3o6 THE CLIMATE OF LOURENCO MARQUES. 



ture are smaller and more regular than during its decrease. This 

 is, perhaps, due to the influence of the rainfall, which retards the 

 rise of tem])erature, thus decreasing it, while, on the contrary, in 

 the month of least rain the decrease takes place more rapidly. 

 It should be noted that these months are less cloudy, and this 

 fact, combined with the great nocturnal radiation caused by the 

 higher duration of the night, must contribute to a quicker 

 decrease. 



The march of the sun bears upon the annual variation of 

 temperature, and the usual retardation takes place. 



Variations of temperaure in day time show very little irregu- 

 larity. The minimum occurs normally before sunrise and the 

 maxinnun at about XT IT. the increase being more rai)id and pro- 

 nounced Ijetween VII and IX ; after this hour, owing probably 

 to the influence of sea breezes, the increase becomes more slow. 



The decrease of temperature at sunset is not so pronounced; 

 from midnight to the hour when the minimum temperature occurs 

 the decrease is generally i^ C. Annual nocturnal radiation seems 

 not to have so great an influence as one might expect, which may 

 be explained by the fact that dew is more fretiuent in less cloudy 

 nights, and this compensates for radiation, owing to the great 

 amount of heat liberated by the condensation of water vapour 

 existing in the atni.osphere. Dew occurs far less frequently in 

 the dry season. The mean daily oscillation, the annual range of 

 which is about io° C. ( i8° F.), attains its maximum — ii° to 12" 

 (20° to 21.5° F.) — from April to August, while in the other 

 months it decreases steadily to 7° C. (12.6" F). This dift'erence 

 arises from the fact that, the mean maximum variation oscillating 

 about 7° C (12.6° F.), the minimum variation oscillates 10° C. 

 ( 18° F".), that is to say, the oscillation of the minimum tempera- 

 ture is much greater than is the case with the maximum tempera- 

 ture, this being the opposite of what happens in Lisbon. I 

 think that this can be accounted for by the coincidence here of 

 rainfall and the greater cloudiness in the hot season. Besides the 

 Iwt ■zi.'inds from N and NNW, caused by the occurrence of 

 atmospheric disturbances, are also felt in the cool season, and 

 cause increases of temperature which are relatively more 

 im])ortant than in the other season. 



An analogous dift'erence exists between the absolute maxi- 

 mum and minimum, which is due to the same causes. 



These temperature extremes are rather interesting for the 

 study of a certain climate, although the development of endemic 

 diseases depends, according to some writers, on the mean tem- 

 peratures. Europeans residing in this climate become much more 

 sensible to the variations of temperature even in respect of such 

 small decreases as cause no impression in Europe, and the longer 

 their term of residence the more they feel such variations, this 

 being possibly due to physical exhaustion and the hydrometric 

 state of the atmosphere. The apparent need for avoiding chills 



