ERUPTIVE ROCKS FROAI MONTANA. 55 



1:3:8:2. Considerino-. however, the exceedingh' small 

 quantitv of substance used for this analysis, the result ev- 

 dently points more closely to analcite than to any other 

 other mineral. 



The first question in the interpretation of this mineral 

 is naturally: Are not the crystals isolated and analyzed 

 secondary products ? It does not seem possible to me to 

 regard them as such: the rock is often fresh and shows 

 no trace of decomposition; even the olivines are usuallv 

 clear and unattacked; the crystals are homogeneous and 

 isotropic, the cleavage is often distinct. Nepheline could 

 of course not be the mineral from which the analcite might 

 be derived ; the form of the crystals prohibits that suppo- 

 sition. Verv little choice then remains: sodalite, haiivne 

 nosean or leucite are the only possible minerals. A decom- 

 position of either of these to analcite could of course take 

 place, but that it could have occurred and left the rock in 

 such a fresh condition, making each crystal a separate 

 individual of analcite seems exceedingly improbable. Be- 

 sides, the form of the crvstal and the lack of inclusions 

 of augite crystals militate strongly against the supposition 

 that leucite could have been the primary mineral. 



The absence of chlorine and sulphur, except sometimes 

 in traces, in the isolated mineral, speaks equally strongly 

 against sodalite, noseane and haiiyne. 



The groundmass in these rocks, as in that section for 

 example, represented in fig. i , consists of small, dark green 

 prisms and irregular grains of augite, a second generation 

 of small analcite crystals and magnetite. There is prob- 

 ably no glass present, though if it were it would be difii- 

 cult to distinguish it from the isotropic analcite. The 

 larger augite crystals are sometimes surrounded b}' a ring 

 of smaller analcites. 



