338 L. V. Pirsson — Geology of Bermuda Island ; 



Mode. — Both the character of the rock and of the sections 

 which can be cut from it forbid any accurate computation of 

 the quantitative actual mineral composition. Judging from 

 what is seen iu the sections and from the analysis presented 

 later, a rough approximation in which only the chief original 

 minerals are considered, might be, iron ore 8 per cent, apatite 

 2 per cent, olivine 5 per cent, pyroxene 35 per cent, melilite 

 35, analcite 15 per cent. 



Classification. — The rock is evidently to be classed asameli- 

 lite-basalt, though from the relatively small amount of olivine 

 it contains it is near the border line of a melilitite if, following 

 the analogy of nephelinite and nephelite-basalt, that name 

 should be applied to rocks of this class without olivine. A 

 near relative of this rock would appear to be the cappaelite of 

 Sabatini in Umbria.* 



Lamprophyric Lavas. 



These rocks occur chiefly in the upper layers of the unal- 

 tered lavas, but appear also in the samples taken at lower 

 levels. Since the casing was discontinued at 785 feet, soon 

 after the firm rock was reached, there must be some uncer- 

 tainty as to their occurrence at the lower levels, since detached 

 fragments from above, to some extent, must have been pre- 

 cipitated into the hole. Their appearance at some levels and 

 not at others is, however, indicative that they belong to some 

 extent in. the lower zones. They have also been found in the 

 upper zone of altered sedimentary deposits, as would naturally 

 be expected, since they were apparently the latest forms of 

 lava ejected. 



As noted by Dr. Thomas, there is a considerable diversity 

 among these rocks, and yet they are so closely related by transi- 

 tion forms and similarity in character of constituent minerals 

 that it seems best to consider them as a single group. They 

 are also closely related to the melilite-basalts, and it is not at all 

 certain that there are not transitions in this direction also, or 

 that they may not occur as varying portions of a single flow. 



One variety, as found in a chip from the 1045-1075 sample, 

 consists of a network of fine interlaced prisms of a brown 

 augite so thickly packed that there seems not much space left 

 for other minerals. It must form perhaps 75 per cent of the 

 rock. In the interstices are analcite, occasional flakes of cal- 

 cite, and a brownish, fibrous substance with aggregate polariza- 

 tion varying from feeble to strong in different spots of the 

 same area. Sprinkled generally through this are small grains 

 and octahedra of perovskite, more or less marginally converted 

 *Boll. E. Comm. Geol. d'ltalia, 1903, p. 376. 



