TUBEROUS BEGONIAS. 



345 



single seedlings, we were ambitious to raise some doubles and began 

 intercrossing the Continental varieties and those of English origin. 

 Our first efforts were very successful, and we obtained about half a 

 dozen varieties that we considered to be distinct advances on the 

 other varieties of the same colours in our possession. 



Our enthusiasm in the work being now thoroughly kindled, Mr. 

 Lascelles procured all the new varieties likely to be of any use 

 either at home or on the Continent. Messrs. Cannell had at this 

 time commenced sending out a batch of new varieties annually, and 

 these we found to possess characteristics differing very widely from 

 those of Messrs. Laing and the Continental varieties, suggesting to 

 us the possibility that a different group of plants was being used as 

 parents. Messrs. Cannell's varieties possessed a stiff habit, round 

 thick leaves and strong flower stems, one of the best examples being 

 the ' Hon. Mrs. Plunkett,' the growth and habit partaking rather 

 of the character of Begonia rosaeflora and B. Veitchii. The varieties 

 sent out by Messrs. Laing, on the contrary, were of a much more 

 drooping habit and with long pointed leaves, showing that B. boliviensis 

 had been the dominant factor in their origin. Others of Messrs. 

 Laing's hybrids had evidently been obtained from B. Davisii, a good 

 example being the ' Marquis of Stafford,' a very valuable variety 

 for bedding on account of its neat upright habit and bright crimson 

 flowers, borne well above the foliage. 



The advantage of having the best examples from three sources, 

 viz. Messrs. Laing, Messrs. Cannell, and from the Continent, will 

 be apparent when we consider that the first essential in cross- 

 fertilization is to possess a number of varieties, each with some desir- 

 able character, but all obtained from more or less distinct sources. 

 This has doubtless contributed largely to the measure of success which 

 1 have obtained in the improvement of the Begonia. As my ex- 

 perience points to the fact that inbreeding is bad, and should be 

 avoided as much as possible, I have depended almost entirely on 

 first crosses, and have not worked at all on what are known as 

 Mendelian lines. 



I have always kept in view, when crossing, the desirability of 

 obtaining two varieties whose colours and habit of growth were 

 somewhat similar, but as far removed as possible in relationship 

 from one another. This necessitates keeping a pedigree book and 

 having a considerable number of stock parents, forming several 

 different families in each group of colours. It is not advisable to 

 make violent crosses, such as crossing yellow with pink or rose, or 

 crimson with white. The best results are obtained from work 

 within well-defined lines by intercrossing only such colours as rose, 

 pink, blush, and salmon in one group, crimson, scarlet, and red in 

 another, and orange and yellow in another. Whilst working generally 

 within these lines I should deprecate a slavish adherence to them, 

 as occasions will arise for making experimental crosses, and these are 

 full of interest, though more often than not very disappointing. 



VOL. XXXIX. 2 A 



