972 CLIMATOLOGY 
churia. It is not possible to reproduce conveniently more 
than this extract and, as would be expected, the com- 
plete specimen year conveys a far better idea of the 
normal change, but even from these three months one 
receives the impression of how the elements vary from 
one day to another. In this diagram, wind, temperature, 
and humidity are illustrated as well as sunshine, pres- 
sure, and rain. In addition, the air-mass classification 
and the fronts which lie in the neighbourhood of the 
station are shown. 
These studies were essentially climatic and contained, 
in addition to the descriptive matter, tables of normals 
to serve as the inevitable background. Here it may be 
emphasised that, however the climatological data are 
discussed, accurate and well-arranged general tables of 
normals must be published if only to serve as a yard- 
stick agaimst which to measure the variability. 
By way of contrast between the description according 
to conventional climatology and that which can be 
built up from the study of synoptic situations, it is of 
interest to compare the description of the winter climate 
of China as given in Kendrew’s Climate of the Continents 
[17] with that given by Hare [1] and Lu [24]. Admittedly 
it may be objected that readers for whom Kendrew 
wrote were less erudite than Lu’s audience or even than 
Hare’s, but if one type of description is read and then 
the other, it will at once be apparent which is the truer 
account. 
In making this contrast between the conventional 
approach and that of the synoptic meteorologist, I do 
not wish to veil the value of the mass of mformation 
which conventional climatology has provided. It has 
indeed formed the basic background from which it 
seems there is now scope for advance in a more detailed 
description of climate, a description based on the physi- 
cal causes which in the aggregate produce that which is 
called climate. 
Since World War II, the procedure which has been 
described above in regard to the use of synoptic charts 
has provided the basis for several reports such as 
Aviation Meteorology of South America [4] for which 
J. S. Sawyer was responsible, and Aviation Meteorology 
of the Azores [3] by L. Jacobs and R. M. Murray. 
This method of approach was the natural outcome of 
the need for a similar type of information in other 
countries. In Germany it would seem that methods of 
air-mass climatology were more or less dropped during 
the war, at any rate in their more detailed application. 
In a number of climatic reports the Germans made use 
of representative types of weather, illustrating them 
with typical weather maps for individual days, but 
there does not seem to have been the systematisation 
which might have been expected. The Germans in their 
reports seem to have relied largely on a great mass of 
‘statistics suitably arranged for aviation and related 
needs. 
In America the stress on securing information as a 
basis for planning during World War II led to a rather 
similar approach under the lead of W. C. Jacobs [16], 
who assembled large masses of data on punch cards for 
the region in which he was interested. By sorting for 
one element (¢.g., air flow) he was able to draw up a 
picture of the probable weather with which each type 
of flow was associated. However, with the advantages 
of the punch-card system it was possible to break down 
the problem still further in whatever direction was most 
advantageous, to calculate, for example, the chance that 
low cloud would occur with a particular general wind 
flow and the chance that the low cloud would affect one 
region when another was clear. 
From the practical point of view these summaries 
could be linked effectively with the weather map, for 
the forecaster could tell at a glance which particular 
type of the numerous arrangements of the data was 
most like the chart with which he was dealing. However, 
the presentation of the data was essentially statistical; 
it masked rather than emphasised the physical processes 
underlying the significant weather. Nevertheless, as a 
practical tool in climatology, the punch card, wisely 
used, will undoubtedly play its part in the final solution 
of the climatic problem. 
The Russian approach to the problem of climatic 
representation is shown in a recent publication by 
Alisov [2]. In his foreword, Alisov is emphatic that the 
study of climatology must not end with the collection 
of data, but that it must be discussed on the basis of 
the physical process by which the climatological phe- 
nomena have been developed. He then proceeds to 
discuss the general climatology of Soviet Russia as a 
whole on the basis of the origin of the air masses and 
their modifications in transport. The remainder of his 
extensive book is devoted to a more detailed description 
of the climates of the different regions, but in each case 
the description is based on an integrated knowledge 
of the synoptic charts rather than on normal values. 
Climate as a Physical and Dynamical Problem 
The climatologist will not be content with the mere 
recording of observations nor with their description, 
even if that description is based on the physical proc- 
esses in the atmosphere. It will be his natural desire to 
seek a physical explanation of climatic change, the 
change from one month to the next, as well as the 
changes which occur from one period of years to 
another. The climatologist has for long been intensely 
interested in long-range forecasting—seasonal fore- 
shadowing as Sir Gilbert Walker termed it. But the 
methods by which the mechanism of climate is investi- 
gated have been sadly lacking in a reasoned approach. 
The methods most frequently adopted are (1) the 
determination of correlation coefficients between a me- 
teorological element in one place and the same element 
(or another) in some other place, perhaps with a time 
lag, (2) the association of the pattern of a mean distri- 
bution of a particular element in a particular month 
with the pattern of the mean distribution of some other 
element in the same or a different month, and (3) the 
association of maps of monthly pressure or temperature 
anomalies with the anomalies of the same element in 
the succeeding month. One should add to these (4) the 
le ee en ee eee en 
