114 INIr. Galbraith on Ike Velocity of Sound. 



On considering the results obtained from this comparison, 

 it may be noticed that the greatest deviation of the formula 

 ti'om observation is about 14 feet, sometimes in excess and 

 sometimes in defect. To what cause to attribute this it is at 

 present difficult to say. 



Future observation will perhaps show how much ought to 

 be attributed to the formula, to the error of observation, to the 

 effects of wind, and to the variable nature and constitution of 

 the atmosphere, which our best instruments are yet unable to 

 detect. This is perhaps countenanced by a comparison of 

 the September and October observations, in which the pres- 

 sure, temperature, and moisture are very nearly the same; yet 

 the velocities differ no less than 24 feet, while the results by 

 the formula differ only about half a foot *. 



This is the season at which the monsoons change their di- 

 rection : and as from an examination of the plan of the grounds 

 accompanying the observations, it may be inferred that the 

 wind in the one case tended to augment the apparent velocity 

 of sound, w hile in the other it tended to diminish it, we may 

 reasonably conclude that this is in a great degree the cause of 

 the discrepanc3\ It is the more likely, as in February and 

 March the same circumstances occur, with nearly the same 

 effects, though in a somewhat slighter degree. 



Mr. Goldingham's allowances for pressure and moisture 

 differ considerably from those stated here, though that for 

 temperature is nearly the same. He makes the variation for 

 1° of the thermometer 1*2 foot, for 1° of the hygrometer 1*4 

 foot, and for x'jjth of an inch of the barometer 9'2 feet ! and 

 the effect of the wind from 10 to 20 feet. 



The allowance for the hygrometer is only one half of what 

 we have made it ; and that for pi-essure ahowtjive times greater ! 

 Perhaps, as the change of barometric pressure is so small in 

 that climate, it is difficult to deduce from observation the pro- 

 per allowance for it. 



I conclude by expressing a hope that some experimenter, 

 provided with all the necessary instruments to ensure the ut- 

 most possible accuracy, will undertake a series of experiments 



* The paper was written thus far before the author knew at what Mr. 

 Goldingham's allowances were estimated, the foregoing being all derived 

 from the abstract subjoined to Dr. Gregory's paper in the Cambridge Phi- 

 losophical Society's Transactions. Since that time Moll's experiments 

 have appeared in the Philosophical Transactions of the Royal Society, em- 

 bracing all the niceties of the subject in a very neat and accurate manner. 

 They might have been susposed to supersede what I have said here on the 

 subject ; but as the paper was drawn up, and may be useful in certain cases, 

 it may perhaps deserve a place in the Philosophical Magazine. 



under 



