Utt 



HXTKOBOLOOT. 



[TCX TU2RMOMETBK. 



Fig. 20. 



o 



space*, Mh of theM becomes a representative of heat or 

 temperature. Supposing we auuuie the temperature at 

 which water freeses to be our starting-point or first 

 limit, and the temperature at which water boils to be 

 our other limit ; and supposing, furthermore, we assume 

 the space between the two to bo divided into any given 

 number of part* say, for example, 180 then we may 

 describe any third body, the temperature of which is be- 

 tween tin- tcnii>erature of freezing and the temperature 

 iling, to be one, two, or any number of parts above 

 the former or below the latter. Thus, by applying these 

 principles, we should have constituted the thermometer 

 r heat-measurer. If the rate of expansion of any one 

 body for given increments of temperature were regular 

 and well determined, there would be no theoretical diffi- 

 culties in the way of making a thermometer ; practical 

 difficulties there would be, but the hypothetical part of 

 the task would be sufficiently easy. It so happens, how- 

 ever, that the number of expansive agents capable of em- 

 ployment for this purpose is limited. 



The earliest thermometer was that of Sanctorini. Its 

 construction is represented in the following diagram. A 

 glass stem, open at one extremity, is terminated at the 

 other by a bulb (Fig. 20). Into the bulb and a portion 

 of the tube is poured a coloured 

 fluid, which being done, the 

 stem is inverted into the lower 

 vessel. By virtue of the ordi- 

 nary laws of hydrostatics, tho 

 level of the fluid in tha stem will 

 remain constant for very con- 

 stant temperature ; but inas- 

 much as every increment of heat 

 will cause the air contained in the 

 bulb to expand, so will it neces- 

 sarily cause the coloured fluid^ 

 which latter merely serves as an 

 index to descend in such a man- 

 ner, that were the ratios of suc- 

 cessive equal linear measures of 

 descent equal, the instrument 

 would be a no less delicate mea- 

 surer of variations of temperature 

 than it is a delicate indicator of 

 tho same. For certain reasons, 

 now to be described, it is not a 

 delicate heat-measurer. Its suc- 

 cessive linear columnar measure- 

 ments are not comparable among 

 themselves ; whence it follows, 

 that the instrument is not a ther- 

 mometer or heat-measurer, but a thermoscope or heat- 

 iiidicator. 



Concerning the reasons wherefore the instrument just 

 described is not a perfect instrument, they readily admit 

 of being made evident. They are immediately referable 

 to the fact, that the coloured fluid, which, according to 

 the necessities of the experiments should be dynamically 

 passive, is really active. Its activity, moreover, is a 

 variable quantity. If the coloured fluid were merely an 

 index, having no dynamical power of its own, then the 

 total increments of expansion and contraction of the air, 

 contained in the bulb and part of the stem, would be 

 proportionate to the increments of heat and cold within 

 no small a deficit of tho truth (see page 1127), that the 

 error need not enter into calculation ; but examination 

 of the structure of the instrument will show that this 

 cannot be so. Actually, the total expansion of the air in 

 tli.- 1'iilb is the resultant of two forces the force of aerial 

 elasticity due to heat, and the force of pressure or down- 

 ward tendency of the columnar liquid. Inasmuch, 

 , as the columnar height of the liquid in ques- 

 tion varies for every temperature ; and, moreover, as the 

 rising and falling of the liquid in the reservoir, or lower 

 receiving-Tewel, also confuse* the result, the heat-deter- 

 mining instrument of Sanctorini is not a correct mea- 

 surer of temiwratore. 



The- Jtiffrmttial Thermometer. Although various 

 forms of air-thermoniaters are occasionally used in 



conducting certain specific experiments, their use is 

 rare. Almost the only form of air-thermometer in fro. 

 iiueiit use is the differential thermometer, an instrument 

 the function of which is to determine the ditferenoe be- 

 tween the temperature of any two adjacent bodies, or of 

 the adjacent parts of any one body, without informing 

 the observer concerning the actual temperature of 

 either. 



The differential thermometer is represented by tho ac- 

 companying diagram (Fig. 21). It consists of a glass 

 Fig. 11. tube having a bulbular ex- 



pansion at each extremity, 

 and joined by a item bent 

 on itself twice at right 

 angles, so that tho two 

 bulbs look upwards. Dur- 

 ing the process of ma- 

 nufacturing this instru- 

 ment, whilst one of tho 

 bulbs is yet unclosed, 

 liquid is poured into the 

 instrument just enough in 

 quantity to reach a little 

 way up the vertical part 

 of the stems ; each of 

 these vertical parts are sup- 

 plied with a scale-division 

 of equal lengths or de- 

 grees. Looking at this 

 instrument, the ob 

 will readily see that, supposing the tension or expansive 

 force of the atmosphere in each bulb to be equal, the in- 

 dex fluid in the stem will stand at precisely the same 

 elevation in both vortical stems ; but supposing tho air in 

 one bulb to become heated to a higher decree, than the 

 air in the other; supposing, in other words, its expan- 

 sion to be greater, the corresponding columnar h< 

 will be diminished, and necessarily, the opposite colum- 

 nar height will be raised. Hence the difference between 

 the two columnar heights will indicate that of the tem- 

 perature of the two bulbs, expressed in equal parts of 

 columnar measurement. 



Thermometer Scale*, and Ordinary Thermometers. Tho 

 liquids ordinarily employed in the manufacture of ther- 

 mometers are mercury and alcohol. The former is pre- 

 ferred to all others, when its use is practicable, on 

 account of the comparatively equal expansion to which 

 it is subject for equal grades of temperature. In the 

 manufacture of thermometers, however, intended to be 

 employed for the measurement of temperatures below 

 the freezing point of mercury, that fluid is necessarily 

 inapplicable. Spirit, or alcohol, has never been fi 

 by the most intense cold yet produced, therefore it is 

 substituted for mercury on such occasions. Wo shall 

 now proceed to detail the successive steps in the manu- 

 facture and graduation of thermometers. 



Under the head of Barometer, the inconvenience was 

 pointed out of using a gloss tube of small diameter, lie- 

 cause of the interference resulting from tho expansion 

 and contraction of mercury by heat. Now that which is 

 a cause of embarrassment in the barometer, is the func- 

 tion on which the action of the thermometer depends. 

 .It follows, therefore, that in proportion as the bore of a 

 thermometer tube is smaller, so is the resulting in- 

 struments more delicate, because greater linear incre- 

 ments of expansion will be generated for given amounts 

 of temperature. However, it happens in practice, that 

 if the diameter of the tube, and the diameter of the mer- 

 curial column contained in the tube, be smaller than 

 tain limits, it is difficult to be seen. It is hardly neces- 

 sary to indicate, moreover, that the length of linear 

 expansion may be increased to any given limit, by 

 increasing tho dimensions of tho corresponding bulb, or 

 mercurial reservoir. Practice alone con determine tho 

 proper relation which should subsist between the bore of 

 a thermometer tube, and its corresponding bulb. 



The following directions, for tho manufacture of a ther- 

 mometer, are not intended to enable tho meteorological 

 student to usurp tho functions of tho mathematical 





