190 



and as, in ascending, portions of the air are 

 continually left beneath, there results from 

 these two causes an average fall of the mercu- 

 rial column strictly corresponding .with the 

 elevations attained ; and on this principle the 

 barometer becomes a convenient and nearly 

 though thus far not absolutely correct means 

 of determining heights. Thus, the average 

 height of the column at the sea level being 

 about 30 inches, a fall of one half this length oc- 

 curs when the elevation of 3.43 miles (Cooke : 

 "Chem. Physics") is reached; at double this 

 elevation the length of column is ^ that at the 

 sea level, and in the like ratio for all elevations 

 whatever ; so that by calculation the elevation 

 can always be deduced with tolerable correct- 

 ness from the barometric indications. Of course, 

 in effecting this, corrections are required for tem- 

 perature, and for any other agency which may 

 cause the atmospheric pressure at the place to 

 vary independently of the mere fact of elevation. 



In the so-called aneroid and metallic barom- 

 eters, several forms of which are now con- 

 structed, the atmospheric tension is allowed to 

 take effect directly on the thin flexible sides of 

 a small metallic case, or upon a compressible 

 tube or chamber only inside the case, and 

 which is acted on with greater force as the 

 pressure is increased, and the reverse ; a con- 

 nection by delicate mechanism from the mov- 

 able wall or chamber, as the case may be, to an 

 index moving over one face of the case, serving 

 to show the pressure ; and the indications of 

 these instruments are readily compared with 

 those of the mercurial, the corresponding values 

 being marked on a scale over which the point 

 of the index is caused to travel. 



Messrs. Home and Thornthwaite, of London, 

 have during the year produced a carefully con- 

 structed aneroid barometer, a new feature of 

 which is that it is intended to show or measure 

 heights by simple inspection. The instrument 

 is of about 2$ inches diameter ; hence, highly 

 portable. The face is graduated in two circles, 

 the outer being the usual scale, and showing 

 the pressure in inches and tenths of an inch. 

 The inner circle is graduated in spaces of 

 course decreasing in length from the greater 

 pressures up which correspond with hundreds 

 and thousands of feet. Supposing the density 

 and tension of the atmosphere not to vary with 

 temperature nor other meteorological conditions, 

 but to remain at all heights precisely that due 

 to height only, the mere observation of the 

 place of the index shows the elevation above 

 the sea level at the time, and a comparison of 

 two such indications shows the difference of 

 heights between two stations. The calculation 

 saved is that showing relation of pressure to 

 elevation. A correction for temperature is, 

 however, required; and a convenient table 

 giving the corrections for degrees of Fahren- 

 heit, and based on a formula of Laplace, is 

 furnished with the instrument. Known fluctu- 

 ations of pressure occurring independently of 

 heat and elevation should also be allowed for ; 



so that it is doubtful whether the instrument 

 can serve more conveniently than any other for 

 the most accurate measurements : though it must 

 still prove very useful for cases in which ordi- 

 narily correct measurements only are required. 



Mr. J. Ball presented before the British As- 

 sociation (1862) a quite abstruse paper on correc- 

 tions required in determining heights by means 

 of the barometer, the general bearings of 

 which, rather than those details suited only to 

 the use of the scientific explorer, will here be 

 given. Serious errors are still involved in the 

 ordinary process of reducing barometrical ob- 

 servations taken for hypsometrical purposes. 

 The process has involved two assumptions 

 neither of them wholly true : 1, that the volume 

 of a column of air unequally heated in its dif- 

 ferent parts is nearly the same as that of an 

 equal weight of air having throughout a tem- 

 perature equal to the mean of the former ; 2, 

 that the mean temperature of the column or 

 stratum of air between two unequally elevated 

 stations is equal to the mean of the readings of 

 the thermometer in the shade at the two 

 stations. The practical error arising from the 

 second of these assumptions only is important. 

 The subject has lately been under investigation 

 by M. Plantamour, the distinguished astron- 

 omer of Geneva. He first ascertained by care- 

 ful levelling the height of the Great St. Ber- 

 nard above Geneva: his comparison of the 

 barometrical observations of 18 years then 

 shows that the mean height deduced from them 

 is less by 14 English feet than the true height. 

 This result he thinks due to an abnormal 

 depression of the mean temperatures of Geneva, 

 owing to contiguity of the lake. The readings 

 of the barometer and thermometer at the ob- 

 servatories of Geneva and the St. Bernard are 

 taken at 9 hours or epochs of each day. Cal- 

 culating the height of the mountain by the 

 elements corresponding to the 9 epochs several- 

 ly, and for the months of June to September 

 inclusive, for successive years, he obtains a 

 series of differing measures of the height 

 those given by the hottest hours of the day 

 being in excess of the true height, and those 

 of the cooler hours falling short. He thereupon 

 formed a table showing the correction neces- 

 sary for each of the 9 epochs at Geneva, for 

 the months named; and introduced a further 

 and more difficult correction for the variation 

 which the temperatures may be undergoing (at 

 the moment of a given observation) from the 

 mean of the corresponding day and hour. The 

 paper will doubtless appear in the publications 

 of the Association. 



Prof. Hennessy read before the Association 

 a paper on the vertical movements of the at- 

 mosphere, as detected by a new form of ane- 

 moscope (ai/e^tor, wind, O-KOTTFW, I observe). The 

 instrument is so constructed as by an index or 

 pencil to show, and also to record, the vertical 

 oscillations of otherwise quiet,or moving bodies 

 of air. Such non-horizontal movements are 

 found to be more prevalent about midday than at 



