ON ATMOSPHERIC WAVES. 121) 
that, by discussing a long series of observations in this manner, a tolerable idea 
of the succession and systems of waves may be formed, and the general fea- 
tures of the normal waves made out. The one under consideration appears 
to have had an interval of fifteen days. The great symmetrical wave of Nov. 
1842 had nearly the same interval, and succeeding waves, possessing a simi- 
larity of character both in interval and curve, were observed about the same 
period of the year in 1843 and 1844. The examination of these recurring 
atmospheric movements forms the subject of the next portion of the report. 
Secrion II. 
Recurrence of Symmetrical Wave. 
The diagram which accompanies this report (see Plate III.) exhibits three 
curves to a great extent similar, at least in so far as there is a general tendency 
in the barometer to rise during the period of the anterior half, and a similar 
tendency in it to fall during the period of the posterior half. From what has 
just been advanced, as well as from the discussions which were reported last 
year, there is great reason to consider the indentations on the anterior and pos- 
terior slopes of the curve of 1842 as distinct secondary and superposed waves ; 
the same may be said of the indentations on the curves of 1843 and 1844. 
Now it is probable that were we to separate the barometric effects of these 
waves, we should obtain a much clearer conception of the form and general 
elements of the normal wave which on the three occasions recorded passed 
London. For this purpose the following steps have been taken. The gene- 
ral contour of the curves indicates that the respective maxima passed about 
the following dates. 
1842. November 18, noon. 
1843. ” 14, 5 
1844. October 27, ” 
These days (noon) are therefore assumed as the axes of the curves, and 
the altitudes at intervals of two hours have been carefully read off from the 
original projections, and a mean of the three taken, from which the follow- 
ing Table has been constructed. The table is arranged in two compartments, 
the first containing the ordinates of the anterior slope, the second those of 
the posterior. The first column in the first compartment indicates the hours 
before the transit of the crest (—) ; the second the mean ordinate correspond- 
ing to any given hour. In like manner, the first column in the second compart- 
ment indicates the hours after the transit of the crest (+ ), and the second the 
mean ordinate corresponding to any given hour after transit. These num- 
bers have been used in the construction of the fourth curve, which exhibits 
to the eye the general form of the normal wave, freed to a certain extent of 
the effects of the superposed waves. 
There are several drawbacks to the value of any conclusions that may be 
drawn from these numbers and projections in their present state :— 
Ist. They are deduced from unreduced observations. The projections of 
the three upper curves are laid down from observations as read off from the 
scale without any reduction whatever, and the mean curve has been obtained 
from these unreduced observations. 
2nd. The observations themselves were made at irregular intervals, so that in 
deducing the mean, the quantities observed have not been used. The altitudes 
at the given hours of the curves drawn through the points indicating these 
observed quantities, are the quantities from which the mean has been obtained. 
3rd. The curves, and consequently the mean, consists of two distinct ele- 
ments, namely, the pressure of the gaseous atmosphere and the pressure of 
4 
i 
