122 Transactions of the South African Philosoiohical Society. 



a, = 3-666 



A, =: 68° 53' 



a, = -535 



A, = 100 53 



^^3 = -200 



A. = 180 



a, = -377 



A4 = 196 42 



The differences T - D are not great so far as Durban, Queens- 

 town, and East London are concerned. It may be suggested that 

 they partly arise because the numbers in M do not stand for quite 

 the same things as those in T : M, indeed, containing the mean 

 monthly values (assumed, without too much warrant, to be equal to 

 the normal value for the middle day of each month) ; whereas T is 

 computed for an actual day. The Umtata differences are greater ; 

 no doubt partly because of the importation of alien conditions, as 

 pointed out a page or two earlier, and in a less degree because the 

 observations are not yet extended enough in years to give the 

 normal curve.'" It is curious how uniformly T - D changes from 

 mimis to jjZ?^s from April to May, the excess of T over D being in the 

 latter month the greatest of all. In this instance the average diffe- 

 rence (leaving out Umtata) is 0°-6. None of the other months reach 

 an average, positive or negative, so great as half a degree. Such a 

 result surely justifies the applicability of the process employed. It 

 is important to observe that the numbers in the columns T could 

 also have been empirically deduced directly from the physical 

 formula (1). This method, however, would not in the least connect 

 them with the numbers in the columns M. The advantage of the 

 actual procedure is that it shows that the maximum temperature 

 upon any day of the year at either a coast or an inland station. may 

 be very well stated by means of an algebraic equation of an 

 invariable form. 



A comparison of the respective values of A and B so far found is 

 interesting. Leaving out the decimal places they are : — 



A B 



East London 16 61 



Durban 24 62 



Umtata 25 56 



Queenstown 42 43 



Aliwal North 53 34 



Bloemfontein 56 33 



Kimberley 73 23 



* And partly perhaps in consequence of some disturbances arising out of the 

 topography of the district in which it is situated. 



