48 BOARD OF AGRICULTURE. [Pub. Doc. 



with every generation; and so we should endeavor to bury it 

 beneath the greatest possible number of generations of the 

 desired type. Again, it has been found that often under the 

 abnormal conditions of cultivation, particularly under glass, 

 there comes an inherited lack of vigor, diminished size and 

 increased liability to disease; and that all of these are coun- 

 teracted by crossing through the flowers with some distinct 

 variety, particularly if it be one nearer the original type of 

 the species. This results not only in increased vigor, but in 

 a great diversity of type; though there are always some of 

 the plants which retain the type of the variety, plus greatly 

 increased vigor, and these can be used for the starting of a 

 new and invigorated line of descent without a change of type 

 or of name, unless it be that it be known as reinvigorated 

 stock of the old variety. 



Again, we often have prepotent plants or those which have 

 exceptional power to transmit their own characteristics, in- 

 dependent of that of the general run of inherited tendency. 

 It is clear, then, that for the production of seed every one of 

 which is likely to produce plants of precisely the same type 

 we must first have not only an exact conception of that type, 

 but that such conception shall be so recorded that we can 

 exactly follow it in our selection for many years, and it is 

 very desirable that this conception should be designated by 

 some universally accepted name; and, second, we must each 

 year select not only plants of that exact type, but of proven 

 prepotency or tendency to produce plants of that type. This 

 necessitates not only the saving of seed from each plant sep- 

 arately, but the testing of its relative prepotency and the use 

 of only those plants which have been proven to be strongly 

 prepotent. This involves the testing of seed of each plant 

 before it is given a place in the hereditary line. Even after 

 by this method we have developed a stock of satisfactory 

 unity of type we cannot safely rest, for such stocks, no mat- 

 ter how carefully grown, tend not only to lose in vigor but 

 to degenerate through accidental crossing ; and the process of 

 building up a stock from prepotent individuals of the right 

 type must be constantly repeated, so that we may have a fresh 

 stock but a few generations from the selected prepotent in- 



