Physical geography. By Mary Somerville ...

226 PHYSICAL GEOGRAPHY. CHAP. XVIT.
the level of the sea: the tide is perceptible in the river of the Ama.
zon 576 miles from its mouth, and it ascends 255 miles ii the Orinoco.
In the temperate zones rivers are subject to floods from autumnal
rains, and the melting of the snow, especially on mountain-ranges.
The Po, for example, spreads desolation far and wide over the plains
of Lombardy; but these torrents are as variable in their recurrence
and extent as the climate which produces them. The inundations
of the rivers in the torrid zone, on the contrary, occur with a regularity peculiar to a region in which meteoric phenomena are uniform
in all their changes. These floods are due to the periodical rains,
which, in tropical countries, follow the cessation of the trade-winds
after the vernal equinox and at the turn of the monsoons, and are
thus dependent on the declination of the sun, the immediate cause
of all these variations. The melting of the snow no doubt adds
greatly to the floods of the tropical rivers which rise in high mountain-chains, but it is only an accessary circumstance; for although
the snow-water from the Himalaya swells the streams considerably
before the rains begin, yet the principal effect is owing to the latter,
as the southern face of the Himalaya is not beyond the influence of
the monsoon, and the consequent periodical rains, which besides
prevail all over the plains of India traversed by the great rivers and
their tributaries.
Under like circumstances, the floods of rivers, whose sources have
the same latitude, take place at the same season; but the periods of
the inundations of rivers on one side of the equator are exactly the
contrary of what they are in rivers on the other side of it, on account of the declination of the sun. The flood in the Orinoco is at
its greatest height in the month of August, while that of the river
Amazon, south of the equinoctial line, is at its greatest elevation in
March.' The commencement and end of the annual inundations in
each river depend upon the average time of the beginning, and on
the duration of the rains in the latitudes traversed by its affluents.
The periods of the floods in such rivers as run towards the equator
are different from those flowing in an opposite direction; and as the
rise requires time to travel, it happens at regular but different periods
in various parts of the same river, if very long. The height to
which the water rises in the annual floods depends upon the nature
of the country, but it is wonderfully constant in each individual river
where the course is long; for the inequality in the quantity of rain
in a district drained by any of its affluents is imperceptible in the
general flood, and thus the quantity of water carried down is a measure of the mean humidity of the whole country comprised in its
basin from year to year. By the admirable arrangement of these
Baron Humtboldt's Personal Narrative.