Eye Movement. 81 



the first left movement of the others is about 40° (usually a little less) . 

 The records in the figure are typical of the availability of this first left 

 movement throughout the series. The first movement to the right 

 which succeeded this, while often definitely less than the 40° amplitude 

 and without correction, was also in 58 per cent of the cases accompanied 

 by a wink, 1 which produces a break in the horizontal line of time dots. 



Since the irregularity beyond the first movement to the left is so 

 prominent, it appears that the only feasible use that can be made of the 

 records, aside from this one left movement, is to measure the total 

 amount of horizontal movement during the 5 seconds and employ this 

 as a general measurement of the speed of eye movements. Large un- 

 corrected movements will of course provide the maximum horizontal 

 distance for a given time interval, as short movements require propor- 

 tionately more time for their performance. 2 



Table 14 contains the data for the two points considered: (1) the speed 

 of the first movement to the left (L) for each record, 3 and (2) the total 

 horizontal distance (H. D.) covered by the eye during the period of 5 

 seconds. The average duration of the left movement on normal days 

 following the control dose is 142 a (see next to the last column, section 

 i of table 14). The durations found in the first period for both groups 

 of days are alike, that is, 140 <r. These durations are somewhat long, as 

 will be seen if comparison is made with the table of normal data col- 

 lected by Dodge and Benedict. (See their report, p. 163, table 26.) 

 The average normal left-movement duration given by them for normal 

 subjects is 101 and 109 a for the first and second normal days, respec- 

 tively. Records were not obtained with all of the subjects for two nor- 

 mal days, however. With these averages there are also errors of 2.7° and 

 2.5°, respectively, which mean that the original long movements did not 

 fully cover 40°, but were nearer 37°. This would naturally shorten 

 somewhat their duration, but even so the results in our series on Subject 

 VI are considerably longer than those previously obtained upon him. 

 Dodge and Benedict give in table 26 the data for the duration of the left 

 movement on Subject VI as an average of 92 a without correction. 4 



The horizontal distance which accompanies the average duration of 

 142 o- in our series of normal measurements is 271 mm., and for the aver- 

 age of first periods of alcohol and normal days we have 293 and 310 mm., 



1 Diefendorf and Dodge (Psychol. Bull., 1916, 13, p. 459) found a tendency to coordinate wink- 

 ing just at the time of saccadic eye movement. 



2 See Dodge and Benedict's report, p, 151, table 24. 



3 There are three instances in which the first movement to the left was complicated by a wink; 

 these were omitted and the next movements left, which covered 40°, were taken in their places. 



4 The difference between 92 and 142 a for the first and second series of experiments is large, and 

 since the eye movements are a good measure of neural condition, this difference must be scruti- 

 nized. The records for October 22, 1913, the one normal day on which this figure 92 a is based, 

 were mostly illegible. However, from them it can be clearly seen that the amplitude of move- 

 ment has been reduced to about 33°. These shorter movements naturally require less time (see 

 Dodge and Benedict's table 24), which would account, at least in large part, for our difference in 

 normal values. 



