
ON THE REDUCTION OF METEOROLOGICAL OBSERVATIONS. 93 
ascent, after which the rapid descent begins. Beyond this I have nothing to 
add to Mr. Birt’s remarks which follow. 
(B.) This group, which extends only from New York 1™ west to Gar- 
diner, Maine 19™ east of Albany, exhibits a general agreement among the 
curves, all of them descending. ‘The group, however, may be divided into 
minor portions, in which different features are well marked. The three 
flattest curves, especially in their anterior portions, were obtained at Provi- 
dence, Boston and Gardiner. The curves at New York, New Haven and 
Middletown, also agree tolerably well with each other, especially the two 
last, New York agreeing more with Flushing, which connects as it were the 
two minor groups. William’s College is nearly similar to the Middletown 
group, and the curve at Montreal differs from the others mostly in the 
depression at 6 and 7 p.m. Gardiner, Maine presents the nearest similarity 
to it. 
(B.) The curves in this sheet present a different phenomenon, inasmuch 
as the increase of oscillation does not appear to be regular; but when we 
separate the curves into groups, as above, we perceive it. Flushing appears 
to be an exception. By including Flushing in the Middletown group, we 
have stations exhibiting smaller ranges between those presenting larger; Wil- 
liam’s College between Montreal and the New Haven or Middletown group ; 
Providence between Middletown and Boston; and Gardiner nearly in the 
same parallel with William’s College. This exhibition of smaller ranges in 
connection with the minor groups of curves is highly interesting, exhibiting, 
as the Middletown group does, the increase of oscillation towards that sta- 
tion. 
(B.) This group is perhaps one of the most interesting, particularly as 
relates to the extent of oscillation at the several stations. By referring to 
the Table of ranges for June and September 1836, it will be seen that the 
oscillations at William’s College were less than those at the neighbouring 
station, Albany. The mean altitude of the barometer is also less at William’s 
College than at Albany. 
(B.) On this occasion (Sept. 1837) the altitude of the barometer was 
greater than on any other at all the stations except at Montreal, which ex- 
hibited the greatest oscillation. By taking a similar view of these as of the 
June observations, there appears to be three centres of oscillation, namely, 
Montreal, Boston and Middletown. The oscillations are accordingly arranged 
in this manner in the Table. 
Dec. 1837.—St. Louis, Western Reserve College, New York, Flushing, 
Quebec, Montreal, Middletown, Burlington, Boston, Gardiner, Bahama, 
St. Blas. 
(H.) All the curves for the above stations, in the order in which they are 
here arranged, down to Boston inclusive, begin by descending to a minimum, 
in the case of New York only coinciding in epoch (4 p.m.) with the afternoon 
minimum; in all the others anticipating of it by 1, 2, or 3 hours, the times 
being as follow :— 
Burlington.... 1 p.m. 21st. BROST OM 16) Sieier Seas ins sain pein oe Baie 
Middletown .. 1 P.M. Western Reserve Coll... 2 p.m. 
St. Louis .... 1 pm. BYISHUD GS, ict arcieuninscaith ccoei ii Pehle 
Quebec ...... 2 P.M. New York .........0204 4: PoMe 
From this minimum they all rise, with more or less subordinate undulation, 
and with degrees of rapidity corresponding to the order of succession of the 
stations above set down. In the Boston, Burlington, New York and Flushing 
