404 AXNALS NEW YORK ACADEMY OF SCIENCES 



3). The function tested was a highly variable one, but it varied irre- 

 spective of the time spent upon the test. 



4). 'No one condition more than another affected detrimentally the 

 accuracy of judgment of handwriting or composition. 



Miss Mulhall said: Experiments were reported concerning, (1) the 

 influence of determination to remember, and (2) the effect of primacy 

 and recency on both recall and recognition. The results showed: (1) 

 Determined recall (recall of material for which there was a determination 

 to remember) differs from undetermined recall more than determined 

 recognition differs from imdetermined recognition. (2) The difference 

 between determined recall and determined recognition is less than that 

 between undetermined recall and undetermined recognition. (3) The 

 factor of determination influences the amount of material remembered 

 which can be associated with other material remembered. (4) Primacy 

 and recency Ijoth influence recall memory. The influence of each on 

 recognition is less than on recall, biit is greater for material devoid of 

 associations and less for material with associations. A detailed account 

 of the experiments will appear in the forthcoming issue of the "American 

 Journal of Psychology." 



Mr. Martin said: The object of this investigation is to discover the 

 transfer effects of practice in canceling a-t words upon certain other can- 

 cellation tests. The subjects were divided into two groups : thirty-six in 

 the practice group and forty in the control group. Both groups were 

 tested with a series of seven cancellation tests, after which the practice 

 group was trained in canceling words containing a and t in English prose. 

 The practice periods were ten minutes long and there were four of them 

 each day for sixteen days. Precautions were taken to impel the practice 

 gi'oup to a maximum of improvement, and also, in the meantime, to keep 

 the remaining forty that constituted the control group interested in their 

 l^art of the experiment. In the training series the practice group im- 

 proved in accurac}'' from 83 per cent, to 96 per cent., and from an initial 

 average ^performance of 10.2 cancellations per minute to a final average 

 performance of 26.6 cancellations per minute. After the conclusion of 

 the practice both groups were reassembled and tested with the same seven 

 tests used before the practice began. 



The results justify the following conclusions : 



1). In the group of tests in which the elements determining the can- 

 celing were the same or partially the same as the elements determining 

 the canceling in the practice series the transfer effects appear as facili- 

 tation. 



