1850.] 



THE CIVIL ENGINEER AND ARCHITECT'S JOURNAL. 



227 



the ivory point; the same action either raises or depresses the 

 column of mercury, and, as the extremity of the ivory point is the 

 zero of the scale, the reading «ill show the real heif;ht of tlie 

 mercurial column above the surface of the liquid metal in the 

 cistern, subject to only two corrections — viz., that of temperature 

 and capillary action. The barometer is attached to a mahogany 

 slab, projecting forward about two inches, and is free to turn on 

 its axis in any direction. In reading off the scale a moveable ring 

 is made to form a tangent to the curved surface of tlie mercury in 

 the tube, a piece of white paper to reflect the light being placed 

 behind it; with very little trouble the temperature and capacity 

 correction may be combined in one, together with a small zero cor- 

 rection to reduce it to the Greenwicli standard. When a table is 

 thus formed, the absolute height of the barometric column may be 

 ascertained from the reading by the application of one coriection 

 only, and that with the accuracy wliich has hitherto been attained 

 by standard barometers at three times the price. As Mr. Barrow 

 makes twelve of these at the same time, there is a fair chance of 

 their readings being identical; and this is reduced almost to a 

 certainty by the pains which .Mr. Glaisher takes in comparing them 

 with the same standard, and supplying the zero correction, which 

 is never more than a few thousandths of an inch. 



An e.\ample from a series of comparisons, lately taken with a 

 barometer by Newman (very excellent of its kind), to determine 

 its zero correction, and one of this construction, will show how 

 much labour is saved in the reduction of observations taken by this 

 improved instrument. 



29375 



— -051 terapcralure + capillarity 



Newman't. 

 29-375 reading of the scale 

 — •018 capacity 



29-357 



+ -032 capillarity 



29-325 corrected Ijeight 



29-389 



— •061 temperature 



29325 corrected height 



With some labour it would be possible to combine all the correc- 

 tions for Newman's barometer in one — a plan 1 am about to adopt 

 for the reduction of three years' observations with the instrument 

 which was made purposely for me. In lieu of the glass cistern and 

 leather bag, a double iron cistern, with a solid bottom, is intro- 

 duced; and, with great simplicity, the mercury is secured for 

 travelling by stopping off the greater portion after the instrument 

 is inverted. It has accompanied me several hundred miles, and, 

 without requiring special care, has returned uninjured. I have 

 applied it to the purpose of measuring heights with great success, 

 and have the utmost confidence in its indications. 



II. — Dry and Wet Bulb Thermometers. 



To render observations of the temperature of different places of 

 any value to science, the instruments employed should be of the 

 most accurate construction — the indications of the thermometers, 

 for example, should be worthy of reliance to the tenth part of a 

 degree: it is difficult, but not impossible, to attain to such accu- 

 racy. Out of twenty-five thermometers made for Mr. Glaisher, in 

 1843, by Watkins and Hill, twenty of tliem agreed witliin one- 

 tenth of a degree at every part of the scale — the extreme dif- 

 ference between the readings of the remainder was half-a-degree. 

 In his experiments on the radiation of heat from the earth at night 

 (Phil. Trans., Part II., 184.7) he mentions his jiossessing at one time 

 upward of fifty instruments, whose extreme difference of reading 

 from the standard was a constant quantity of half-a-degree in one 

 thermometer, and of 0-2° in three others, the remainder being 

 absolutely free from error. 



These facts are mentioned to show the possibility of obtaining 

 perfect instruments, but especially as a caution against the pur- 

 chase of thermometers which have not been tested, or whose 

 performance has not been guaranteed by a maker of high character: 

 it is evident that no comparison of tlie mean temperature of dif- 

 ferent places can be made with instruments whose errors are not 

 reduced within very narrow limits. Consideration of certain 

 sources of error in constructing thermometers will at once show 

 that those exhibited in shops, and sold for a trifling sum, have the 

 name by courtesy and not by desert. To ensure delicacy in the 

 construction of a thermometer, many niceties demand attention- 

 two especially may be worth mentioning. 



]. The bore of the tube should be uniform throughout: if it be 



not, the length of the degrees will not be the same at every part of 

 the stem. 



