846 



Broad hanging drops cannot be obtained ; it is clear tiiat it is im- 

 possible to make drops hang from a wide tube, but also the drops 

 which are formed on the iinder-side of a horizontal moistened plate 

 do not attain large dimensions. If for instance a horizontal plate is 

 immersed in water and then lifted out, the liquid which adheres 

 to the plate collects in one or more drops which grow (sometimes 

 flow together), drop off, grow again etc. but the width of the drops, 

 finishing up tangentially to the plate, is in no way commensurate 

 with the size of the plate. 



The meridional section of the capillary surface in that case (as 

 long as the drop is not yet constricted in the middle) consists of an 

 undulating curve (fig. 4), the waves of which become lower and 



Fig. 4. 



lower, as they are further away from the axis) ') ; the suspended 

 drop represents the part (JAB comprised between the axis and the 

 first maximum. Curves drawn according to Kelvin's graphical method 

 seem to show, that the distance of the successive maxima and minima 

 from the axis of rotation increase with the radius of curvature at 

 the top; however, there is a limit to this increase: even for very 

 flat drops the distances remain moderate, as may be shown in the 

 following manner. 



When the drop is very flat, the angle ff may be considered as 

 very small everywhere and the shape of the section is determined 

 by (9), or putting k =z — k' by 



^" + - + fc'z 



(9") 



This equation passes into (9') by the substitution xV^k' = 'i, 

 2 = ?j, therefore 



') See WiNKELMANN, loc. clt., p. 1146. 



