448 ON THE IMPROVEMENT OF PLANTS. 



Doubtless there are limits ; I have tried on various 

 occasions and in various ways to obtain a hybrid between 

 the Pyrus spectabilis and the Cydonia japonica, two 

 different genera, but without success, and I announce this 

 because to acknowledge a failure is sometimes as useful to 

 others as to record a success. On the other hand, I have 

 succeeded in raising hybrids between the Moss Rose (Rosa 

 centifolia) and the Alba Rose, two different species ; also 

 between a Hybrid China Rose (R. gallica x R. indica) 

 and Rosa multiflora. Then, again, I have often tried in 

 vain to obtain crosses between certain varieties. So that 

 it does not seem to be exactly a question between species 

 and species, or between variety and variety as these are at 

 present classified by botanists. I have found, however, as 

 a rule that the more nearly allied the varieties the more 

 certain is the cross. The Tea-Scented and Noisette Roses 

 are found to cross readily, but then, although these form 

 separate floricultural groups, they belong to the same 

 species R. indica. But the Hybrid Bourbon and Moss 

 Roses also cross readily, and here we have intermixed 

 three different species according to most botanists R. 

 indica, R. Gallica, and R. centifolia The grandest and 

 the most beautiful of all the floral groups of the Rose the 

 Hybrid Perpetual has a curious genealogy : R. gallica 

 crossed with R. indica produced Hybrid China ; the latter 

 crossed with an autumn flowering variety of the R. dama- 

 scena produced Hybrid Perpetual. 



My first attempt at crossing the Rose was made in 

 1843. I crossed three flowers of the Tea-Scented Rose 

 known as Goubault with the Bourbon Rose Souchet with 

 the view of obtaining a dark -coloured Tea-Scented Rose. 

 Let me call these flowers C, D, and E. C produced a very 

 large pod, which ripened perfectly and gave ten large 

 seeds. D produced a medium-sized pod with nine seeds 

 of unequal size, one being very large, four large, and four 

 small. E produced a small pod which contained thirteen 

 small seeds. Of these thirty-two seeds four only ger- 



