386 
mean temperature of the earth’s surface by 11C, the 
approximate difference of mean temperature which most 
geologists accept as differentiating between a glacial 
and an interglacial climate. However, this temperature 
differential is computed on the assumption that total 
precipitation remains unchanged during the tempera- 
ture cycle, a totally unjustified assumption. 
Simpson [20, 21] pointed out that a general lowering 
of the earth’s mean temperature by a decrease of the 
solar constant would inevitably decrease both the mois- 
ture content of the air and, by reducing the latitudinal 
temperature contrast, the intensity of the general cir- 
culation. These two effects would cause a drastic world- 
wide reduction of precipitation, which precludes the 
possibility of an ice age, rather than favoring it. No 
adequate answer to this objection has been offered by 
the proponents of a reduction of the solar constant as 
the primary cause of an ice age. 
Simpson’s conception of glaciation by increased solar 
heating rests on the assumption that the effect of this 
heating is reflected primarily by an intensification of 
atmospheric circulation, cloudiness, and precipitation. 
The mean temperature of the entire earth’s surface is 
not decreased, but heating in certain regions (sub- 
tropical high-pressure belts) and cooling in others (zones 
of convergence, storm tracks) is a necessary part of the 
intensification of the general circulation. Glaciation 
occurs in regions of increased storminess and localized 
cooling. Simpson recognizes that if the increased solar 
heating proceeds far enough, the temperature of the 
entire atmosphere must rise enough to reverse the 
trend towards glaciation abruptly, with a quick reaction 
to a warm, very wet (pluvial) interglacial period. He 
considers that the Pleistocene Epoch consisted of two 
solar maxima, each of which corresponded to a double 
glacial maximum interrupted in the middle by a short 
pluvial interglacial. The long second interglacial pre- 
sumably represented a dry period of reduced solar 
constant. 
The principal objection to Simpson’s hypothesis of 
glaciation lies in the paradox of warmer sun and cooler 
earth, at least locally. Most geologists find it very 
difficult to accept the possibility of an ice age without 
a substantial lowering of the earth’s mean temperature 
such as that caused by a decreased solar constant. 
However, there are several considerations that defi- 
nitely favor Simpson’s hypothesis over the opposite 
point of view, in particular the following: 
1. The increased intensity of the general circulation, 
with the great increase of precipitation not only in 
glaciated areas, but also in the lower middle latitudes 
and in the tropics, which is noted to characterize a 
glacial (in contrast to an interglacial) climate, certainly 
fits Simpson’s conception of the effect of an mcreased 
(as opposed to a decreased) solar constant. 
2. The statistical verification of Starr’s concept [23] 
of the low-index pattern (relatively high pressure in the 
polar latitudes) as a necessary condition for increased 
poleward heat transport definitely conforms to the low- 
index character of the glacial as opposed to the inter- 
glacial weather pattern. By Starr’s hypothesis the gla- 
COSMICAL METEOROLOGY 
cial low-index pattern manifests the necessity of an 
increased poleward transport of heat, hence presum- 
ably also the occurrence of increased solar heating. 
3. Two advantages for Simpson’s Ice Age theory, 
including a partial removal of the primary objection 
to it, ensue from the assumption that the necessary 
variation of the solar constant occurs primarily in the 
ultraviolet, increasing with an increase of the other 
solar phenomena discussed above (see pp. 379-381). 
These advantages are: 
a. Contrary to the lack of any observational evidence 
of significant variation in the visible spectrum, there is 
much direct and indirect evidence of highly erratic 
fluctuation in the ultraviolet roughly paralleling the 
sunspot cycle. This erratic sunspot fluctuation appar- 
ently runs in irregular cycles which increase in ampli- 
tude with the time range and which to some extent 
parallel the correspondingly irregular weather cycles 
(see below). 
b. The increase of solar ultraviolet does not entail 
the same increased heating of the earth’s surface and 
entire atmosphere as must a significant increase in the 
entire visible spectrum. The total absorption of the 
increased ultraviolet in the higher atmosphere may alter 
and intensify the pattern of the general circulation 
without sufficient heating of the troposphere to inter- 
fere with glaciation as Simpson conceives its occur- 
rence. 
Climatic (Postglacial) Fluctuations. There has been 
little effort made to explain the very considerable post- 
glacial fluctuations of climate. Of the currently ac- 
cepted explanations of the glacial-interglacial fluctua- 
tions, only irregular solar variability can conceivably 
be applied to the similar postglacial fluctuations of 
relatively short period. Furthermore, there are a few 
scattered observations since the twelfth century which 
indicate that solar disturbances (sunspots) have been 
exceptionally active during periods of exceptional cli- 
matic stress (13th and 14th centuries), and exceptionally 
inactive during the period of minimum climatic stress 
(later 16th and earlier 17th centuries). However, in 
general, solar observations previous to 1750 were en- 
tirely inadequate to make possible any satisfactory 
correlation with climatic conditions. On the other hand, 
the marked similarity of the postglacial to the geological 
climatic fluctuations suggests the operation of the same 
factor of climatic control (probably variation of solar 
ultraviolet radiation), and there is no alternative ex- 
planation which has been suggested. 
Secular (Intra-Century) Fluctuations. Only the secu- 
lar fluctuations of the climatic changes are short enough 
in period so that extensive comparison with observa- 
tions of variable solar activity is possible. Many studies 
of the secular variations and trends of climate have been 
undertaken, but only solar activity survives as a plau- 
sible primary factor of control, and that control is far 
from proved. Certainly it can be said that the secular 
fluctuations of the general circulation unmistakably 
show the influence of the sunspot cycle, but it is equally 
certain that there exists no demonstrable strict cor- 
respondence between solar variability (sunspots) and 
