15-1 University of California Publications in Zoology. [Vol. 4 



shortening- the time required, or by adopting better methods 

 in later tests. Glaser found that there was no improvement: 

 the later tests require as much time, and involve as many varied 

 movements as the first ones. In the same way Glaser finds that 

 there is no improvement in many trials in the righting; reaction 

 of the brittle star. After righting itself many times, the animal 

 requires as much time and employs as many movements as at 

 first. Glaser therefore concludes that the brittle-star shows in 

 these experiments no indication of intelligence. 



In these experiments of Glaser the animal is of course left to 

 itself while "solving the problem", the object being to discover 

 whether it will of itself under natural conditions "improve" in 

 the performance of a complex act in consequence of repetition; 

 the results are negative. I carried on a considerable series of 

 experiments of the same sort, my results confirming throughout 

 those of Glaser. I first tried Preyer's experiment of fastening 

 the starfish by means of broad-headed tacks placed in the angles 

 of the rays (see p. 150). I found, as Preyer had done, that the 

 animal under these conditions performs many movements of 

 most varied character, and that by one or the other of these it 

 always finally escapes. But in a long series I was unable to 

 confirm Preyer's statement that there is improvement in a given 

 individual after it has been repeatedly subjected to the experi- 

 ment. The length of time required to escape does not decrease, 

 nor is there a decrease in the number and variety of the move- 

 ments before escape is effected. The starfish does not discover 

 that one method accomplishes the result more readily than an- 

 other, and thereafter use that method. 



In the same way I kept precise records in a long series of 

 experiments on the righting reaction. After repeated experiences 

 by a given individual, there was no improvement in the per- 

 formance of this reaction, either in the time taken, or in the 

 number of movements employed in accomplishing the righting. 



Thus my results so far confirm throughout those of Glaser. 

 When left to themselves, these animals do not, as a rule, improve, 

 nor form habits as a result of the repetition of a complex act. 



As Glaser well points out, this negative result may be due, 

 not to any actual incapacity for habit-formation, but to what 



