112 ABNORMAL VARIATIONS AT TORONTO. 



the six observation hours. If the annual means alone be rep^arded, 

 there is nothing to warrant the belief that one hour is to any great 

 extent more subject than another to thermic disturbance ; but on refer- 

 Ting to the hourly table for the separate months, from which table III. 

 is derived, and comparing the numbers in the columns for 10^ i2^ IS**, 

 and 20"* with those for 2"* and 4^ it was found that in the six winter 

 months,(October to March,) the former group were, in nearly every case, 

 number for number, greater than the latter group, and that exactly the 

 reverse occurred in the other six months. Table III. shews that the 

 winter half-yearly means are in every case greater, and the summer half- 

 yearly means less, at each of the hours 10^ 12''. 18^ and 20'' than at 

 2^ and 4''. Hence it appears, from both series, that there is in winter 

 a greater uncertainty respecting the temperature during the night and 

 morning than during the hours of the day ; whereas in summer the 

 warmer hours are more subject to irregularity ; or it may be briefly 

 stated that the warm hours are most subject to disturbances of tempera- 

 ture in the warm months, and the cold hours in the cold months ; the 

 difference in the extent of the disturbance for each season being about 



Table IV. gives the mean abnormal variations with their proper 

 signs, arranged according to the direction of the wind at the instant of 

 observation. In the results for the years 1853 to 1859 the variations 

 are arranged in seventeen groups corresponding to the sixteen principal 

 directions of the wind, and to light winds with a velocity not exceeding 

 half a mile per hour, which it was the custom formerly to regard as 

 calms. In the later series the variations are arranged in nine groups 

 only, corresponding to the eight principal directions and to absolute 

 calms. 



From the observations of the first seven years, if each point be 

 considered as including an angular space of 11°. 15 on each side of it, 

 it appears that the teqiperature is above or below the normal according 

 as the wind blows from a point lying to the South or to the North of a 

 line drawn from N.E.b.E. to S.W.b.W. The greatest depression, 3°. 58, 

 accompanies a wind from N.N.W., and the greatest elevation 3.61 

 occurs with a wind from S.S.W., giving a range of 7° 19. 



Tliat the variations that accompany the N. E, and West winds have 

 different signs in the two series is partly owing to the proximity of 

 these points to the line, which, in the earlier series, is found to separate 

 the relatively warm from the relatively cold winds. Another cause of 



