On Sal Nitriun and Nitro- Aerial Spirit 129 



quently press less on the water which lies under them. 

 Whence it is that the water rises gradually from the 

 inner edge to the centre of the vortex, the water 

 rising more in that part of the vortex where the 

 rotation is more rapid, as is shown in the figure. 



I remark here that it is probable that at the base 

 of the mass of rising water, the water, as shown in the 

 figure, is somewhat depressed (although this, perhaps, 

 can scarcely be seen by sailors at a distance). For 

 when the air surrounding the first cylinder, a^ a, h^ 3, 

 is thrust from a to / and from h to g^ in the manner 

 aforesaid, the result is that the air at i and g^ being 

 much compressed, undergoes considerable condensa- 

 tion, and in consequence the water beneath will be 

 somewhat depressed by the greater weight than usual 

 of superincumbent air. 



With regard to the fog that rests upon the pile of 

 waters, I think it is caused in this way. Thus since 

 the air about the surface of the rising water rotates 

 very rapidly, the water, at the outer parts at least, is 

 carried round along with the air ; whence it comes 

 about that small particles of the water, receding from 

 the centre of their motion, are dispersed in all direc- 

 tions and borne upwards, just as would happen if a top 

 with its upper surface spherical were wetted with 

 water and made to spin. 



But the reason why those vapours, as shown in 

 the figure, are bent and at last unite in the torrent at 

 ^, e^ seems to be this, that the nearer the rotated air 

 is to the centre of the aerial vortex the more it is 

 rarefied and thinned, as has already been shown. For 

 hence it is that while the aqueous particles driven away 

 from the aforesaid mass are carried upwards and out- 

 wards, they are at every instant of their progress bent 

 inwards by the air, which gradually becomes denser 



I 



