Anticyclones and S.A. Weather. 59 



turns and comes west, causing a gradual rise of the barometer readings, 

 and as the centre passes to the south during May the mean variation 

 of pressure is greater than in any other month in the year. 

 That this rise of pressure .is due to the anti-cyclone is further 

 confirmed by an examination of the tracks of the hurricanes 

 which have come into the vicinity of Mauritius, and have been 

 recorded at the Observatory for a great number of years. I am not 

 aware that any connection between the two has ever previously been 

 pointed out, but it must be apparent that if the anti-cyclone has a 

 regular progressive movement towards and away from Mauritius, the 

 number of hurricanes recorded must decrease month by month as the 

 anticyclone approaches, and must cease when it lies near enough to 

 ward them off. This decrease is just what is found to take place, 

 and from June to October they cease altogether. Moreover, the 

 tracks the hurricanes will follow each year will be controlled by that 

 of the anti-cyclone, and they will take a straight or parabolic course, 

 according to the lie of the isobars of the anti-cyclonic system. In 

 these charts we possess a direct indication of the position of the 

 system each month, and a very strong confirmation of all the views 

 that have been advanced regarding its progressive movements from 

 Australia to Africa and back again. 



We have previously seen that the Atlantic system is fast 

 approaching the East Coast of Africa during the month of May, and 

 merging with the Australian. This must also affect the barometric 

 pressure at Mauritius, and as the two systems become established and 

 migrate north with the sun up to the June solstice, the barometer must 

 continue to rise. (Appendix VI.) The chart of the isobars for July, 

 issued by authority of the Meteorological Council, to which reference 

 has already been made, places the centre of the high pressure prevail- 

 ing over South Africa, not far from Mauritius. The mean monthly 

 barometer readings for July and August are in consequence the 

 highest of the year. From September to about February the anti- 

 cyclone is receding, and the barometer falls gradually. The 

 Mauritius records show just the rise and fall of barometer which 

 our theory requires, and are all in favour of it. But there is a further 

 test that can be applied. In the records of the Natal Observatory 

 at Durban we possess from 1885 onwards meteorological observations 

 of the greatest importance in this inquiry. They indicate that 

 secondary anti-cyclonic systems frequently pass from west to east 

 across Natal, especially from May to August, and from observations 

 made in the Sydney Government Observatory such systems are 

 believed to reach that place after an unbroken journey. This 

 supports the view that the atmospheric conditions from end to end of 

 this track are favourable to the progressive movements of anti- 

 cj'clonic systems. 



During the 10 years 1890-99, the mean moisture in the 

 atmosphere at Durban during the several months in grains per foot 

 cube was as follows : — 



Jan. Feb. Mar. April May June July Aug. Sept. Oct. Nov. Dec. 

 Tl ;-5 7-3 6-3 5-2 4-5 4-4 4-9 5-4 5-9 6-5 /'O 



