124 



of the machine. The deviators thus represent causes the 

 effect of which is proportional to the distance from a 

 flxed Hne; just as the causes of wealth may be said to be 

 roughly proportional to the absolute dimension of this wealth. 



Àfter passing through several rows, therefore, the grains 

 of sand must be distributed in a skew curve. This curve 

 is rendered apparent as soon as ail the sand is collected 

 in the compartments which hâve been constructed below 

 the last row. 



In order to get workable dimensions for the deviators 

 and the channels, the absolute dimensions hâve to be 

 somewhat considérable. The total height of the outside 

 of the frame is 104 centimeter and, even so, a few of 

 the very narrowest channels had to be slightly widened 

 to let the sand freely pass (this of course has no influence 

 on the resulting curve). 



Practical détails» The deviators were made very exactly 

 of equal thickness. They were glued to the glass which 

 forms the front of the machine. The bottom is of tracing- 

 cloth resting on a sheet of flannel, which is firmly pressed 

 against it by a deal board. In this way the escape of 

 the sand between the deviators and the bottom and front 

 of the frame is pretty well avoided. 



The beam on which the sand rests after having passed 

 through the machine, can be taken off, in order to remove 

 the sand for later experiments. 



13. Questions raised by skew frequency curvcs. 

 We hâve reached the resuit that normal causes are 

 generated under the influence of causes which act with 

 equal force on small and big individuals, whereas skew 

 curves are generated under the action of causes which 

 produce déviations depending on the size of the individual. 



We are naturally led to the questions: 



Given the observed skew frequency curve of the 

 quantifies x, 



