ii.] "BLAZE-CURRENT' 25 



cornea (positive), is ultimately directed from cornea to fundus 

 (negative). We shall get a better general view of this gradual 

 change by plotting a curve of readings taken at regular intervals 

 (Fig. 9). I think you will agree with me that the entire curve, 

 falling with diminishing rapidity towards and beyond the zero 

 value, is badly named when it is called a current of rest, and 

 that it cannot possibly be only a subsiding current of injury at 

 the transverse section of the optic nerve. I think and speak of 

 it as a declining manipulation blaze caused by the unavoidable 

 mechanical disturbance that occurs in the most careful possible 

 preparation of the eyeball, for the effect can be at once 

 reproduced by slightly compressing the eyeball, and the current 

 aroused by such intentional manipulation declines just like the 

 current unavoidably aroused in the preparation of the eyeball. 

 So much for the normal or accidental current, which is de- 

 creasingly positive and increasingly negative, at least during 

 the first hour or two of observation. 



14. The principal fact. We may now proceed to the 

 principal part of the experiment as first made by Holmgren, 

 and subsequently repeated by Dewar and MacKendrick, Kuhne 

 and Steiner, Fuchs, and many others. 



The eyeball current is steady, the galvanometer spot is 

 sinking with imperceptible slowness to your left, and I expose 

 the eyeball to a brief flash of light by means of an ordinary 

 photographic shutter timed at about -^ second. The galvano- 

 metric spot makes a small excursion to your right (positive), 

 indicative of a current through the eyeball from fundus to 

 cornea in response to the flash of light. The direction and 

 character of this response, which are perfectly normal, are 

 nearly sufficient to assure me that we have under our observation 

 a normal and well-prepared eyeball, but I shall not be quite 

 sure of this until I have seen how the galvanometer spot 



faces east, so that the N terminal of the galvanometer is to the left and the 

 S terminal to the right. As far as possible the connections are arranged so 

 that currents in a "positive" or "outgoing" or "ascending" direction shall 

 give deflections to the right, " negative " or "ingoing" or "descending" 

 currents deflections to the left. 



