xii ELECTROMOTIVE ACTION IN THE EYE 473 



Klihne and Steiner (I.e.) at first experimented in a dark room 

 divided into three parts. The two farthest contained the 

 galvanometer and telescope, while the eye was prepared in the 

 third room by a lamp. In the later experiments the galvano- 

 meter was placed in a light room, while the electrodes and 

 illuminating arrangements were in an adjacent and absolutely 

 dark chamber. The stimulus was made by an Argand gas- 

 burner at a distance of 50-75 cm. from the preparation. The 

 lamp was turned up and down by an assistant at a signal, so 

 that the retina was suddenly illuminated or darkened. The 

 lead-off from the inner and outer surfaces of the retina was 

 effected by specially constructed clay electrodes, covered (Engel- 

 mann's method) with frog's lung. Each adequate and sudden 

 illumination with blue, green, yellow, red, or white light then 

 produced a considerable complex varia- 

 tion of the retinal current, with or 

 without the presence of visual purple. 

 The typical effect (Fig. 282) in a retina 

 containing the purple is a positive 

 variation at the moment of illumina- 

 tion (b c] rising rapidly to its maxi- 

 mum, and then passing quickly into the 

 negative variation. This phase reaches 

 its maximum (cl e) during the impact 

 of light, is delayed some time at this 



point, and then declines very gradually to zero, even during con- 

 stant illumination. At the moment of darkness there is again a 

 sudden positive effect (e f), which must be regarded as the result 

 of a second stimulus due to the disappearance of light. The 

 mode of excitation is therefore in a measure comparable with 

 that from an electrical stimulus. As in the latter, the impact 

 and duration of the current on the one hand, and its disappear- 

 ance on the other, act as a stimulus, so with the impact of light 

 upon the retina, where the effects are visible on the galvanometer 

 as an initial diphasic (positive then negative) and a second 

 simple (positive) variation of the rest-current. The presence or 

 absence of visual purple appears from Klihne and Steiner to 

 be of essential importance to the intensity of the retinal " current 

 of action." Not merely does the magnitude of the variations 

 differ in the two cases, being greater in the unbleached retina 



