370 ) 



moon's phases. The result of this calculation is given in 

 Table TV. 



Erom this table it appears that the mean daily range of the 

 temperature is high er on the days about f uil moon than on 

 other moon's phases. 



In Table V. the results contained in Tables I. to tV. have 

 been put together. The numbers contained in this table seem to 

 indicate : l st that the mean temperature of the twentyfour hours, 

 the mean temperature of the hours of the day (7 a. m. to 

 5 p. m.) and the mean daily range of the temperature are high- 

 er on the moons phases (4) and (5), or on the days about 

 full moon, than on the other moon's phases; ~ (I that, on the 

 contrary, the mean temperature of the hours of the night (7 

 p, m. to 5 a. m.) is lower on the moon's phases (4) and (5), 

 or on the days about full moon, than on the other moon's 

 phases ; 3 d that this variation in the temperature of the air de- 

 pend ent on the moon 's phases, which is to be considered as 

 one phenomenon, cannot be a direct effect of heat radiated from 

 the moon's surface towards the earth, but that it is very likely 

 a secondary effect caused by a variation in the clearness of the 

 sky, dependent on the moon's phases. 



Being unable to give a complete historical di<cussion of the 

 qnestion, ï refrain from all considerations of this kind, giving 

 only the mere result deduced from the hourly ob.^ervations of 

 temperature of the air made at Batavia during the decade 

 1866 — 1875, a result which, however, wants to be corroborated 

 by another series of ten vears observation. 



