Quoted by T. Quick, 'Disciplining Physiological Psychology: Cinematographs as Epistemic Devices, 1897-1922', Science in Context 30 (4), pp. 423-474.
Description:

'Such enthusiasm reflected a new emphasis within physiology on the study of nerve junctions themselves as sites at which nerve conduction might be interrupted. In his Textbook of Physiology of 1900, Edward A. Schäfer had suggested that impulses were 'momentarily arrested at these places of contact of nerve cells with one another', which he referred to as 'synapses' (Schäfer 1900, 608).'

 

Relevant passage from Schäfer:

 

'Every nervous path is formed of a chain of nerve cells. The chain of the simplest reflex process may be conceived as composed of only two cells, the sensory and the motor, or afferent and efferent, root-cells, and in such a chain there would be only one place of adjunction, where the central process of the sensory cell comes in contact with the dendrons or the cell body of the motor cell. On the other hand, a path which includes any of the higher nerve centres or any complex nerve processes, must have a chain of several cells, with a synapse at the place of contact between each two links in the chain. There is reason to believe that the additional delay ("lost time"), which is characteristic of the passage of nervous impulses through the nerve centres, is due to a block at each synapse; that, in fact, the nervous impulses are momentarily arrested at these places of contact of the nerve-cells with one another. And it is not improbable that the relative number of these blocks will furnish a key to the differences which are found to obtain in the reaction time for different reflexes and psychical processes. [note: 'The relative strength of the blocks in different individuals may also be the physical cause of the individual differences in reaction time for the same stimulus. As is well known, these differences gave rise amongst astronomers to the term "personal equation" being applied to the difference in noting the time of an observation which is found to obtain, with a considerable degree of constancy, between eny two observers.'] The differences of reaction time are too great to be accounted for simply by the fact that the nervous impulses are sent along paths of different length in the various cases; it is more than probable that they pass through an increasing number of nerve units (i.e. nerve cells), according to the increased complexity of the psychical process involved. And this, according to the theory of isolated units (neurone theory), would mean the passage across as many synapses, which are probably the parts of the nerve chain where relative blocks occur.' (608-609)