259 
1914-15.] The Keflexes of Autotomy in Decapods. 
resistance is necessary in most forms, but in Portunus and Hyas the leg 
may snap off at the breaking-plane without having a point d’appui. 
In Hyas and InacJms plucking by the chelipeds may be the direct 
cause of autotomy when it is delayed, but the co-operation of the 
autotomising muscles at the limb-base is necessary. 
Discussion. 
The descriptive accounts, in the preceding pages, of autotomy in various 
decapod forms show that there is much ground for fruitful comparative 
study, both from morphological and physiological view-points, in this process 
of self-amputation of limbs. A brief summary of the leading points is 
given below. 
It has been recorded by Przibram (9) that if the leg of an amphipod 
{e.g. Gammarus) be damaged or cut, the animal at once proceeds to bite 
the stump down to the level of the first segment. This closely resembles 
the behaviour of the spider {Tarantula), which also bites off a damaged 
leg down to the coxa at the base (Wagner (14)). In the decapod 
Crustacea this autophagy does not exist as a normal reaction to injury, 
but it takes place under certain conditions. 
The common prawn, when seized violently by a leg, extends the basal 
segment, and by a violent tail contraction tears the limb off at the free 
joint between the second and third segments. 
In the lobster and crayfish the same reaction takes place, but the 
rupture takes place at the level of a groove in the proximal part of the 
third segment. Previous to the tail contraction taking place, however, a 
flexor muscle of the third segment weakens the limb at the level of the 
groove by pulling inwards one part of the ring of calcareous integument 
of the third segment central to the groove. There is a definite time- 
relation between the various elements of the reflex, and autotomy can 
occur when the nerve cord to the brain is cut. The reflex is plurisegmental, 
but the part of it vrhich causes weakening of the limb at the breaking- 
plane is confined to one segment of the nervous system. If the limb be 
cut cleanly off and immediately freed, movements resembling autophagy in 
the lower forms result. The limb may be cut off by the chelate walking- 
legs, and in the case of the crayfish by the chelae themselves. 
In the hermit crab, the normal process of autotomy is the result of a 
unisegmental reflex ; but if the crab be removed from its shell, plucking 
with the chelae is necessary. Thus the change of conditions has necessi- 
tated the reinforcement of the unisegmental reflex by arcs of higher levels, 
i.e. the reaction is plurisegmental. If the extensor muscle to the second 
