ame
earlier; summer commenced earlier and continued longer; autumn held off
later, and cold weather, when it came, was uniform and severe. This season
the transit has seemed to be less than for several years.[10] The spring
was backward; the summer cool, but exceedingly regular; the autumn thus
far without extremes, and the whole year healthy and productive. It is the
normal period of the decade, between the irregular heat of the first part,
and the irregular cold of the last; and it has been normal in character,
and conformed beautifully to its location. If the transit of 1854 was
further north than the mean, as it seemed to be over this country, that of
itself would convey the showers which follow up in the western portion of
the concentrated trade, on the east of the mountains of Mexico, and cause
them to precipitate further north, over New Mexico, and thus, rather than
from a diverted trade, they may have derived their unusual supply of
moisture during the summer of 1854. On this subject I can but conjecture,
and leave to future observation a discovery of the truth.

Enough appears, however, to show the importance of taking the location of
the year in the decade, and even the character of the decade itself, into
the account.

But whatever the remote cause of the difference in the seasons, the
character of the seasons is directly influenced by the character of
storms, or periodic changes. Sometimes the tropical storms are most
numerous; at others the polar waves; and at others the irregular local
storms, or general tendency to showers. The seasons when the polar waves
are most prevalent, are the most regular, healthy, and productive. Those
where the tropical tendency is greatest, are irregular; and so are those
where the other classes predominate. These differences in the character of
the storms, are but the varying forms in which magnetic action develops
itself. I have said that there was a decided tendency to cirrus without
cumulus, in mid-winter, and cumulus without cirro-stratus or stratus, in
midsummer, and during the intermediate time an intermediate tendency. But
there is a difference between spring and autumn. Dry westerly (not N. W.)
gales prevail in March, and N. E. storms in April and May, but violent S.
E. gales are not as common. On the other hand, the dry westerly gales of
March are comparatively unknown in autumn, and the violent, tropical,
south-easters are then common.

Snow-storms occur during the