STUDIES ON POLYCHET LARVE. 615 
further suggestions. Claparéde’s division, Metachetz, is 
found to coincide almost exactly with the sub-order Spioni- 
formia, exclusive of the very highly modified Chetopteride, 
which form a distinct group to themselves, and differ from all 
other known Polychet larve in being mesotrochal. But the 
respective characteristics, 1f there be any, of the larve 
belonging to the remaining Polychet orders are still unknown. 
Within the orders Spioniformia and Nereidiformia, to which 
my own observations have up to the present been almost 
entirely confined, certain characters can, however, be selected 
which appear to be diagnostic of the larve belonging to 
individual families, and the contrast between larve of the 
Nereidiformia and Spioniformia may be found in other struc- 
tures besides the sete. 
Before going on to discuss these characters it will be well 
to consider some interesting points bearing on the phylo- 
genetic values to be assigned to the different larval structures 
that appear to be most useful for systematic purposes. 
Hacker (1896, pp. 93-100) has done good service in showing 
the effect of environment upon the form of the larva. He 
points out that larvee which pass through the early part of 
their development in a gelatinous capsule tend to pass through 
a protrochophore stage ; and that larve which pass into the 
later stages of development thus protected omit the true 
trochophore stage from their life-history, the broad band of 
short cilia characteristic of the protrochophore being retained 
by the metatrochophore; such worms may, moreover, com- 
mence their lives on the sea-bottom at an early period with- 
out ever having developed a true prototroch. Hacker (loc. 
cit., p. 98) further finds that larve developed in a brood- 
pouch appear at the surface of the sea with a fully developed 
powerful prototroch. In connection with this Malaquin’s 
work on Syllid larve may be noted. He finds (1893, pp. 424, 
425) that hastened development causes the suppression of the 
anterior paratrochs, and in extreme cases of all the larval 
ciliated bands, including even the prototroch. It is also 
noted by Hacker (loc. cit., pp. 97, 98) that the stage at which 
