A.—MATHEMATICAL AND PHYSICAL SCIENCES 39 
but I consider that this is not good enough and that we have been too 
ambitious. Also while the approximate prediction formula of 1908 has 
stood the test of time with credit, the later ones of 1924 for north-west 
India and the Peninsula separately, although certainly better in theory, 
have not, in the short period of trial, proved so successful. The contrast 
between the working of the formule before and after their date of 
preparation will be seen in Fig. rr. 
Happily in Southern Rhodesia, which in 1922 adopted statistical 
methods similar to those of India with only twenty-four years of data to 
work upon, the results have been eminently satisfactory. Out of eleven 
1924 FORMULAE 
eee oOavawa—wmyua=s=—' 
a 
PENINSULA  sune-Serr. ‘Limit’ = -g42Vi-eo = 3-48 \ 
CALCULATED [| 
s* Ny pm Ni ai 
JL WSVAVANIEL ATIVAN 
AGAR AMITNMTAVU AIA NGHINA 
ACTUA 
Peis 
N.W.INDIA_ sune-Sepr. ‘Limit’= 3h. 
PERT 
CALCULATED 
5: pe | IN | ! e 
os ASEAN NECA TAY SAVIN NAL ADS 
SNAIL TEDL AVALT TT 
ae INV AVI 
VV, 
1875 
Fig. 11.—Calculated and actual Indian rainfall. 
years since publication was begun, there have been eight in which a 
departure of over 3 in. was given by the formula, and in seven of these 
the character was correctly indicated (Fig. 12). 
At Batavia the efficient Dutch observatory under Braak started in 
1909 to issue forecasts founded on the simple rule that low pressure 
from January to June was followed by abundance of rain from July to 
December. The rule demanded a more complete presistence of pressure 
than actually prevails, and in 1927 Berlage adopted a formula based on 
three local conditions, together with data of the rare rains of northern 
Peru: this gives, on paper, a relationship of over 0°8. 
In Australia calamitous failures in the rains have long demanded 
forecasts, and these led to the production of weather cycles, which broke 
_ down so frequently that their use was discarded. In spite of this 
experience, however, Hunt, the Commonwealth Meteorologist, put 
