33 



that point upwards expands in the same ratio as it previously- 

 contracted, a fact which complicates finer researches. 



Considerable difficulties have been experienced in obtaining 

 consistent results in verifying the bushel. The chief seems to 

 be the large area of the surface (viz., 18J in. diameter), and 

 the consequent difficulty of getting a true water surface. The 

 drawing in Plate XII. shows a water-line indicator attached to a 

 bushel measure. This is, I understand, now used in preference 

 to the glass plate at the Board of Trade. The reasons of 

 errors in these verifications of the bushel measure are as 

 follows : — 



1st. Its difference of specific gravity caused by absorption 

 of air, &c, from the atmosphere. 



2nd. The evaporation of the water during the time neces- 

 sarily occupied by the weighings. 



3rd. Variations of temperature during the weighings. 



4th. Difficulty of obtaining an exact plane of the surface of 

 the water. 

 The importance of the last will be seen from the fact that a 

 difference of '001 inch in the height of the water line makes a 

 difference of 31 grains. In view of these facts it has been pro- 

 posed by Professor Miller (as the process of verifying a bushel 

 measure is most troublesome on account of the great weight of 

 water, and the necessity of filling the standard exactly up to 

 the brim) that a form of standard measure more resembling the 

 cubic foot bottle, used in gas measurement, would answer 

 better. 



GAS MEASTJEIXG. 



The Sales of Gas Act of 1859 provides that certain instru- 

 ments deposited with the Board of Trade should be regarded as 

 standards for the sale of gas, chief among which is the cubic 

 foot bottle, as shown in drawing. And the unit is thus speci- 

 fied, Clause II. : — " The only legal unit for the sale of gas by 

 meter shall be the cubic foot, containing 62'321 pounds weight 

 of distilled water, weighed in air at the temperature of 62° 

 Par., the barometer being at 30 in." And the next clause pro- 

 vides for the construction of this and derived standards, and 

 their deposit in the Standards Office. The mode of using the cubic 

 foot bottle is by placing counterbalance weights upon the chains. 

 The tank containing water is raised, and the measure is thus im- 

 mersed, and the air contents expelled through the outlet at the 

 top (Plate XII.). As soon as the water makes its appearance 

 at the neck (seen through glass plates) the outlet cock is shut 

 and a cubic foot of air has been expelled. The secondary 

 standards take the form of small gasholders of five or ten feet 

 capacity, carefully made and provided with a scale of cubic 

 feet, divided into tenths and hundredths. The principal diffi- 



