Il8 HYBRIDISING AND CROSS-BREEDING. 



fection that is to say, decorative or flowering plants should be 

 grown on to the flowering stage before any just and right record 

 can be written of their collective characteristics ; and the same 

 rule holds good with foliage-plants and fruit-trees. In selecting 

 plants for scientific experiments, it is highly necessary that those 

 only be used which retain their natural fixity of character, and 

 which consequently reproduce themselves true from seed when 

 fecundated with their own pollen. Many plants which have 

 been long in cultivation are in such a highly variable state that 

 no reliance can be placed on any .experiments in hybridising 

 wherein they are used. Among plants of this character may 

 be named Hippeastrums, Camellias, Roses, Crotons, variegated 

 Dracaenas, Agaves, Primulas, tuberous-rooted Begonias, and 

 many others. As I have said, the want of definite information 

 as to hybrid or cross-bred offspring of plants is very great ; but 

 although the parentage of many of the hybrids enumerated 

 in this work is open to question, the great and practically 

 important point still remains, that they are not only distinct 

 from their supposed parents, but also quite different from all 

 other known plants. Nevertheless, as we trust to the hybridist 

 at home to give us proof of the natural or accidental hybridisa- 

 tion of plants abroad i.e., in their native habitats we must 

 also insist on his being a trustworthy and skilful operator 

 before we place faith in his records, even though those records 

 be the supposed hybrid offspring themselves ; since seminal 

 variation, the sudden development of latent characters or 

 " sports," or the increase or decrease of health, as in the case 

 of seedling variegated plants, may each or a combination of all 

 produce the change attributed to hybridism. For example, 

 the flowers of the seed-bearing plant must be carefully emas- 

 culated before there is the shadow of a chance that pollen 

 can have escaped from the anthers. The female plant must 

 then be isolated from all other plants at all related to it, and 

 the pollen of the intended male parent must be applied to the 

 stigma in its pure state, and not with a camel's-hair pencil 

 which has been used indiscriminately for applying all sorts of 

 pollen. It is not enough to isolate a seed-bearing plant in a 

 greenhouse, unless due precautions are taken against the pos- 

 sible entrance of wind-wafted pollen from allied plants in con- 

 tiguous houses or from adjoining gardens. Hence comes it 

 that botanists and intelligent observers are always suspicious of 

 the parentage of hybrids and varieties as given by the majority 

 of cultivators who raise florists' flowers. For botanical pur- 

 poses hybrids raised between different genera or species of 

 Orchids or Asclepiads, where the pollen is in firm, heavy 



