160 PHYSIOLOGY OF NERVK [CH. XJV. 



to be shorter; the difference between the time intervals in the two 

 experiments will obviously measure the time occupied by the impulse 

 in traversing a stretch of nerve equal to the distance between his toe 

 and his knee. 



Another method, largely employed by Bernstein, is to take the 

 electrical change as the indication of the impulse. A stimulus is 

 applied to one end of a long nerve, and the change in the electrical 

 condition of the nerve is recorded by a galvanometer connected to 

 the other end of the nerve. The time between the application of 

 the stimulus and the galvanometric reply is measured. 



The velocity of the nerve impulse has by such experiments been 

 found to vary with temperature, and to be approximately the same 

 in both motor and sensory nerves. In cold-blooded animals it is 

 thus slower than in warm-blooded animals. In the frog, for instance, 

 at ordinary room temperature it averages 27 metres per second. In 

 man, at normal body temperature it is 120 metres per second. In the 

 case of non-medullated fibres the velocity is much slower; these 

 observations have been chiefly made on invertebrate animals ; in the 

 non-medullated nerves of the lobster it is 6, and in the octopus only 

 2 metres per second, and values lower than these have been recorded 

 in other cases. 



Direction of a Nerve Impulse. 



Nerve impulses are conducted normally in only one direction : in 

 efferent nerves from, in afferent nerves to, the nerve-centres. But 

 there are some experiments which point to the conduction occurring 

 under certain circumstances in both directions. 



Thus, in the galvanometer experiment just described, if the nerve 

 is stimulated in the middle instead of at one 

 end, the electrical change (the evidence of an 

 impulse) is found to be conducted towards both 

 ends of the nerve. 



Kuhne's gracilis experiment proves the same 

 point. The gracilis muscle of the frog (fig. 

 163) is in two portions, with a tendinous in- 

 tersection, and supplied by nerve-fibres that 

 branch into two bundles; excitation strictly 

 limited to one of these bundles, after division 

 of the tendinous intersection, causes both por- 

 tions of the muscle to contract. 



Another striking experiment of the same 

 " IG>1 'lfte?wa] 1 Je S r ) ffrog - kind can be performed with the nerve that 

 supplies the electrical organ of Malapterurus. 

 This nerve consists of a single axis cylinder and its branches ; stimu- 

 lation of its posterior free end causes the " discharge " of the electrical 



