61 
1916-17.] Experiments and Observations on Crustacea. 
animal, with dark posterior and lighter anterior portion, may readily be 
mistaken for one that has just undergone posterior moult; indeed, it was 
some little time before I appreciated the distinction. By testing the 
rigidity of the posterior part with the fingers the possibility of error is 
avoided. 
The change in appearance of the dorsal aspect is detectible at least 
five days before posterior moult. How long the chalky whiteness on the 
ventral aspect is present, I cannot say ; Herold gives a period of two to 
three weeks for other Oniscoidea. 
In captivity an animal about to moult apparently seeks to retire from 
the others, preferably to a damp spot. Moulting individuals were invari- 
ably to be found on the piece of wet cloth kept in the box that served for 
habitation, and this even at times when the other animals had clustered 
together in a different region. Dampness is certainly essential. On one 
occasion a Ligia that had just moulted its posterior part was placed in a 
small dry tin in a warm room. When the tin was opened an hour later 
the animal was apparently dead, all reaction to stimulation having ceased. 
It was now replaced in the tin on a strip of cloth wetted with tap-water. 
Half an hour afterwards it responded feebly to stimulation. When 
uncovered seven hours later it ran about actively. Moulting may occur 
equally well in sea-water and in an air-medium. 
In posterior moulting (which, as well as anterior, I have observed on 
many occasions) a complete transverse separation takes place between the 
fourth and fifth (free) thoracic segments, while a longitudinal split occurs 
dorsally and laterally on each side of the fifth, sixth, and seventh segments ; 
no longitudinal or other splitting occurs in any of the abdominal segments. 
The animal, remaining almost stationary, loosens itself from the cuticle by 
a series of writhing worm-like movements ; finally, it walks away, leaving 
the old integument with dirty, black gill-lamellae behind. 
A period of four days having elapsed (temperature from 10 5° to 14'5° C.), 
the creature proceeds to moult the anterior portion. Two longitudinal 
cracks, this time involving the second, third, and fourth thoracic segments 
(no such split occurs in the first segment), appear dorsally and laterally. 
The limbs are early withdrawn, and the animal by active movements, 
usually in the forward direction but also slightly backwards (a rare move- 
ment in Ligia), gradually frees itself from the remaining covering. In 
this process, owing to the softness of the underlying cuticle, the head 
remains characteristically depressed, while the passively trailing integu- 
ment of the haif-moulted antennas may be inadvertently stamped upon by 
the already free and rapidly hardening front limbs. As the antennary 
