ON THE IMPROVEMENT OF PLANTS. 447 



sports seem desirable here, and with that I shall close this 

 part of my subject. We have seen that variations or 

 sports are, when reproduced from seed, liable to vary 

 greatly in character, with a general tendency to revert to 

 the type. From the first sowing there will probably be 

 but few like the variation, but by successive repetitions of 

 the process of sowing and selecting the variation is at 

 length fixed. It is thus that many of our races of annuals, 

 as Asters, Stocks, Clarkias, Candytufts, Tropaeolums, and 

 the like, not readily propagated otherwise than from seed, 

 have been obtained. But the process of fixing variations 

 or sports from the leaf-bud is almost certain to proceed 

 without interruption. You graft, bud, or make cuttings of 

 the sported branch or plant, and the plants so propagated 

 usually remain unchanged. 



I will now proceed to the second part of my subject, 

 namely, the improvement of plants by hybridising and 

 cross-breeding. 



I have said that in hybridising and cross-breeding we 

 start with a conception which we labour to convert into a 

 fact. For example, we find existing in two separate flowers 

 or fruits certain high qualities as size, colour, form, flavour, 

 mien, &c., which we seek to combine in one and the same 

 individual. We cross or hybridise these two plants, sow 

 the seeds, and after few or many attempts one out of many 

 is likely to give the desired result. As a guide in hybridis- 

 ing and crossing it should be noted that our best observers 

 are of opinion that hybrids and cross-breeds derive their 

 form and habit in great part from the female, the colours 

 of the flowers from the male, while the constitution may be 

 acquired from either parent. 



By hybridising we understand the crossing of two 

 different genera or species ; by cross-breeding the inter- 

 mixture of two different varieties of the same species. 

 Practically speaking the work is the same. It would be out 

 of place here to enter into a lengthened description as to 

 how far the different species of plants may be hybridised. 



