358 Mr. J. P. Harrison on the Relation of [Feb. 14, 



And the monthly maxima of solar radiation (in the next Table) accord 

 with the higher mean tensions of vapour in almost every instance. 



Table III. 



Monthly Mean Maxima of Solar Radiation. 



Year. 



May. 



June*. 



July. 



August. 



September. 





o 









o 



1842. 



82-1 



98-9 



92-4 



102-2 



81-7 



1843. 



80-9(o) 



85-1 



9G-9(^) 



93-0 



92-6(c) 



1844. 



87-5 



950 



94-1 



87-2 



87-0 



1845. 



75-0 



93-1 



90-3 



(91-7)t 



81-3 



1846. 



87-1 



103-1 



97-4 



93-3 



90-3 



1847. 



86-3 



85-7 



93-0 



89-0 



77-6 



(fl) Mean amount of cloud 8-0. (b) Cloud 8-5. (c) Cloud 4-5. 



In those cases where the values of radiation and vapour do not agree, the 

 exceptions will, it is believed, be sufficiently accounted for ; thus the oc- 

 currence of the second maximum of solar radiation in September 1843 in 

 place of July, though tension was higher in the latter month, is explained 

 by the obscuration of the sun, and the unusual quantity of rain in July. 

 The tension of vapour in September was -06 higher than in June+. 



Heat and Vapour in Canada. — The dependence of the maximum tem- 

 perature in the day on the quantity of moisture in the air, in winter, at To- 

 ronto, though not directly connected with the present inquiry, is closely 

 allied to it, and may be referred to with advantage in the absence of ob- 

 servations of solar radiation, to show the effects of slight variations of va- 

 pour in that country. 



General (at that time Colonel) Sabine, in a paper in the Transactions of 

 the Royal Society " On the Periodic and Non-Periodic Variations of Tem- 

 perature at Toronto" §, pointed out the fact that the period of minimum 

 heat in the year, both hourly and daily, at that station occurs between 

 the 7th and 1 7th days of February, on the days when vapour is also at its 

 minimum. The means of temperature and tension on the ten days 



* In June the maxima of radiation are, as a rule, found to be accompanied by less 

 vapour than in August and September. Thus it will be seen in Tables II. and III. 

 that the difference between the mean tensions in June and August 1846 is -013 ; the 

 excess is in favour of August ; but the maximum solar radiation is in June. In 1843 

 tension was at its maximum in July — but not radiation, in consequence of the abnormal 

 quantity of cloud and rain in that month. 



t Six observations only. 



J A still more remarkable instance of monthly maximum solar radiation in Sep- 

 tember occurred in I860, when the excess over the mean of the five preceding Sep- 

 tembers was 21°-7, and the increase of contemporaneous vapour-tension -086. This 

 appears to be accounted for by the rainfall in August ; the weather in September 1865 

 was, in fact, very like that which follows the rainy season in India. The tension of vapour 

 was -084 higher than in June. 



§ Anno 1853, p. 148. 



