OV ARTS AND SCIENCES. 187 



" A series of experiments was first made with this aneroid barom- 

 eter, to determine the whole range of the instrument. It was placed 

 for this purpose, first under the receiver of a common air-pump, and 

 afterwards under the receiver of a condensing pump. In this way, it 

 was found capable of indicating a change of atmospheric pressure, 

 which would move the column of mercury in an ordinary barometer 

 from about twenty inches up to about thirty-one inches. From the na- 

 ture of its construction, the index is incapable of moving beyond the 

 point which corresponds to twenty inches of the mercurial barometer, or 

 beyond that which corresponds to thirty-one inches of the same. How 

 accurately its march between these limits agrees with that of the mer- 

 curial barometer will appear from an examination of Table I. The 

 pressure of the air in the receiver of the pump was derived from 

 the pump-gage, which was supplied with common mercury, and cor- 

 rected for level and capillarity. This table shows that, while the index 

 of the aneroid barometer continues to move, it moves farther than the 

 column of mercury. As it approaches its lower limit, it will begin, 

 of course, to move more slowly, and afterwards the differences be- 

 tween its indications and those of the mercury change sign. It is 

 obvious that, in this instrument at least, and with large ranges, similar 

 changes of pressure are not marked by equal quantities of motion in 

 the index, in all parts of the scale. This might be expected in an 

 instrument where no consideration is given to the distinction between 

 the potential and the apparent leverage. Besides this error, which we 

 may call the instrumental error, there appears to be an irregularity in 

 the motion of the index, arising from friction, bending, or some other 

 cause, which would interfere seriously with the accuracy of this ba- 

 rometer, even if the arc over which the index moves were so graduated 

 as to indicate the true pressure. 



" At the meeting of the British Association, in 1848, it was stated by 

 Mr. Lloyd, that one of his friends had made a similar experiment to 

 that I have described, and that the indications of the aneroid barometer 

 agreed with those of the pump-gage to within .01 of an inch. Such 

 is the statement in the London AthencBum, although no mention is made 

 of the subject in the Report of the Association for that year. As we 

 are not informed to what amount of diminished pressure the aneroid 

 barometer was subjected in this case, and whether the difference above 

 mentioned was the result of a single observation or the mean of many. 



