appear in the tergite, and divide 
it into three parts—a median and two lateral. Now if the 
original form of the tergum in the Diplopoda was originally, 
as in the Scolopendridz, composed of three parts it would be 
easy to understand that in most forms the three parts of the 
tergum have fused in accordance with the law just mentioned, 
while in the forms known to have the pentazonal character, the 
original three pieces have persisted. ‘This law or generalisa- 
tion seems to apply to the fusion of meromes in adjacent somites 
only, but I think that it can be extended to the fusion of bilater- 
ally symmetrical meromes in the same somite. The neuromeres 
ina single segment of Peripatus; the two neural cords are 
completely separate, and we have every degree of fusion in 
