1865] WRITINGS OP JOSEPH HENRY. 5 



clouds are seen to exhibit the appearance of giving out rain 

 though none falls to the earth, the whole being entirely 

 evaporated. That the air should ever be under-saturated 

 during rain is at first sight a very surprising fact; it may 

 however be accounted for on the principle of capillarity. 

 The attraction of the surface of a spherical portion of water 

 for itself is in proportion to the curvature or the sraallness 

 of the quantity, and hence the tendency to evaporate in a 

 rain-drop ought to be much less than in an equal portion of 

 a flat surface of water. 



If the diminution of quantity of rain at the upper station 

 depends principally on eddies of wind, then the effect will 

 be diminished b}' an increase in the size of the drops, which 

 will give them a greater power of resistance; and the size of 

 the drop will probably be influenced by the intensity of the 

 electricity of the air, as well as by its dryness. The former, 

 as well as the latter, will tend to increase the evaporation 

 from the surface of the drop. 



It is a well-established fact, which at first sight would ap- 

 pear to be at variance with the results of observations on 

 towers, that a greater amount of rain falls in some cases on 

 high mountains than on the adjacent plains. For example, 

 the amount of water which annually falls at the convent of 

 St. Bernard is nearly double that which falls at Geneva. 

 This effect however is due to tlie south wind, loaded with 

 moisture, ascending the slope of the mountain into a colder 

 region, which causes a precipitation of its vapor. From what 

 is here said, it will be evident that the subject of rain is one 

 which involves many considerations, and which still pre- 

 sents a wide field for investigation. 



A series of observations has been commenced at the Smith- 

 sonian Institution on the quantities of rain at difierent ele- 

 vations, as well as on gauges of different sizes and forms, 

 the results of which will be given in one of the meteorologi- 

 cal reports. 



