202 SECTIONAL ADDRESSES 



because it is simple and easily intelligible. It is indeed the explanation 

 of visual perception which any person informed of the nature of the retinal 

 image and of neural transmission of impulses would come to by his own 

 reflection. The objections to it he would be unlikely to realise by 

 everyday experience, since these depend on laboratory experiments and 

 observations of which those who do not work in laboratories have, in 

 general, no knowledge. 



The transmission theory is easily intelligible because it can without 

 difficulty be explained by a physical analogy. Photographs might be 

 transmitted telegraphically by forming an image on a plate made up of a 

 large number of small photo-electric cells each of which was connected 

 by a wire with a corresponding reproducing cell at the other end. This 

 is not, of course, the method actually used for the telegraphic transmission 

 of photographs, but it is physically a possible one. If the receiving 

 electric cells are replaced by the retinal organs, the transmitting wires by 

 the fibres of the optic nerve, and the reproducing cells by the nerve cells 

 of the visual centres of the cerebral cortex, we have a perfect analogy to 

 the physiological process of vision on the transmission theory. 



Yet this advantage of simplicity and easy intelligibility must be given 

 up if the transmission theory does not fit the facts. We have so far 

 criticised it only in connection with one experiment. Perhaps this will 

 be a convenient place to summarise the whole case against it. 



First, there is a physiological difficulty as to the mechanism of trans- 

 mission. Such a method of transmission as is suggested by the above 

 analogy would require a number of wires equal to that of the receiving 

 cells. This condition is not fulfilled by the visual system since the number 

 of retinal end-organs is two hundred times as great as the number of 

 fibres in the optic nerve. 



Secondly, a breach in the transmitting part of such a system would lead 

 to a corresponding gap in the received picture. This expectation is not 

 fulfilled in vision. We might explain away on the transmission theory 

 the fact that we do not see a gap in the part of the monocular visual field 

 corresponding to the blind spot, but Fuchs has shown that similar comple- 

 tion may take place over a blind area of the retina caused by an acquired 

 destruction of part of the optic nerve. 



Thirdly, if this theory were true it would be necessary that differences 

 in the picture at the sending and at the transmitting end should always 

 accompany one another. The experiment already discussed has given 

 one example of that not being the case, since the impression of a circular 

 shape may be given either by the circular retinal image given by a circular 

 object at right angles to the line of vision, or by a retinal image which is a 

 flattened ellipse if this is made by an object which is itself an elongated 

 ellipse viewed at a suitable angle of inclination. 



There are plenty of other examples of this in visual perception ; indeed, 

 except in those conditions of simplified perception which were character- 

 istic of the early investigation of visual ' sensations,' exact correspondence 

 between the details of the retinal image and of what is perceived is the 

 exception rather than the rule. In Rubin's reversible figures, for example, 

 we may have a pattern which is seen either as a row of black T's on a white 



