house till rooted. All cuttings of Cactuses may be
treated in this way. If anything proves destructive to these cuttings,
it is excessive moisture in the soil, which must always be carefully
guarded against.

Grafting.--The object of grafting is generally either to effect certain
changes in the nature of the scion, by uniting it with a stock of a
character different from its own, which usually results in the better
production of flowers, fruit, &c., or to multiply those plants which are
not readily increased by the more ordinary methods of cuttings or seeds.
In the case of Cactuses, however, we resort to grafting, not because of
any difficulty in obtaining the kinds thus treated from either cuttings
or seeds, as we have already seen that all the species of Cactuses grow
freely from seed, or are easily raised from cuttings of their stems, nor
yet to effect any change in the characters of the plants thus treated,
but because some of the more delicate kinds, and especially the smaller
ones, are apt to rot at the base during the damp, foggy weather of our
winters; and, to prevent this, it is found a good and safe plan to graft
them on to stocks formed of more robust kinds, or even on to plants of
other genera, such as Cereus or Echinocactus. By this means, the
delicate plants are raised above the soil whence the injury in winter
usually arises, and they are also kept well supplied with food by the
more robust and active nature of the roots of the plant upon which they
are grafted. Grafting is also adopted for some of the Cactuses to add to
the grotesqueness of their appearance; a spherical Echinocactus or
Mamillaria being united to the columnar stem of another kind, so as to
produce the appearance of a drum stick; or a large round-growing species
grafted on to three such stems, which may then be likened to a globe
supported upon three columns. As the species and genera unite freely
with each other, it is possible to produce, by means of grafting, some
very extraordinary-looking plants, and to a lover of the incongruous and
"queer," these plants will afford much interest and amusement. Besides
the above, we graft Epiphyllums, and the long drooping Cereuses, such as
C. flagelliformis, because of their pendent habit, and which, therefore,
are seen to better advantage when growing from the tall erect stem of
some stouter kind, than if allowed to grow on their own roots. By
growing a Pereskia on into a large plant, and then cutt