14 



ON THE VARIATIONS OF TEMPERATURE AT TORONTO. 



[1803 



hammer take all apart in a few minutes, and put them together 

 in the same space of time. 



There will be several sizes of these rams, from No. 1 up to 

 No. 5; capable of throwing from 1 pint to 8 or 10 gallons per 

 minute. 



On the Variations of Temperature at Toronto. 



To the Editor of the Canadian Journal. 



Provincial Magnetic Observatory, 



Toronto, August 2, 1853. 



Sir,— 



Having lately received through the kindness of Col. Sabine, 

 a paper read by him before the Royal Society, (Feb. 10, 1853.) 

 it has appeared to me of so much importance and interest, to 

 Canadians especially, that I venture to lay before you the follow- 

 ing account of it with a few accompanying comments. The 

 title of this paper is 



" On, the periodic and non-periodic variations of the Tem- 

 perature at Toronto, in Canada, from 1841 to 1852, 

 inclusive. By Col. Edward Sabine, Treas. and V.P.R.S. 



It may be regarded as an anticipation of the 2nd Volume of 

 Observations at the Toronto Observatory, which has been pre- 

 pared for publication by Col. Sabine, and is now forthcoming 

 in a few weeks. 



The terms periodic and non-periodic, though familiar enough 

 to scientific ears, may require a little explanation for the general 

 reader. The temperature at any place is understood to be that 

 of the air as indicated by a thermometer placed at a certain ele- 

 vation above the ground, usually four or five feet, at the observer's 

 station : the thermometer must be screened from the Sun's rays, 

 and distant from any object capable of reflecting or radiating heat ; 

 it should also be exposed to a free current of air on all sides, and 

 must moreover be sheltered from the rain, snow, and in short as 

 much as possible from all extraneous causes, which might mix 

 up their effects on the temperature with that of the air. Strictly 

 speaking, the instruments should also be protected from the heat 

 radiated by the earth, but this is not the usual practice, and it is 

 preferred to make separate observations of this element ; no error 

 can arise from this cause, the custom being universal. It is evi- 

 dent that the fulfilment of all these conditions is extremely diffi- 

 cult, and no plan of exposing thermometers has yet been discovered 

 which is free from objection. The thermometers at the Toronto 

 Observatory, are situated on the North end of the buildino-, and 

 sheltered by a double shed of Venetian blinds, with many pre- 

 cautionary details which need not be entered into, but after all, 

 the position cannot be deemed perfectly satisfactory ; in the. day- 

 time they are read by a telescope through a window, but at night 

 the observer is compelled to approach them with a light and read 

 with the naked eye : this plan, which is very objectionable, it is 

 intended to alter, the effect of it has undoubtedly been to make 

 the night-t3mperatures higher than they should be, the thermom- 

 eters being so sensitive that the heat of the observer's person 

 during even the short period necessary for reading them, causes 

 the_ mercury to rise sensibly. With regard to the difficulties 

 which arose in an early stage of the observations, from the me- 

 chanical imperfections of the instruments, I would refer to an 

 excellenl article by Captain Lefroy, in the 2nd and 3rd numbers 

 of the Canadian Journal : until very lately, those made by Eng- 

 lish makers were in general altogether untrustworthy, and in 

 fact until the establishment of the Kew Observatory, where 

 standard thermometers can now be obtained at the low rate of 

 §5 each, there could not be said to exist any English standard. 

 I would here reiterate Capt. Lefroy's warning to all who may 



be inclined to make meteorological observations, to put no confi- 

 dence in thermometers where the graduation is not cut on the 

 glass stem itself, and to remember that for the purposes of scien- 

 tific investigation, the common cheap thermometers are worse 

 than useless. 



It should also be remarked that the day in all the following 

 remarks is reckoned from 6 A. M. to 6 A. M. : the reason of this 

 is to be sought for in the original plan of the Colonial observa 

 tories. Their especial object was to procure magnetic values in 

 different parts of the world, at the same instant of absolute time, 

 and the astronomical day at Gottingen from noon to noon was 

 chosen as the day of reference for all the stations. As a connec- 

 tion between the magnetic and meteorological changes at the 

 different stations was anticipated, the latter were made simulta- 

 neous with the former, and the noon of Gottingen happening to 

 coincide very nearly with 6 A. M. at Toronto, our day is made 

 to commence with that hour. 



Coming now to the general question of variations of tempera- 

 ture, some of them i.re so obvious and familiar, as to require no 

 pointing out, such as the increase in the morning and decrease 

 towards night : by watching, however, a sensitive thermometer 

 continuously for a short time, the column of mercury is seen to 

 be almost constantly in a state of fluctuation, ascending and de- 

 scending rapidly and repeatedly under the observer's eye, still on 

 the whole, after the lapse of some minutes, sensibly altering its 

 mean position in the tube, so that we may consider it to be 

 always in a state of oscillation about a mean value which is itself 

 in a state of ascent or descent, exactly resembling the small waves 

 of the sea which alternately adv; nee and retire while the great 

 tide wave is steadily progressing. It is evident therefore, that 

 noting the indication of the thermometer at one instant of time, 

 will not give accurately the general temperature that pervades a 

 sensible period including that instart; by actually watching the 

 thermometer unceasingly, and recording all the changes that 

 occur, we should evidently be able to eliminate the small and 

 irregular variations, and deduce with certainty the mean temper- 

 ature of the period, or that temperature which sustained uniform- 

 ly throughout that period would be equivalent to the amount of 

 heat which has been expended in producing the irregular changes 

 actually occurring during the period. Such a mode of observa- 

 tion would not however be practicable, nor luckily is it necessary. 

 By noting the temperature at certain equidistant periods of time, 

 and with a frequency which experience has shown to be sufficient 

 we may be certain that the effects of the irregular variations will 

 sensibly counterbalance each other and leave us a mean result 

 freed from their influence. Thus, experience has shewn that 

 observations made every hour, or every two hours, will give 

 with accuracy the mean temperature for the day ; and the mean 

 of all the daily means thus obtained in a year, will give with 

 accuracy the mean temperature of that year, and the mean of a 

 few years will give the mean annual temperature of the place. 

 So also the mean of these annual temperatures at a considerable 

 number of places on the same parallel of latitude gives the 

 annual temperature for that latitude, and by extending the same 

 process over all the parallels, we obtain finally the mean annual 

 temperature of the whole surface of the earth, and can watch the 

 changes that occur in it from year to year. When, however, we 

 descend to the changes at the same place during the year, and com- 

 pare them with those occurring at other places, the above process is 

 no longer applicable. For this purpose the mean daily tempera- 

 tures observed for a considerable number of years, are collected into 

 monthly averages, and the intermediate temperatures are interpo- 

 lated between these by Bessel's formula, so as to obtain the mean 

 or normal annual march of the temperature. Again when the 

 daily observations are taken at fewer times than hourly or two- 



