54 ROBERT M. YERKES AND CHESTER E. KELLOGG 



reversed and the apparatus, as a whole, is quite as adaptable as 

 is Professor Watson's. 



General discussion of methods. The above devices for obtaining 

 graphic records of maze-reactions yield less satisfactory results 

 than would a good photographic device, and we recommend 

 them simply because they are less expensive in construction and 

 operation. All are so simple that detailed description is needless. 

 We shall, however, in concluding this article, call attention to 

 certain important points which experience in the use of the 

 graphic method has brought to our attention. 



In the first place, although it is perfectly possible to get along 

 with a simple lens, especially if one is working with small 

 mazes, a much more satisfactory image may be obtained by the 

 use of a compound lens. Second-hand portrait lenses are 

 available and wholly suitable, but even such a lens is likely to be 

 much more expensive than a simple lens. 



Each of the four devices which we have described has its 

 obvious advantages and disadvantages, and it is clear that 

 choice of a method should depend upon the relative importance of 

 these in each particular case. On the whole, it is likely to be 

 more convenient for most experimenters to have their drawing 

 board slightly inclined toward them. This is possible in methods 

 III and IV. Method II necessitates the use of an inclined 

 drawing board, but unfortunately the observer must sit at one 

 end of this board and work in a somewhat awkward position. 

 So far as the position of the drawing surface is concerned, methods 

 III and IV would seem slightly more satisfactory than methods 

 I and II. 



In those devices in which the light falls upon the record sheet 

 from above, the shadows cast by the experimenter's hand and 

 pencil are disturbing, sometimes rendering it difficult to follow 

 accurately a swiftly moving animal. Other things being equal, 

 it is therefore preferable to have the light come from beneath 

 the drawing surface, as in methods I and IV. 



In the first few trials with a given animal, it is extremely 

 important for the experimenter to be able to change record 

 sheets quickly, since the animal is likely to traverse the alleys 

 of the maze rapidly and repeatedly. If the image comes from 

 above the record sheet, it is possible to have the sheets made 

 up in the form of a tablet or block with two edges glued. The 



