MICHIGAN ACADEMY OF SCIENCE. 
29 
of as made up of atoms and its character is determined by the nature 
and arrangement of the atoms of which it is composed, so the character 
of living matter is settled by its determinants. Most visible char- 
acters of plants are compound and made up of several or many units, 
how many we do not know, but in hybridizing the true nature of a 
character may come out and not a few unit-characters have been isolated. 
A mutation is, therefore, a congenital change in the determinants. It 
is by aid of this theory of unit characters that much of the best work 
in plant breeding is now being done. 
The history of the origin of our fruits will at once convince us that 
the great majority have arisen, not by repeated selection but by the 
isolation of a single plant. Thus the Tyson pear was a native seedling 
found in a hedge near Philadelphia. The Sec-kel, a single tree on waste 
land four miles from the same city. The Sheldon, an accidental seedling 
on a farm in Wayne county, New York. The Concord grape, a vine said 
still to exist at Concord, Mass., and so on with most fruits whose his- 
tory we happen to know. Our flowers tell the same story. The “crimson- 
eye” hybiscus was found in a swamp in New Jersey and is probably 
a mutation of the common pink species from which it differs also in 
tint of leaf and shape of seed-pod. From time to time a peculiar dwarf 
plant appears among ordinary sweet peas — this low bushy form is 
known as the “Cupid” and is evidently a mutation. It breeds true and 
no amount of selection of the shorter sweet peas will ever develop the 
“Cupid.” Ordinary variation will not reach it, mutation must occur. 
Doubtless many a genus of plants, both wild and cultivated, is a com- 
plex of species, mutations, and varieties crossed and recrossed in genetic 
tangle. Such are probably our asters, our sunflowers, our grains and 
others that are either a constant nightmare or a perennial delight 
according to the temperament of the taxonomist. Our wheats origi- 
nated in Eurasia, the various tribes carried different species with them 
into Western Europe. Our modern wheat is a mixture of these species 
with the results of thousands of years of cultivation, variation and 
selection superimposed. Races are mixed and improvement has only 
been maintained by. continuous rigorous selection. 
Now it is evident that if an improved race appeared as a mutation 
and it were isolated and multiplied so that no crossings occurred we 
would have at once an improved strain that would breed true, evidently 
a surer and a quicker method than continuous selection. 
The first men to use the principle of initial selection with isolation 
were Le Contour of Jersey and Patrick Sheriff of Scotland. As early 
as 1810 Sheriff had noticed a single plant of wheat of fine quality and 
much branched. The seeds were saved and sown separately. In two 
generations he had enough seed to offer it to the trade as the “Mungo- 
swell” wheat. To this day it is grown in Scotland, England and France 
and recognized as one of the better wheats. Sheriff continued his work 
for over half a century bringing out about a dozen new varieties of wheat 
and oats and in 1872 he published a little book for private circulation 
describing his method of work. 
Sheriff had no theory to offer and perhaps for that reason his 
practice had little weight. The old method was still used, more perfect 
in detail, more extensive in application. The Germans especially carry- 
