47 
numerical results, which Mendel obtained, can be readily 
explained on mathematical grounds by what we know of 
the various combinations which are possible between egg- 
cells and pollen-grains when differing in two characters. 
Of course, if the parental forms differ in several characters 
the possible combinations are much more numerous and the 
numerical chances of complete reversion to the parent forms 
is much smaller. On the other hand, while we obtain a 
large number of hybrid forms which will show reversions 
to the original parents, we shall find some new combina- 
tions which are pure forms and therefore breed true. These 
we can isolate from the rest by the rejection of all strains 
which show reversion, and thus we can obtain new and 
permanent varieties. This no doubt has been the means 
adopted by plant breeders in the past from practical 
experience, but thanks to Mendel and those who have fol- 
lowed up the path shown us by his investigations, a scien- 
tific basis has been laid io the practice of hybridisation. 
In the course of these scientific enquiries some remark- 
able facts have come to light. It has, for instance, been 
discovered in crossing a white and a yellow variety of the 
Marvel of Peru (Mirabilis) that the hybrid produced was 
not intermediate in character, that is, of pale yellow 
colour, neither was it like either of the parents, but of a 
pink colour and marked with red stripes. The white form 
must evidently have possessed some chemical factor which 
changed the yellow colour into red, while the character 
producing striping, which could not show itself in the 
white form, became visible in the hybrid by reason of the 
coloured sap. The offspring produced by self-fertilisation 
of this hybrid were of twelve different kinds: five with 
different shades or striping in yellow, five corresponding 
forms in red, and two white forms, which, though re- 
sembling each other externally, differed in constitution as 
could be seen from their progeny. 
A somewhat similar and equally remarkable result was 
obtained by crossing two white Sweet Peas belonging to 
the variety, Emily Henderson, the offspring having a 
partly-coloured flower, red with blue keel, probably very 
like the ancestral form. Of the two white forms evidently 
each one contained a special factor, which combined with 
that of the other white form caused the formation of a 
coloured sap. The offspring of the coloured hybrid were 
mostly coloured but in different degrees, while white forms 
