comparisons; in addition they assisted in maintaining standard tests,, 

 Generally, however, reliance was placed on the observer's judge- 

 nneni of a response as gained from visual observation^ At the close 

 of each experiment, the strength of a response was recorded in one of 

 several categories (or a fraction thereof I which are designated as 

 follows; 



Often a reaction would be indicated by X(X^ or XX{X), 

 with numerical equivalents of 1-1/2 and 2-1/2 respectively. 



Although admittedly subjective and not well-suited to 

 statistical analysis, the foregoing measure is simpler and at times 

 more nneaningful than that dependent on the tinning and counting 

 alone. The two observers, by working together, attempted to 

 standardize their judgennent of the strength of a reactionc The ob- 

 served reactions, measured in the above categories^ are included 

 m table 2 for the series of experiments analyzed in the preceding 

 paragraphs. They may be compared with the summed time and count 

 differences in the second to last column of the table. There is gen- 

 eral agreement between the two measures in that the stronger ob- 

 served reactions (XX and XXX) are associated with the greater 

 differences between control and test quantitative d«ita. However, 

 there are also notable discrepancies as^ for exannpie, in compeiring 



23 



