GLOXINIAS AND THEIR ARTIFICIAL FERTILISATION. 33a 
Thus, in connection with the colour, I have said that the Gloxinia 
crassifolia had only three very inferior colours at the commencement, 
viz. dark vinous red, dark blue slightly silvered on the edges, very pale 
violet and rose. It was necessary to select from these colours the best. 
and it was by continual sowings that one day I obtained much purer 
colours, viz. pure carmine, pure violet with a slight white border. Then, 
on experimenting again, I obtained the whole series of selfs edged with 
white, which were destined to become such favourites under the names of 
‘Le Progrés,’ carmine-red edged with white ; and‘ Patrie,’ dark blue edged 
with white. I obtained at the same time a plant of a magenta-lake 
colour, which, fertilised by a red variety, produced an amaranth-coloured 
flower, which also had a white margin. I at last obtained the pure white 
Gloxinia crassifolia, together with the whole series of ground colours; so 
that in a few years I created from a stiff, rigid plant, impossible to pack, 
with very inferior flowers, a type having an abundance of rich-coloured 
flowers, and perfect both as regards habit and inflorescence. In such a 
matter the greatest attention must be paid to the choice of the seed- 
bearing plants. 
With regard to the colours, it is very important to observe that the 
hybridiser must proceed exactly as the water-colour painter does; for 
instance, if he is desirous of obtaining fine distinct colours, he must 
only use, for the male parent, plants havivg pure, well-defined colours. 
Consequently, a carmine-red plant should be considered as a cake of 
paint of that colour; and if a dark blue variety is fertilised with that 
plant, four-fifths of the plants will be of a fine purplish-violet colour. 
If one takes a rose variety for the female and a dark carmine for the 
male, an intermediate colour of a bright pink will be the result. 
If, on the contrary, a plant of a mauve or magenta colour is fertilised 
with a bright carmine, one will obtain the amaranth colour known under 
the name of ‘ magenta.” 
The mingling of colours, or their intensifying or lessening, is thus 
a matter of absolute certainty. It is like a palette at the disposal of 
a painter. It is, however, necessary to be well acquainted with the 
genealogy of the plants, to have studied them for a long time, for 
without, one will only get unexpected and curious reversions. 
A red plant may have been fixed as the result of selection, but if it 
have blue or magenta in its blood (horticulturally speaking) it would 
not be surprising if, having this origin, its progeny should come out half 
red, half magenta—or worse. 
Some types have been fixed in such a way that it becomes practically 
impossible to alter them. I have obtained a Gloxinia named ‘ Boule 
de Feu,’ the very brilliant carmine-red of which has defied the inter- 
mixture of every other colour. 
This plant, which had for its great-grandfather the Gloxinia ‘ Mina ’ 
of Van Houtte, has served us as a mine of colours for some years, the 
richness of its tints being inexhaustible. Everything I fertilised with 
this plant assumed superb colours, blues, pinks, magenta, violets ; all 
these plants, impregnated by the Gloxinia ‘ Boule de Feu,’ preduced 
remarkable colours and of incomparable brilliancy. 
But when I desired to fertilise my ‘Boule de Feu’ by other 
