JO 



NATURE 



[January 5, 1922 



stands in their way is the lack of faith and of 

 courage on the part of the present directing agencies. , 

 Probably the most practical suggestion of immediate | 

 value that has been made is that the various scientific 

 societies in any town should arrange meetings for 

 study and discussion — the discussion which seeks 

 agreements and does not emphasise differences — and 

 that the underlying, but not obtrusive, object of the 

 meetings should be the progressive connecting of 

 science with individual and corporate conduct. 



The second question has many factors in common 

 with the first. The British Association has suffered 

 from the mental reaction which set in a quarter of 

 a century ago. It is a much smaller .factor in the 

 thought and life of the age than it was a generation 

 ago. I think the first helpful change desirable is the 

 recognition of a new principle in the selection of a 

 president and in the making of his annual address. 

 Above all things, the president should stand for the 

 unifying of the sciences, and his address should make 

 some definite contribution to that unity, even when 

 it is built largely on the recent achievements of one 

 section of knowledge. 



It is only through a conception, becoming ever 

 clearer, of this unity that science can become the 

 "chief formative factor of modern life." The yearly 

 appeal of a president may do much, but more would 

 be achieved if a day were set apart for the study or 

 discussion of the thoughts and facts he has com- 

 municated — a study or discussion, I say again, which 

 should emphasise agreements and not differences. 



Progress in this direction might receive a healthy 

 impetus from the universities. They, too, have lost 

 a considerable amount of influence. They are not, 

 at home or abroad, leading humanity. The note of 

 real universality is departing from them. Here again 

 the first practical improvement will come from the 

 manifestation of greater care in the selection of the 

 principals of the attached colleges. They ought to be 

 something more than skilled administrators. They, 

 too, have a great unifying function. 



This unifying work might be facilitated if there 

 were periodical meetings of the various professors and 

 lecturers for the study of unifying problems. That 

 does not, I admit, promise to help them immediately 

 to overcome the financial difficulties which are now 

 laming them to a terrible extent and driving them 

 to seek greater support from an overburdened State. 

 But if the universities, encouraged by a steadily in- 

 creasing enthusiasm for science locally and centrally, 

 were themselves to become again great inspirers of 

 thought, they would soon cease to be troubled by the 

 lack of pence. W. Robertson. 



Middlesbrough, December 29. 



Cohesion. 



The theory of cohesion put forward by Dr. Herbert 

 Chatley in Nature of August 18 is logically based on 

 those of other investigators, and, consequently, does 

 not involve any new element. In all these theories 

 cohesion is made to depend on centrally directed 

 forces which follow either the inverse square law of 

 gravitation or electrical attraction, or that of some 

 other inverse power higher than the second. Dr. 

 Chatley says : " It is difTicult to conceive of one force 

 having all these properties, but perfectly simple to 

 imagine an attraction and repulsion combined that 

 will do so, provided that the attraction decreases more 

 slowly with separation than the repulsion." He 

 takes the ground that the force of cnhpcion as s^a^pd 

 by him is related to those following the inverse square 

 law, and that the question of the relation between 

 them is of great importance. 



Now it is a matter of common observation that two 

 NO. 2723, VOL. 109] 



free liquid spheres on coming into contact with each 

 other always coalesce. The force which causes this 

 is evidently a force enveloping their masses, and not 

 a force attracting them. This enveloping property of 

 surface tension was noticed by Maxwell and others ; 

 but the theory which makes it depend on molecular 

 attraction renders it impossible to conceive of such a 

 force as enveloping molecular masses. 



The present writer has adduced {Phil. Mag., June, 

 192 1) very strong, if not conclusive, evidence that the 

 same force which causes liquid spheres to coalesce 

 also causes the free molecules of a gas to coalesce or 

 cohere. It cannot, therefore, be explained by mole- 

 cular attraction. The alternative is that it is an 

 elemental force acting, not in lines, hut over areas. 

 As such it is a universal property of the surface of 

 both liquid and solid mass extending to molecular 

 dimensions. 



Fortunately, however, there is veiy definite and 

 easily verifiable evidence that cohesion, and adhesion 

 also, is due to a surface force, whatever its nature 

 may be, as can be seen from the following simple 

 experiments which will be published later in fuller 

 detail. 



Spheres of mercury, ranging from 0-05 mm. to 

 15 mm. in diameter, were hung from a drop of water 

 wetting a glass surface above. Each one fitted into 

 an inverted hemispherical cavity in the water, with 

 a well-defined angle in the contact circle where the 

 water surface joined the mercury surface. With a 

 specially adapted microscope the diameter of the 

 sphere, the width of the contact circle, and the angle 

 between its water arm and the vertical were 

 measured. From these measurements, W. the weight 

 of the mercury sphere, and T, the vertical component 

 of the pull of the water surface on the mercurv, were 

 calculated for a large number of spheres. The re- 

 sults showed that for small spheres T greatly ex- 

 ceeded W, but tended to become equal to W when 

 the spheres were at the point of falling off. The 

 ratio T/W decreased gradually from about 6 to i, 

 thus showing that the surface force of the water 

 pulling on the mercury in the periphery of the contact 

 area was more than suflficient, except in the limit, to 

 support the weight. 



Similarly, mercury spheres, with diameters from 

 005 mm. to 225 mm., w^ere suspended from a hori- 

 zontal glass surface. They were attached to the glass 

 either directly or by suspending them from water as 

 before and allowing the water to evaporate. The 

 mercury surface was joined to the glass surface in the 

 periphery of a wide circular contact area, and formed 

 a definite angle with the j/lass' surface. Measurements 

 were made as before, and W and T (for mercury) were 

 calculated. The results showed that, as the spheres 

 increased throughout the range, the ratio of T/W de- 

 creased from the surprisingly large number of more 

 than 6000 to about 2. Had T, however, been calculated 

 from o- = 27o instead of <r=547, the decrease would 

 have been from about 3000 to i as before, and hence 

 270 may be regarded as an approximate value of the 

 surface tension of glass. The increasing values of 

 T/W for the smaller particles would account for the 

 persistence with w'hich molecules of a gas condense 

 on a glass surface. 



Further, small particles of anv insoluble solid be- 

 come attached to any surface above bv the evapora- 

 tion of a connecting water drop; or, if the particles 

 be clean and small, they become attached to any clean 

 surface by simple contact wi^h it. This is amply con- 

 firmed bv extended observations. 



Now there is no reason to think that the force of 

 cohesion Is not of the same nature in the case of two 

 solids as it is in the cases of a liquid and a solid and 



