OBTAINING VARIETIES FEOM SEED. 37 



propagate for future crops. There are methods of even 

 crossing the different varieties, by hybridizing them, and 

 thus forming new and improved sorts ; but this process is 

 attended with considerable trouble, and requires some skill, 

 and indeed comes more under the attention of the gardener 

 than of the farmer, unless he happens to live within reach 

 of any large centres of consumption, where the early 

 potato crop, well managed, always produces very remu- 

 nerative returns. Sometimes it will be found that plants 

 which produce their tubers very early do not blossom at 

 all, and thus produce no seed. When this is noticed, and 

 the seed is desired, the soil round the plant should be 

 carefully removed, and the tubers that are formed picked 

 off. By thus preventing the plant from increasing its 

 underground stem, it is stimulated to throw out leaves 

 and flowers, producing seed in great abundance. In this 

 way Mr. Knight, the President of the Horticultural So- 

 ciety, succeeded in procuring seeds from potatoes which 

 had never before blossomed, and from these seeds he 

 raised several excellent varieties some being early, some 

 late, and some of a more vigorous and of a hardier nature 

 than those known before. Sargeret found that out of 

 300 plants of different varieties raised from seed, he had 

 not a single plant exactly like the original one from which 

 he obtained the seed, and that out of the whole number 

 he only found three worth keeping. 



If new sorts, or an entirely new stock be desired, 

 they may probably be obtained by this method, se- 

 lecting those only which appear likely to suit the pur- 

 pose intended, and carefully continuing their cultiva- 

 tion for three or four years, until the tubers produced 

 have reached the usual size. If, however, as under the 

 ordinary circumstances of farming, a crop be required at 

 once, the only plan to adopt is to make use of the ma- 

 tured tuber as seed, which in due course will reproduce 



