CONTRIBUTIONS TO MARINE BIONOMICS. 397 
of conspicuous serrations on these margins of the carapace is func- 
tionally related to the exigencies of respiration when these animals are 
buried in sand. 
The marginal teeth are perhaps best developed and most conspicuous 
in crabs of the family Portunidee (Swimming-crabs). As M. Alphonse 
Milne-Edwards has remarked: “Je ne connais aucun Portunien ot 
le bord latéro-antérieur de la carapace soit entier ou armé d’épines 
arrondies ou de tubercules obtus.” (1860, p. 202.) 
In Bathynectes longipes there are five sharp-pointed teeth on each 
of the antero-lateral borders. These teeth increase in size regularly 
from before backwards, and the posterior tooth is a particularly stout 
and sharp structure. This crab is almost invariably an inhabitant of 
sandy areas (¢.g., Mounts Bay in Cornwall); and the individual whose 
habits I am about to describe was also dredged upon a bottom of fine 
sand in the neighbourhood of the Eddystone. 
In an aquarium containing sand the crab burrows into the sand just 
beneath the surface, leaving its eyes and the transverse slit-like aperture 
of the buccal frame exposed. The crab is actually imbedded up to the 
anterior edge of the external maxillipeds; but it pushes away the sand 
in front of it by means of these appendages, and when at rest maintains 
these appendages in a sloping posture, so that they act as a quadrangular 
sieve-like fence in front of the buccal area. This happens both in very 
fine siliceous sand and in fine shell sand. The crab was not seen at any 
time to go completely beneath the surface, though I do not mean to 
imply by this that the crab never buries itself entirely. This may 
or may not be the case. Atelecyclus heterodon is another sand-burrowing 
crab, whose habits I have studied for a much longer period; and this 
crab has very diversified habits. It may remain partially imbedded at 
the surface of the sand, with its eyes and a broad funnel formed by the 
second antenne alone protruding, or it may disappear completely beneath 
the sand to a depth of several inches. 
When the crab (Bathynectes longipes) is partially imbedded in the sand 
as above described, it may be noticed that the chelipeds are flexed and 
approximated to the under side of the antero-lateral regions of the 
carapace in an attitude precisely similar to that assumed by Afelecyclus 
heterodon, or the Oxystome crab Matuta, under the same conditions 
(1897). The position of the cheliped is such that the marginal teeth 
of the antero-lateral region of the carapace exactly overhang the slit-like 
orifice between the distal half of the cheliped (carpopodite and propodite) 
and the pterygostomial fold of the carapace. There is thus produced on 
each side of the crab, between cheliped and carapace, a channel similar 
to that which would be produced by the approximation in parallel 
planes of two flat plates. This channel communicates below with the 
