40 Meteorological Observations on ParisnatJi Hill. [No. 1. 



If the view here developed be correct, this will in no way oppose 

 the assumption that the mass of watery vapour contained in the at- 

 mosphere in a vertical direction is in most cases, on an average, 

 proportionate to the tension found near the surface of the ground. 



According to the foregoing, the mode in which the watery vapour 

 chiefly affects the barometer must be two-fold : in the first place by 

 its tension or elasticity, and secondly by its weight. The tension 

 is found by the Hygrometer and its action has just been considered. 

 The weight can only be determined accurately after finding the 

 mean tension throughout the atmosphere, or within the limits of a 

 space for which the weight is required. Directly a palpable effect 

 of the changes in the weight during the daily period can hardly be 

 observed, as its variation must be a very small part of the daily 

 barometrical variation. I will, however, not omit to notice a circum- 

 stance which may perhaps partly be traced to this cause. 



In the regular daily course which the hygrometric state of the air 

 undergoes, the quantity of moisture is, as a general rule, found smal- 

 lest at sunrise, when the temperature of the day has reached its mini- 

 mum, and when the vapour contained in it is nearest to its point of 

 saturation. This minimum of moisture is caused by the condensa- 

 tion of a part in the shape of dew and fog. The quantity of dew de- 

 posited varies with the nature of the surface, — being greater, where 

 the ground is covered with vegetation than where it is bare, and 

 varying for different kinds of soil. I have seen it in the month of 

 February under a group of trees near Poonah, equal in its effects 

 to a slight shower of rain, coming down from the leaves. Its daily 

 quantity will be regulated besides, as well as that of the fog by the 

 mean degree of saturation, the mean temperature of the day (the 

 tension of saturated vapour at the mean temperature) and the 

 daily range of temperature. In tropical climates, not far from the 

 sea-coast, where the mean temperatures and the state of saturation 

 in the different months change comparatively little, the influence of 

 the daily range would probably the most prominent. 



If a merely local diminution of the water suspended in the atmos- 

 phere, confined to a small space, such as for instance is caused by a 

 shower of rain, is not always observed to effect a perceptible change 

 in the barometrical column, this is easily accounted for by the in- 

 stantaneous restoration of the equilibrium which must take effect as 



