On a new Theoretical Determination of the Velocity of Sound. 449 
3. The effect of the tide caused by the sun and moon is ren- 
dered insensible at the surface by the great viscosity of the 
liquid at the bounding surface, which can only be called a fluid 
by courtesy. This viscosity distributes and destroys the pressure, 
Let the foregoing hypotheses be assumed, which are quite as 
likely as any hitherto adduced, and it follows easily that the 
earth’s crust, if it have one at all, need not exceed ten miles in 
thickness. 
In conclusion, I would observe that in this controversy I have 
a natural advantage of position, which I am not prepared to re- 
linquish. I deny our knowledge of the interior of the globe; 
on this subject I maintain that our ignorance is absolute and 
necessary. If Archdeacon Pratt possesses peculiar sources of in- 
formation on this subject, let him give us the benefit of his 
knowledge ; but he may rest assured that something more is 
necessary than reiterated assertion, and that to accuse an oppo- 
nent of a fallacy which has no existence but in his own miscon- 
ception of a mathematical principle, neither convinces others, nor 
advances his own cause. 
I am, yours sincerely, 
Trinity College, Dublin, SamureL Haueuton, 
May 8, 1860. 
‘LXI. Ona new Theoretical Determination of the Velocity of Sound. 
By the Rey. Samurt Harnsuaw, M.A., Sheffield. 
To the Editors of the Philosophical Magazine and Journal. 
GENTLEMEN, 
| AM perfectly aware the problem of the propagation of sound 
is considered to have been solved; but notwithstanding this 
I venture to offer the following new solution to the notice of the 
philosophic world; because it not only Jeads to a numerical result 
quite different from any before obtained from theory, and agree- 
ing better with experiment, but likewise furnishes some new 
results of an unexpected character, and affords besides a glimpse 
into a department of nature which has hitherto remained her- 
metically sealed. Laplace’s ingenious suggestion of a change of 
temperature due to asound-wave, brought the result of theory so 
very near to that of experiment, that it has been thought un- 
reasonable to require a closer agreement. But it is confessed 
that the experiment by which the effect of a change of tem- 
perature is obtained is one that is remarkably difficult to manage, 
—one also in which errors of observation are greatly magnified 
intheresult: this is shown to be so, from the great differences 
between the results of different experimentalists ; and I think I 
may say that the requisite value of the coefficient (commonly 
