114 



DR. J. W. EVANS ON THE ROCKS OF THE [Feb. 1906, 



Here, as at the Caldeirao do Inferno, the river separates into a 

 number of rocky channels which, after a fall of 4 feet, reunite in 

 the pool below the cataract, where the steamers that ascend the 

 river from the Amazon are moored. 



The rock (M 13) is a coarse and handsome granite, with large 

 red felspars occasionally showing Carlsbad-twinning and traces of 

 perthitic structure. They sometimes measure 2 centimetres or more 

 in diameter. There are smaller crystals of white felspar, abundant 

 grey quartz, and a little black mica. 



Under the microscope, the predominant felspar is seen to be 

 a microcline-microperthite, in which the albite occurs either in 

 more or less linear enclosures, or in broader rectangular or irregular 

 patches connected by fine veins. Independent albite is found, and 

 orthoclase is also present. 



Quartz occurs in allotriomorphic masses constituting about a 

 fifth of the rock. In some places, in the intervals between the 

 larger crystals, there is an aggregate of small rounded grains, most 

 of which are albite, although a little quartz also occurs, and the 

 whole has a granulitic facies. There are a few crystals of biotite, 

 and a little green hornblende with low angle of extinction and low 

 birefringence ; it sometimes occurs in micrographic relation with 

 the orthoclase. 



At one point are some small 

 crystals of a tetragonal mineral with 

 very low birefringence. They show 

 well-marked lines of growth, and 

 may be altered zircons. There is 

 00 ^S^\ X a ^ s0 au obliquely -cruciform twin 



^^ of another mineral, apparently 



monoclinic (see fig. 3), which I 

 believe to be allanite (orthite). 

 At some points it is dark brownish- 

 yellow and more or less turbid, 

 at others it is pale yellow. The 

 distribution of the colours is ir- 

 regular, but the latter usually occurs 

 as a border to the former. The 

 angles seen in the section are near 

 those of the combination of the 

 forms 001, 100, and 101 of allanite. 

 The plane of twinning appears 

 to be the face 101. l This is 

 approximately the direction of vi- 

 bration of the wave-front (a') with 

 greater velocity, and the position of extinction is almost identical 

 in the two portions of the twin. The same direction is also that 

 of greatest absorption, but the pleochroism is very feeble. The 



Fig. 



3. — Section of twin- 

 crystal of allanite (x51 

 diameters). 



001 



[Some of the smaller faces are im- 

 perfectly developed in the original.] 



I belieye that twinning on this face has not been previously recorded. 



