CROSS-EXAMINING A CRAB 87 
nected with the third and fourth pairs of legs, but that 
these had anything to do with the branchial cavity was 
not suspected. ‘hese two pairs are pressed more closely 
together than the rest. The opposed surfaces of their 
basal joints, that is, the binder surface in the third, and 
the front surface in the fourth pair, are flat and smooth, 
and their margins are clos:ly fringed with long, sheeny, 
peculiarly formed hairs. Milne-Edwards, who compares 
them to articular surfaces, as their appearance warrants, 
thinks that they serve to diminish the friction between 
the two legs. On this supposition the question arises why 
precisely in these crabs and only between these two pairs 
of legs such a provision for diminishing friction is neces- 
sary, not to mention that it leaves unexplained the singular 
hairs, which must augment instead of diminishing friction. 
While, then, I was bending to and fro in ever so many 
directions the legs of a large Sand-crab, in order to see 
in what movements of the animal friction occurred at the 
place in question, and whether perhaps these were move- 
ments often recurring and of special importance to it, L 
observed, when I had stretched the legs far apart, a round 
opening of considerable size between their bases, through 
which air could easily be blown into the branchial cavity 
or even a slender probe be introduced. The aperture 
opens into the branchial cavity behind a conical tubercle, 
which stands above the third foot at the place of a bran- 
chia which is wanting in Ocypoda. It is laterally bounded 
by ridges which rise above the articulation of the legs and 
to which the lower edge of the carapace is applied. Also 
outwardly it is overarched by these ridges with the ex- 
ception of a narrow slit. Over this slit extends the cara- 
pace, which just at this point projects further downwards 
than elsewhere, and so a complete tube is formed. While 
Grapsus always admits water to its branchiz only from in 
front, in Ocypoda I saw it also streaming in through the 
just described aperture.’ 
For its details about one particular crustacean such a 
passage is interesting, but it is far more important asa 
lesson in scientific observation. There are numbers of 
8 
