= = + 077406, and to 210 27-y b (13'724 − h) + c(13'724 h)2 + d (13·724-h)", where b = - 1°0326, c = + 43006 and d express the other equations in a similar manner. The calculated and experimental values of y are : From a comparison of theory with experiment it appears that the probable error per cent. on the average weight calculated is 1'48. Here, also, it would doubtless have been fairer to omit Section 1. The point h=3'1 indicates the theoretical position at which the chemical effect of the unexposed trunk as a mineral carrier is at its lowest. It originates, doubtless, with the sudden demand for ash by the large lowest branches which occur near there. The upper critical point h=13724 depends on a similar demand due to the exceptional vitality of the extreme summit of the tree. It that this demand is so sudden as to be to a slight extent appears oversatisfied. PART II.-LIME. By EDMUND J. MILLS, D.Sc., F.R.S., and The hydrochloric extract of the silica in Part 1 contained, of course, the lime of the ash. This was determined by the usual methods, and doubly precipitated in order to ensure the absence of magnesia. As in the case of the silica, the determinations were all in duplicate, and we report the means of these. If these weights be consecutively added (y), the relation of their sums to the corresponding heights (h) can be fairly represented by the equation. 252'49-y (14 h)b + (14 − h)2 + (14-h)3d, where b = '89571 c = = *20992, d h. The probable error of a single comparison is 2'71 = about 1'7 per cent. on the average quantity determined. Much of this, however, may (as in the other cases) depend upon casual phenomena otherwise connected with the growth of the tree. A drawing of this curve is given in Fig. VI. = 3 be put in the form = 2'5, a critical lower point somewhat below the silica lower point. Taking the corresponding values. d == (h−2·5)2 + d (h-25)3 where a= 58°497,b= 44'470, c=28389 031315; which could give h 30 219 as an upper critical point. It is obvious that this point is quite of an ideal character, being above the actual summit of the tree. Although we have indicated lower and an upper critical points in the bole only, it would seem reasonable to suppose that similar points must occur in all branches, and even twigs. Professor Mills afterwards showed his Artiograph or Curvesmoother an instrument whereby the smoothing of experimental curves is much facilitated, and which may also be used for drawing. A description of this is reserved for a future occasion. curves. X-On the Thermal Conductivity of Substances of very low Conductivity. BY VINCENT J. BLYTH, M.A. [Read before the Society, 7th March, 1900.] It is the purpose of this short paper to describe a series of experiments undertaken at the suggestion of Lord Kelvin, with a view to the determination of the thermal conductivity of certain sub stances whose conductivity is low, and to give some of the results obtained. Since the comparative experiment of Ingenhauz, in 1789 much has been done by Forbes, Tait, Kelvin, Angstrom, and others in the subject of thermal conductivity, but most of these experimenters have directed their attention to metals and highly conductive substances; and not so much has been done in the investigation of materials of low conductivity or "heat insulators” as they have been called. In 1873 Stephan (1) determined a result for the conductivity of air which closely agreed with the value obtained by Maxwell from calculations based on the kinetic theory of gases. Within more recent years researches conducted by Lamb and Wilson (2) have determined results for air, paper, felt, asbestos, and a heat insulator known as "kapok"; and in an elaborate paper, by Milner and Chattock (3), on the conductivity of water, a result for cotton wool is given. Still more recently an investigation of the conductivity of vulcanite has been made by Osgood Pierce (4). It was primarily for the purpose of determining the insulating efficiency of kieselguhr that the experiments to be described were (1) Jour. de Phys., tom. II. p. 147. (3) Phil. Mag., July, 1899. |