SOMK MERCURY STANDARDS OF RESISTANCE, KTC. 61 



difficulties, much valuable information is contained in u paper by Dr. G. J. PARKH!,* 

 " On the Thickness of Liquid Films formed by Condensation at the Surface of a 

 Solid." It appears that, in all wises where condensation of moisture takes place at 

 temperatures not below the dew-point, the thickness of the surface film of water 

 varies with the substance employed and the conditions of temperature and pressure. 

 Tin* thickness of the water film on glass in saturated vapour at 15 C. is about 

 13'4 X 10~ centim. If, at the commencement, the glass be absolutely dry, an 

 interval of 15 or 16 days elapses l>efore the film attains its maximum thickness, 

 although at the end of 12 hours its thickness is half this maximum. This, it should 

 be remembered, results when the glass is exposed to a saturated atmosphere. To 

 completely remove the film, a very high temperature (probably near 300 C.) is 

 required, but the proportions removed after heating to different temperatures has not 

 apparently been investigated. It appears certain, however, from the data given in 

 the paper, and more especially from the conditions requisite to completely remove the 

 film, that a skin constant in thickness to 1 X 10~ centim. should be ensured by the 

 operation of the same cycle of temperature. Thus a glass tube chemically cleaned 

 and dried by heating to a temperature approximating to 80 C. (a current of dry air 

 simultaneously passing through the tube) should, on cooling, have a liquid film 

 condensed on its surface, equal in thickness to those of other tubes treated similarly. 

 After the introduction of a mercury column, the film will separate the mercury from 

 the glass, but, since its constancy throughout all operations may be anticipated, it is 

 no longer a source of trouble. This anticipation amounts to almost a certainty if 

 taken in conjunction with the determinations of resistance and of cross-sections dealt 

 with in this paper. 



Respecting difficulties (1) and (2), these may be simultaneously disposed of by 

 filling the tube at some definite temperature, and ensuring plane termini by the 

 pressure of pieces of plate glass at the ends; or (1) may be overcome by calculating 

 the length of the column equivalent to the meniscus. When the column is supported 

 horizontally, the meniscus is a portion of a curved surface, which is spherical only if 

 the tube be so narrow that the effect of gravity may be neglected. The radius of 

 nine of the tubes employed averaged 0'43 millim., that of the other two approximated 

 to 0'59 millim. Even with these latter, with which the gravitational effect is more 

 marked, and the extreme point of the meniscus does not therefore lie on the axis of 

 the tube, the equivalent length may be obtained!, by assuming the surface as 

 spherical, and the sagitta of the curve as that length of the meniscus observed when 

 looking down at the tube. The calculation is simple. 



Let ABD (fig. 2) represent a section of the meniscus by a vertical plane through 

 the axis xy of the tube, the distance CD being the meniscus length or sagitta. Then 

 the volume of the spherical segment is 7r/~OD + TrOD 3 , and hence the length of 



* ' Proceedings of the Physical Society,' vol. xviii., 1903, p. 410. 

 t R. T. GI.A/EBROOK, ' Phil. Trans.,' A, 1888. 



