136 DE. L. A. BORRADAILE ON THE 
function in this case is obvions. But it also happens at intervals when the 
water is quite clear and pure. It probably then drives out particles which 
have gradually accumulated under the gills while the current was flowing in 
the forward direction and which would in course of time prove harmful. 
The reversal of the current^ which was first studied by Garstang (8, 9) in 
Corystes and other genera, has been examined in Carcinus by Bohn (2, 4), 
who has shown that it always takes place, but varies in frequency and 
duration with the foubiess of and amount of matter in suspension in the 
water, and that when the water is chemically foul the crab will raise the 
normally exhalent opening to the surface so that it draws in either the better 
oxygenated water of the surface or air, which passes out in bubbles at the 
opening of Milne-Edwards after oxygenating the water in the chambei'. 
These results I can confirm from my own observation. The crabs will live in 
exceeding!}' foul water if they be permitted to raise the front of the body 
out of it, but may be asphyxiated by preventing this. Bohn (4) thinks that 
the reversal of the scaphognathite has an advantage in resting its muscles. 
I am not clear that it is necessary to suppose that it has any other function 
than that of cleaning the chamber and enabling the animal to obtain a 
better supply of oxygen. 
3. In feeding, the crab severs the food into morsels which it swallows 
without finely dividing them. It always seizes the food with the chelse. 
The third maxillipeds then part, and the food is placed by the chelse between 
the mandibles. If the mass of it be very bulk}', one or more of the legs 
may be brought into action to assist in lifting it and thrusting it towards the 
mouth. The mandibles part to receive the food, and then close upon it. 
They do not cut it by a slicing action or chew it, but unless it be soft enough 
for them at once to sever a morsel of it by pressure, they hold it firm wdiile it 
is being divided by the action of other organs. If the food be very soft, 
they may even sometimes be seen to be held wide apart while it is thrust into 
the mouth by the action of their palps and of the nose of the labrura. 
Usually, however, the food (I have fed my crabs upon various parts of the 
body of fishes and upon meat) requires the assistance of other orgaijs for its 
severance. In this case it is torn by being pulled outwards from the 
mandibles, much as a crust may beiorn hj being held with the teeth while 
the hand wrenches a jDart of it away. The outward pull is sometimes given 
by the chela?, especially if the mass of food be tough, and they often keep a 
hold upon it when they do not appear actually to be pulling ; but most often 
the work is done by the third maxillipeds, which grasp the food by pressing 
upon it with the toothed inner edges of their ischiopods and at the same time 
press downwards upon it with their palps, somewhat as the digit of a sub- 
chela is closed. These limbs do not cut or chew the food, and rarelj' pass 
it towards the mouth, but bj' pulling it outwards tear from it the portion 
which is held by the mandibles. The second maxillipeds work in various 
ways, and their action, like that of the organs in front of them, is often hard 
