6 THE VOYAGE OF H.M.S. CHALLENGER, 



while those parallel to /3 are almost black ; the shape of these sections is nearly that of 

 elongated parallelograms. This mineral is generally well crystallised; some sections 

 perpendicular to the lamellae are hexagonal, two of the sides being very elongated, the 

 other much smaller, and the presence of the latter shows that the biotite has crystallised 

 in this rock with pyramidal faces but little developed, truncated by the pinacoid OP. 

 The lamellse parallel to this last face remain dark through a complete rotation between 

 crossed nicols, and in convergent light show a black cross. Sections of magnetite are 

 wedged into the sides of the biotite, which sometimes appears to be altered, and in that 

 case magnetic iron surrounds the decomposed sections. This association appears to 

 show that the magnetite might very well have been derived from the alteration of 

 biotite. In the ground-mass, small plagioclases predominate : these often assume a 

 radial spherulitic arrangement which is repeated with a certain constancy. This rock 

 might be classed among the micaceous augite-andesites. 



A rock collected in the middle of the Canadas is a basalt ; it is massive, with more 

 or less nodular blackish grey fracture ; it does not show any macroscopic minerals. 

 Under the microscope it is seen to be made up of plagioclase, rather large and irregular 

 sections of olivine, and little crystals of augite ; some black mica is also present. 



A rock from the top edge of the Canadas has a compact appearance and is blackish 

 in colour ; with the naked eye very elongated crystals of sanidine, embedded in a 

 homogeneous mass, are to be seen. Microscopic examination shows a ground-mass 

 composed of small microliths of felspar more or less radiated, large sections of prismatic 

 and colourless felspar, whose sharp outlines stand out clearly in the surrounding mass. 

 These microporphyritic sections are twinned according to the Carlsbad law, and ought 

 to be referred to sanidine ; plagioclastic striae are never to be observed ; the trace of the 

 composition plane divides the section from end to end without ever deviating from a 

 straight line. The usual fissures of sanidine traverse both the twinned individuals as if 

 they formed but one. The extinction of these crystals may be noted here : — Sometimes 

 a section with sharp outlines shows the trace of the twinning reduced to a simple line, in 

 which case it may be assumed that the section is cut nearly perpendicular to the plane 

 of symmetry v. z. I. in the zone P : k ; the extinction of these sections shows that one of 

 the individuals extinguishes nearly always in a direction parallel to the trace of the 

 twinning, and the other at a greater or less angle. Sections, which do not show 

 this sharpness of outline and the trace of the composition plane, ought to be con- 

 sidered as cut obliquely to the zone P : Jc. In this case, if the extinction angles are 

 large and symmetrical, the section is in the zone of the prism ; if, on the other hand, 

 they are unsymmetrical and the angular difference very large, it is probable that the 

 section is in the zone P : M of one individual, and in the zone P : x of the other, i.e. in a 

 zone in which the extinction of one of the individuals increases but slightly (zone P. : M), 

 and that of the other rapidly (zone P : x). If the extinctions are very small it tends to 



