462 THE GEOLOGY OF THE NEW HEBRIDES, 



Examination of various types of this andesite discloses the 

 fact that miixh of the fine mo.gnetite which is present, darkening 

 the base, is produced by secondary processes,* and has resulted 

 from the disintegration of pre-existing silicates rendered unstable 

 by succeeding physical conditions. 



Such operations are no doubt widely experienced in the 

 volcanic types of rocks, more especially in the basic varieties, 

 and account for much of the second generation magnetite. 



(d) In the hornblendes of the Santo andesites, as has already 

 been hinted, a certain relation between thickness of altered 

 border and basicity of the hornblende seems to prevail, whereby 

 the more basic types have, in comparative degree, a broader 

 peripheral zone of secondary magnetite and pyroxene. 



In some cases, notably the andesite from the Wai Bubo, this 

 alteration has continued so far that cojuplete magmatic paramor- 

 phisrii of the hornblendes has resulted. When highly magnified, 

 tiny rod-like microlites of augite, arranged parallel to the hornblende 

 cleavage, can be seen amongst the magnetite dust. In the centre 

 of one of the crystals (Plate xxiii., fig. 2) is still preserved a 

 remnant of the former mineral which, with its light yellow 

 colour and faint pleochroism, is decidedly more acid than was 

 expected. 



7. Glomeritic Basalt Porphyrite. 



L o c a 1 i t y. — From the island of Mau, taken from the western 

 spur at an elevation of about 600 ft. 



Macroscopic Characters. — Colour light grey. 

 Fracture uneven. Texture microcrystalline porphyritic. Con- 

 stituents : the only distinguishable minerals are abundant, some- 

 what rounded, faintly lighter-coloured, porphyritic ])lagioclases 

 and occasional dark- stained olivines. 



Microscopic Character s. — Texture : Crystallinity, 

 holocrystalline porphyritic. Grain-si::e and Fabric \ two distinct 

 crystallisations — (a) An elder generation of much larger indi- 



* Subsequently it was discovered that Washington had aheady noted this 

 phenomenon. Journ. Geol., iv., 273. 1S96. 



