86 



ON HYBRIDIZATION AMONGST VEGETABLES. 



lieve, true generic features. If they should be found deficient 

 in any species, a thing which I do not much expect, it will be 

 one of the many instances in which a feature becomes obsolete 

 without affecting the general character of the type, as in the 

 section Azalea five of the stamens of the genus (Rhododendron) 

 are oftener deficient than present. Cross-breeding shows that 

 such a deficiency is not original, but an incident. By true 

 generic features I mean features that belonged to the type of the 

 genus, whether that was the created type, or whether the genera 

 themselves had diverged from fewer created individuals ; an in- 

 quiry which would be superfluous and vain. 



The facility with which the larger Rhododendrons intermix 

 with some sorts of Azalea is now well known, and it is quite 

 clear that Azalea is a condition of the same genus, in which half 

 the anthers are usually deficient (half in the perfect form being 

 of inferior power), and the leaves disposed to be deciduous, 

 although they are deciduous also in Rhod. Kamchaticum, Dahu- 

 ricum, and one or more Chinese species. Some seedlings of the 

 yellow Pontic Azalea frequently produce seven, eight, or nine 

 anthers, and Indica with evergreen leaves has usually a fuller 

 complement, but is much more unwilling to cross with the ever- 

 green Rhododendrons, showing clearly thereby that the feature 

 which had been assumed for the generic character, is even of less 

 importance than some secret difference which we cannot discover 

 amongst the species. Az. (so called) squammata has ten stamens. 

 From Rhodora Canadensis impregnated by A. Pontica I obtained 

 a plant (A. Seymouri, Bot. Reg.) which would everywhere pass 

 muster as Azalea. It is therefore preposterous to look upon 

 Azalea and Rhodora as other than strongly marked and defective 

 sections of the genus Rhododendron. It was my original notion 

 that the cross-bred produce from such plants, if placed alone in 

 a soil and climate perfectly congenial to their fructification, 

 would perpetuate the cross-bred type. I am satisfied that in 

 many cases that would occur, but perhaps not where the diversity 

 is as wide as between Azalea and Rhododendron. The crosses 

 between the Nepal, Pontic, and American Rhododendrons are 

 fertile, and doubtless would perpetuate themselves, though with 

 a good deal of sporting. I formerly mentioned that the result 

 from the impregnation of the shrubby Calceolaria integrifolia by 

 Calceolaria plantaginea, which is quite as humble and herbaceous 

 as a plantain, was remarkable, the whole produce having a 

 similar aspect, that of a very dwarf plant, with long serrated 

 leaves on very short stiff branches, the inflorescence being ex- 

 actly intermediate between that of the two species. It appeared 

 at first to be sterile, but last year I obtained a pod from it, and 

 it has reproduced itself as perfectly as if it were a natural species 



