On the Devitiificalion of Glass* 265 



of the paste of the glass, and which tend to collect them- 

 selves into stars of different forms. These stars are all nearly 

 of the same size, and about two or three millimetres in 

 length. V^hen this crystallizalion is effected of itself in 

 large masses at the boitom of the furnaces, the colour of 

 the glass becomes darker, inclining to black, bv the pre- 

 sence of a certain quantitv of ashes mixed with it. The 

 striated st?.''rf here mentioned become the more numerous 

 as they recede ruriher from the side in contact with the fire. 

 Crystallizations first insuKui^.d are soon succeeded by a niuss 

 entirely crystallized, in which nothing of the vitreous cha- 

 racter is d:stmguished. 



Such arc the most usual characters of crystallization; 

 l)Ut one often sees others which are certainly owing to 

 chance, snd which deserve to be observed on account of 

 their variety*. 



I have son:^ pieces of glass containing crystals of so great 

 tenuity that they can scarcely be seen by the help of a mag- 

 nifying irhi^s, llie\ are prisms diverging from tiie same 

 centre and forming stars which are often not more than a 

 mHiini Ire in si^e : ihtir unjon seems to be a slight obscu- 

 141 . lu iLl- paste of the glass. 



;i'oma pi' ces exliibii the aspect of a saline crust applied 

 en p. i-reign bo:ly with which thp glass was in contact. In 

 s .: <c, this crust, composed of striated paps, seems :' ad- 

 vance more and more, and to gain on the glass. 



There Is still aaoiher variety more curious than an\' of 

 those already mentiofied ; it is that observed m the middle 

 pf the paste of thi; glass; centres of crystallization like peas, 

 and almost sin>iLir to grains. Thev are small globules flat- 

 tened at the two cads, with an umbilicus in the middle of 

 each dcpvt ssion. The sides have ribs like the seeds of the 

 capucin, and these ribs are always six in number. 



Wy Intention is, when I can procure a sufficient number 

 of these singular crystals, to analyse them, in order to dis- 

 cover what is the. earth which affects so extraordinary a 

 form. 



Such, in a few words, are the principal facts bv which 

 the precipitation and crvstallization of i^lass are character- 

 ized. It is seen that tlicy arc of the same nature as ihose 

 produced by the cementation indicated by Reaumur, and 

 that the result is always a devitrification more or Icis abso- 

 lute. 



• M. Sage lias "in liis pos<!es>.if)n n piece nf gl'ss crystallised into Sa- 

 la'ilC prism* of six jittnci anil cn-ircly tl(.\i:u'tkil. 



VVhca 



