her
intertropical regions; the quantity of rain-fall being considerably less.
It begins in November and ends in April. During our first season in that
place, only a little over nineteen inches of rain fell. In an average
year, and when the crops are good, the fall amounts to about thirty-five
inches. On many days it does not rain at all, and rarely is it wet all
day; some days have merely a passing shower, preceded and followed by hot
sunshine; occasionally an interval of a week, or even a fortnight, passes
without a drop of rain, and then the crops suffer from the sun. These
partial droughts happen in December and January. The heat appears to
increase to a certain point in the different latitudes so as to
necessitate a change, by some law similar to that which regulates the
intense cold in other countries. After several days of progressive heat
here, on the hottest of which the thermometer probably reaches 103
degrees in the shade, a break occurs in the weather, and a thunderstorm
cools the air for a time. At Kuruman, when the thermometer stood above
84 degrees, rain might be expected; at Kolobeng, the point at which we
looked for a storm was 96 degrees. The Zambesi is in flood twice in the
course of the year; the first flood, a partial one, attains its greatest
height about the end of December or beginning of January; the second, and
greatest, occurs after the river inundates the interior, in a manner
similar to the overflow of the Nile, this rise not taking place at Tette
until March. The Portuguese say that the greatest height which the March
floods attain is thirty feet at Tette, and this happens only about every
fourth year; their observations, however, have never been very accurate
on anything but ivory, and they have in this case trusted to memory
alone. The only fluviometer at Tette, or anywhere else on the river, was
set up at our suggestion; and the first flood was at its greatest height
of thirteen feet six inches on the 17th January, 1859, and then gradually
fell a few feet, until succeeded by the greater flood of March. The
river rises suddenly, the water is highly discoloured and impure, and
there is a four-knot current in many places; but in a day or two after
the first rush of waters is passed, the current becomes more equally
spread over the whole bed of the river, and resumes its usual rate in the
channel, although continuing in flood. The Zambesi water at other times
is almost chemically pu