IN THE TROPICS. 
365 
It is unfortunately true that in the case of all these experiments 
no two generations were strictly comparable one with another as 
regards the external conditions under which they were grown. 
Moreover external conditions have unmistakably a large influence 
upon the growth of the plants, an influence which, as De Vries has 
pointed out, probably extends over more than one generation. It 
is nevertheless clear that a marked improvement in the vigour and 
prolific habit of the plants was arrived at by the process of crossing 
and then selecting the strongest plants from among the progeny of 
the cross-breds. Natural selection also played a part in determining 
the later generations, as may be gathered from the above account. 
Simple selection of uncrossed strains, in conjunction with rigorous 
natural selection, did not lead to any advance of a comparable 
kind. 
The effect of the climate of Peradeniya upon pure strains of 
English peas next demands attention. The exact result seemed 
to vary in different cases. Thus, when sown in November, French 
Sugar Pea grew to a considerably greater height than was to 
be expected from the description accompanying the sample. On 
the other hand Earliest Blqe Pea was much more dwarf. The 
particular effect in different cases will be described later on, but in 
this place I would emphasize the fact that the wdiole result appeared 
at once in the plants grown at Peradeniya from English seed, and 
that subsequent generations grown from the seed which these pro¬ 
duced closely resembled this first generation. The following races 
retained for five generations the characters distinctive of the first 
generation, as grown at Peradeniya :—Telegraph, Telephone, Ring¬ 
leader, French Sugar, Fillbasket, Satisfaction, and Sutton’s Earliest 
Blue Pea. The last-named, which was also the most modified, may 
be more particularly described by way of example. 
The first generation (English seed) was sown on November 29th, 
1902. 426 seeds were sown in a bed manured as evenly as possible. 
The seeds were planted in two rows 4 inches apart and 2 inches 
apart in the rows. The seeds were ripe on January 20th (seven weeks 
and three days after sowing), and 398 plants were then gathered 
8(9)05 (2) 
