roop, and the whole plant wilts and in bad attacks dies. It
has yearly proved a more serious danger in Texas and other parts of the
south-west of the United States, and the damage due to it in Texas
during 1905 was estimated at about L750,000. No remedy is known for the
disease, and cotton should not be planted on infected land for at least
three or four years.

"Boll rot," or "Anthracnose," is a disease which may at times be
sufficiently serious to destroy from 10 to 50% of the crop. The fungus
which causes it (_Colletotrichum gossypii_) is closely related to one of
the fungi attacking sugar-cane in various parts of the world. Small
red-brown spots appear on the bolls, gradually enlarge, and develop into
irregular black and grey patches. The damage may be only slight, or the
entire boll may ripen prematurely and become dry and dead.

Many other diseases occur, but the above are sufficient to indicate some
of the principal ones in the most important cotton countries of the
world.

_Improvement of Cotton by Seed Selection._

In the cotton belt of the United States it would be possible to put a
still greater acreage under this crop, but the tendency is rather
towards what is known as "diversified" or mixed farming than to making
cotton the sole important crop. Cotton, however, is in increasing
demand, and the problem for the American cotton planter is to obtain a
better yield of cotton from the same area,--by "better yield" meaning an
increase not only in quantity but also in quality of lint. This ideal is
before the cotton grower in all parts of the world, but practical steps
are not always taken to realize it. Some of the United States planters
are alert to take advantage of the application of science to industry,
and in many cases even to render active assistance, and very successful
results have been attained by the co-operation of the United States
Department of Agriculture and planters. With the improvement of cotton
the name of Mr Herbert J. Webber is prominently associated, and a full
discussion of methods and results will be found in his various papers in
the _Year-books_ of the U.S. Department of Agriculture. The principle on
which the work is based is that plants have their individualities and
tend to transmit them to their progeny. Accordingly a selection of
particular plants to breed from, because they possess certain desirable
characteristics, is as rational as the selection of particular animals
for b