24 Proceedings of the Royal Irish Academy. 



per animm from an area of one square metre, according to the experi- 

 ments. 



Such calculations can of course give only approximate evaluation 

 'of surface. We possess no definite knowledge as to the depths to 

 which solutions might in such cases penetrate heneath the surface of 

 the various minerals and exert a solvent action. If this depth is 

 considerahle, which does not appear probahle, the gain in siu-face 

 obtained by reducing the material to very fine particles is more 

 apparent than real. Doubtless, the rate of abstraction of dissolved 

 material is in any case much greater at the actual surface. If we 

 assume such abstraction of material to only go on for molecular 

 distances beneath the surface, or at least distances small compared 

 with the diameters of the particles, the calculations afford at least a 

 definite basis of comparison with purely solvent processes in Nature, 

 for here also a similar penetration of solvent influences occurs. In 

 each case, too, we may assume a somewhat similar protective effect due 

 to residual materials. The conclusion to be drawn is that under- 

 estimate of the surfaces exposed — even on the assumption of cubic 

 particles — is more probable than o^?^r-estimate. 



As regards chemical attack over the surface of glass exposed in 

 each vessel, it may be stated here that this can only cause small error. 

 The glass used throughout (save the U -tubes in the basalt experiments) 

 was Jena glass. This glass has been the subject of tests made at the 

 Technical Institute of Charlottenburgh, which the makers have 

 published, and which apply to the amount of ]S'a20 liberated under 

 various conditions. These show that the area of 500 sq. cms. exposed 

 in the flask, even if the full rate of solution for water at 20° C. 

 continued for three months, would only liberate 0"00017 grammes of 

 NajO. As it is not to be supposed that the primary rate of extraction 

 would continue, and as the mean temperatures were not higher than 

 12° C, in any case the error may be considered negligible. The fact 

 is the area of glass exposed ,is small compared with the areas of the 

 mineral particles. 



The Chemical Results. — The chemical analyses were carried out in 

 the Chemical Laboratory of the Eoyal Dublin Society, under the 

 supervision of, and in part by, Mr. R. J. Moss, V.C.S., Chemical 

 Analyst to the Society, to whom my best thanks are due. Mr. Stone, 

 the assistant, bestowed the utmost pains on the very difficult task of 

 evaluating the small quantities available for estimation. 



Unfortunately the estimate of the alkalies in the case of the sea- 

 water solutions, owing to the indirect methods available and the 



