2G 



REPORT 18G7. 



0)1 the Errors of Aneroids at various Pressures. 

 B>/ Balfour Stewart, LL.D., F.B.S., Superintendent of Keiv Observatory. 



At the request of tlie Meteorological Committee experiments have lately been 

 made at Kew Observatory, with the view of ascertaining to what extent an aneroid 

 may be considered as a reliable instrameut when exposed to considerable changes 

 of pressure, such as occur in moimtain ascents. 



In order to make these experiments, a large receiver had attached to it a standard 

 barometer, of which the accuracy had been previously ascertained. By means of 

 an air-pump, the aneroids, when placed in this receiver, might be subjected to any 

 jn-essure, the exact amount of pressure being noted by the standard barometer. 

 An arrauo-ement devised by Mr. Beckley, mechanical assistant at Kew, enabled the 

 aneroids to be tapped Avhile in the receiver, so as to imitate, as well as possible, 

 the tapping of the hand, to which these instruments are usually subj ected previous 

 to their readings being taken. 



For the aneroids, to which I shall immediately refer, observations were made for 

 every inch of pressure between 30 inches and 10 inches, ten minutes being occu- 

 pied in going troni one stage to the next, and the instruments being always tapped 

 at evei-y stage. When they had reached their lowest pressure, they were kept at this 

 for an hour and a half, and were then raised in stages of 1 inch every ten minutes 

 until the ordinary atmospheric pressure was finally reached. The instruments 

 themselves were obtained from the best-known makers, w^ho kindly lent aneroids 

 for the purpose of this experiment. 



The following Table denotes the average behaviour of these instruments _ so 

 treated, eight sets of experiments having been made, and the instruments being 

 one half large instruments, diameter 4 inches, and one half small instruments, 

 diameter 2 inches. 



Supposing the instruments were quite right at starting at the pressure of 30 

 inches, then their behaviour while the pressui'c was being lowered is represented 

 by the following Table : — 



in. 



At 24 inches, error — ^02 



Froiii this Table we may learn the following facts : — 



1. If we compare an aneroid with a standard barometer before beginning our 

 observations, in order to ascertain its index error, and if we then gradually lower 

 the pressure, using the above index error, we shall find that the instrument lags 

 behind or reads rather too high Aovm to 2G inches, at which point its behaviour 

 appears to be reversed, and it falls thereafter too fast. 



2. The instrument is, however, tolerably accurate down to 24 inches, or through 

 a range of 6 inches. 



3. If we compare the aneroid -^vith a standard at the end instead of at the be- 

 ginning of the observations, we shall get much less reliable results. 



Suppose now that the instrument is allowed to remain an hour and a half at the 

 lowest pressure, and that it starts from this pressure of 19 inches, going upwards, 

 beiniT quite right at starting, as compared witli a standard barometer, then the 

 average behaviour will be represented by the following Table : — 



From Table 11. we may learn as follov^^s : — 



1. If we start from a low pressure (19 inches) and compare our aneroid wiih a 



