WHITE ROCK OF SQUAKK BUTTE. 417 



It is closely related to l)arkevikite, as shown by the analysis (quoted later 

 in this article. 



The orthoclase occurs in lath-shaped forms and in irregular grains. 

 Those abutting against sodalite show crystal faces. Associated with the 

 orthoclase is a triclinic feldspar referred to albite. The sodalite is found 

 in irregular grains between the feldspars, allotriomorphic in regard to 

 them, idiomorphic against analcite. The latter, which is in considerable 

 amount, was along with the sodalite separated and analyzed. The anal- 

 cite is thought to be derived from the albite. The rock is calculated from 

 the analysis (given later in this paper) to consist of (3() parts of feldsi)ar, 

 23 of hornblende, 8 of sodalite and 3 of analcite. 



In addition to these fiicts we have only to add that in the additional 

 material studied by us we have detected a small amount of nephelite, 

 which is being changed by borders, bays and tongues of analcite eating 

 into it and thus suggesting an additional origin for the analcite; also 

 considerable natrolite is sometimes present. Its fibrous masses are sec- 

 ondar}' after sodalite and at times it completely replaces it. 



GENERAL PETROLOGY OF SQUARE BUTTE. 



The facts which have already been given in regard to Scjuare butte show 

 it to be one of the most remarkable and interesting occurrences of an 

 igneous rock that has been described and from a petrologic point of view 

 one of the most important; for while the differentiation of a molten 

 magma as a factor in the formation of igneous rocks is now regarded by 

 the majority of petrologists as an established fact, it is also true that the 

 theory has been founded almost entire!}' upon inferential proof and by 

 the exclusion of other hy})otheses. The direct j)roofs which have come 

 under ol)servation have not been all that could be desired, and some of 

 them indeed, as in the case of mixed dikes, have had more than one 

 interpretation. 



In the case of S(|uare butte. however, the proof of differentiation is 

 unequivocal and direct, for in no other rational way, we believe, would it 

 be possible to explain the dis|)ositi(jn of the rock masses, the cone-in- 

 cone arrangement of the two differing masses of intruded igneous rock, 

 HO unlike in chemical and mineral composition, yet geologically a unit 

 and absolutely homogeneous in granularity and texture and so i)erfect 

 in continuity of stru(;ture and platy j)arting. 



It is therefore a matter of interest to compare the chemical and min- 

 eral composition of these two rocks, the syenite and shoidvinite, with one 

 another and see, so far as possible, how and uiKh r wliat conditions the 

 differentiation has taken place. For this ])Urpose the analyses of the two 

 rucks are here compared : 



