January i, 1920] 



NATURE 



451 



and the surface; and (4) the difference between the 

 surface-soil minimum and that of the air above it. 

 These determinations are necessary because : — (i) The 

 radiation from the soil on calm, clear nights is a 

 function of the relative humidity (A. Angstrom, 

 Smithsonian Misc. Coll., 65 No. 3); (ii) the radia- 

 tion from the soil can be accounted for in balancing 

 the upward conduction and the latent heat of 

 freezing, the residue only cooling the soil ; and (iii) the 

 temperature of the surface soil rapidly falls sufficiently 

 below the temperature of the 4-in. depth to make the 

 conduction from this depth balance the radiation ; after 

 this the surface tempyerature falls no faster than that 

 of the 4-in. depth. 



Franklin has also noted rapid changes in under- 

 ground temperature (6-in. depth) after heavy rain. 

 This he attributes to the receding water drawing hot 

 or cold air into the soil, but it is scarcely possible 

 that the volume of air thus drawn in could supply 

 or abstract sufticient heat to account for the observed 

 differences. These latter are in the direction which 

 would be expected from the time of the day when 

 the readings were taken ; and while the re-aeration 

 of the soil may have slightly increased the tempera- 

 ture changes, it is unlikely that it had a predominating 

 influence. B. A. Keen. 



LUBRICATION AND LUBRICANTS. 



THE meeting of the Physical Society held on 

 November 28 was devoted to a discussion on the 

 subject of lubrication, suggested by a pa]5er on " Oili- 

 ness and Lubrication" by Mr. R. M. Deeley, sent by 

 the Committee on Lubrication. Other contributors 

 included Dr. Stanton, Principal Skinner, Messrs. 

 VV. B. Hardy, Dorman, Southcombe, Martin, .'Vrch- 

 butt, Edser, and Dr. H. S. Allen. 



In the consideration of lubrication problems a clear 

 distinction must be made between two prevailing 

 conditions, viz. : (i) Those in which the solid sur- 

 faces are completely separated by a film of oil, as 

 occurs in the lubrication of cylindrical journals and 

 their bearings working under moderate pressures, 

 revolving at high speed, and supplied with abundance 

 of oil ; and (2) those in which, owing to the shape or 

 condition of the surfaces, the kind of motion, high 

 bearing pressure, low speed, or an inadequate supply 

 of oil, the oil-film cannot form completely, or becomes 

 broken, and the solid surfaces come into contact. 

 In (i) the friction is entirely due to the viscosity of 

 the oil, as proved bv Beauchamp Tower and Osborne 

 Reynolds. Engineers have Reynolds's theory to guide 

 them in the desigfn of bearings ,so as to secure fluid 

 friction, and it is possible by determining the viscosity 

 and chemical characteristics of an oil to form a sound 

 opinion as to its suitability to meet the required condi* 

 tions. In (2) the circumstances are quite different. 

 Lubricating value is then found to depend upon some 

 property which is quite distinct from viscosity, and 

 has been called "oiliness." 



In opening the discussion Dr. Stanton referred to 

 experiments recentlv made at the National Physical 

 Laboratorv for the Lubrication Committee with the 

 Lanchester worm-gear testing machine, in which 

 working pressures of several tons per square inch 

 are developed, showing that in the case of all the 

 mineral oils tested the lubrication at a certain tem- 

 Derature suddenly became imperfect, the friction 

 irret?ular. and the eflficiencv of the gear fell off. With 

 fixed oils no such "breaking-down" point was ob- 

 served under the conditions of the tests, and it was 

 found that by adding quite a small percentage of 

 fatty oil to a mineral oil the breaking-down point, 



NO. 2618, VOL. 104] 



though not obliterated, occurred at a higher tem- 

 perature. • 



Mr. Deeley described and exhibited a small hand- 

 driven machine which he had invented for the purpose 

 of measuring the oiliness of lubricants under condi- 

 tions of metallic contact. Three flat-ended metal 

 studs, each 5/32" in diameter, secured concentrically 

 as feet to a metal disc, rested upon another metal disc 

 which could be slowly rotated. The upper disc could 

 be weighted as desired, and actuated a spindle to 

 which a spiral spring and a recording finger were 

 attached. The lower disc, when rotated, carried the 

 studs and upper disc with it by friction until the stress 

 in the spring- caused the surfaces to slip, when the 

 pointer gave the frictional resistance, oscillations being 

 damped by gearing the spring and pointer to a train 

 of wheels. Experiments made with this machine 

 showed that the static friction depended upon the 

 nature of the metal surfaces in contact as well as 

 upon the oil, and the fatty oils which in practice are 

 found to be the best lubricants gave lower static 

 coefficients than the mineral oils. Mr. Deeley 's view 

 is that the oil, or some constituent of it, enters into 

 physico-chemical union with the comparatively rigid 

 rrietallic surface, forming a composite film having the 

 yielding nature of velvet-pile, and that the best lubri- 

 cants are those which produce the most easily sheared 

 contact films. 



Dr. Allen directed attention to the important work 

 of Irving Langmuir on surface films, and suggested 

 that the property of "oiliness" depends upon the 

 chemical forces called into play between the active 

 part of the oil molecule and the solid surface of the 

 bearing, and not only on the nature of the lubricant, 

 but also on that of the solid surfaces with which the 

 liquid is brought into contact. 



Mr. W. B. Hardy referred to the work carried' on bv 

 his son and himself, and discussed by them in the Philo- 

 sophical Magazine of July last. In the apparatus they 

 used there was onl/ one point of contact between the 

 solid surfaces, which consisted at first of a curved 

 watch-glass on a flat glass plate, and latterly of 

 similarly shaped surfaces of bismuth. The force 

 measured was that necessary to cause the curved 

 surface to slide over the plate when lubricated by the 

 films formed by individual chemical substances repre- 

 sentative of various groups of chemical compounds. 

 The authors concluded that the true function of a 

 lubricant was to reduce the energy of the surface, 

 and thereby to reduce the capacity for cohesion and 

 the resistance to slip when two composite surfaces 

 are applied one to the other. This, in the opinion of 

 the present writer, is the conclusion to which all 

 recent work leads, and the problem before the chemist 

 is to determine in what manner the molecular struc- 

 ture of the chemical compounds in lubricants deter- 

 mines their "oiliness." 



The practical side of the problem cannot, however, 

 be neglected, and it is necessary to experiment with 

 commercial lubricants acting between such dissimilar 

 metal surfaces as are used in the construction of 

 machines, in order to obtain the information which 

 engineers require to guide them in the lubrication cf 

 machinery. Mr. Southcombe's observation that the 

 interfacial tension between oil and water and oil and 

 mercurv is much higher with mineral oils than with 

 fatty oils, and that the addition of a very little free 

 oleic acid to a mineral oil greatly lowers the surface 

 tension, is of great importance. It appears that by 

 adding i per cent, of free fatty acid to a mineral oil the 

 lubricating power is increased as much as by adding 

 a very much larger percentage of fatty oil. The 

 emulsifving properties can also be materially modified 

 bv the addition of fattv acids. L. A. 



