446 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



necessarily follow. Post hoc certainly, but ergo propter hoc by 

 no means. This palpable fallacy is further exposed when we find 

 that many reverse crosses are not at all different ; for instance, 

 Mr. Seden, for Messrs. Yeitch & Sons, crossed C. Schlimii $ with 

 C. longifolium £ and C. longifolium $ with Schlimii $ , which 

 produced the well-known hybrid C. x Sedenii. " No perceptible 

 difference was observed between the plants raised from the two 

 separate crosses : they agree in habit, foliage, structure, and 

 colour of flower ; in fact in every particular." (Veitch, Man. 

 Orch. Part X. p. 143.) Instances like this completely dispose 

 of the theory that reverse crosses are intrinsically different. 



A careful inquiry into the variations occasionally seen in 

 reverse crosses shows that they are mostly due to a different 

 variety being used as a parent in each case, the offspring varying 

 accordingly. 



In the few cases in which the same variety has been used in 

 each case the variation is no doubt due to the variability of the 

 parent's history, which is duly handed on to the offspring, and if 

 but one or two plants be raised from each cross the chances are 

 that the survivors will be widely different. 



Theoretically, if every possible variety could be raised 

 between A $ x B $ and between B $ x A $ on comparison the 

 varieties of the former would be found to correspond with the 

 latter exactly. It would be difficult to demonstrate this in 

 Orchids, because out of thousands of individual seeds in a pod so 

 few plants are raised which arrive at the flowering age ; but it 

 has been demonstrated in plants other than Orchids by the 

 experiments of careful observers. (Kerner and Oliver, " Natural 

 History of Plants," ii. p. 557.) 



To sum up the ordinary facts of hybridisation : — 



(1) Hybrids of the first generation between two distinct 

 species are generally intermediate between their parents. 



(2) This applies to their inner structure as well as to their 

 outward characters. 



(3) They are so far intermediate between their parents as to 

 be quite distinct from either. 



(4) All hybrids between the two same species are specifically 

 alike, but like species differ in varietal characters. 



(5) These varietal differences are found to correspond with 

 the varietal differences in the parent species. 



