30 
AS) 
Inches 
MAy 2, 1884.] 
of the atmosphere, and is found to have a very 
definite position relative to the centre of the 
eyclone; and this directly confirms the ex- 
planation given by Mr. Ferrel of the persistent 
left-handed rotation of the tornado, as well as 
of its regular direction of advance. There is 
no better example than this, of the successful 
deductive study of meteorology. 
There are, of course, other theories of tornado 
action still held. The electrical, or, as it may 
be called, the vague theory is one of the most 
popular; but fortunately it is condemned by 
SS er ee 
eet 
8 
SCIENCE, 
D931 
it makes a determined resistance. There it 
survives for a time as a curiosity, a relic of 
by-gone days. 
LETTERS TO THE EDITOR. 
et Correspondents are requested to be as brief as possible. 
The writer's name is in all cases required as proof of good faith. 
Atmospheric waves from Krakatoa. 
I NOTICE, in your publication of the 14th of March, 
an account of an atmospheric wave which took place 
soon after the eruption of Mount Krakatoa. Think- 
ing that it may be of interest, I enclose a copy of a 
sheet taken from a self-registering barometer that is 
under my charge. The fluctuations shown by the 
barometric line upon this sheet are very unusual; but, 
Aug. 29. 
27, 1883, TO 9 A.M., AUG. 29. THE RECORD 
as I was in Europe at the time they occurred, I can 
12 co OSS TO SGRES kN i ala: 
Aug. 27, 1883. Aug. 28. 
RECORD OF SELF-REGISTERING BAROMETER, PROVIDENCE, R.I., FROM 8 A.M., AUG. 27, 
IS ONE-HALF LARGER THAN THE ACTUAL VARIATIONS. 
electricians. The theory of descending winds, 
or of commotions beginning in the upper air, 
and then descending to the ground with actual 
downward currents, has had not a few sup- 
porters, but now seems to be defended only 
by Mr. Faye of the French bureau of longi- 
tudes. The last Annuaire of this bureau con- 
tains a brief repetition of his Défense de la loi 
des tempétes of 1875, in which he persists in 
regarding tornadoes and storms in general as 
down-cast draughts of air, and, strangely 
enough, finds proof of his statements in the 
descriptions of western tornadoes published by 
the signal-service, which make mention of the 
‘descent of the tornado cloud.’ It is quite 
time that the downward ‘ growth’ of the cloud 
should no longer be misapprehended, and that 
the real meaning of this significant appearance, 
so long ago well explained, should be generally 
understood. But, before an error finally dis- 
appears, it is natural enough to find it restricted, 
like an organic species on the verge of extinc- 
tion, to a small habitat, like the Island of 
Mauritius, or the Bureau of longitudes, where 
only say that the sheet must explain itself, and that 
the barometer is a very sensitive and reliable one. 
EDMUND B. WESTON. 
Providence, R.I., April 16. 
Your correspondent, ‘ S.,’ in Science, No. 63, would 
seem to be wrong in attributing to the atmospheric 
waves following the Krakatoa explosion any thing 
like the character of the rapid waves of compression 
and expansion which cause sound; for this would 
be the kind of disturbance referred to as following 
the explosions of powder-mines, which disturbance 
generally takes the form of shattering glass windows, 
and is probably due to the suddenness and unusual 
amplitude of the first wave of compression, or per- 
haps to the shivering vibrations set up in the window- 
sashes, or in the whole sides of wooden buildings. 
None of these waves could, on account of their fre- 
quency, show themselves at all on barometric traces. 
In the Krakatoa waves the barograph traces, com- 
bined with the velocity of transmission, show that 
these waves must have been long, smooth swells (vary- 
ing from fifty to five hundred or six hundred miles 
in length, with the shorter waves sometimes super- 
posed upon the long ones) something like the ground- 
swell of the ocean, only with the waves much longer 
than the latter, and tr avelling in an elastic medium 
whose density and pressure vary from that at the 
earth’s surface up to zero. 
For the cause of such an unusual condition of the 
atmosphere, we must examine the results of the new 
hydrographic survey of the vicinity of Krakatoa, as 
published in Nature, 1884, Jan. 17, p. 268 (also, in 
part, in Science, No. 44, p. 211), and also the data 
