face _towards_ the quarter from which it is
advancing--S.W., he will find that with a _falling barometer and S.E.
wind the current passes him from the left to the right hand_; but if at
a barometric _minimum_ he place himself in the same position with his
face directed to the quarter from which the N.W. current is advancing
laterally, also S.W., he will find that with a rising barometer _and
N.W. wind the current passes him from right to left_. Now the two
classes of phaenomena are identical, and it would not be difficult to
show that, had we an instance of a rotatory storm in the northern
hemisphere moving from N.W. to S.E., it would present precisely the same
phaenomena as to the direction of currents passing from left to right and
from right to left with falling and rising barometers, increase and
decrease in the force of the wind, &c., as the oppositely directed
aerial currents do which pass over western central Europe.

In the absence of direct evidence of the production of a revolving storm
from the crossing of two large waves, as suggested by Sir John Herschel,
although it is not difficult to obtain such evidence, especially from
the surface of the ocean, the identity of the two classes of phaenomena
exhibited by the storms and waves as above explained amounts to a strong
presumption that there is a close connexion between them, and that a
more minute investigation of the phaenomena of atmospheric waves is
greatly calculated to throw considerable light on the laws that govern
the storm paths in both hemispheres. The localities in which these
atmospheric movements, the waves, have been hitherto studied, have been
confined to the northern and central parts of Europe--the west of
Ireland, Alten in the north of Europe, Lougan near the Sea of Azov, and
Geneva, being the angular points of the included area. It will be
remarked that the greatest portion of this area is _inland_, but there
is one important feature which the study of the barometer has brought to
light, and which is by no means devoid of significance, viz. that the
oscillations are much greater in the neighbourhood of _water_, and this
appears to indicate that the junction lines of land and water form by
far the most important portions of the globe in which to study both the
phaenomena of storms and waves. It is also very desirable that our
knowledge of these phaenomena should, with immediate reference to the
surface of the ocean, be increased, and in this