716 
NATURE 
[FEBRUARY 26, 1914 
sphere. The. information is given on the back 
of the daily bulletin, and the Weather Bureau 
is to be congratulated upon being the first to 
publish a map showing the distribution of pres- 
sure and temperature over a hemisphere on the 
day of issue. 
It rests with the bureau, or with some still 
more enterprising institute if there be one, to 
add the available observations from the southern 
hemisphere, and realise what everyone who thinks 
about the subject knows to be the most sure basis 
for the study of the daily weather, viz., a daily 
map of the main features of the distribution of 
pressure and temperature over the globe. 
Practically no lines are drawn on these maps 
for latitudes lower than 25°, and it is interesting 
to speculate as to what sort of characteristics a 
synoptic chart of the equatorial regions would 
show if it could be drawn. North of 25° the 
rotation of the earth makes it possible for pres- 
sure differences represented by “parallel isobars” 
to be sufficiently permanent to be charted, while 
ordinary centrifugal action makes “circular” 
isobars also equally possible. Hence on a chart 
for temperate and polar regions, isobars may 
take any shape between the small circle of a 
cyclonic depression and the great circle of 
“straight”? isobars; but in the equatorial region 
there is no place for “parallel isobars,” as they 
are understood further north, because the influ- 
ence of the rotation of the earth is too feeble; 
the winds required to balance isobars such as 
those to which we are accustomed would be 
prodigious. Consequently a pressure distribution 
sufficiently permanent to be mapped could only 
be made up of “circular” isobars, and therefore 
a chart of isobars for part of the equatorial region 
ought to be a collection of small circles with what- 
ever may be necessary to. represent the diurnal 
variation. It would be interesting to have this 
conclusion verified, and the transition between 
the region of circular isobars and the region of 
straight isobars carefully explored. 
Variations of pressure, small in magnitude, but 
associated with weather changes, are shown as 
irregularities in the course of the well-known 
diurnal variation, on barograms for equatorial 
regions, and the translation of a collection of baro- 
grams into synoptic charts is an attractive 
problem. It would presumably tell us what the 
meteorological conditions would be if the earth 
were fixed and the sun went round it in twenty- 
four hours as the ancients used to suppose. 
One of the striking features of the maps 
now issued by the Weather Bureau is that for 
the first time in the history of official meteoro- 
logical institutions, c.g.s. units of pressure and 
the absolute scale of temperature are used for a 
daily issue of charts. The isobars are figured for 
every five millibars, and the isotherms for every 
ten or five degrees on the centigrade scale 
measured from 273° below the freezing point of 
water. 
This is indeed a remarkable step towards the 
unification of the methods of expressing pressure 
NO. 2313, VOL. 92] 
over the globe, and it has been immediately fol- 
lowed by the Meteorological Office in the corre-_ 
sponding charts which are published in the weekly 
weather report. The office’ figures the centibars, © 
while the bureau figures the millibars, but that 
is only a matter of decimal point. ; ; 
Millibars are in future to be used, though not 
exclusively, for the international publication of the 
results of the investigation of the upper air, so that 
while it now seems likely that before many years 
are passed we may see a daily synchronous chart 
for the globe, and really begin to study weather as 
it ought to be studied, we may at the same time 
expect to take leave of the inch and the milli- 
metre as measures of pressure. They certainly 
have had a very long innings on a side to which 
they did not properly belong, and it will be in-— 
teresting to see how the more scientific measure 
of pressure in pressure-units will adapt itself to 
practical requirements. The Meteorological 
Office is to make use of c.g.s. units of pressure 
for the Daily Weather Report on May 1 of the 
current year, and the preparations for that event 
have already placed some well-known facts in a 
curious light. The task which during the last 
sixty years we have been setting to British 
instrument-makers is as _ follows :—‘‘ Construct 
a barometer which will give a true pressure 
reading when the whole instrument is in lati- 
tude 45°, the mercury at 273° A., and its brass 
case at 290° A.” Continental makers have had 
a problem that sounds simpler, viz. to.construct 
a, barometer which will give a true pressure read-— 
ing when the instrument and its case are in lati- 
tude 45° at 273°A. The figures show that if 
instrument-makers were to make a barometer 
which was correct at the equator at the freezing 
point of water, it would be correct in latitude 45° 
at the ordinary air-temperature of 289° A. (61° F.) 
and at the poles at 305° A. (89°6° F.). So for each 
latitude there would be a temperature within the 
common range for which the readings were true 
pressures. At other temperatures, of course, a 
correction would be required. WiriNpos 
THE RECENT SEISMOLOGICAL DISTURB- 
ANCES IN SOUTH JAPAN. 
Gs the accompanying figure is reproduced, on 
about half the original scale, an interesting 
seismogram received from Prof. A. Belar, of 
Laibach, through the courtesy of the foreign 
editor of the Daily Mail. The earthquake in 
question occurred on the morning of January 12, 
and was recorded by a Galitzin seismograph with 
electromagnetic damping. The times indicated 
on the diagram are referred to mean time of 
central Europe, which is one hour in advance of 
Greenwich time. In a second letter, Prof. Belar 
gives toh. 4om. 35s. as the time of arrival at 
Laibach of the first preliminary tremors, and 
he estimates that the earthquake occurred at 
6h. 29m. 2s., p.m., Japan time.1 According to 
1 According to the data given by Prof. Belar in a more recent letter, it 
would seem that the time at the origin should be 6h. 3om. 13s. (mean time of 
135 E.) p.m., which does not differ materially from that given above. 
