■28 ROBERT M. YERKES AND JOHN B. WATSON 



The matter deserves as serious attention as does the method 

 of presenting the stimuli which are to be discriminated. 



(3) That preHminary, problem-defining investigations of vision 

 be carried on with rough and ready methods, and that by means 

 of the " brightness " apparatus systematic and detailed studies 

 be made of the visual capacities of one, or a few, individuals 

 of know^n life-history. Thus our results may be rendered inten- 

 sive, accurate, and valuable for comparison with the data on 

 human vision. 



(4) That the visual stimuli be presented in an apparatus 

 whose construction is in essentials like the one described. 



(5) That light perception be investigated by the presenta- 

 tion of tw^o visual areas (circular in form and either 5 or 6 cm. 

 in diameter) whose photic values are accurately controlled and 

 measured by the experimenter. 



(6) That size perception be investigated by the presentation, 

 similarly, of two visual areas w^hose only constant difference is 

 in size. The stimulus adapter might carry, for example, a 5 

 cm. circle in the middle window% and a 6 cm. circle in each of 

 the end windows. To the animal there could be presented 

 either 6 cm. circle on right and 5 cm. circle on left, or the reverse, 

 as the experimenter desired. Discrimination could be made to 

 depend wholly upon ability to react to the size difference of 

 the areas. 



(7) That form perception be investigated by the presentation 

 of two visual areas whose only constant difference is in form. 

 The form plates have been so chosen that the two forms may 

 be either equal or unequal in area. For example, a square 

 and a circle both measuring 28.2743 sq. cm. may be used; or 

 with the same circle may be used a scjuare which may be inscribed 

 in the circle. 



IV. METHODS OF INVESTIGATING COLOR VISION 



I. Sources of stimulus 



The statements concerning the advantages and disadvan- 

 tages of sources of light made on pages 4 to 10 apply, with 

 certain qualifications, in this connection. In experiments on 

 color vision we may use either a light capable of yielding, in 

 fair intensity, any of the hues of the solar spectrum, or a light 



