650 KEPORT— 1887. 



far above it. The stopcock was then cautiously closed and the small portion of 

 liquid above it, removed by means of a capillary pipette, was submitted to appro- 

 priate tests for the substance in solution in the contents of the body of the tiask. 

 In contrast with a sample of equal volume taken from the flask itself, the portion 

 which had been slowly driven up by expansion was found to contain a diminished 

 amount of material in solution, often a very notably diminished amount, and in 

 two or three instances practically none. 



All tbe solutions tried were filtered beforehand through four thicknesses of fine 

 close filtering-paper so as to remove suspended particles of solid matter. No film 

 consisting of, or rich in, the material dissolved in the bulk of the liquid could be 

 detected on the inner surface of the tube in its upper part, so that the separation 

 could not well be attributed to surface adhesion. 



The extent of exposure to the air on the small cross-section of the tube would 

 hardly allow of an explanation being foimd in chemical change of the dissolved 

 material. 



The term apantlesis (d-TdvrXrja-is) might be used to signify a draining away of 

 some of the molecules of the solvent undergoing expansion from amongst those of 

 the colloid solid in solution, yuch drainmg away would seem to connect itself on 

 to change in the opposite direction, but leading to the same result, when the colloid 

 begins to separate out by gelatinising on cooling. 



The conditions which seemed most to influence the production and the distinct- 

 ness of the phenomenon were — flrst, the proportion of the colloid solid in solution ; 

 and second, the time occupied in the rise of temperature of the liquid. In regard 

 to each of these conditions there appeared to be a certain point at which the 

 separation was most notable, above or below which it became gradually less 

 distinct, and sufficiently removed from which it was not observable at all. 



Sub-Section B.— CHEMICAL SCIENCE. 

 1. The Chemical Structure of some Natural Silicates. Bij F. "W. Clarke. 



The common impression that the silicates are exceedingly complicated is pro- 

 bably erroneous. The complexity is apparent, not real. Isomorphous mi.xtures 

 e.xist, impurities occur, and ine.xact analyses are published ; and these causes 

 account for the prevalent belief. The natural silicates are generated under con- 

 ditions which preclude great complexity, such as conditions of high temperature, 

 &c. They are stable, and therefore presumably simple, and they are moreover few 

 in number. Only five or six hundred are known as natural minerals ; whereas, if 

 they were as complicated as many organic bodies, thousands should be commonly 

 found. 



Eliminating the errors due to isomorphism, impurity of material, &c., it is 

 found that all double silicates may be represented as substitution derivatives of 

 normal silicates. The forraulre so developed well represent the natural associations 

 and alterations of mineral species, particularly among the aluminum salts. Pos- 

 sibly the same generalisation maybe extended beyond the silicates, so as to include 

 all double salts, although the double acetates, formates, haloids, &c., oflPer diffi- 

 culties. 



2. Apparatus for Measuring the Volume of Gas evolved in various Chemical 

 Actions, with or without the Application if Heat, with prnjoosecl Ex- 

 tension to Organic Analysis, and to the Continuous Determination of 

 Abnormal Vapour Densities. By P. W. Watkix, M.A. 



The apparatus consists of generating tubes, flasks, and U measuring tubes, in 

 direct communication with the generatiug tubes ; the measuring tubes contain 

 water, and readings are taken with the water at the same level in the two 



