ndicated by
the fact that although no species of _Gossypium_ is native to the United
States of America, that country now produces over two-thirds of the
world's supply of cotton. Under normal conditions in warm climates many
of the species are perennials, but, in the United States for example,
climatic conditions necessitate the plants being renewed annually, and
even in the tropics it is often found advisable to treat them as annuals
to ensure the production of cotton of the best quality, to facilitate
cultural operations, and to keep insect and fungoid pests in check.

Microscopic examination of a specimen of mature cotton shows that the
hairs are flattened and twisted, resembling somewhat in general
appearance an empty and twisted fire hose. This characteristic is of
great economic importance, the natural twist facilitating the operation
of spinning the fibres into thread or yarn. It also distinguishes the
true cotton from the silk cottons or flosses, the fibres of which have
no twist, and do not readily spin into thread, and for this reason,
amongst others, are very considerably less important as textile fibres.
The chief of these silk cottons is kapok, consisting of the hairs borne
on the interior of the pods (but not attached to the seeds) of
_Eriodendron anfractuosum_, the silk cotton tree, a member of the
Bombacaceae, an order very closely allied to the Malvaceae.

_Classification._--Considerable difficulty is encountered in attempting
to draw up a botanical classification of the species of _Gossypium_.
Several are only known in cultivation, and we have but little knowledge
of the wild parent forms from which they have descended. During the
periods the cottons have been cultivated, selection, conscious or
unconscious, has been carried on, resulting in the raising, from the
same stock probably, in different places, of well-marked forms, which,
in the absence of the history of their origin, might be regarded as
different species. Then again, during at least the last four centuries,
cotton plants have been distributed from one country to another, only to
render still more difficult any attempt to establish definitely the
origin of the varieties now grown. Under these circumstances it is not
surprising to find that those who have paid attention to the botany of
the cottons differ greatly in the number of species they recognize.
Linnaeus described five or six species, de Candolle thirteen. Of the two
Italian botanis