166 SECTIONAL ADDRESSES 
a poor copy of the nervous mechanism, for it reacts jerkily and is often 
damaged in the process, but the ground-plan of the mechanism is the 
same. 
Thus in the activities concerned in the rapid conduction and in the 
setting up of rhythmic trains of impulses, it does not appear that the 
cells of the nervous system have properties not shared in some degree 
by other tissues. 
So far we have only considered what happens in nerve fibres. We 
can tap the messages which pass along the wires between the front line 
and headquarters, but this does not tell us how they are elaborated there. 
A great deal has been found out already by the analysis of reflexes—i.e. by 
sending in a known combination of signals and finding what signals 
come out to the muscles; indeed, the great part of Sherrington’s work on 
the spinal reflexes and Pawlow’s on the brain has been carried out in 
this way. An account of the central nervous system which does not 
include a full discussion of such important work is like the tragedy of 
Hamlet without the Prince of Denmark; but the results are so well known 
that I shall deal instead with a recent line of attack of an entirely different 
kind. This relies on the fact that nervous activity, in the central grey 
matter as in the peripheral nerves, is accompanied by electric changes. 
They seem to be a reliable index of the underlying activity, and by 
recording them we come a step nearer to the main problem. The chief 
difficulty is to interpret the records. In the cerebral cortex, for instance, 
very large electric oscillations are constantly occurring, except in the 
deepest anzsthesia, but they vary from moment to moment and from 
place to place, and it is only in the visual cortex that they are under a 
fair degree of experimental control. Here they can be produced by 
shining a light in the eye (Fischer and Kornmiiller) or stimulating the 
optic nerve (Bartley and Bishop), and the prospects of analysis are more 
hopeful. But at the moment the most significant feature of these records 
from the brain lies in the appearance of the waves. Whenever a group 
of nerve cells is in action, in the cerebral cortex, the brain stem or the 
retina, and whether the nerve cells in question belong to a vertebrate, 
or an insect, the waves are alike in general form. Instead of the abrupt 
spikes which appear in a record from a nerve fibre when a train of 
impulses passes down it, we have more gradual potential changes which 
form a series of waves of smooth contour. In the simpler structures 
where most of the neurones are acting in unison the waves may have a 
regular rhythm (5 to go or more a second), which rises and falls when 
the stimulus changes in intensity. It is often possible to make out 
both the abrupt nerve fibre impulses and the slower nerve cell waves, 
and to show that they occur together. In the cerebral cortex of an 
anzesthetised animal there is much more variety and less orderly repeti- 
tion ; the waves usually occur at irregular intervals ; they vary in size and 
duration, and some of them may last for half a second or even longer. 
Nerve cell waves may be the wrong name, for they are probably due 
to the branching dendrites and not to the cell body of the neurone ; but 
there can be no doubt that they represent a characteristic activity of the 
structures which make up the grey matter. They show that the same 
