A. B. Howe—Gmelinite from Nova Scotia. 273 
analyses with those given by Rammelsberg, it will be seen that, 
like the majority of those, they give an excess of silica over 
that required by the ration 1:1:4. This excess is very 
marked in the Nova Scotia varieties, but in these it may be 
partly accounted for by a slight decomposition which may have 
taken pla ergen Hill crystals, however, were ve 
evidently sound and undecomposed, and the excess in this case 
can not be accounted for in that way. 
Both the Nova Scotia varieties had apparently undergone a 
slight alteration. The crystals were somewhat porous and hol- 
low, and thus would be exposed to easy decomposition ; or this 
structure may have been the result of alteration. Thin sec- 
tions examined under the microscope with polarized light 
afforded no evidence of free quartz. If the bases had been 
chlorie acid, decomposition with separation of finely-divided 
Silica took place, but there were no traces of gelatinization. 
_ Several experiments were made on the loss of water by heat- 
Ing, but with no satisfactory results. The mineral, even when 
also found at 160°-180° corresponding to a loss of 17°02 per 
be Lagat others were found at 200°-210° and at 250° equal 
19-42 and 20-40 per cent respectively. At every tempera- 
‘ure mentioned the loss was rather rapid at first, peasnally 
ming slower and slower till a constant weight was obtained. 
sar nerently the mineral possesses the property of giving a con- 
t weight at any temperature, which may account for the 
Th various results which other experimenters have obtained. 
d € water is driven off completely at a red heat, and the pow- 
on sintered together into a mass but without absolute 
Itis interesting to notice the fact that the habit of erystalli- 
