604. Fr. H. GRAVELY. 
summer evenings. The young specimen was, it is true, 
taken in a weighted tow-net sunk on a long line in water 
whose depth was probably about 10 fathoms, but the presence 
in the net of the earlier stages only of the Pluteus of He hino- 
cardium cordatum shows that the material amongst which 
this specimen was found was brought up from a level some 
distance above the bottom; also, as our object was simply 
the collection of material and not a comparison between 
surface and deep-water plankton, no precaution was taken to 
prevent specimens from getting into the net on its way 
through the surface waters, so this little worm may in reality 
be a surface form. 
Many Syllids, including those of the genus Odontosyllis, 
develop special natatory sete and take to a pelagic existence 
at the time of sexual maturity. In these cases each tuft of 
natatory sete springs from a broad muscular dorsal lobe of 
the parapodium which is supported by several stronger sete, 
or by a definite aciculum, and the parapodia bearing them 
usually show a more or less definite relation to the segments 
in which the sexual cells are produced, occurring only either 
upon or (as in the male Autolytus—*“ Poly bostrichus”’) 
behind them. It has been pointed out by Malaquin (1891, 
also 1893, pp. 427-450) that when special natatory sete 
appear at the time of sexual maturity on appendages of those 
Syllidee which normally bear a ventral cirrus, then these 
appendages show the complete normal structure of a Poly- 
cheet parapodium ; and he has further pointed out that the 
essential parts are always developed in the following order— 
ventral ramus, dorsal cirrus, ventral cirrus, dorsal ramus, 
and that when any of these are suppressed it is always in the 
reverse order. ‘The only exception to this rule that he finds 
is that forms devoid of ventral cirrus before maturity may 
develop the dorsal ramus without it. 
It would appear from his account, too, that in no known 
case is the dorsal ramus developed until the time of sexual 
maturity. It seems, therefore, that the primitive Syllid 
stock was characterised by the loss of the dorsal ramus. 
