Manchester Memoirs, Vol. xlviii. ( 1 904), No. 34. 7 



(5). A Device for explaining MendeVs theory. 



I think this theory may be made clearer by a device 

 which has been useful to me : some have imagined that it 

 is intended merely as an instance of the application of the 

 theory of probability : this, however, is not the case ; its 

 value, such as it is, lies in the way the process is managed, 

 which has nothing to do with probability. 



All that is needed is some red and white counters. 



Mendel's conception of the gonad of a hybrid as an 

 organ containing germ cells, 50% of which bear the 

 dominant character, and 50% the recessive, can be easily 

 imitated by a bag containing equal numbers of red and 

 white counters : in fact the production of the hybrid (or 

 rather its gonad) may be imitated by pouring equal 

 numbers of red and white counters into some convenient 

 receptacle. Now let us pair two such imitation 

 hybrids, using for this purpose two bags or hats each 

 containing equal numbers of red and white counters. 

 Two vertical lines are drawn on a large sheet of 

 paper ; the space between the lines being reserved 

 for pairs each consisting of a red and a white 

 {R W) ; the space on one side of the two lines for two 

 reds (J^R), and that on the other for two whites {WW). 

 A counter is taken at random out of one hat ; then 

 another out of the other hat ; and the pair, according to 

 its character {RR, RW or WW), is assigned to the 

 column, on the paper, prepared for it. We should, of 

 course, expect in a large number of trials that there would 

 be 25% RR, 50% RW 2ind 25% WW: this is in fact what 

 happens. Now in playing this game there is only one 

 rule, to be observed, which is " When a red is drawn with 

 a white the red shall be placed on the top of the whiteP (It 

 is this rule that confers whatever usefulness there is in 

 this device). Let us make fifty draws and place the 



