68 ELECTRO-PHYSIOLOGY CHAP. 



With excitation peripheral to the ganglion 0'123 sec. (average of 148 experiments) 

 ,, central 0'087 sec. ( 97 ) 



Difference . . 0'036 



V. Uexkiill's recent experiments on the function of the stellate 

 ganglion in Eledone moschata (16) bear on the same question. A 

 large number of nerves (stellar nerves) radiate laterally from 

 this ganglion, and supply the muscles of the skin and mantle. 

 " On exciting the stellar nerves the near muscles first come 

 into action, and then the more distant, in ratio with the con- 

 ductivity of the nerves. With excitation above the ganglion 

 the contraction of the near muscles is delayed, as is expressed in 

 the gentler rise of the curve. The apex is, however, steeper, 

 thus showing that the total effect on all the muscles is com- 

 pressed into a shorter time. The ganglion stellatum would thus 

 appear to correct the slower conductivity, since it enables the 

 muscles of the mantle to perform more synchronic and therefore 

 more effective movements." 



If the difference in reflex time, on stimulating at different points, 

 thus affords a gauge of the influence exerted by the ganglion- 

 cells interpolated along the nerve-fibre, upon the time-relations 

 of the excitatory process, the same is no less evident from the 

 character of the reflex period itself. A reflex movement (i.e. a 

 motor impulse in the muscular apparatus in consequence of a 

 centripetal stimulus) can only occur when the centripetal path, 

 which is first traversed by the stimulus, is connected with the 

 efferent path by means of the central nervous system. The 

 solitary ganglia "of invertebrates, the spinal cord and bulb in 

 vertebrates, are more especially the seat of these nervous pro- 

 cesses. In 1855 Helmholtz first pointed out that the time 

 between the impact of a stimulus and the corresponding reflex 

 movement of a striated muscle was 1012 times longer than the 

 time required to conduct an impulse in a peripheral nerve of the 

 same length. This assumes the rate of conductivity to be approxi- 

 mately equal in motor and in sensory nerves, as does in the 

 above experiments appear to be the case. 



The duration of an entire reflex process may be summed up 

 in three factors: (1) the time occupied by conduction in a 

 centripetal nerve, from the point of stimulation to the central 

 end; (2) the time which elapses between the arrival of the 

 excitation at the centre, and its transmission to the central end 



