436 LANDES, [VOrE. 1x. 
Two weeks after the operation, the anterior and posterior 
division moved in unlike rhythm, and at times cramp or inter- 
rupted movements appeared for short intervals in the posterior 
division. 
(ec) The cord was cut both anterior to the first and third 
ganglia (I and III, Fig. 1). The gill-plates are thus divided 
into two divisions, both of which are separated from the brain. 
The first and third ganglia were affected by the operation, and 
the corresponding gill-plates move only passively. Often the 
first two plates have cramp movements, while the others con- 
tinue at their usual rate. Observations made during the two 
weeks following the operation showed that the first division 
moved in a rhythm differing from that of the other, and was 
not affected by cramp or interrupted respiratory movements of 
the latter. Moreover, the posterior was not influenced by the 
behavior of the anterior division. 
Experiments (@) and (e) demonstrate, that when the gill- 
plates are divided into two divisions, one or both of which are 
separated from the brain, each division has its own peculiar 
rhythm, and is not influenced by the behavior of the other. 
We can conclude, therefore, that each has its own rhythmical 
centres. 
(f) The second ganglion was extirpated (II, Fig. 1). By its 
removal, the second gill-plate lost its sensibility and motility. 
It is moved by the action of the water and neighboring plates. 
All the other plates continue their activity. If we hold the 
second plate away from the others, we see that each division 
continues to move in its own peculiar rhythm and extent. 
During the two months that observations were made, no 
marked change appeared in the respiratory movements. At 
times the posterior division was either motionless or in cramp 
movements, while the anterior behaved normally. At other 
times the latter had cramp movements, while the former did 
not, and usually the two divisions moved in different rhythms. 
When the second ganglion is cut through transversely the 
results are practically the same as those above. The same 
results were obtained with other Limuli. This experiment also 
showed that destruction of the ganglion at once arrests the 
