65 
1916-17.] Experiments and Observations on Crustacea. 
rim ” which “ runs round the under side of the anterior portion of the 
cephalic shield.” In the moulting decapod transverse splitting occurs 
between thorax and abdomen. In the amphipod, as in the isopod, the 
transverse splitting is intrathoracic, not however between the fourth and 
fifth (free) segments, but between the first and second true segments, 
i.e. between the apparent head and thorax. 
The functional separation of the isopod into two locomotor areas at 
the time of moult temporarily unmasks for us a primitive distinction be- 
tween two groups of walking limbs. When discussing function of the 
limbs I shall return to this subject. In the meantime my only aim is to 
establish a case for the study of moulting in its relation to systems of 
classification. 
Physiological. 
In the case of amphipods and decapods moulting takes place in one 
single act. A Gammarus or a lobster, however, withdraws the anterior, 
not the posterior, part of its body first. In these animals the gills have 
a thoracic and not as in isopods an abdominal attachment. One might 
in • consequence be tempted to suggest that the order of withdrawal of 
the body is determined by the position of the gills — an arrangement quite 
comprehensible from the physiological standpoint. The fact that the 
Brachyura (also with thoracic gills) are said to withdraw the abdomen 
first, may or may not upset this supposition — cf. also Limulus. 
One or two interesting facts observed during experiment upon Ligia 
fall to be recorded under the heading of Physiology. 
As we shall find later, the walking limbs of Ligia react in a very 
specific manner to sensory stimulation. It is remarkable that, immediately 
after either posterior or anterior moult, the freshly emerged limbs do not 
respond to stimulation of any part of the body. Whatever be the point 
at which the interruption occurs, the normal reflex mechanism is for the 
time being out of gear. One can scarcely refer the interruption to softness 
of the cuticle, for on such a supposition the muscles are discounted. The 
condition lasts only for some minutes. 
Another curious fact observed in the case of the animal mentioned 
on page 61 as having been restored to life after virtual death by drying 
is that the earliest motor reaction then elicited was one of the 'posterior 
and not of the anterior limbs, which failed to react. A hypothesis formed 
at the moment was that all the available internal fluid, essential for 
function of the reflex mechanism, was temporarily monopolised by the 
hinder half of the body. In a half-moult, such as occurs in isopods, we 
VOL. xxxvii. 5 
