THE DEPTH AND MARINE DEPOSITS OF THE PACIFIC OCEAN. 25 



the plane (010), a determination quite reliable in the case of such highly basic 

 felspars. Good results are also obtained by the use of liquids of known re- 

 fractive index, as advocated by Schroder van der Kolk. The plagioclase 

 is present as (1) niicrolites, (2) fragments of relatively large crystals, and 

 (3) rarely in well-defined crystals with crystallographic outlines. 



Orthoclase was not met with in the samples studied from this cruise, 

 either through its being really absent from the mother-rocks, or through 

 its having been quite decomposed. On the other hand it was found in a 

 few deposits collected during the 1899-1900 cruise. It may not be out of 

 place to draw attention once more to the fact that potassic felspars weather 

 at a remarkably high rate under the action of sea- water. 



Volcanic glass is one of the most typical constituents of the area surveyed 

 by the " Albatross." It is quite colorless in certain cases, whilst distinctly 

 brown and smoky in others. It is always present as sharp splinters of vari- 

 ous sizes, from microscopic fragments to pieces quite visible to the naked 

 eye. 



Pumice. Large fragments of pumice were not often dredged during this 

 cruise; on the other hand minute particles of finely comminuted pumice are 

 of frequent occurrence, even at long distances from volcanic centres. 



Palagonite is in certain cases extremely abundant, forming large cakes of 

 palagonitic tuff. It forms angular grains of variable size (from 0.1 mm. to 

 2 or 3mm.), yellowish or light brown in color, having a resinous lustre iind 

 conchoidal fracture. It is quite opaque unless in very thin sections, in whicli 

 case it is slightly pellucid ; it is also quite amorphous, exhibiting no trace of 

 birefringence. As to its origin, it is, in the case of the material studied here, 

 undoubtedly due to the decomposition of volcanic glass; in fact, certain 

 specimens were observed showing the gradual transformation of the glass 

 into palagonite. The latter is often accompanied by other products of 

 decomposition, usually yellow or yellowish green, and exhibiting crypto- 

 crystalline structure ; decomposition is too far advanced to permit the 

 determination of the original minerals. 



Hornblende was only seldom met with. 



Chlorite., the name being used in a broad acceptation, is fairly common, 

 but it was found impossible to determine it specifically, owing to its crypto- 

 crystalline structure and absence of well-defined optical properties. It is 

 quite possible that this substance may be the product of the transformation 

 of more than one kind of ferro-magnesian mineral. 