2. The observer should be fully persuaded of the accuracy of 

 the scale which is adapted to the instrument, both as regards the 

 zero point and the uniformity of the divisions. The zero point is 

 ascertained by plunging the thermometer into melting ice, when 

 its reading ought to be 32°. But it is a singular fact, that this zero 

 point may be found to change, and hence it should be tested from 

 time to time, .ind an index error allowed should any alteration 

 have taken place. The thermometers used by the observers whose 

 returns are published quarterly by the Registrar-General, have, for 

 the most part, been examined by Mr. Glashier, and compared with 

 a standard; and thus, as far as possible, uniformity of result is 

 provided for: a character is given with the instrument, consisting 

 of a series of comparisons with the standard, which is thus accom- 

 plished. The two thermometers are plunged into water of the 

 temperature of the highest reading of the one whose character is 

 the desideratum, and a series of comparative readings is taken as 

 tlie temperature lowers. The differences will supply corrections to 

 be applied at the various temperatures to reduce the reading! to 

 the standard; but in those supplied by Barrow, after Mr. Glashier's 

 comparison, the differences are so minute that they may be safely 

 disregarded in practice. 



The wet and dry bulb thermometers are simply two thermome- 

 ters side by side, which are presumed under the same circum- 

 stances to give similar indications. The dry thermometer, of 

 course, shows the temperature of the air ; the wet thermometer 

 has its bulb surrounded with muslin, and from it lead a few inches 

 of lamp-wick into a small vessel of water: the reading of this 

 latter will in general be below the dry, and a comparison of the 

 two will supply data for ascertaining the hygrometric state of the 

 .atmosphere. Without entering too far into the subject, a few 

 words of explanation may be desirable. 



Under general circumstances the atmosphere will take up the 

 vapour of water: the drier it is the more rapidly will evaporation 

 proceed, and the more slowly as its condition approaches that of 

 complete saturation. When in that state no more moisture is 

 capable of being held in suspension. Now, as evaporation proceeds, 

 heat is absorbed by the conversion of the water around the bulb of 

 the thermometer into vapour, and the mercury in the wet bulb 

 will fall a greater or less number of degrees according to the dry- 

 ness of the atmosphere. When the air is saturated, the readings 

 will be the same. In Mr. Glaisher's ' Flygrometrical Tables' the sub- 

 ject of the wet and dry bulb thermometers is fully discussed, and, 

 by their assistance, various interesting particulars may be deduced 

 from the simple record of the different readings of the two. The 

 most important of these deductions is the temperature of the dew- 

 point, or that degree at which the atmosphere will part with its 

 moisture, or will be cooled down to the point of saturation. The 

 capacity of air for holding aqueous vapour in suspension diminishes 

 with the abstraction of heat. The dew-point is that degree of 

 temperature at which saturation is attained and moisture deposited. 

 The difference between this and the temperature of the air has 

 been investigated, and formula; have been given by it which may 

 be deduced from observations with the wet and dry bulb thermome- 

 ters; but in Glaisher's 'Hygrometrical Tables' the dew-point may be 

 found at sight from the readings of the wet and dry bulb; and I 

 believe I am correct in stating that these results have been obtained 

 from observation and not from theory. 



The elegant hygrometer of the late Professor Daniel gives the 

 dew-point by inspection; but, as it is attended with some incon- 

 venience and expenditure of time, it has net come into general use. 

 IMany observers, like myself, occasionally verify, by means of 

 simultaneoiis observations with it, the deductions from the dry and 

 wet bulb thermometers. 



The phenomenon of the dew-point may be illustrated by refer- 

 ence to the affairs of common life. A bottle of wine, to be rendered 

 more agreeable, has been iced before its appearance in the dining- 

 room. You will find that the bottle will be covered with a coating 

 of dew the moment it enters the room; the temperature of its con- 

 tents being far below the point of saturation, the watery vapour 

 from the atmosphere will be condensed on the surface. — ^I visited 

 some time since the observatory of a distinguished astronomical 

 friends— every instrument in it was streaming with moisture. "The 

 great drawback to my position," said he, "is the neighbourhood of 

 yonder piece of water — see the effect." Upon inquiry, I learnt that, 

 on the preceding evening, the observatory had been open; the 

 instruments had been cooled down to the night temperature, and 

 the day chancing to be much warmer than usual, they had not had 

 time to get heated above the temperature of the dew-point, and 



31* 



