640 PROFESSOR E. RAY LANKESTER, 
some forms of structural features which were the distin- 
guishing characteristics of their immediate ancestry, we find 
that frequently genealogical groups do not admit of strict 
definition in terms of structure. And, further, we find that, 
even in order to arrive at a clear notion with regard to the 
relationships of a limited portion of a large group—such a 
portion as are Arachnida in regard to the Arthropoda—it is 
necessary to consider the genealogy of the whole series 
included in the larger group. 
The Arthropoda form a very large branch of a great 
phylum to which I have applied the name ‘ Appendiculata’— 
celomate animals with more or less distinct metameric seg- 
mentation of the body and possessed of lateral lobes or 
processes of the body itself which serve primarily as loco- 
motor organs. Besides the Arthropoda the phylum Appen- 
diculata includes the Rotifera and the Chetopoda. Each 
of these three great branches of the Appendiculata has its 
special developments, but it seems to be probable that they 
all started from a common ancestry which had characters 
intermediate to those of such a Rotifer as Pedalion and of 
such a Chetopod as Syllis. Probably the Arthropoda were 
developed from an ancestry resembling the Chetopoda, but 
devoid of the chetz carried by the appendages of the 
latter. 
The distinguishing motive of the development of the 
Arthropod branch of the Appendiculata is the adaptation of 
one or more pairs of the appendages proper to the segments 
succeeding the mouth, to the purposes of the prehension and 
mastication of food. Hence it would be well to substitute 
the term Gnathopoda for Arthropoda. All Arthropoda are 
not arthropodous, that is to say, do not exhibit a jointing of 
the exo-skeleton of the appendages. Peripatus though truly 
a Gnathopod is not an Arthropod or Condylopod. The dis- 
appearance of such jointing in connection with a softening 
of the integument and a scavenger mode of life amongst 
rotten wood, is one of those changes which it is probable 
might occur as an adaptation, and accordingly it is very 
doubtful whether we should regard the non-arthropodous 
condition of Peripatus as a retention by it of the soft-bodied 
character proper to the Chetopod-like ancestry of the 
Arthropoda. 
The structure of its eye, the presence of two lateral nerve- 
cords in place of a double ventral cord, the limitation of the 
jaw-feet to a single pair, the existence of paired nephridia 
in each segment of the body, the peculiar histological struc, 
ture of the muscular tissue,seem to me to be conclusive 
