196 ROUND THE YEAR 



sides of the vessel. Why is this ? The water rises 

 on the sides of the discs and also on the inside of the 

 vessel, forming in each place an ascending capillary 

 curve. Descending capillary curves are also to be 

 met with. If a lump of soot or a greased disc is set 

 floating on water, it will be surrounded by a descend- 

 ing curve. Mercury in a glass dish or a barometer- 

 tube has its free surface bordered by a descending 

 curve. We can easily change the ascending curve of 

 water in a glass vessel to a descending one. Take a 

 glass or cup half-full of water ; the water creeps up 

 the side in an ascending curve. Add water until it 

 rises level with the brim ; then we say that the vessel 

 is full. Though it is full, we can still add a consider- 

 able quantity without spilling, if we do it steadily. 

 The vessel can be made over-full, when the flat 

 surface of the water will be visibly bounded by a 

 descending capillary curve. 



We have next to remark that if two small floating 

 objects are surrounded by capillary curves, they will 

 either attract or repel one another according to circum- 

 stances. If the curves are alike, both ascending or 

 both descending, the objects will attract one another ; 

 if they are unlike, one ascending and the other descend- 

 ing, they will repel one another. A theoretical proof 

 of this can be given, 1 but the fact can be directly 

 established by experiment. Take a small vessel 

 nearly full of water, and place in it a small disc of 

 cork or wood. If the side of the vessel and the disc 

 are both wetted by water, there will be an ascending 



1 I have tried to put the proof in the simplest possible way 

 in my Object Lessons front Nature , Part II., p. 150. 



