KEMARKS ON THERMOMETRTC REGISTERS. 



[1852. 



iiig with tlie boiler, is a secouj tube (f) of the same height, hut 

 lai-o-e enough to have a space of au inch or more clear between 

 the two, at the bottom of this outer tube is a small pipe fur tlie 

 escape of the steam ; it may be three-quarters of an meh in dia- 

 meter at the orifice. Lastly, a cap tube [d) closed at top, is 

 made to fit the exterior tube, and slide upon it tightly enougli to 

 stand at any height at which it may be set. In the centre of 

 the top of this latter tube is made an orifice large enough to re- 

 ceive a common quart bottle corlt, and a short neck soldered on 

 to hold it firml}-. The boiler then being half or two-thirds filled 

 with hot rain or snow water, the thermometer is passed through 

 a cork, and slipped down through the orifice just mentioned, un- 

 til the ball nearly, but not quite, dips in the water, the exterior 

 tin tube being at the same time drawn out so that no more 

 of the tube of the thei-mometer is abox'e the cork than is absolute- 

 ly necessary. The apparatus is then set over a braizer's furnace 

 or something of the same nature, until the water boils briskly 

 and the steam by degrees expels the whole of the air, and escapes 

 freely by the pipe left for the purpose. If the size of this escape 

 pipe is properly propoitioned, which does not appear to be a 

 matter of much nicety, the vessel will now be filled with steam 

 of an elasticity precisely represented by the Barometric pressure 

 at the moment; and the mercury in the thermometei', rising just 

 above the cork, will stand at the boiling point. It is desirable to 

 ascertain whether the escape tube is rightly proportioned by par- 

 tially stopping it; if two or three such alterations of its size have 

 no ™ible effect upon the reacUng of the thermometer, we may 

 be satisfied. In finished instruments a mercurial syphon guage 

 is attached. 



The follomng table, calculated by M. Regueault, contains the 

 true temperature of steam, of elasticity corresponding to the bar- 

 ometric pressures annexed; in other words, the temperature 

 which should be indicated by a thermometer plunged in the 

 steam of .such a vessel as is described above, when the elasticity 

 of such steam, (me:rsured by the exterior Barometer) is e(pial to 

 the pressure stated. The English equivalents of the French mea- 

 sures are given for convenience. 



Suppose then that the mean of several reading's of the ther- 

 mometei' over the boihng water is 2 1 1 "^ .50 ; the Barometer, i-e- 

 duced to 32 ° , giving a pressure of 29.826 at the time. The 

 temperature of steam, of elasticity equal to 29.826, we see by 

 the above table, falls between 211^.82 and 212°. 00: it will 

 be precisely 211 ° .84, which is, therefore, the true temperature, 

 and the thermometere reads ° .34 too low : this is not, how- 

 ever, the true bulling point, the pressure being less than llio 

 .standard pre.5.sure ; tlie reduction is — 



0.i8-f-s!;i;=').i6 

 The true boiling point u]ion tins therniometor, is, Iherefore, 

 211 ° .66 instead of 212 ° , showing, !is before, an em>r of gra- 

 duation of — ° .34. As one-tentli of a degree is a very .sensible 

 quantity in the scale of tliese thermoinetei-s, and the )ierfe<'t fixity 

 of the mercury in the steam, as long as the pressure remains tlic 

 same, enal)les an observation to bo made with great precision : 



the above apparatus can be employed for determining differences 

 of level, upon occasion — allowing 511 feet for the first degree, 513 

 feet for the second, 515 feet for the third — from 212 ° down- 

 wards : but the observer must be careful in this climate to choose 

 very settled weather for the purpose, or a change of the Barome- 

 ter may introduce an error larger than the quantity to be mea- 

 sured. 



The freezing and the boiling points of the standard thermome- 

 ter should be vi.-iltii'il o,-.'n<ionall3', and all other thermometers 

 carefully compni'."l witli it, at several points on the scale. So 

 little h:i\(! even tin- best instruments of the best makers, hereto- 

 fore justified in all eases, their title, or their cost, that instances 

 have been recently adduced of the Standard Thermometere of 

 more than one Observatory, being a degree or two in error at the 

 extremes of their scales. That of Toronto was found to be 1 ° .8 

 too high at — 10°, that of Makeretoun to be ° .97 too high at 

 90 ° , it cannot be therefore too strongly insisted on at the out- 

 set of Meteoi-ological Observations, that the accuracy of the in- 

 strument demands the first care. 



II. Position of T hermometers. — The object of the register 

 being to obtain the true temperature of the air of the locality ; 

 the thermometer must be guarded, first, from inffuences which 

 affect the insti'ument more than they affect the air; its power 

 of railiatidu and absm-jitiou being greatei'; secondly, from causes 

 which make the air in which it is immersed, an unsuitable ex- 

 ample of the temperature of the neighbourhood. As when, for 

 example, it is placed in a narrow court yard, with buildings soine- 

 times reffe;,'ting heat, sometimes evaporating moisture, all round 

 it. It should be placed on the north side of a building to javoid 

 reflected heat; but where there is a free circulation of air, other- 

 wise such a situation is apt to be damper, and therefore colder 

 than is natural : at the same time it should itself be secured from 

 wind. It shoultl b.- di'taeh.'d six iiuh.'s (d least, from the wall 

 or other support, and liscil, iiit liuii^-, upon a bracket. The 

 almost universal juaciic;' of Eugli^li observers is to place the bulb 

 at four feet above the ground. The official instructions to the 

 observers in Prussia, (Regent's Reports, 1850,) direct, however, 

 the height to be not less than twelve or fifteen feet, — a first floor 

 window is therefore not inadmissible, and has indeed been select- 

 ed in several instances, where local circumstances made it con- 

 \-enient, by the able superintendent of the State Meteorological 

 Observations of New York and Massachusetts, Professor Guyot. 

 The decrea,se of mean temperature, as we ascend, being only one ' 

 degree for 280 feet or thereabouts, a diftereuce of ten or twelve 

 feet would .be entirely insignificant, were it not that the air with- 

 in a few feet of the ground is w^armed by day and cooled by night, 

 by causes Avliich vary sensibly within that range. Thus a ther- 

 mometer fully exposed to the sky, at one foot abo\e long grass, 

 was found \i\ Mr. Glai.sher in his elaborate experiments on radi- 

 ation, ( Phil, trans. 1847,) to read at night 1 ° .68 lower than it 

 would have done, but for the absti-action of heat, by the g;'as.s, 

 from the air in contact with it, to compensate for its own loss of 

 heat by radiation into space. 



At two feet, it read 1.32 lower, 



four " " 1.18 lower, 



six " " 1.03 lower, 



eight " " 0.73 lower, 

 than 1h:^ standard, .and ho was led to the important conclusion, 

 that " if .a tiiermometer be freely suspended in the air with its 

 bulb at tlie height of thirteen feet above the soil, and far from 

 any obji'ct to reflect iieat to it, its readings will represent the true 

 temperature of the air at the time, and much more truly than 

 tli'isc of any one ]>lac.xl near the gi'ound, or within a few feet of 

 walls or building.s." Tlie thermometer in this situation is not 

 supposed to he jirotocted from the sun or from rain. It luus also 



