ON THE FORMATION OF CRYSTALLINE ROCKS. 283 



prevented me from multiplying results as it Avould have been desira- 

 ble to do. The facts, however, already discovered are conclusive, 

 and show the fecundity of this method of experimenting. 



The principal difficulty consists in finding walls and fastenings that 

 ■will resist for a sufficient length of time the enormous tension which 

 steam acquires when the temperature is raised towards a dull red 

 heat. Water and the matters that are to react are placed in a glass 

 tube, which is then sealed. This glass tube is next introduced into a 

 very thick tube of iron, which is closed in a forge at one of its ends. 

 The other end is often closed by means of a screw, having a square 

 head, which can be turned with a wrench. Between the head of the 

 screw, which should be made with great precision, and the end of 

 the tube a Avasher of very pure copper is placed; it should be thin 

 enough to be crushed by the pressure when the tube is closed, and 

 penetrate into the grooves made for that purpose. For closing the 

 second extremity, however, I have now adopted another plan in prefer- 

 ence; I introduce hot a very strong bolt, which, if ihe welding is 

 skilfully done, becomes part of the tube. A workman must be very 

 skilful to succeed in this operation; for it is essential that the greater 

 part of the tube remain cold, in order that the water in the interior 

 may not, by evaporating, hinder the operation. To counterbalance 

 in the interior of the glass tube the tension of the steam, which 

 might burst it, water is poured around it, between its sides and those 

 of the iron tube which surrounds it. In this way the principal strain 

 is put upon the latter tube, which offers the most resistance. This 

 apparatus, like those which de Senamont used, is laid on the dome, 

 or in the conduits of the retort-furnace of a gas-works in contact 

 with the masonry, which is at a dull red heat, and is buried beneath 

 a thick bed of sand. At a temperature which is a little below nas- 

 cent red heat water reacts very energetically on certain silicates. 



In this manner ordinary glass, at the expiration of a few days, 

 gives two, and often three, distinct products : 1. A white and alto- 

 gether opaque mass, which results from the complete transformation 

 of the glass; it is porous, sticks to the tongue, and would look like 

 kaolin if it were not for its very decidedly fibrous structure. The 

 substance has lost a large part of its weight, nearl}'^ half its silica, and a 

 third of its alkali; anew silicate has been formed which has fixed the 

 water, and belongs, by its composition, to the family of the zeolites.* 

 2. An alkaline silicate which has been dissolved, carrying \vith it 

 the alumina. 3. Often there are developed, in addition, numerous 

 perfectly limpid and colorless crystals, which have the ordinary 



had no fault, and we suppose that it preserves at 300° centigrade the same tenacity as whea 

 cold, such an explosion would indicate an interior pressure of nioie than 1,000 atjQospheres. 

 It is to be noticed that, before bursting, the tube swells out into the form of aa 

 ampulla of one and a half to two inches in length, and opens in its middle so as strikingly 

 to call to mind the gihbi-sity of the profile of Etna, with the central notch of the Val de 

 Bove, the origin of which was long since, by de Buch and Elie de Beaumont, attributed to aa 

 expansive force of the same kind. 



« Bulletin de' In SmU G£ulo(/iqite de France, 2d series, vol. xvi, p. 588. At a higher tem- 

 perature an anhydrous bilicate, which appears to resemble woUabtonite, is obtained. 



