322 H. M. JOHNSON 



brass plates in which windows of the desired size and form have 

 been accurately cut and beveled. The windows may conform 

 to any plane figure. Circles, squares, triangles and hexagons are 

 most commonly used. 



Tests of brightness-sensitivity are made with forms geo- 

 metrically equal but differently illuminated. The apparatus is 

 adequate for this purpose within the limitation of the problem 

 itself. This limitation will be remarked below (page 327). 



In testing size-discrimination two circles of different diameter 

 may be employed. The circle used as a standard is usually 

 6 cm. in diameter, the diameter of the comparison-circle varying 

 between 3 cm. and 9 cm. For form-discrimination the stimulus 

 windows may be a circle 5 cm. or 6 cm. in diameter, used as the 

 standard; the comparison-form being the circle's equivalent, 

 inscriptible or circumscriptible square, triangle or hexagon. As 

 applied to the study of size- and form-discrimination, certain 

 limitations of the method should be noted. 



In the study of size-discrimination, two constant stimulus- 

 characteristics other than size must be controlled. These are the 

 luminous intensity of the respective forms, and their respective 

 brightness. Since these two terms are somewhat variously used 

 in physiological and psychological literature, and since in this 

 paper they have a restricted meaning, I beg the reader's indul- 

 gence while the present meaning of the terms is being specified. 



In the accepted photometric nomenclature, if the light coming 

 from a source produces on a surface the same illumination as 

 that produced by a standard source at the same distance, both 

 sources are said to have the same luminous intensity in the 

 directions respectively given. The standard measure of luminous 

 intensity in America is the candle. The brightness of a source is 

 taken as the luminous intensity divided by the area of the source ; 

 it is usually measured in terms of candles per square meter or 

 derivatives thereof. 



The opal glass screens limited by the windows therefore serve 

 (1) as sources, each having a certain candle-power; and (2) as 

 visual objects, each having a certain brightness and area. Both 

 screens being equally illuminated and the transmission and diffu- 

 sion coefficients of the one being respectively equal to those of 

 the other, the brightness of the larger and the smaller surfaces 

 would be equal: but their luminous intensity, or candle-power, 



