26 CONDITIONS FOR PRODUCING A CONTINUOUS FLOW. 



top of the tube, but will not Jlow over, as one might perhaps have expected. In an or- 

 dinary capillary tube, therefore, nothing like a constant flow can take place, but the 

 liquid having attained its highest possible elevation, remains there. 



74. That a continuous flow should take place, all that is necessary is, by any proper 

 means, either by evaporation, chemical action, or other processes, to remove away the 

 superficial portions of the elevated liquid when they stand at the extremity of the tube. 

 An illustration will show how this is accomplished. The wick of a lamp is nothing 

 more than an extensive system of capillary tubes, tubes which are formed by the juxta- 

 position of the cotton fibres. As common experience satisfies us, such a wick may be 

 immersed in a reservoir of oil for months, or even years, without any sensible portion 

 of that liquid being removed ; but if the lamp is lighted, the process of combustion dis- 

 sipating the oil as fast as it reaches the upper portion of the wick, fresh quantities are 

 furnished from beneath, and a continual flow takes place until all the oil is gone. So 

 also in a spirit lamp, as long as the extinguisher is over the wick, and no evaporation 

 of the alcohol can take place, there is no flow ; but the moment the extinguisher is remo- 

 ved, so that evaporation into the atmosphere can be accomplished, there is a constant 

 upward flow until the alcohol is dissipated. These are results which have been long 

 recognised. 



75. From these elementary considerations, therefore, it is clear, that although, in the 

 ordinary use of a capillary tube, continuous movement along it is not witnessed, that 

 movement readily sets in as soon as the proper conditions are fulfilled. In the two 

 cases we have used as illustrations, combustion in the one, and evaporation in the other, 

 gave rise to a continuous motion. 



76. In the same way a variety of other ordinary causes may produce these move- 

 ments. Suppose,^. 109, we had a vessel A, containing water, and another vessel B, 

 containing alcohol, and between them a narrow capillary tube, C, passing. Let us 

 farther suppose this tube C to be previously filled with water. At its extremity which 

 opens into the vessel B, containing alcohol, it is clear that the water will be brought 

 in contact with the alcohol, but in this liquid water is soluble ; and, therefore, as fast 

 as that water can be presented, the alcohol will dissolve it, take it up, and remove it 

 away. It is plain, therefore, that there will be a constant flow of water from the ves- 

 sel A to the vessel B, a flow which arises from the circumstance that the water is dis- 

 solved from the end of the tube as fast as it presents itself, by the affinity of the alco- 

 hol for it. This, therefore, is nothing more than a repetition, under another form, of the 

 cases already used above (74) as illustrations. 



76. But an attentive consideration of the facts will convince us that the affinity thus 

 existing between the two liquids, and which thus gives rise to a flow from A to B, 

 ought, after a short time, to cause portions of the alcohol to find their way through the 

 water in the tube C, and present themselves in a certain quantity at its entrance into 

 the vessel A ; there, in their turn, they are exposed at once to the pure water, which 

 takes up and dissolves alcohol just in the same way that alcohol takes up and dissolves 

 it. There must, therefore, be a constant flow of alcohol, from B to A, along the tube, 

 for the very same reason that there is a simultaneously constant flow of water from A 

 to B, in the contrary direction. 



