SCIENCE- GOSSIP. 



163 



being also considered unhealthy. The lowest 

 temperature given is 3° C, which occurred in 

 August 1879, but I have myself seen frost in July 

 in Pretoria, and of course down country it is much 

 colder. June, also, is a cold month. The eleva- 

 tion of Pretoria is given as 4,471 feet above the 

 sea level, and, roughly speaking, it is 200 miles 

 south of the tropic of Capricorn. The site of the 

 town is in a depression between hills, having an 

 area of approximately ten square miles. 



The water supply is good. A small stream is 

 carried through the town in all directions by so- 

 called " sluits " or gulleys, and the water is used 

 for irrigating gardens. The sluit can be closed by 

 a small gate : the water passes through a drain 

 made for the purpose into the garden, and over 

 the ground under cultivation. 



During the hot weather the air in the Pretoria 

 valley became heated and dry, and appeared to act 

 as a buffer against storms which approached, 

 deflecting them on to the surrounding hills. The 

 drier the air, the greater this effect. 



Tlie storms in South Africa are very remarkable 

 phenomena. Frequently of small diameter, they 

 blow with terrific violence and travel over large 

 stretches of country. They vary in size from the 

 "dust devil" to a storm a couple of miles in 

 diameter. I am told that their violence is at times 

 sufficient to whirl sheets of corrugated iron into 

 the air as if they were pieces of paper. I have 

 never encountered any equal in power to this, but 

 I have seen many trees levelled to the ground by 

 them in Pretoria. 



The storms which visited Pretoria during my 

 residence there varied in size from 100 yards to 

 about a mile in diameter. Usually they were 

 accompanied by thunder and rain, the whirlwind 

 coming in front and raising a pillar of dust, sand, 

 papers, branches of trees, etc., which were some- 

 times carried a couple of hundred feet or more into 

 the air. In the dry weather the whirl pillar may 

 or may not be accompanied by rain and thunder. 

 The lightning from the dry storms is said to be 

 very dangerous, as the lightning discharge is very 

 intense. The whirl pillar appears to be at an angle 

 of 80° or 90°, with the ground sloping backwards 

 from its line of progression to the cloud. The 

 usual speed appeared to be from seven to nine 

 miles an hour 



The dust and dirt carried in these storms is evi- 

 dently one of the causes of the spread of disease. 

 I have heard people attribute an attack of fever to 

 having been overtaken by one of them. I was 

 informed that a nutrient gelatine plate exposed for 

 a second in one of these storms develops thousands 

 of colonies of bacteria. 



The barometric readings in Pretoria showed a 

 marked regular diurnal variation in fine weather. 

 On the approach of a storm this became modified, 

 and a steady fall heralded its advent, as may be 

 £een by the observations on March 18th and 24th. 



In Februarj', before the wet weather commenced, 

 there was a general depression in the mercury. 



With regard to the storms themselves they always 

 came against the direction of the surface winds, 

 but when about a mile distant the wind veered 

 completely round, taking the direction of the 

 storm; that is to .say, an easterly wind shifted to 

 the west when the storm was about a mile distant 

 from the point of observation. 



I had an exceptional opportunity of observinf;- 

 this phenomenon, as there was a very tall chimney 

 over the Electric Supply works belching forth large 

 volumes of smoke, which proved a very delicate 

 wind vane. The direction of tlie wind as mentioned 

 above sliowed very clearly the depression near the 

 storm, and the belt of high pressure immediately 

 surrounding the whirlwind. Tlie greatest depres- 

 sion appeared to be about a mile in front of the 

 storm, and to it the air flowed from all points, in- 

 cluding the storm itself. The wind flowing from 

 the storm was shown by the shift of wind on if.s 

 near approach. Before the storm ariived the air 

 was always sultry and oppressive ; but when it broke 

 the temperature fell several degrees, as will be seen 

 by the records of March 18th and 29th. 'J'o us the 

 air always felt hot and oppressive on the approach 

 of a storm, and when it broke the air seemed 

 excessively cold — more so than the number of 

 degrees of actual descent would lead us to expect. 



The contour of the country appears to be well 

 adapted for the production of storms of this nature. 

 The extensive plains and valleys di.splay facilities 

 for heating large bodies of air. The hilla are 

 always fanned in fine weather by cool breezes. It 

 was most refreshing to travel from the valley of 

 Pretoria to the top of one of the surrounding hills, 

 which rise to the height of a few hundred feet 

 above the town. Here the air was cool and 

 bracing. 



Judging by the direction of the higher clouds, 

 I was led to assume that a fairly constant breeze 

 from the west or south-west existed. 



The medium-sized storms, and even the small 

 ones, appear t'o aft'ect the barometric pressure for 

 long distances, as in my observations tliere was 

 depression of the mercury for several hours before 

 their arrival, while the sky was still quite clear. 

 In tlie rainy season the area affected appears to 

 be very lai'ge, and is ])robabl3' of the nature of a 

 monsoon. 



The storms usually came from the west or south- 

 west, turning somewhat to the north at the east 

 end of the valley of Pretoria. In the really wet 

 weather the rain-clouds came from the north-east. 



The whirl pillars appear to consist of numerous 

 secondary ones, which, besides rotating on their 

 own axes, travel also round the axis of the primary 

 storm. When fences of corrugated iron, houses, 

 or obstacles of the kind are encountered, the axes 

 of the secondaiy whirls frequently became trans- 

 verse, and. as it were, rolled along the giound. 



G 2 



