SEED SAVING AND HYBRIDISATION. 95 



other sources, will be beneficial, and in practice this is found to be the case, 

 provided only that the newly imported "strain" is not in any way inferior 

 to the stock already in hand. In large collections, where many of the indi- 

 viduals are far removed from one another in character and descent, this point 

 is not of quite so much consequence as where the number of plants is limited ; 

 but even the largest collection will probably be benefited more or less by the 

 infusion of fresh "blood" at intervals. 



But in some cases, as where, for instance, it is desired to produce any 

 particular colour or other characteristic, it is not possible to select types differing 

 in all respects for breeding from, and what may be termed " in-breeding " 

 must to some extent be resorted to. For instance, supposing an improved 

 form of a white flower to be desired, it would be a very slow way of going 

 to work to cross the best white in the stock with one of any other colour, 

 for hardly one in fifty of the seedlings would come white, the majority of 

 course reverting to the more natural or normal colour of red or pink. Theo- 

 retically, perhaps, such might be the best way of obtaining an improved 

 form, and even in practice the progeny would probably possess more vigour 

 than the result of a cross between two closely- related and more or less similar 

 white flowers. But the shortest way to go to work is to select two plants of 

 the same colour, or nearly so, but differing in other respects, and if possible 

 obtained from different sources. For instance, supposing one parent to be a 

 large but lovely-formed flower, perhaps tinged or tinted with pink, we should 

 select for the other parent a plant bearing a well-shaped, purely white bloom, 

 even if it were somewhat small. 



Among the progeny of such a cross would be almost certain to be found 

 one or more seedlings in which the good qualities of each of the parents were 

 blended, to the exclusion of their faults, and thus an improvement would have 

 been effected. Next season, starting again with these blooms, faults still 

 remaining Avill be eliminated and good qualities still further developed, and 

 thus a constant progress towards perfection is effected. 



In hybridising, a good general rule is to select for the second parent a flower 

 possessing those qualities that the first is deficient in. For instance, supposing 

 a plant with large but rather loosely-shaped pink blooms to be taken in hand, 

 we should select for the other plant a well-shaped i.e., circular bloom of a red 

 or crimson colour, with good substance and habit, and among the produce would 

 be sure to be found at least a few plants decidedly superior to either of the 

 parents, combining the good points of both. It does not greatly matter which 

 way the cross is made ; on the whole, perhaps, the female parent has generally 

 more influence upon the character of the seedlings than the pollen-bearing 

 parent, though at times we have noticed that pollen from an exceptionally fine 

 bloom of a distinct character produced a remarkable advance in the quality and 

 number of the seedlings produced, some of them being far superior to the 

 female in all respects, and one or two even ahead of the male. But, as a matter 



