76 L. ^Y. SACKETT 



confused his rats and that restoration to the original position 

 again confused them. Rotating 90 degrees had a similar effect 

 but shifting the table and maze so as to change direction of 

 lights and air currents had no apparent influence. In explana- 

 tion of these phenomena Watson postulates "static sensations, 

 or some non-human modality of sensation." Miss Vincent (44, 

 page 89) in the same laboratory observed that tilting the maze 

 two and one-half inches did not disturb the reaction of the 

 rats to any great degree but this has never been satisfactorily 

 controlled and tested. Small (37) has also pointed out that 

 his rats seemed to have a strong "sense of position." The Peck- 

 hams (31) resorted to the same explanation in interpreting the 

 behavior of wasps ; but Bethe was probably the first to mention 

 this fact in connection with his observation on the efi:ect of 

 rotating the beehive. The present study so far is defective in 

 that there are two factors entering: (i) The sense of direction, 

 if there be such a thing either organically or through environ- 

 mental stimuli; (2) the kinaesthetic difference arising from the 

 change in the slope of the paths when the maze was rotated. 

 Either one of these factors would, no doubt, be sufficient to 

 confuse the porcupine and efforts to separate the two have so 

 far been abortive. 



Dark Room Experiments. — Another variation was made with 

 the maze in the barn. Porcupines Nos. 3 and 6, both of which 

 knew it perfectly, were used in some dark-room experiments. 

 Windows were blinded carefully and most places where light 

 could enter were closed. In addition, operations were post- 

 poned until long after dark so that the room was absolutely 

 dark to the human eye, even after two hours' adaptation. A 

 person who knows the maze well, however, finds no difficulty 

 in keeping trace of the animal's movements even in total dark- 

 ness. No. 3 was started in the maze in the usual manner. He 

 made four abortive starts in getting past b. iVfter passing that 

 point he went wrong in j but only for a moment and had a 

 similar experience in 2 immediately after. After that he righted 

 himself and went without error through the remainder of the 

 maze entering the free end near 5 according to his habit with- 

 out difficulty. His time was 8 minutes 47 seconds, chiefly 

 consumed in the start (G plate IV). His second trip was iden- 

 tical with his first except that he had less difficulty in starting 



