PRIMARY INHIBITION OF CILIARY MOVEMENT 433 
with a specially differentiated cylindrical ‘sensory epithelium’ 
(Chun, p. 159; Hertwig, p. 22). 
Additional evidence in favor of the existence of a nervous 
inhibitory mechanism in Beroé can be obtained from a number 
of the experiments carried out with poisons. I found that the 
best means for narcotising Beroé was chloral hydrate. <A solution 
isotonic with the sea-water is prepared. So much of this is added 
to one quantity of sea-water that the proportion of chloral 
hydrate is 0.1 per cent and to another quantity of sea-water so 
much that the proportion of chloral hydrate is 0.2 per cent. In 
both cases the mixture is well stirred. The animal is first trans- 
ferred to the 0.1 per cent solution and, in the case of small speci- 
mens of animals as in my experiments, after about twenty min- 
utes to the 0.2 per cent solution. With this method of procedure, 
after quite slight symptoms of excitation, which affect the mus- 
culature to a remarkably slight extent, the chloral hydrate gradu- 
ally extinguishes both the neuromuscular excitability and the 
ciliary activity in Beroé. 
Experiment 4 shows that there exists a stage of incomplete 
chloral-hydrate narcosis, when the inhibitory effect on touching 
the edges of the mouth has disappeared, but the ciliary activity 
in the plates still continues in the aboral part of the rows. In 
the animal in this stage an oro-central current of density 2—2.93 
m.amp./em.?, which in an intact animal produces total inhibition, 
causes an acceleration of the movements of the swimming plates. 
In experiment 3 it can also be observed that with the progress 
of the chloral-hydrate poisoning there is an increase in the thresh- 
old of the current density with which inhibition is reached. If 
it were a fact that the inhibitory effect of the current directly influ- 
ences the stimulus-conducting connection between the swimming 
plates by a direct reduction of their ability of conduction, one 
would expect that, in an animal which is to a certain extent 
poisoned by chloral hydrate and where the conduction of the 
stimulus is already weakened because of the poisoning, a weaker 
current would be sufficient to produce total inhibition than under 
normal circumstances. When one finds that, on the contrary, a 
stronger current is necessary to attain total inhibition, one must 
