CANADA PORCUPINE 



45 



and so on until he reached the end and then look around the 

 end. He would then begin with the end one which he had just 

 examined and take them in regular succession until he found 

 the circle, provided he did not turn and make a third and even 

 a fourth exploration of those forms and compartments which 

 he had just examined twice.. As will be seen in the first hori- 

 zontal column of results (A) in table VI, in no series of 20 ex- 

 periments each, did he make more than eight correct choices. 

 On the 1 8th series he made only four correct choices and on 

 the 19th and 20th series five each with an average of 5. 7 ±1.5 

 correct choices for each of the 20 series. 



TABLE VI 

 Form Study. Porcupine No. 3 



A = Six-wise. 



B = Triangle and circle with base down and apex down in alternate series. 



C = Circle and triangle with base down. 



D = Circle and circle-triangle with curve down. 



E = Circle and 'circle-triangle with apex down. 



F = Circle and circle-triangle with curve down. 



These results were regarded as negative and the problem made 

 simpler but still the circle was used to mark the food recep- 

 tacle. The circle was presented pair-wise with the triangle as 

 the latter had stood lowest in the former experiments. By- 

 using the triangle, apex down, in one series of 20 experiments 

 and the base down in the next series an attempt was made to 

 determine whether he was observing the triangle as a whole or 

 only some part or aspect of it. The second line of results (B, 

 table VI) shows a steady increase of ability up to the ninth series 

 where there is rapid disintegration of the association when the 

 base of the triangle was down. In the following 20 series (C, 



