






































APRIL I4, 1923] 
Letters to the Editor. 
(Zhe Editor does not hold himself responsible for 
opinions expressed by his correspondents. Neither 
can he undertake to return, nor to correspond with 
the writers of, rejected manuscripts intended for 
this or any other part of NATURE. No notice is 
taken of anonymous communications.) 
_ The Cause of Anticyclones. 
In reply to Major Goldie’s letter in Nature of 
March 31, p. 429, the following figures may be of 
_ interest. 
Defining an anticyclone as a region where the 
barometric pressure exceeds 30°20 inches (1022°7 
mb.), there are 52 cases of anticyclones in the British 
Isles during the years 1909 to 1922 inclusive in which 
the results of registering balloon observations are 
available. 
Expressed as a departure from the mean for the 
height and date, these 52 cases give +1°6° F. at 1 km., 
>+4°5° F. at 2 km., and +5°6° F. at 3 km. for the 
mean departure in an anticyclone. At 1 km. there 
are 18 negative values, at 2 km. there are 16, and at 
3 km. there are 13. Ten instances, or only about one 
n five, show negative departures at each height. 
_ For cyclones with a barometer reading below 29-40 
hes (995°6 mb.) the corresponding mean figures are 
e611 F., —83° F., and -9°7° F., with 3, 2, and 1 
positive signs respectively, out of 27 cases. 
_ The above figures, the results of observations pub- 
ished by the Meteorological Office, do not seem to 
e to point to the conclusion that in England anti- 
cyclones are formed by pockets of cold polar air, but 
nother test is available. 
As I understand the theory of the Polar Front, 
lar air should be cold and have a low relative 
humidity ; cold because it comes from a colder lati- 
de, and dry because it is gradually warming up 
without a fresh supply of water and hence has a 
decreasing humidity. Conversely, equatorial air 
should be warm and nearly saturated. Where, then, 
P olar air lies under equatorial air and forms a dis- 
‘continuity, the inversion of temperature should be 
ssociated with an increase of relative humidity in the 
npper layer. Actually, just the reverse is the case. 
have examined the published figures for the 
continent, where records of the humidity are available 
or the years 1910 and 1911 separately; both years 
are consistent, and the combined result of nearly 300 
_ observations is as follows :— 
Percentage value to the nearest digit of observa- 
‘tions in which an increase of relative humidity 
accompanies a temperature inversion, o per cent. 
ases of inversion with no appreciable change of 
humidity, 4 per cent. Cases with distinct decrease, 
36 per cent. Cases with very distinct decrease, 22 
= cent. In these results both surface inversions 
and those over 3 km. are excluded. 
_ Since the figures are not published on a homo- 
_-geneous plan, classification is difficult, but ‘‘ No appre- 
_ ciable change’ means 5 per cent. or under, and 
_ “Distinct decrease ’’ means a fall of 20 per cent. or 
more. 
One solitary instance of an increase of humidity 
exceeding 5 per cent. (it was 6 per cent.) was found ; 
-it occurred at Vienna on December 6, 1911. 
- The continental figures are fully supported by many 
-hundreds of kite ascents in England, and they prove 
‘that cases of warm damp air overlying colder and 
drier air are practically non-existent. 
_ Thus it appears from the abundant observational 
‘material available that the lower layers of the atmo- 
sphere are almost always cold in a cyclone, and are 
NO. 2789, VOL. 111] 
NATURE 
495 
usually (three cases out of four) warm in an anti- 
cyclone ; also that when an inversion of temperature 
occurs it is nearly always associated (58 cases out of 
62) with dry rather than damp air above. 
W. H. Dives. 

Benson, Wallingford. 

Hypotheses of Continental Drift. 
In many recent discussions of Wegener’s theory 
and of other geological hypotheses, the assumption 
has been freely made that any force, however small, 
can deform the earth to any assignable extent if 
only it acts long enough. To declare that this 
assumption is incorrect is scarcely possible, in the 
absence of much more accurate knowledge of the 
physical conditions within the earth than we at 
present possess; but there is substantial evidence 
against it. 
We believe that mountains have stood for millions 
of years, indicating that the rocks at their feet can 
endure for that time stress-differences equal to the 
ressure at sea-level in the middle of a mountain. 
he strength of rocks at depths of 200 to 4oo km. 
is almost certainly less, but no geodetic observations 
indicate that the strength is insufficient to support 
an uncompensated hill 200 metres in height; in- 
equalities greater than this appear on the whole 
to be compensated, but the unexplained gravity 
anomalies remain almost the same whether we suppose 
that inequalities less than 200 metres in height are 
compensated or not. Other data, however, indicate 
that isostasy is not always perfect: Dr. Morley 
Davies has pointed out one, and I have shown that 
several otherwise uncoordinated data can be co- 
ordinated on the hypothesis that the rocks in the 
asthenosphere, though weaker than those near the 
surface, have a finite permanent strength. Accord- 
ingly, the hypothesis that the asthenosphere can be 
deformed to an unlimited extent by any small force 
acting for a long time is one to be regarded with 
great suspicion, and not to be accepted until it has 
been proved that it will account for more than 
appears to be explicable on the contrary hypothesis. 
n conjunction with this hypothesis another is 
often utilised, which can be definitely stated to be 
inconsistent with physical knowledge: namely, that 
such a small force can overcome a much larger 
force acting for the same time in the opposite direc- 
tion. In Wegener’s theory, for example, not only 
is a small force supposed to have moved America 
across the Atlantic, but also the resistance of the ocean 
bottom to deformation is supposed to have caused 
the elevation of the Rocky Mountains. Now, given 
a sufficiently weak asthenosphere and enough time, 
it would be possible to twist the outside of the earth 
over the inside to any extent. So long as the layers 
of equal density remained symmetrical about the 
polar axis, no elevation or depression of rocks 
taking place, deformation could proceed undisturbed, 
America going steadily on its way without mountain- 
building or any other phenomenon observable by 
geologists. In order that mountain-building may 
take place, however, energy must be supplied to 
taise and lower the rocks affected against gravity ; 
and to keep them in position, in spite of the tendency 
of gravity to restore the symmetrical form, the force 
required must be enough to overcome gravity and 
the strength of the surface rocks. The minimum 
stress needed to account for mountain-building is 
therefore greater than the pressure due to the weight 
of the mountain. Tidal friction and differences 
between the values of gravity at the tops and bottoms 
of continents are capable of producing stresses of 
