12 INDIVIDUAL DIFFERENCES AND FAMILY EESEMBLANCES 



probable error calculated for each point in the curve. In accord- 

 ance with a plan proposed by Professor Cattell, the limits of the 

 probable error are shown by the broken lines. The chances are even 

 that with a greatly increased number of cases the time would have 

 remained between these limits, and a nearly smooth curve can be 

 drawn within them. When the gates were changed at the eighteenth 



TABLE II 

 COMPLETE EECOEDS FOE THE WHITE FAMILY IN THE MAZE TEST 



trial, an interference effect occurred that resulted in a rise of the 

 time curve to 118 seconds at the first trial. This was just half the 

 number of seconds it took the average mouse to go through the maze 

 for the first time. The rest of the interference test showed an aver- 

 age time curve that was above the curve for the last few trials of 

 the initial learning test, except for the sixth day of the interference 

 test when the curve dropped to 40 seconds. The curve based on the 

 average for the interference test began to follow the usual course of 

 learning until the sixth day when the maximum speed record was 

 reached, but at that point, for some unknown reason, a retarding 

 factor occurred that caused a decided rise in the curve from then to 

 the end of the test. There are two possible explanations for this 

 phenomenon; first, that at the lowest point of the curve the mice 

 had reached their maximum speed and efficiency and thereafter they 



