Gregory O
Gregory, Olinthus, "An Account of some Experiments made in order to
determine the Velocity with which Sound is transmitted in the Atmosphere",
in 'The Philosophical Magazine', 30 June 1824, Vol. 63, No. 314: pp. 401-5
(Written at Royal Military Academy, Woolwich, on 25 October 1823)
Investigating speed of sound, as previous results differed widely: eg.
Feet per second
Gassendi - 1569
Roberts - 1300
Boyle - 1200
Walker & Duhamel - 1338
Mersenne - 1474
Florence Academy- 1148
Cassini de Thury - 1107 (Paris Academy of Sciences, 1738)
Meyer - 1105
Derham - 1142
Muller - 1109
Pictet - 1130
Arago (1822) - 1106.32 - most recent before Gregory's own
(Actual is 1086.9 ft per second at 32 degrees F - obtained in 1986)
Gregory deduced from this that current theoretical formula then in use,
which includes variations of temperature, can only be regarded as
approximate, and before better one available, only thing to do was to
"augment the number of recorded facts, as they result from accurate
experiments, in order that at some future time a cautious investigator may
so select, compare and classify them, as to deduce a more comprehensive and
accurate theorem than is yet known."
To this end, Gregory did experiments in 1823: object was to ascertain
velocity with which the sound passed over the surface of land and water;
under different temperatures; in still and windy weather; by day and by
night; the velocities of direct and reflected sound; and the velocities of
sounds of different intensities and produced by different means. Instrument
he used to measure intervals of time "was one invented and made by Mr Hardy"
- allowed measurement of interval accurately to 10th of a second, and
approx. to 20th of a second. Velocity of wind was measured by use of an
anemometer, and also used barometer and thermometer. Detailed notes of
results of repeated experiments taken.
3 Jan, 1824: Did experiments from battery near the Royal Artillery Barracks
on Woolwich Common - part of Royal MilitaryAcademy and Royal Arsenal
complex that moved here in 1806 into building by architect James Wyatt -
William Congreve of Congreve rocket was comptroller of the Royal Laboratory
of Woolwich Arsenal at this time. Musquet fired from the battery, and
interval between flash and sound was observed at 2 different distances on
the mortar-range - 3600 ft and 2100 ft distance, while also measuring
temperature, atmosphere. Musquet then used on 18 August, with Gregory
stationed on the Essex bank of the Thames perpendicularly opposite the large
storehouse on Roff's Wharf at Woolwich.
9 Jan, 23 May (windy day) & 7 August 1824: Followed these with use of
six-pounder cannon (using charge of 8 oz of powder) on Shooter's Hill
Station, between Severn-Droog Castle and the 8-mile stone on the Dover Road,
by permission of General Ramsey of the Royal Artillery, Commandant of the
Garrison. Selected three other stations from which gun could be seen, using
a good theodolite telescope - one of which on Blackheath (13,440 ft), one in
Kidbrook Lane (8820 ft), and third on turning from Dover Road into Charlton
Lane (6550 ft). Sergeant-major directed men to commence firing at a certain
minute by synchronized watches, and then to fire regularly a certain number
of rounds at intervals of 2 minutes. Similar exps with cannon then done on
18 August, with six-pounder gun placed upon the wharf by side of Thames in
the Royal Arsenal, with Gregory taking up station at opposite extremity of
the Gallion's Reach, not far from mouth of Barking Creek (9874 ft): during
this exp., gun was fired both towards and away from Gregory, and results
taken.
17 Feb 1824: Used bells on the mortar-range of Woolwich Common, lying nearly
north to south. A bell rung at the north station was heard by a soldier at
the south station, who immediately rang another bell, having his arm
elevated for purpose. Gregory stood by the soldier who rang first bell, and
measured interval of time between the sound of the first bell, and the sound
of the second bell when transmitted from other station.
21 August 1824: Mortars were being fired from the Royal Artillery battery,
and Gregory stationed himself 3900 ft south of it.
From all these exps, Gregory gave velocity of sound for 27 degrees = 1094.2;
33 degrees = 1099; and 35 degrees = 1102, and worked out theorem for
working out speeds at different temperatures. Not bad for accuracy: said new
evaluation "may be useful, 1st, to the military man in determining the
distance of an enemy's camp, of a fortress, a battery etc; 2nd, to the
sailor, in determining the distance fo another ship, etc; 3rd, to the
land-surveyor in ascertaining the length of base lines, etc in conducting
the survey of a lordship or county; 4th, to the philosophic observer, in
appreciating the distances of thunder-clouds during a storm." But admitted
that in all of these, must be regarded as approximate only, because few
people able to measure smaller intervals than tenths of seconds, and error
of 10th of second will produce mistake of between 37 and 40 yards in
estimate of distance: however, mean of five or six exps will give much more
accurate reading.
However, was able to conclude that i) sound moves uniformly, at least in a
horizontal direction; ii) that difference in intensity of a sound makes no
appreciable difference to its velocity; iii) difference in instrument from
which sound is emitted makes no difference to velocity; iv) that wind
greatly affects both intensity and velocity of sound; v) that when direction
of wind concurs with that of sound, the sum of their separate velocities
gives the apparent velocity of sound; whereas when the direction of the wind
opposes that of the sound, the difference of the separate velocities must be
taken; vi) that in the case of echoes, the velocity of the reflected sound
is the same as that of the direct sound; vii) that, therefore, distances may
frequently be measured by means of echoes; viii) that a rise in temperature
occasions a rise in the velocity of sound, and vice versa. Then details what
other questions his exps raise, before signing off.