e
Pacific by Cape Horn, to continue the three-hourly observations until
the 90th meridian is passed.

Before quitting the Atlantic Ocean it may be well to notice the marine
stations mentioned in my Third Report on Atmospheric Waves,[5] as being
particularly suitable for testing the views advanced in that report and
for tracing a wave of the south-westerly system from the most western
point of Africa to the extreme north of Europe. A series of hourly
observations off the western coast of Africa has already been suggested.
Vessels staying at Cape Verd Islands should not omit to make
observations at three hours' interval _during the whole of their stay_,
and when circumstances will allow, hourly readings. At the Canaries,
Madeiras, and the Azores, similar observations should be made. Vessels
touching at Cape Cantin, Tangier, Gibraltar, Cadiz, Lisbon, Oporto,
Corunna, and Brest, should also make these observations while they are
in the localities of these ports. At the Scilly Isles we have six-hourly
observations, made under the superintendence of the Honourable the
Corporation of the Trinity House. Ships in nearing these islands and
making the observations already pointed out, will greatly assist in
determining the increase of oscillation proceeding westward from the
nodal point of the two great European systems. We have already mentioned
the service surveying vessels employed on the coasts of Ireland and
Scotland may render, and the remaining portion of the area marked out in
the report may be occupied by vessels navigating the North Sea and the
coast of Norway, as far as Hammerfest.

In connexion with these observations, having especial reference to the
European system of south-westerly atmospheric waves, the Mediterranean
presents a surface of considerable interest, both as regards these
particular waves, and the influence its waters exert in modifying the
two great systems of central Europe. The late Professor Daniell has
shown from the Manheim observations, that small undulations, having
their origin on the northern borders of the Mediterranean, have
propagated themselves northward, and in this manner, but in a smaller
degree, the waters of the Mediterranean have contributed to increase the
oscillation as well as the larger surface of the northern Atlantic. In
most of the localities of this great inland sea six-hourly observations
may suffice for this immediate purpose; but in sailing from Lisbon
through the Straits