My correction is as follows: “The first effect of the combustion was to expand the air in the vessel, and this expansion it was that caused the accident.”
On which the author, after quoting my correction, replies, “Now you will allow me to say that here you have fallen into an error; I am perfectly correct in saying that the accident arose from the external pressure of the atmosphere; for remember that the vessel contained a mixture of oxygen and hydrogen gases, which, by combustion, immediately combined and formed water, leaving an almost perfect vacuum in the interior.”
If any one entertain a doubt as to which of us is correct, I would suggest his filling a small bladder with the proper mixture of oxygen and hydrogen, and exploding it by electrical means; as I did nearly sixty years ago. The bladder will be destroyed; but, according to Dr. Parris’s view, it should simply collapse.
But even men of unquestionable scientific knowledge are not always correct. The late Professor Phillips, in his able and interesting Address as President of the British Association in 1865, after noticing Foucault’s recent admeasurement of the velocity of light, proceeded as follows:—“By this experiment the velocity of light appears to be less, sensibly less, than was previously admitted; and this conclusion is of the highest interest. For, as by assuming too long a radius for the orbit of Jupiter, the calculated rate of light-movement was too great; so now, by employing the more exact rate and the same measures of time, we can correct the estimated distance of Jupiter and all the other planets from the Sun.”[371]
Professor Phillips’s great forte was geology, not astronomy. To any one familiar with the means by which Römer determined the velocity of light, it is unnecessary to point out that, although his observations were made on the satellites of Jupiter, the radius of Jupiter’s orbit has nothing to do with the problem. The only material facts are, first, the difference between the maximum and minimum distance of Jupiter from the earth,—that is to say (disregarding eccentricity) the diameter of the earth’s orbit; and, secondly, the effect which this varying distance has on the times at which the eclipses apparently take place. This effect Römer found to extend to about 16 minutes—and he thence concluded that light occupied 16 minutes in travelling across the earth’s orbit.
With the view of rendering the above intelligible to those not familiar with the subject, I offer the following illustration:—Suppose it to be known that about a certain hour a gun will be fired at a remote spot, the direction of which, but not the distance, is known, and that two persons (A. and B.) arrange to avail themselves of the opportunity for ascertaining, approximately, the velocity of sound; then, each being furnished with a good watch marking seconds, A. places himself at a certain spot, and B. at a known distance—say a mile—from A., and in a direction opposite to that of the gun, so that B.’s distance from the gun shall be a mile greater than A.’s—the actual distance in either case is unimportant.
Each now records the exact moment at which he hears the report; and if the gun be fired repeatedly, several such records are made, in order to give a more accurate result.
A. and B. then meet and compare notes. They, of course, find that A.’s time is in each instance earlier than B.’s. The average of the several differences would be about 4¾ seconds—showing that sound travels a mile in that time.[372]
The mode of procedure here described is, of course, not that actually adopted for determining the velocity of sound, but it is a practicable mode, and is selected because it is analogous to that adopted by Römer for determining the velocity of light.