236 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 



key to the differences which are found to obtain in the reaction time 

 for different reflexes and psychical processes." 



In the following pages I have attempted to make use of the fragmen- 

 tary data which exist, in estimating the time of an optical reflex con- 

 ducted through the usual paths of the spinal cord. I then attempt a 

 comparison of this time with the time necessary to conduct an optic 

 reflex over Reissner's fibre. In doing this I lay myself open to criticism 

 for introducing estimates which are subject to so large a percentage of 

 error. But the method I believe promises much, and a beginning must 



be made somewhere. 



a. Central Delay. 



"Wundt ('76) found that the reflex time in the frog, when stimulated 

 at the lower part of the spinal cord, was 0.008 second less tliau when the 

 sensory cells were stimulated ; that is, 0.008 second was the time occupied 

 by the passage of the impulse through the central nervous system. In 

 the case of a crossed reflex this was increased 0.004 second, that is, to 

 0.012 second. Now if this crossed reflex was produced by the interposi- 

 tion of a single neuron, then the delay due to one synapse would be 

 0.004 second, but if by two neurons, it would be only 0.002 second. 

 Assuming the first conjecture to be correct, the reflex in its transmission 

 through the brain and cord must pass through a chain of (0.012 -^ 0.004) 

 three neurons. 



Exner ('74) found that the reduced reflex time of the closure of the 

 eyelid on direct excitation was 0.047 second, hence this more complex 

 reaction must involve a chain of many more neurons than in the preced- 

 ing case. He also found that the ' lost time ' between excitation of the 

 frog's cerebrum and the response of the leg muscles is 0.05 second. In 

 the cerebrum and optic lobes about 0.025 second is lost, between the 

 bulb and the lumbar enlargement only about 0,015 second, and in the 

 sciatic nerve and muscle plate still less. Exner concludes, as the result 

 of many experiments, that the time of block in a synapse of the brain 

 maybe 0.008 second. This time is accepted by Schafer (:00) as a 

 reliable working value. Gotch (:00, p. 483) says "the central delay 

 for apparent ending in the spinal cord is at least 0.006 second and may 

 be considerably more;" and he had previously ('87, p. 510) shown that 

 this delay must be regarded as due to retarded transmission in 

 the arborizations. 



LangendorfF und Krawzoff ('79), working with the frog by electric 

 stimulation and the graphic method, found that stimulation of the cere- 

 brum resulted in a contraction of the leg muscles 0.036 second later 



