688 
Transactions of the Society. 
being forced into position through the upper portion. Fig. 5 is a 
diagram showing a side view of the same organ, with other grains 
travelling up inside to he attached in due course. 
Any long, thin, cylindrical particles, such as spines of Echini , 
spicules of sponges, &c., are utilized for fringe-making, and are de- 
posited separately. The filaments huilt under my observation consisted 
of single grains of sand cemented one above the other in a straight or 
curved line, but those found on some tubes are delicate columns con- 
sisting, in cross section, of two or three grains placed side by side 
and cemented together. It is probable that in these cases a column 
is first huilt of single grains and afterwards strengthened or thickened 
by additional material. 
The fringe-building operations appear to be chiefly carried on 
during the night, and many hours and even days may, and frequently 
do, elapse between each favourable opportunity for observation. 
M. A. de Quatrefages,-* writing in 1865, and Messrs. J. T. Cun- 
ningham and Gr. A. Ramagef in 1887, describe the filaments of the 
fringe as hollow “ tubules,” and say that “ when the head of the worm 
is protruded the tentacles are partly contained in them, and so pro- 
tected.” This is certainly not the case in any specimen which I have 
seen, and I have examined a very large number. The filaments are 
solid columns, along which the tentacles do frequently stretch them- 
selves, but which it is impossible for them to penetrate, for on ex- 
amining the tube internally no opening to these “ tubules ” is visible. 
Moreover, it seems to me that were the hitherto prevailing idea 
correct, these “ tubules ” would, so far from protecting the animal, 
prove rather a source of danger by impeding its hasty retreat. The 
object of these filaments it is impossible to divine with certainty, hut 
three or four advantages occur to me as possibly connected with them. 
(1) They may act as snares to catch food ; I have several times seen the 
worm pass its upper lip and tentacles along the filaments. (2) They 
may form favourable vantage points from which the animal can fish 
around with its wonderful tentacles, supplying, as it were, rods to the 
fishing lines. (3) These filaments or fringes are set upon two semi- 
circular flaps, facing each other, and falling together when the water 
leaves them ; so they may be a protection by thus closing the tube. 
Or (4) it may be a case of protective mimicry, the resemblance to the 
surrounding growth of coralline being very close. 
I have found that the tubes can be preserved as museum specimens 
by treatment with sea- water saturated with calcium chloride, and sub- 
sequent drying, during which the fringe should be carefully lifted and 
picked out with a needle. I have one such specimen with a spray of 
coralline attached (just as taken from the pool), in which it is almost 
impossible to distinguish the one from the other (fig. 7). 
* Quatrefages, A. de, ‘ Hist. Nat. d. Anneles Marins et d’Eau douce,’ ii. (1865) 
p. 346. 
t Trans. Eoy. Soc. Edinb., xxxiii. (1888) p. 664. 
