IN THE TROPICS. 305 
variation by a number of instructive examples. The recent 
advances in knowledge following upon a return to the 
experimental method were in one place explicitly foretold. 
A number of further important contributions towards a 
solution of the problems of variation and inheritance will 
be found mentioned in the list of literature at the end 
of this paper. 
I.—MENDEL'S OBSER VA TIONS. 
It will now be necessary to consider those remarkable dis¬ 
coveries of Mendel, which were independently rediscovered 
and confirmed in the year 1900 by de Yries (67), Correns 
(11), and Tschermak (58). An essential point in Mendel’s 
treatment of the subject of hybrids is the separate study 
which he applied to separate pairs of differentiating or 
alternative characters (one member of such a pair being 
exhibited by one of the parent forms, the second by the 
other, e.g.y the presence of coloured flowers upon one, and of 
white flowers upon the other). Following the terminology 
of Bateson and Saunders (6) we may call such characters 
allelomorphs. A heterozygote is formed by the union of a 
pair of opposite allelomorphic gametes (sexual cells), a 
homozygote by the union of gametes in which the allelo¬ 
morphs are identical. 
Bateson describes the immediate offspring of a cross as the 
filial generation F 1} subsequent generations being F 2 , F 3 , &c. 
The parents of the cross-bred plant belong to the generation 
Pi, its grandparents to P 2 , and so on. 
Mendel’s discovery may be at once briefly stated, using 
these more modem terms. Having applied to certain varie¬ 
ties of peas the method of artificial cross-pollination with 
subsequent separate raising of all the offspring obtained, and 
that of recrossing with the parent forms, Mendel came to the 
conclusion that the male and female germ cells (or gametes ) 
of a heterozygote contain the pure parental allelomorphs com¬ 
pletely separated from one another, and that the segregation 
of characters takes place in such a manner that the 
