o that fresh horses have to be brought for this purpose from La
Plata at a great expense. The reduced size of the horses bred on both
southern and northern islands, and on several mountain-chains, can hardly
have been caused by the cold, as a similar reduction has occurred on the
Virginian and Mediterranean islands. The horse can withstand intense cold,
for wild troops live on the plains of Siberia under lat. 56 deg.,[118] and
aboriginally the horse must {53} have inhabited countries annually covered
with snow, for he long retains the instinct of scraping it away to get at
the herbage beneath. The wild tarpans in the East have this instinct; and,
as I am informed by Admiral Sulivan, this is likewise the case with the
horses which have run wild on the Falkland Islands; now this is the more
remarkable as the progenitors of these horses could not have followed this
instinct during many generations in La Plata: the wild cattle of the
Falklands never scrape away the snow, and perish when the ground is long
covered. In the northern parts of America the horses, descended from those
introduced by the Spanish conquerors of Mexico, have the same habit, as
have the native bisons, but not so the cattle introduced from Europe.[119]

The horse can flourish under intense heat as well as under intense cold,
for he is known to come to the highest perfection, though not attaining a
large size, in Arabia and northern Africa. Much humidity is apparently more
injurious to the horse than heat or cold. In the Falkland Islands, horses
suffer much from the dampness; and this same circumstance may perhaps
partly account for the singular fact that to the eastward of the Bay of
Bengal,[120] over an enormous and humid area, in Ava, Pegu, Siam, the
Malayan archipelago, the Loo Choo Islands, and a large part of China, no
full-sized horse is found. When we advance as far eastward as Japan, the
horse reacquires his full size.[121]

With most of our domesticated animals, some breeds are kept on account of
their curiosity or beauty; but the horse is valued almost solely for its
utility. Hence semi-monstrous breeds are not preserved; and probably all
the existing breeds have been slowly formed either by the direct action of
the conditions of life, or through the selection of individual differences.
No doubt semi-monstrous breeds might have been formed: thus Mr. Waterton
records[122] the case of a mare which produced {54} successively three
foals without ta