HYBRID ODONTOGLOSSA. 263 
produce more wonderful results, not possibly contemplated by Nature. 
Foremost among these will be a new race of ‘ Red Odontoglossa,’ by 
crossing Odontioda with Cochlioda, selecting the largest and best forms 
for such a purpose. To this it may be objected that the recross will 
tend to reduce the size of the Odontioda x Cochlioda; but even if it 
does so in the majority, there will always be appearing among them 
giants, such as have already appeared in Hallio-crispwm ‘Theodora’ 
Fic. 82.—O. crispum ‘ WinutAm Stnvens.’ (Orchid Review, 1905, 48.) 
(fig. 86), and amabile ‘ Royal Sovereign’ (fig. 88), these two being greatly 
in excess of the whole of the remainder of the plants of the same 
parentage. 
REVERSION. 
This is the most potent adverse factor we have to deal with and guard 
against; it will always trend towards the most powerful parent in the 
hybrid’s genealogy. Each batch of seedlings raised from two species 
shows this, and each lot of plants, the result of more complex and hybrid 
parentage, proves it even more clearly, and from continued study of the 
matter, I think the aboriginal forms of all our so-called species were self- 
coloured flowers. 
What process Nature employed to produce those species which are 
now parti- or multi-coloured I think only conjecturable; but we may 
