190 ANNUAL OF SCIENTIFIC DISCOVERY. 



concentrates the sounds which enter by its base, and the vibrations of the 

 membrane thus produced are written by the pencil upon a paper coated with 

 lamp-black, which is uniformly passed under the pencil by clock-work. The 

 traces thus produced may be copied and preserved (magnified if necessary) 

 by photography. 



When the common accord was sounded on different instruments, the 

 figures formed were very different both in form and dimensions, according 

 as wind instruments, stringed instruments, or the human voice were used. 

 The same differences were seen when the record of singing was compared 

 with that of unmusical noises. M. Scott established this curious fact, that 

 the series of vibrations formed by the sound of an instrument or voice was 

 more regular, even, and consequently more nearly isochromous, in propor- 

 tion as it is more pure and agreeable to the ear. In shrill cries, and harsh 

 sounds of instruments, the waves of condensation are irregular, unequal, 

 and not isochromous. In one experiment, it was shown that, in the impure 

 sounds of the voice, two, and sometimes three, secondazy sets of vibrations 

 could be detected, combined with the principal. 



THE AIR-BALANCE: A NEW FORM OF BAROMETER FOR WEIGHING 

 ATMOSPHERIC FLUCTUATIONS. BY J. B. JAMES, M. D. 



Reflections upon the action of a barometer said to have been in use for 

 some time in the observatories of Liverpool and Rome, which has that por- 

 tion of the tube not immediately immersed balanced upon a lever, has sug- 

 gested an arrangement, which, it is believed, is new, and by which minute 

 changes in the pressure of the atmosphere may be weighed with great 

 accuracy. Let The barometer tube be fixed over a platform, with its lower 

 end freely exposed; let the platform also support a balance, one arm of 

 which, being placed directly under the lower end of the tube, supports the 

 reservoir containing the surplus mercury; in this the lower end of the tube 

 is immersed in the usual manner. If, now, the other arm of the balance be 

 weighed to counterpoise the reservoir and the mercury not supported by the 

 atmosphere, the alteration which change of atmospheric pressure renders 

 necessary in the counterpoise, indicates the change of pressure which has 

 occurred. 



Several circumstances require attention, besides delicacy in the balance 

 and accuracy in the weights, to secure accuracy in the results. 1st. The 

 tube must be of uniform capacity throughout its length, excepting so much as 

 is immersed in the mercury; varying capacity in the different portions of the 

 tube will be productive of error by the alterations which change of tempera- 

 ture produces, even if the vacuum chamber be uniform. 2d. An alteration 

 of either the temperature of the mercury or the atmospheric pressure, will 

 require a readjustment at each successive observation, by elevating or de- 

 pressing the support of the balance or the tube in such manner as to cause 

 the immersion of the tube in the mercury to the same point to which the 

 first adjustment was made; the immersion of more or less of the tube will 

 cause an error in the result. For this purpose two balances may be used; 

 upon one the tube may be balanced in the manner first alluded to, by which 

 the point of immersion is determined, the reservoir-balance being fixed in 

 its normal position for that purpose. The tube-balance being now fixed, the 

 other is freed, and brought to the same position by weights ; thus the mercury 

 need not be seen in any part of the operation.- This adjustment would prob- 



