6o Report S.A.A. Advancement of Science. 



These figures clearly indicate the decrease in moisture in the winter 

 months, and, coupled with the absence of rain from Natal during the 

 winter, they strongly support the view that an anti-cyclonic system is 

 present, and is impressing its well-recognized characteristics upon the 

 district. An analysis of all the records for the 15 years 1885-99 

 shows that a ridge of high pressure is first established in April, and 

 out of the last 10 years of that period only two are exceptions to 

 the rule that there is an increase of pressure in April over March, 

 and in May over April. The exceptions are in 1891 and 1893. The 

 mean monthly barometer readings for the period are : — 



Jan. Feb. Mar. April May June July 



29-980 30-008 30-047 30*ii9 30-172 30-229 30-232 



Aug. Sept. Oct. Nov. Dec. 



30-214 30-140 30-092 30-023 30-000 



The mean pressure differences from month to month reach a 

 maximum in March- April and in August- September, and it is in these 

 months that the arrival and departure, respectively, of the anti- 

 cyclonic systems belonging to the Atlantic and to the Pacific are 

 making their influence felt upon the barometric pressure. The 

 differences amount to o.o72in. between March and April, and to 

 o.o74in. between August and September. These are large variations 

 when W'C find that during 15 years the greatest departure of any 

 annual mean from the mean for the whole period is only o.o43in., and 

 that in 15 years the range of the barometer, as shown by the 

 maximum and minimum records, was only i.33iins. The highest 

 reading ever recorded in the annual reports is 3o.838ins., and the 

 lowest 29.507ins. It is quite in conformity with these results that 

 the dry season in Natal commences when the anti-cyclone becomes 

 established about the end of April, and ends when it leaves about 

 the beginning of September. In Buchan's charts for August and 

 September we see that the secondary anti-cyclonic system, which has 

 persisted over Natal from May to July, has gone, and that the 

 Australian and Atlantic systems are beginning to separate, and to 

 depart to their summer position. The gradual fall in the mean 

 monthly pressure at Durban is the result. The valuable records kept 

 at this Observatory of rainfall, winds, pressure, and other 

 meteorological elements are full of details supporting the theory of 

 the progressive movements of the two systems, but we must leave them 

 and pass on to the final point to which space will permit a 

 reference. (Appendix VII). In Buchan's " Rainfall of South 

 Africa " he discusses the observations made from 1885 to 1894, and 

 publishes maps which i\Ir. Struben reviews in the Report of the Cape 

 Meteorological Commission, with the fuller information available up 

 to 1897. The facts that are given, and the conclusions arrived at. 

 are made quite intelligible when looked upon as the direct results of 

 cyclonic systems which are controlled by the steadily-moving Atlantic 

 and Australian anti-cyclones. It was found that the rains borne 

 from the S.E. are at a minimum in August, and are confined 

 practically to the coast, while the rains borne from the S.W. at the 

 same period are at their maximum. After August the S.E. borne 



