J. BE. Willet on Indian Summer. 343 
servations of “smoky,” “smoky atmosphere” and “ very smo- 
ky” are recorded on the following number of days. 
1853. No. of days 1854, No. of days 
Aug. 3 Jan 1 
Sept. 6 Feb. 1 
Oct. 10 March 1 
Nov. 5 
Dec. 3 
The register of Lieut. J. M. Gilliss at Santiago in Chili is still 
More conclusive; for no month, observed during the years 
1850-1-2, is without “ slightly smoky,” “smoky” or “ phe 
smoky” days. The number of such days in each month will 
et in the following table, which has been collected from the 
eteorological Observations, vol. vi. 
1850. 
1851. 1852 
Jan. 6 9 23 
Feb. 7 8 24 
March, 21 15 26 
April, 18 6 16 
ay, 5 7 10 
June, 4 4 4 
July, 5 1 9 
Aug. 3 4 9 
Sept. 4 3 6 
Oct. 2 3 
Noy. 2 2 
Dec 2 6 
i n ell as in H 
and in other parts of Europe. That of 1783 had remarkable 
intensity and “ presented the following phenomena :—its thick- 
hess was such, that in some places objects at the distance of five 
kilometers (3 miles) could not be distinguished ; they sometimes 
the Altai. In 1834, Kaemtz noticed others in May, July and 
: August ; the one areal to be transferred from place to 
* 
