iv GENEBAL PHYSIOLOGY OF NEKVOUS SYSTEM 261 



cellulifugal path. After isolating the cephalic ganglion that 

 innervates the second pair of antennae, Bethe frequently succeeded 

 in completely removing the part which contains the bodies of the 

 ganglion cells, so that not one of them was left connected with the 

 neuropile and the peripheral nerve processes. Twelve to twenty- 

 four hours after the operation he found that the tone of the muscles 

 of the antennae was normal ; the reflexes excited by contact were 

 carried out normally, or even exaggerated ; and on applying re- 

 peated slight stimuli which were individually insufficient to induce 

 a reflex, they summated, and eventually discharged a reflex. On 

 the second day from the operation, however, the reflex excitability 

 was diminished, the movements of the antennae became smaller and 

 slower ; finally, on the third to fourth day they ceased altogether, 

 even to the strongest stimuli, the antennae remaining drooping and 

 relaxed as if their nerve had been divided. From these results 

 Bethe concluded that the ganglion cells, i.e. the nucleated portions 

 of the neurones, are not essential to reflex phenomena ;. and that 

 muscular tone, co-ordinated reflexes, and summation of stimuli, 

 may persist even after removal of the ganglion cells, as if the 

 excitations passed directly from the neuropile to the motor nerve 

 of the antennae, as indicated by the arrow on the diagram. The 

 early disappearance of functional activity after removal of the 

 ganglion cell is due to the loss of its trophic action upon the 

 entire neurone. 



In the unipolar neurones of vertebrates, as in those of the spinal 

 ganglia, the cell body appears to be a collateral appendage to the 

 paths of physiological conduction, and there is reason to doubt 

 whether the excitations naturally pass through it, and if it is inter- 

 calated on the paths followed by the physiological impulses. This 

 hypothesis, already raised by Nansen and by Eamon y Cajal, seems 

 probable not only from Bethe's experiments, but also from those of 

 Steinach, who endeavoured to bring about the degeneration of the 

 frog's spinal ganglia by cutting off their blood-supply. Under such 

 conditions he observed that reflexes could be obtained on exciting 

 the sensory nerves as long as ten to fourteen days after the 

 operation, although under the microscope it could be seen that the 

 ganglion cells had undergone a more or less profound degeneration. 

 But Verworn rightly points out that this experiment is of no great 

 value because the exact degree to which degeneration must be 

 pushed before the cells are rendered incapable of conducting has 

 not been determined by histological examination. Steinach's experi- 

 ment does not therefore exclude the possibility that the impulses 

 normally pass through the ganglion cells. 



Greater importance must be attached to the experiments on 

 whether the afferent impulses conducted by the sensory nerves 

 are delayed in passing through the spinal ganglion or not. Exner 

 (1897) was the first to state that there was no delay. His experi- 



