494 
of those cases, downward currents appeared to prevail. During the fine 
weather at the close of January, I observed so little of vertical currents, 
that I laid aside my journal; but on the two days preceding the disas- 
trous storm of Saturday, February 9, some circumstances attracted my 
attention, and induced me to resume my observations. On the 7th, I 
observed, at about 4 Pp. m., in a part of a wide street, bordered only by a 
wall, a gentleman’s hat lifted off his head to a height of at least five feet. 
The hat dropped back again, without being transferred horizontally to any 
appreciable extent. On Friday, the 8th, at2 in the afternoon, my attention 
was called to the anemoscope, by its shifting round, through N., towards 
N. E., with decided and frequent downward plunges. It appeared as if 
showers of cold air were descending ;* the temperature was also falling. 
At this time the horizontal motion of the air was comparatively little, 
while the vertical convection was apparently highly developed. Next day, 
during the storm, the disk of the vane was in a state of constant oscil- 
lation; but no marked prevalence of upward or downward motions 
could be observed, and nothing resembling the plunges noticed on the 
preceding day. 
Although we should scarcely expect to be able to observe the influ- 
ence of such disturbances in a mercurial barometer, they may become 
manifest, were we to use a liquid column of much greater height. 
During the short period that a column of water was so employed by Pro- 
fessor Daniell at the apartments of the Royal Society, he appears to have 
noticed numerous and rapid oscillations during storms. The diminution 
of such oscillations, both in frequency and intensity, enabled him to pre- 
dict the approach of fine weather. During windy weather, the regular and 
continuous motion of the water-column resembled the action of respira- 
tion. This remark corresponds with what I have stated, both here and 
elsewhere, as to the oscillations of the anemoscope during gales of wind. 
The inference which I drew from the comparative regularity of such oscil- 
lations is, not that vertical currents prevail with strong winds, but that the 
motion of the air is essentially undulatory. The irregular and plunging 
motions of the air observed by me before storms, and sometimes even be- 
fore rain, are essentially different phenomena, and they do not appear to 
have been noticed by Mr. Daniell in any of the appearances exhibited by 
his water-barometer. 
One of the most remarkable phenomena connected with the storm of 
the 9th February was the rise of the barometer a short time before its 
commencement. A fall had taken place at a preceding period; and, in 
accordance with the usual empirical rules, the rise would be considered 
* Dove refers to the precipitation of cold air during whirlwinds, and to the storm 
assuming the form called by the Greeks exvediac.—* Board of Trade Meteorological 
Papers,” No. III., p. 22; and “‘ Taylor's Scientific Memoirs,” vol. iii., p. 215. 
+ In extraordinary cases the mercurial barometer has always exhibited such fluctua- 
tions: thus Dove refers to a remark of Hoskiaer, that during the great hurricane of 
August, 1837, at St. Thomas, the mercury sank two lines at each gust, and then imme- 
diately rose to the same height as before. 
