THE CLI!\rATF. OF LOl'RENCO MARQUES. 305 



with those of the Western Districts, owing to the large difference 

 in altitude and- to the position of Lourenco ^Manjues on the 

 littoral : it is an easy matter, however, to make such a comparison 

 with other j)laces on the coast. 



From such a comparison (see Table No. II j one concludes 

 that the increase in temperature between Durban and Lourenco 

 ]\Iar(jues is smaller than one might expect from the dift'erence of 

 latitude (234 miles) existing between the two places. 



Thus the annual means at Lourenco Marques exceed Dur- 

 ban only by 0.6'' C. (1.1° F.), the means for the hottest month 

 being i.i" C. (2° F.j, while the tem])erature in the coldest month 

 is practically the same in both towns. 



The extreme temperatures — absolute maxima and minima — 

 are higher at Lourenco Marques respectively l)y about 1.3° C. 

 (2.y F.) and 2.1^ C. (3.8° F.). 



if we compare Durban with Fast London, at a distance of 

 KjO miles in latitude, we find much greater differences, and the 

 same occurs between Lourenco Marques antl Inhambane. where 

 ihe latitudinal difference is 124 miles, while it should be noted 

 that Inhambane differs considerablv as to the absolute maxima 

 and minima, owing to its diverging to a great extent from the 

 meteorological conditions of the South African littoral. 



These comparisons are favourable to Lotirengo Aiarques as 

 10 the increase in temperature from south to north, and show the 

 siuMlaritv of the climate as to temperature in the towns of 

 Lourenco Ahir(|ues and Durban, a larger difference being noted 

 in the aljsolute minima, which are lower in the latter place. 



This circumstance nmy be partly due to the configuration of 

 the coast to the South of Zavora Point, and the extent of the Bav. 

 which compel the Mozaml)i(iue current to keep' away from 

 Lotu'enco Marqties, thus preventing the increase of temperature 

 there, but this can only be definitely ascertained by means of a 

 more detailed study of the currents. 



From the analysis of temperatures in the various months, 

 one finds that the temperattire in the months from May to Octo- 

 ber is rather agreeable, the percentage of relative humidity also 

 being smaller in these months. The temperattires being higher, 

 the climate from December to ]\larch becomes more tropical, 

 especially in the last two months., which are the dampest. In the 

 remaining month, which may l)e called transitory, the tempera- 

 ture is more pleasant. 



The four seasons are not well marked by special character- 

 istics. The transition months cannot be called " Spring " and 

 " Autumn," and it is more correct to divide the year into two 

 seasons, that is, the warm and rainy season from October to 

 ]\Iarch, and the cool and practically dry season from April to 

 September. 



In Table No. Ill the variations of temperature are regis- 

 tered, with the monthly means of temperature and rainfall. 



Diff'erences in the same period during the rise of tempera- 



