TFKKI^SrklAI. I'llNSHS. 4,^5 



two of the most interesting and characteristic jnirts of the 

 atmos])here. Hitherto, so far as I am aware, nothini^- has been 

 (lone in Sonth Africa excei)t tlie ordinary routine surface obser- 

 vations of rainfall, barometer, wind, wet and dry bull) thermo- 

 meter, and the very valual)le surface work of Air. Sutton ai 

 Kimberlew 1 doubt whether we even know the ordinar\- c\cl. mic 

 tracks in the lower atmos])here. There is therefore a \ery wide 

 and o])en field for observation and discussion in this branch of 

 meteorological jjliysics. a held likely, when cultivated, to yield 

 a rich harvest 'jf important results. 



4. The comparative simj^ilicity of the South African climate 

 suggests the possible fruitfulne^s. here if anywhere, of another 

 line c(f research which has fascinated many investigators — 1 mean 

 the enquiry into possi:ble periodicities in atmospheric climatic 

 conditions. Mr. D. C. Hutchins devoted a good deal of atten 

 tion to the (|uestion of the periodicity of rainfall, and 1 lia\c 

 made some analyses, l)y I'ourier's method, of the rainfall ai the 

 Royal ( )l)servatory. Cai)et(jwn. of which I read a ])reliminar\ 

 account before the .Soiuh African Philosophical Societv a good 

 many years ago. . The results reached may i)erha])s be regarded 

 as at least justifying the expenditure of some more labour, more 

 es]>ecially a^ a good deal has lately been done to shorten the 

 necessaril\- lengthy time needed to carry out the com|)Utation.> bv 

 the contrivance of harmonic analyzers and other counting de- 

 vices. 1 have just had constructed a single working model of a 

 harmonic analyzer on what 1 think is a ne\v principle, and hope 

 to attack the (|uestion ot rainfall periodicitx' with its liel]). 



5. riiere i> one more held of geo])hysical research in wlnCh 

 international co-operation is essemial, and for which, 1 think, lite 

 evennos of otir coast-hnes gi\es u> special natural facilities, 

 namely, the investigation of ocean and earth tides. In regard 

 to the former the (-l)servations (if Forel have made us realize 

 more fully than ever bef(^re that each ocean has its own par- 

 ticular type and am])litude of forced tidal oscillations. The cal- 

 culation of the tides even in an isolated ocean of sim],)le t\])e 

 ha-^ hitherto prtn-ed itself beyoud the ])owers of our anahsis.- 

 and it seems tmlikelv that we shall e\er be able to account b\ 

 mathematical methods for the tides in the actual oceans, with. 

 their intercommtmications, although d\namical reasoning is or 

 great importance in indicating the directions which observational 

 work should follow. There is therefore room for much work 

 on the classic lines laid down by I'^orel. For this purpose tick- 

 records would be needed at points well distributed along the 

 coasts of the chief oceans. A tide-recording machine at I'on 

 Nolloth or Swako])mund w(Aild be hel])ful in the study of the 

 tides of the Atlantic; one at Cajjc Agulhas, or other similarb.- 

 placed ])oint on the South Coast would be essential in the ;uial\ - 

 sis of the tides of the Antarctic oceanic belt, while one well U|) 

 the Kast Coast would help in regard to the Indian Ocean. 



To turn to earth-ii(le>, I should like personally lo be alilo 

 to re])eat in .South Aifrica tlie recent ])rilliantly conceived an<l 



