1853] 



ON THE VARIATIONS OF TEMPERATURE AT TORONTO. 



17 



tual observed values are entered in a column of the register, thus 

 shewing- at a glance the non-periodic variation at that time. In 

 this form the monthly meteorological reports will iu future ap- 

 pear in the Upper Canada Medical Journal. 



The mean monthly temperatures on which these calculations 

 are founded are as follows :- — ■ 



January, 24-97 May, 



February, 23-40 June, 



March, " 30-23 July, 



April, 41-14 August, 



51-18 September, 58-02 



61-05 October, 44-93 



66-41 November, 36.51 



66-16 December, 26-75 



July is consequently the hottest mouth, and February the 

 coldest; there are 6 months above and 6 months below the mean 

 of the year (44°-23), and the mean temperature is passed through 

 in April and October. The hottest month is 22°-lS above, and 

 the coldest month 20 o- S3 below the mean, the difference between 

 the hottest and coldest months being 43°-01. 



Col. Sabine remarks that " February and March appear to be 

 months most liable to extreme variations ; July and August the 

 least so." But it may be doubted whether the series of years is 

 sufficiently long to warrant this conclusion. On a comparison of 

 the iso-thermal curves over the whole surface of the earth, Dove 

 states that September is " the season when the distribution of 

 temperature over the globe is most regular, even America form- 

 ing no exception. Then begins the Indian Summer, the time 

 which the Great Spirit of the Red-Skin sends to him that he 

 may follow the chase. Nature falls gently asleep in Autumn, 

 and awakens with feverish starts in Spring." 



On the yearly mean, Col. Sabine states : — " The mean annual 

 temperature derived from the whole body of the observations in 

 twelve years is 44°-23; and on the supposition that no constant 

 errors, instrumental or observational, or occasioned by insufficient 

 protection or defective exposure of the thermometer, are involved, 

 and that the variations of the temperature in different 3-eais may 

 be regarded strictly as accidental oscillations round a mean value, 

 and of equally probable occurrence in every year, the probable 

 error of this result is +0°-18. The probable variability of a 

 single year is +0°-63, shewing that there is an equal chance 

 that the me a n temperature of any one year will fall within the 

 limits of 43°-60 and 44°-86, as that it will exceed those limits; 

 a conclusion which, perhaps, would scarcely have been anticipated, 

 considering the great range of the thermometer in the course of 

 the year, and the magnitude of the non-periodic variations in 

 short intervals. The climate of Toronto p>resents a remarkable 

 combination of great regularity in the annual temperature with 

 great variability occurring in the course of the year. 



The mean temperatures of the several years differed from the 

 average mean temperature as follows : — 



00 o 



1841.... —0-31. -.1845.... +0-35....1849.... —0-14 



1842-... —0-27. ...1846.... -4-213 1850 +0-22 



1843.... —1-88.. ..1847.... — 053. ...1851 .... —0-25 



1844.... +0-25....1848.... +0-85....1852.... —0-39 



The excess of cold in 1843 (l°-88), was due chiefly to the oc- 

 currence of very low temperatures in February and March of that 

 year; the excess of heat in 1846 (2°-13), was more generally 

 diffused throughout the year, all the mouths except February and 

 October being above their average." 



On comparing the mean monthly temperatures with the nor- 

 mal temperatures clue to the geographical latitude of Toronto, it 

 appears that " every month of the year at Toronto is colder than 

 the normal temperature of the parallel in which it is situated, the 

 mean annual temperature being nearly 7° below the normal. 

 The thermic anomaly is least in July and August (between 2° 



B 



and 3°), and greatest in February, when it exceeds 11°. Its 

 sudden increase in October and decrease in November are de- 

 serving of notice. In viewing the bearing of the thermic anomaly 

 at Toronto on the more general question of the thermic anomaly 

 in the part of North America in which it is situated, it is neces- 

 sary to bear in mind that the thermometer at Toronto was about 

 342 feet above the sea level, equivalent, as usually estimated, to a 

 diminution of rather more than 1° of Fahrenheit, on account of 

 vertical elevation. Dove's normal temperatures are all reduced to 

 the sea-level, and when the monthly temperatures at Toronto 

 have undergone the same reduction, the thermic anomaly indi- 

 cated by them is diminished to about 1° in July and August, 

 but in February still reaches the large amount of 10°; in both 

 respects, therefore, confirming Dove's conclusion, that the sum- 

 mers of North America are not warmer than is due to their lati- 

 tude, whilst the winters are much colder." All these results are 

 embodied in the accompanying plate, which is copied from that 

 in Col. Sabine's paper. 



The black curve line indicates the annual normal march of the 

 temperature, its vertical ordinate being- the number of degrees 

 Fahr., corresponding to the period of the year indicated by the 

 horizontal one. 



The fainter straight lines give the actual observed averages of 

 five-day periods for the twelve years, and the vertical distance be- 

 tween the two corresponding points of the faint and dark lines is 

 the non-periodic variation at that period. The points .surrounded 

 by small circles are the actual monthly means of the twelve years 

 from which the normal Curve is computed, and these are con- 

 nected by vertical lines with the corresponding monthly means 

 (similarly distinguished), of the temperature due to the latitude 

 of Toronto as calculated by Dove, the march of the latter being 

 denoted by the c'otted curve, and the vertical distances between 

 this dotted curve and the dark one are the thermic anomalies for 

 that period of the year. The normal geographical curve is com- 

 puted for the sea level, but the normal curve itself is formed from 

 the temperatures uncorrected for vertical elevation, so that in 

 geographical comparisons of climate the thermic anomalies must 

 be each diminished by about 1° Fahr. 



An examination of these curves will prove highly interesting. 

 I may especially direct attention to the rapid descent of the nor- 

 mal curve between November and December, its prolonged pause 

 throughout most of January, (the January thaw) its sudden fall 

 in February and its equally rapid rise in March. 



As I have often heard it remarked that a cold winter is follow- 

 ed by a hot summer, I have thrown together the monthly tem- 

 peratures into the usual seasons (December, January and February 

 for winter, and so on), in the following table : — 



Year. 



Spring. 



SrarsiER. 



AUTUMN. 



"Winter. 



1841- 



1842- 



1843- 



1844- 



1845 







39-1 

 42-7 

 37-1 

 44-1 

 42-4 

 44-2 

 399 

 413 

 40-2 

 38-4 

 41-7 

 39-1 







65-0 



62-0 

 63-1 

 63-4 

 65-0 

 66-6 

 63-8 

 65-9 

 66-0 

 66 7 

 62-6 

 64-5 







46-0 

 44-7 

 44-8 

 42-4 

 46-4 

 498 

 46-1 

 45-0 

 48-7 

 46-9 

 46 8 

 47-2 







27-8 

 22-6 

 25-4 

 26.9 

 2° 7 



1847 



1848 



24-1 

 28.5 

 22.4 



1849 



27-4 



1S50 



1851.. 



24-9 

 21-1 



1852. . 



26-3 







Mean. 



40-9 



64-6 



46 5 



250 



