280 The Twenty-Ninth General Meeting. 



of the cavity. These act as a diaphragm for the dialysis of the 

 imprisoned liquid, a film of silica deposited on them in a gelatinous 

 form would be the first step in the dialysing pi'ocess ; and this will 

 be one of exfiltration of the ingredients other than the silica, which 

 remains behind, these ingredients passing through the walls of the 

 cavity, while the silica remains in a colloid form ; and this will have 

 subsequentlyshrunk and passed by the ultimate depravation of its water 

 into a delicate layer of chalcedony, or what we may term incipient 

 chalcedony, for its conversion into that mineral will need enormous 

 time. The repetition of the process, sometimes more slowly, some- 

 times more rapidly, would cause the deposition o£ the agate material 

 in successive layers, some with more and some with less of the 

 foreign substances, to which varieties in colour would be due. In 

 Bome of these layers quartz has been deposited, in the so-called 

 mycrocrystalline condition, that is to say, in microscopically minute 

 crystals disseminated through the still colloid chalcedony. In all 

 of them some of the silica has changed from the soluble to the 

 insoluble and denser condition. Perhaps the nature and the character 

 of the other ingredients present and extruded from the crystals 

 regulate the amount of the colloid silica that passes into the in- 

 soluble crystalloid form or becomes actually crystallised. Colloid 

 silica heated to a strong red heat passes into the insoluble variety. 

 But the problem of the conversion at ordinary temperatures of 

 colloid silica into crystallised quartz, or into the crystalloid form of 

 silica no longer easily soluble in alkalis, is one for the solution of 

 which we must look to analogies only ; for we have no actual ex- 

 perience as witness of the change. The spontaneous conversion of 

 transparent barley sugar (a colloid form of sugar) into opaque and 

 in fact coryptocrystalline sugar is a case in point. Arsenious acid 

 will also, after the lapse of many years, spontaneously develop a 

 crystalline form in the substance of its glassy and colloid mass. 

 And probably it is in this way that the insoluble silica which alkalis 

 do not attack in flint and agate — and the crystals of quartz which 

 often line the cavities in both substances — have taken their origin. 

 I have dwelt on the case of agate as being one in which we can 

 foreshadow the method — however incompletely we can explain it— 



