Cole and Long, Visual Discrnniuation iii Raccoons. 679 



equal in briglitness for the raccoon. We now have to ask whether 

 there is any evidence to justify this assumption. 



After training our animals to select Gray 5 from the colored papers 

 of Group 5 we substituted for the latter Grays 3, 4, 6, 7 and 8 of 

 the Hering series. Eaccoon No. 2 was given only sixty trials with 

 these papers. The results appear in Table 23. 



TABLE 23. 

 Gray 5 in Grays 3-8 inclusive. 



It appears from this table that Raccoon No. 2 confused Gray 4 

 with Gray 5 in the first thirty trials and showed, in the second thirty, 

 that he was learning to discriminate Gray 5. This animal was not 

 hungry, as he had just been given ninety trials on Grays 5, 10, 15, 

 20, 25, and 30, wdiich series presented no difficulty of discrimina- 

 tion. Table 23 shows that Raccoon No. 3 made very few mistakes. 



It seems evident, when these records are compared with those of 

 the color-discrimination tables, that the brightness differences between 

 the colors of any group were less for the raccoon than the brightness 

 differences between any two of these consecutively numbered gray 

 papers. This agrees with human vision, for Professor Titch- 

 ener wrote us as follows : ''We tried to arrange the four equivalent 

 disks in a scale or order of apparent brightness, by the eye alone ; 

 and we got into gTeat difficulties at once. I do not think that any 

 two of us would have taken the same arrangement except by chance. 

 We all felt pretty sure that VT2 was the lightest of the five, though 

 it is true that suggestion (from experiments I have mentioned) may 

 have played a part here." 



