METEOROLOGY. 409 



A few notable instances which bear somewhat upon the preceding 

 subject, and which at first sight seem not to be in accordance with the 

 principles of radiation, we find, on the 2d, 6th, and 7th of November, 

 where, with a light haze (not fog) but otherwise cloudless sky and a 

 calm atmosphere, a great reduction of temperature is seen against that 

 of the city. On the 2d of November, at 7 a. m.,the difference of tem- 

 perature amounts to eight degrees ; on the 6th, at 9 p. m., it amounts 

 to five degrees ; and on the 7th, at 7 a. in., to nine and a half degrees. 

 These facts might be brought forward against the doctrine of the inter- 

 ference of smoke; for it may be asked, "Why does the haze not inter- 

 fere with radiation in these cases?" 



To this we reply that the dry haze, consisting, as it does, chiefly in 

 minute floating particles of smoke, has slowly settled down to the lower 

 strata of the atmosphere, in consequence of the powerful radiation and 

 cooling of its oivn particles, and may therefore be considered as the 

 result of radiation, and not as an interfering medium! But when, after 

 continual additions, the haze has accumulated to such a depth as to 

 prevent the penetration of the radiant heat, then the case will be quite 

 a different one. An instance of this kind we find also in our register, 

 illustrating in a striking manner what we have just said. For while, 

 at 7 a. m. on the 7th of November, with a light haze, a sky free from 

 clouds, and a scarcely stirring breeze from southeast, the temperature 

 fell nine and a half degrees lower than that of the city ; by 9 p. m., 

 the haze had grown so thick as to screen the sky completely ; and al- 

 though with a calm atmosphere prevailing at this time, the temperature 

 was not in the least cooler, but, on the contrary, was warmer by three 

 degrees than that of the city. 



Class No. 2, namely, that of calms and clouds. 



That with a completely clouded sky and a calm atmosphere it should 

 be warmer in town than in the country, simply from accumulation of 

 heat, can easily be conceived ; but why, in many instances, the reverse 

 should take place, and, as shown by the figures in class No. 2, the 

 temperature of the town should sometimes be cooler than that of the 

 country, could not be so easily accounted for if the register itself was 

 not consulted. In class No. 2 we have twenty-three instances where 

 the temperature near the city limits is warmer than that of the city. 

 By reference to the register we find, however, that of these twenty- 

 three cases eighteen have had only partial cloudiness, and therefore 

 offering open passages to the rays of the sun as well as to partial radi- 

 ation ; circumstances capable of modifying the result to a considerable 

 extent. But five have entire cloudiness ; and in four out of these re- 

 maining five instances we see them associated with rain and thunder 

 storms, while the fifth and last stands between fog and mist. 



In regard to most of the instances comingunder the head of the fifth 

 class, namely, all those of a clouded sky, accompanied by brisk and 

 violent winds, we find that the mean numbers of the temperature of the 

 city limits differ but very little from those of the central part of the 

 city. This is what may be expected ; for the rush of strong winds 

 beneath a clouded sky allows no great accumulation of heat in the streets, 



