400 NATURAL SCIENCE {December 
external divisions must be considered as secondary rings and not as 
true segments. 
The correctness of this suggestion seems to me to be entirely 
borne out by the abdominal segmentation of Protocaris in which 
this secondary ringing has not taken place. If a body-ring were 
marked round the abdomen of Apus for each of the pairs of limbs, 
leaving out the most- minute at the posterior end (cf. figs. 1 and 6), 
we should get a condition, at least for the limb-bearing portion of 
the abdomen, not unlike that shown in Protocaris, in which a multi- 
tude of very small segments (though not diminishing quite so clearly 
as they would in Apus) are in striking contrast with the anal seg- 
ment. 
It should, however, be noted by the way that this interpretation 
of the segmentation of Apus which receives such unexpected and 
welcome support from Protocaris was greatly complicated by the 
presence in Apus of a varying number of limbless segments in front 
of the anal segment. These still require explanation; at present I 
am inclined to look upon them as secondary reduplications of the 
anal segment. 
This difficulty must not, however, be thought to stand in the 
way of my interpretation of the segmentation of Apus. That inter- 
pretation has already received abundance of support from the fact 
that the same fixation of rudimentary segments is found in the 
Trilobites. Inasmuch, however, as the relationship between Apus 
and the Trilobites is still a matter of discussion, it is especially 
welcome to obtain direct evidence from a fossil whose close affinity 
with Apus cannot be for a moment doubted. 
The second point arises from the peculiar shape of the shield. 
Mr Schuchert describes it as subquadrangular, and quotes Clarke’s 
suggestion that it has probably been subjected to some horizontal 
distortion in the shale, The longer I contemplate the figure of this 
shield the more convinced I am that it has simply been flattened 
out, and that in its original shape it was folded down at the sides 
of the body. Not only do the two anterior lateral projections of 
the shield suggest this, but the absence of the usual spikes at the 
postero-lateral corners of the shield are quite in accord. In Apus 
these spikes are turned up somewhat on to the back (fig. 6), and in 
the Trilobites they are spread out wide of the body in the horizontal 
plane. If the shield were folded down at the sides, these spikes would 
be a serious danger to the limbs and abdomen, and would be sooner 
or later dispensed with. Whether, again, the dotted lines running 
along the shield, shown in Mr Schuchert’s figure (reproduced in fig. 
1), lend any support to this view, I should not like to say, because 
we have no means of getting at their true meaning, but they cer- 
tainly suggest to my mind a dorso-ventral flattening. 
