FOURTEENTH REPORT. 
26 
qualities, whatever they may be, are all expressed in figures. The 
amount of bookkeeping is very great. In the year 1900, according o 
de Vries, some 3,109 varieties were in culture. The records embrace 
the complete botanical description of each new sort from iis germina- 
tion till the time of harvest, with all the details required for controlling 
its constancv and uniformity, and for the study of all those qualities 
on which the introduction into general agriculture will ultimately 
depend. . . ,. , 
It is evident that the chance of a fit variety turning up is proportional 
to the number of plants tested, the larger the number of rejected unfit 
the greater the chance of rapid progress. Thus to produce the Bur- 
bank' Paradox berry 40,000 blackberry and raspberry hybrids were 
produced and grown until the fruit matured. Then from the whole lot: 
a single plant' was chosen as the best, all others were uprooted and 
burned. Tn the trial grounds of Dixon Bros, of Belfast, Ireland, oi a 
total of about 1,300 choice hybrid seedling roses raised each year per- 
haps ten new kinds worthy of preservation may be expected and of 
these onlv one mav be suited to the American trade. At such a cost 
have the "Ki Harney,” the "Betty'’ and a score more of our best roses 
been produced. 
Even the chemist is now an aid to the -selector. Our annual prod- 
uct of 2,500 million bushels of corn is used for a great variety of pur- 
poses, stock feeding is the chief but over 100 different commercial 
products and about fifty kinds of food are derived from corn and its 
various constituents. For many of these purposes a special part of the 
o^rain is desired. For fifteen years the experiment station at the Uni- 
versity of Illinois has been breeding corn for high and for low oil con- 
tent and for high and for low proteid content. In 1908 the races thus 
produced vielded respectively 7.19% and 2.39% of oil, 13.94% and 
S.9 qc ; 0 f proteid. Thus the possibility of breeding for chemical content 
has been demonstrated. 
But all the improvements in methods and extent of selection would 
be of no avail unless there was variety to make it effective. Till the 
variation occurs man is helpless. He cannot as yet make plants to 
order. When we speak of plant creations we mean nothing of the kind. 
All man can do is to select, sort, arrange, combine, the qualities 
have to be created for him through variation. It is evident that the 
key to the situation lies here and that future progress depends on an 
increase in our knowledge of the nature of variation. As difference from 
parents constitute variation and as the preservation of a valuable strain 
depends on resemblance to parents, the whole problem of heredity is 
involved. 
As an important part of practice we must here notice that in the 
majority of our perennial plants Negative propagation can be used. So 
grafts, buds, stem, root, and leaf cuttings, layers, runners, bulbs or 
tubers and other methods are used to propagate and the uncertainties 
of generation avoided. Many a clone has arisen from a single individual 
and is si ill genetically one, tho grown and disseminated as thousands of 
separate plants. 
What then is variation? It used to be defined as a departure from 
the type of ihe species but when we ask what the type of the species 
mav b<>. we find we are talking of an abstraction. Brob- 
