48 ZOOLOGY 



be reached experimentally, without going to the trouble 

 of raising plants. Take a half-dollar and toss "heads 

 or tails." On the first toss you are equally likely to get 

 heads or tails ; so also on the second toss, the first hav- 

 ing no effect on the second. So half the first tosses of 

 each pair will be heads (H) and half will be tails (A). 

 Now, after tossing heads, half the second tosses (on the 

 average) will be heads and half tails. The same after 

 tossing tails. Hence, after a large number of successive 

 pairs of tosses, you get this result, \HH, \Hh, \hH, 

 \hh. The tosses, like the Mendelian combinations of 

 determiners, follow the "laws of chance." In a small 

 number of cases the proportions will not be likely to 

 agree exactly with the theory, but the larger your 

 statistics, the closer the agreement. 



Explanation 12. Returning now to the visible results, the Ft 

 three-to-one g enera tion from our cross between the tall and dwarf 

 ratio peas gives us jTT, \Tt, \tT, \tt. The first is homozy- 



gous or pure-bred for tallness, and will of course be tall. 

 Crossed with others like itself, it will give only tails - 

 provided that by "like" we mean not merely in ap- 

 pearance, but in actual constitution. Tt and tT differ 

 only in that one got its tall factor from one parent, the 

 other from the other. As this makes no difference, and 

 as T is dominant, both will be tall. Finally, tt is 

 homozygous for dwarfness, the recessive character, and 

 will therefore be dwarf. It is what we call an "ex- 

 tracted recessive," and when crossed with others like it- 

 self can give only dwarfs, in spite of the fact that both 

 its parents and one of its grandparents were tall. We 

 now see how the three-to-one ratio is explained theo- 

 retically ; given the facts, it seems very simple, but it 

 is hard to exaggerate the credit due to Mendel for first 

 detecting the law governing inheritance. 



