ii Transactions of the South African Philosophical Society. 



before you further considerations on this matter of a South African 

 Standards laboratory. The problems I wish to speak of to-night are 

 more particularly connected with the physics of the earth's surface 

 and atmosphere in South Africa. 



The International Scientific Catalogue groups these subjects 

 under the title Meteorology. This science in its narrower sense 

 is more or less familiar to all of us, and I wish to direct your atten- 

 tion to it for a few minutes. The meteorologist must in the first 

 instance be a good physicist. The world for him is a great physical 

 laboratory in which the forces are mostly beyond his control. The 

 principles, however, are throughout physical, and advance can 

 only be made through a knowledge of, and the correct application 

 of, them. 



In considering meteorological data we may proceed in the follow- 

 ing way. We consider at one place the change in any one element 

 —such as the temperature — as time goes on. The first and natural 

 period is the day. In a normal day the temperature varies in a 

 manner which is so regular that the changes may be represented by 

 fairly simple mathematical formulae. There is for this and for each 

 of the other meteorological elements a well-marked daily variation ; 

 the diagrams before you show this variation for Cape Town in 

 temperature, humidity, and barometric pressure. 



Instead of taking the day as our unit of time we can take the 

 year. The phenomena looked at from this point of view show a 

 second well-marked variation. This is the annual variation. 



The second set of diagrams shows this again for Cape Town. 



Still confining ourselves to the meteorological data at one place, 

 we may now inquire whether there are other variations extending 

 over longer or shorter periods, such as the so-called eleven-year 

 period ; and, finally, we may try to ascertain whether the data give 

 any indication of a secular change. In other words, is it possible to 

 decide whether Cape Town is becoming hotter or colder, drier or 

 wetter. The immense practical value of this subject now becomes 

 evident. With sufficient data it might be possible to say whether or 

 not the average yearly rainfall at any given place was increasing or 

 decreasing, and to determine whether the change — if any — is con- 

 nected with any other change, such as an increase or decrease of 

 vegetation in the neighbourhood of this place, and whether or not it 

 would be possible by human agency to control the change. 



I pass on to consider two subjects which in some of their mani- 

 festations are closely related to, meteorology — the subjects of earth 

 magnetism and of atmospheric electricity. A magnet suspended 

 freely at its centre of gravity takes up at a given place, and at a, 



