2G2 MR. 0. A. DEIlBY ON NKPHELINE-KOCKS IN BRAZIL. 



oceniTonce of these aggregations in phonolite. It represents a half- 

 buried mass of foyaite near the eastern end of the Santo-Antonio 

 aqueduct, traversed by a dyke of plionolite about a metre wide. 

 Some specimens of similar rock are found abundantly at various 

 ])oints. Irregular holocrj-stalliue aggregations, without definite 

 form, also occur. 



Large phenocrysts of orthoclase, angite, hornblende, titanite, and 

 in one case of ])Iagioclase (the only instance in which this mineral 

 has been observed in the Tingua rocks besides certain pseudo- 

 crystals) are frequently seen half-embedded in the pseudo-crystals, 

 half in the surrounding phonolitic groundmass. Similar crystals 

 form a part of the mass of the former, and in the case of the large 

 aggregates in foyaite it seems to be the rule that an unusually 

 large phenocryst of orthoclase has served as a nucleus *. The 

 crystallization of the larger part of the mass, however, seems to 

 have been coincident with that of the phonolitic groundmass, por- 

 tions of which, as Dr. Hussak has observed, are sometimes included 

 in the pseudo-crystals. While it is in general quite impossible to 

 distinguish accurately between the elements of the first and second 

 generation, it is probably quite safe to refer to the latter such parts 

 as occupy the position which should, in a normal crystal, however 

 rich in inclusions, be occupied by the leucite molecule. Such is the 

 marginal layer, rarely absent, about a millimetre thick, composed 

 of lancet-shaped orthoclase (with nepheline according to Herr Graeff), 

 disposed normally to the faces of the pseudo-crystal. A lucky 

 fracture revealed to Dr. Hussak a skeleton formation, of like compo- 

 sition and aspect, disposed in the interior of the mass according to 

 the faces of the octahedron. A similar parallel (or radial) arrange- 

 ment of lancet-shaped crystals is frequent throughout the mass of 

 the pseudo-crystals, more particularly in the outer members of the 

 large compound groups in the foyaite. From this it may be con- 

 cluded that a large part of the mass of the pseudo-crystals (at least 

 so far as the orthoclase and nepheline are concerned) are of a second 

 generation, subsequent to the phenocrysts of the general mass of 

 the rock, which probably tend to become concentrated in these 

 centres of crystalline activity, and thus become embedded among 

 the later-formed elements which complete the form of the aggregate. 



A zonal structure is sometimes seen in the pseudo-crystals in the 



* This is the cnse with a similar foyaitic aggregate, without crystalline 

 boundaries, attached to the side of a large orthoclase in the phonolite of Fer- 

 nando de Noronha, the only specimen of foyaite known from that locality 

 (i?ee Eosenbusch, 'Mikr. Phys. Gesteine,' pp. 91. 628 ; AVilliam?, op. cit. p. 18o). 

 Large crystals of orthoclase project from the faces of the deco)uposed crystals 

 from the P090S de Caldas locality, referred to analcime in my former paper 

 (p. 470). I am now convinced that there are also pseudo-crystals in the form 

 of leucite, which mineral has been found in fragments of phonolite from the 

 same cutting. The only test applicable was that of crushing and washing the 

 totally decomposed mass. The lieavy residue obtained, consisting of altered 

 fragments of titanite (?). was altogetlier too abundant to be regarded as inclu- 

 sions in a simple crystalline mass. Undoubted pseudo-crystals, as perfect as 

 those in the Tingua phonolite, occur in a porphyritic foyaite near the tunnel 

 on the Caldas railway. A chance fracture revealed a portion of the outer surface 

 of one of these masses showing three faces of the leucitohedron. 



