THE CONDUCTION OF WATER. II 6 5 



mercury in the barometer, are at first sight surprising, because they appear 

 to contradict the lessons learnt from the Torricellian vacuum. How can these 

 facts be explained, and such heights be theoretically accounted for ? 



By employing an air-pump in place of the evaporating block of gypsum, and 

 filling the glass tube with air from the beginning, it would certainly be possible 

 to produce a vacuum on attaining a height comparable to that produced by 

 atmospheric pressure. In order to obtain this result in our experiment how- 

 ever, the adhesion between water and plaster of Paris, between the water 

 and the wall of the tube, as well as the cohesion of the water particles them- 

 selves, must, first of all, be overcome. It is well enough known that the force 



'gs. 15, 16. From DETMER'S Smaller Practical Physiology (Figs. 77 and 78). 



of adhesion is very great, but, taking as a basis the older physical experiments 

 on the subject, it is quite obvious that we have considerably underestimated 

 the force with which the water particles cohere. ASKENASY and DIXON have 

 performed a very great service in showing how immense that force really is. 

 Detailed estimates of the force of cohesion are as yet wanting, still we may for 

 the present be content with the results arrived at by DIXON and JOLY (1895, b, 

 p. 570), according to whom a pull equivalent to at least seven atmospheres 

 is necessary to tear asunder a column of water. In all probability this is an 

 under- rather than an over-estimate of the force of cohesion. If it can only 

 be arranged that no air passes into the gypsum, a column of water 70 m., or 

 even more, or a column of mercury, 5^ m., can be held in suspension by an 

 evaporating block of gypsum. According to the detailed statements of 

 REINGANUM (1896) and NERNST (1900) columns of water of considerably 

 greater length must be supported by transpiration. 



JOST 



