58 Transactions of the American Institute. 



mercury. We find here, therefore, that original instrument was very- 

 simple, and yet it demonstrated to all eyes that the air has weight, 

 and that it could give the measure of that weight. The value or 

 weight of the column of mercury (if you supposed the surface exposed 

 to be about one square inch) gives about fifteen pounds to the square 

 inch as the weight of the atmosphere. This, therefore, is the original 

 instrument which afterwards was modified into the form here repre- 

 sented. All these contrivances are simply to measure the difference 

 between the surface outside the column and the surface at the top of 

 the column. (The Professor here pointed to the diagrams of the vari- 

 ous forms of the barometer, and gave explanations in regard to the 

 construction of each.) Of late years there have been introduced 

 into this country a number of what are called "Aneroid Barometers." 

 The name indicates what the instrument purports to be. It is a box 

 which is supposed to be void of air, and therefore very sensitive to 

 the pressure of the atmosphere ; as the pressure of the atmosphere 

 increases the box becomes flatter ; but as the pressure becomes less the 

 box expands again. This motion is shown by means of an index. No 

 mercury is used, but merely a metal box. Another form of barometer 

 is a double band of metal ; it uncoils when the pressure increases, and 

 coils as it is removed, and, therefore, by means of a hand, it indicates the 

 weight of the air. This instrument is also very sensitive, and would be 

 excellent if the expansion of the metal were not a constantly changing 

 element. We are never sure of what we are doing with an instru- 

 ment of this kind. The variations between one and another of these 

 instruments are very great ; every instrument must be tested for itself, 

 both for temperature and otherwise. The general correction for the 

 reading of the common barometer, which must be made for the 

 expansion of the mercury, is about the 3,000th of an inch for every 

 degree of temperature, but this cannot be applied here, and that is 

 the great difficulty with the Aneroid barometer. And besides, it may 

 be deranged by a simple knock, or by being exposed to a great and 

 sudden change of temperature. And it does not tell you that it is 

 out of order, as the other form of barometer does. You see at once 

 when the mercurial barometer is out of order ; and here you cannot 

 see. It wants to be constantly checked by the mercurial barometer, 

 and therefore it is not an independent instrument. I carried one of 

 them to the top of Mount Washington, but I found it nearly useless ; 

 in fact I have discarded them altogether. The Aneroid is excellent 

 at home, in the street ; but it is a bad traveler. Now, how shall we 



