become finer, then indistinct, and finally leare tbe delicate flair-like tip smooth. The
shafts of all are striated longitudinally, and are also crossed at intervals by transverse bars,
which, however, do not affect the outline. In the first tuft of bristles, the two kinds are
more nearly of equal length, and in the last tuft the tips of the simple forms are more
gently tapered as well as often fractured, whilst the great length and tenuity of the tips
of the second type cause them almost to equal the length of the stronger. In transverse
section these bristles are rounded (not circular).
The hooks (Plate CXXIV, fig. 3 a) have a short horizontal shaft, a rounded .crown,
six teeth along the front edge in lateral view, then a broader part, which at first sight
looks like a seventh tooth, but which really is a series of more minute teeth, as in the
typical Pectimi-ia belgwa; finally, the keel below shows a convexity, a hollow, and a small
knob at the edge.
The caudal process recalls the condition in the Opheliidae, jnst as the head, buccal
region and the first body-region do those of the Hermellidse. Two segments without
bristles follow the last bristle-tufts, and then a constriction, the anal process sharply
curving ventrally thereafter. In outline it is HfysosfomitTO-shaped, having a convex,
obliquely striated ventral surface, and a concave, transversely striated dorsal surface like
a sucker, surrounded by a rim, which is notched and papillose, whilst it terminates distally
in a differentiated flap ventral to the anus, and another freely moveable flap of the same
length dorsally. At the origin of the caudal process three to four hooks (Plate OXXIV,
fig. 3 d) occur on each side of the median dorsal groove. They have short, stout, striated
shafts and acutely curved tips, a few transverse strias also being present here and there
on the shaft, especially at the base. One or two developing forms accompany the former.
In a small variety from Norway (dredged by Canon Norman) the dorsal flap has a distinct
papillose margin, a condition also seen in those from Naples, - -
The anal funnel is, when the animal is removed from the tube, carried at an angle,
usually greater than a right augle, to the caudal region, is rounded ventrally, flattened
dorsally, and with a spatulate valve hinged dorsally at the tip. The dorsal edges of the
process are somewhat scalloped at the base, one deep fissure being present, and each
edge has four small clavate papillaa. The dorsal surface of the organ is often expanded
into a wide sucker with an obliquely ridged centre and a free erenated edge. The
apparatus would seem to act as a powerful ejector, and the external parts also, according
to Arnold Watson, as an instrument in repairing the tube when broken. He saw both
the scapha and its bract at work, “ the latter at times apparently pressing down the edge
of the material which, mingled with mucus, forms a dome which closes in and diminishes
the wide opening to a pin-hole ” (in lit.). He thought that the internal structure of the
scapha suggested a respiratory function.
Young examples apparently of this form occur frequently in the bottom nets at the
end of June and in July in St. Andrews Bay. ' They occupy little transparent tubes
about 1 mm. in length, nearly straight and tapered posteriorly, both ends being open.
This areolated tube is composed solely of secretion and mimics the adult’s tube of sand-
grams; the free edge being formed of somewhat larger spaces, the double edge under
pressure being seen to the right. The posterior end of the tube presents a clear, transparent;,
margin, then a granular belt, which is followed by somewhat smaller reticulations than in
front. The tube is further chequered by a series of larger reticulations which cause it to
resemble crocodile leather.
Mr. Watson mentions (in lit) that Mr. Chadwick at the Port Erin Laboratory
observed the metamlorphosis of the larval form into the post-larval, viz., the settling down
of the former on the bottom of a vessel ; it rotated rapidly for an hour and a half and then
#as found to have secreted the membranous tube.
The perseverance and ingenuity of Mr. Arnold Watson have lately enabled him to
observe the building of the sand-tube by a post-larval form in July. Thus one which had
cast its membranous tube endeavoured to manipulate minute sand-grains toward what
should have been- the edge of the tube, and he subsequently was enabled to watch the
early post-larval form, which has only two buccal tentacles, by means of which minute
sand-particles (tooo to 3^0 inch in size) are collected and passed to the mouth which
opèns at their bases. What is required for food is swallowed, but those grains suitable
for building purposes are rolled over in the mouth and then deposited on the edge of the
tube (Plate CXIII, fig. 6). The annelid then advances slightly in its tube, and for four
or five seconds applies to that sand-grain the secretion of the cement-gland. The post-
larval membranous tube is thus the structure to which the first sand-grains are attached,
so that further use is made of this protection.
Lagis Koreni is one of the most skilful artificers in the family of the Amphictenidæ,
fashioning a tube like a slightly curved horn, composed of minute pebbles or large grains
of Sand, carefully selected and dexterously fixed to each other by the usual secretion.
In placing the grains together in the tube there is no haphazard, but angle fits angle, as
'in a skilfully built wall, and no excess of cement hides slovenly masonry. The tube is
perfectly round and dilates from the narrow end to the wider, which is often the lower
end in the sand, for the animal carries it thus, and its symmetry and workmanship are
faultless ; yet the architect is devoid of eyes, and depends on its exquisité Sense of touch
and remarkable instinct. The tubes harmonise with their surroundings, the wall being
formed of similar proportions of pale sand-grains, of yellow, brown and black, probably
because in such a selection the average colours are fairly represented. The tube is more
or less transparent in life, so that the position of the occupaut can be determined from
the outside, the posterior end, however, being opaque from an accumulation of soft sand,
with a screw-like tunnel leading to the posterior end. Rarely à coarse tube is öbserved
at St. Andrews, the minute pebbles projecting from the surface like thoSè of PectinaHd
granulata, and to one a gelatinous capsule containing mucus was attached externally.
Much, therefore, depends on the nature of thé ground inhabited by the specimen.
Habits.—When weak the animal drops from its tube. A healthy example removed
from its tube and placed in sea-water constantly extends and contracts its body, the
tail moving more freely than the anterior énd. It usually lives in the sand with the
wide end of the tube inferior, as first pointed out to me by the late Mr. David Robertson,
öf Cumbraé (1889), who found the specimens bétwéen tide-marks, wherèas at St. Andrews
they are only tossed on shore by storms.
The great numbers of this species found in the stomachs of tbe cöd and the haddock
off St. Andrews Bay show that in all probability they are swept in thousands from their
sites in the sand, and thus come in (he way of fishes which feed on the bottom. Both