1894. PROCEEDINGS OF THE NATIONAL MUSEUM. 673 



Assiiniing' that all the chlorine belongs to the socialite, the above 

 analysis indicates the rock to contain nearly six per cent of the mineral. 



The composition of the two rocks just described, and their intimate 

 association even in widely separated areas, are peculiarly interesting- 

 in the present state of petrographic knowledge. It is evident that on 

 structurj^l grounds, such may be best classed with the lamprophyres, 

 though they differ from any thus far described in many important par- 

 ticulars. This is eminently true with regard to the more basic one of 

 the two, and it seems impossible to give it a specific name without 

 coining one entirely new, a proceeding which, in my present frame of 

 mind, is quite objectionable. Considered as a lamprophyre it would 

 seem to stand distinct from the monchiquites as described by Prof. 

 Eosenbusch,* in the presence of a feldspathic rather than a glassy 

 base, though such a distinction can S(!arcely be considered an essential, 

 since such might result from merely slight differences in rates of cool- 

 ing. On purely chemical grounds it is further separated from this 

 group by the high percentages of silica and magnesia, and the fact 

 that the potash preponderates over the soda. From other members 

 of the lampro])hyre group, as described by Chelius,t Goller,| J. F. 

 Williams, § Harker, || Doss,1] and others, it differs in equally important 

 particulars. Its closest homologue so far as shown by existing litera- 

 ture, appears to be among the rocks forming the "exceptional dikes and 

 flows in the Absaroka range" of Wyoming, as described by Iddings, 

 and to which reference has been made above. These are regarded by 

 Professor Iddings as forming a part of a series "grading into the nor- 

 mal basalt of the region." So ftir as the Boulder Creek locality is 

 concerned, there is nothing to suggest any such transition. It is, how- 

 ever, very probable that a further comparison of these with the rocks 

 of Fort Ellis (p. G41), Cottonwood Canon (p. 666), Flathead Pass (p. 643), 

 and Horse Shoe Bend (p. 649) might throw more light on the subject. 

 Their general similarity in composition as well as association is cer- 

 tainly very suggestive. 



At South Boulder the eruptives occur in the form of successive sheets 

 of which the lower is a compact hornblende andesite which is succeeded 

 by a semiglassy hypersthene andesite, and this, in its turn, by basalt 

 followed by a small sheet of rhyoUte. 



* Min. u. Pet. Mittheihmgen, 1890, p. 445. 



+ Neiies Jahrbuch fiir Min., etc., 1888, ii. Band, 1. Heft, p. 67, 



X Ihid., 1889, VI. Beilage-Band, p. 485. 



§ Vol. II, Ann. Rep. Arkansas Geological Survey. 1890. 



II Geological Magazine. May, 1890, p. 199. 



H Min. u. Pet. Mittlieilungeu, xi. Baud, 1. Heft, p. 17. 



Proc. N. M. 94 43 



