47 



(4) Cutting of the nerve to the limb at its point of emergen'^e 

 from the central cord prevented it. 



The process was further investigated as follows : — The tendon 

 of the extensor muscle of the second segment was cut, and its end 

 attached to a heart lever writing on an ordinary physiological 

 drum, the lobster being secured on its back on a frog-board. A 

 record of the reflex contraction, when the cut end of the limb was 

 electrically stimulated, . shows a latent period of -13 sees, and 

 presents the appearance shown in fig. 9. 



The tail, or properly speaking the abdomen, which flapped 

 violently after each stimulation was also connected up to a lever, 

 and a record of its contraction made along with the contraction 

 of the extensor of the second segment. A time relation of about 

 four seconds exists between the two events (fig. 10). 



Owing to the inaccessibility of the tendon of the autotomiser, 

 I was unable to record the time of its contraction and its latent 

 period, but I believe that it contracts after the extensor of the 

 second segment and before the tail-flapping begins. 



The following conclusions can be drawn from the above study 

 of the process of autotomy of walking legs in the lobster. 



The break occurs at a definite plane in the limb base. It 

 passes partly through brittle calcareous material at the region 

 of the groove in segment three, but also partly involves the tough 

 arthrodial membrane between segments two and three. A 

 nocuous stimulus appHed to the limb causes extension of the second 

 segment, and at the same time, contraction of the autotomiser 

 tends to draw the portion of segment thr.e central to the groove, 

 inwards. This causes weakening in the plane of the groove, and 

 if the traction set up by tail flapping cannot be resisted, division 

 of the limb occurs (figs. 6, 7 and 8). 



The. records of muscle contraction in the basal parts of the 

 limb and in the tail establish the fact that the reflex is due to 

 definite contractions, co-ordinated in time and degree. The reflex 

 is pluri-segmental, for autotomy does not occur when the nerve 

 connection to the tail is severed. If the influence of upper neurons 

 be removed the reflexes are more brisk. This demonstrates that 

 even in an animal with so little cerebral development as the lobster 

 the inhibitory action of the upper ganglia is manifest. 



