ARTIFICIAL SELECTION 127 
fect the plant as a whole, or any part, with reference to size, 
form, flavor, proportion of chemical constituents, time of 
appearance or ripening, hardiness, and so on. Since in plants 
which are raised from seed, the special peculiarities of the 
parent are found to reappear in its offspring to a greater or 
less degree, it becomes possible for the farmer to preserve in 
future crops the peculiarities which please him, by taking 
his seeds from those individual plants which satisfy him 
best. Thus if early ripening is the quality desired, the earliest 
seeds are the ones chosen year after year, until in the course 
of several, or it may be many generations, furnishing earlier 
and earlier plants, the offspring finally produced from these 
selected seeds are found to ripen their product so much sooner 
than any other sorts that they are recognized as a new 
variety. 
With many plants, such as strawberries, the seedlings are 
apt to vary so widely from the parent and from each other 
that the varieties are said to be “not true to seed”: and in 
these cases it is the practice when once a seedling possessed 
of desirable qualities has been obtained, to propagate it by 
means of “‘cuttings” or similar detached portions of the 
parent plant instead of by seed. 
Occasionally important differences appear among the in- 
dividuals raised from cuttings or the like, and these may 
similarly form the basis of new varieties. 
50. Artificial selection. Besides these principal ways in 
which cultivated varieties arise, there are some others the 
consideration of which must be deferred to a later chapter. 
What at present concerns us is the general truth that to a 
very large extent hwman or artificial selection exerts a control- 
ling influence either upon the development or the perpetua- 
tion of varieties, and frequently upon both. Since the longer 
and more widespread the cultivation of a given plant has 
been, the more extensive and more varied must have been 
this influence, we should expect in general that the number of 
varieties of a cultivated plant would be proportional to the 
time and area of its cultivation; and this expectation we 
find to be justified by the facts. 
