140 THE INCLUSIONS OF THE VOLCANIC ROCKS 



formed into the coarse-grained rocks we now see. The negative evidence against this 

 view is, however, stronger than the positive, and it is perhaps best, in the present state 

 of our knowledge, to consider them also as fragments of deep-seated rocks of close genetic 

 connection with the trachyte magma. 



The materials for the study of the felspathic inclusions of the kenytes is, unfor- 

 tunately, very meagre, but it is of interest to find that both sanidinites and microtinites 

 can originate in kenytic magmas. The most noticeable mineralogical difference from 

 the kenytes lies in the presence of biotite, a mineral not recorded at all from the volcanic 

 rocks. It is well known that a dry melt of orthoclase and biotite in certain pro- 

 portions will recrystallise under slow cooling as a mixture of olivine and leucite. The 

 two last minerals are found occasionally as phenocrysts in the kenytes, so that it 

 appears that the biotite, and therefore the rocks containing it, crystallised under 

 different conditions of pressure from those which prevailed during the formation of 

 the phenocrysts of the kenyte. The inclusions cannot therefore be simple aggrega- 

 tions of phenocrysts. 



There is a close correspondence between the natures of the hosts and of the inclusions. 

 Sanidinites are found in the acid kenyte of Cape Royds which contain anorthoclase as 

 phenocrysts, while microtinites are found in the basic kenytes of the Delbridge Islands, 

 which contain phenocrysts of oligoclase. The augite of both types of inclusion is titani- 

 ferous and not R'girine- augite as in the sanidinites of the trachytes. From this 

 correspondence it may be safely inferred that the inclusions originated from the same 

 magmas as the rocks enclosing them. Further inquiry as to whether they are fragments 

 of already consolidated portions of the magmas or foreign fragments that have been 

 reconstituted by the influence of the magmas cannot be profitably followed on the 

 evidence available. The microlitic structure of the ground-mass points to the former 

 view. 



III. HORNBLENDIC INCLUSIONS IN THE TRACHYTES 



Inclusion in the Trachyte of Observation Hill. Prior has already noticed the rounded 

 brownish enclosures in the trachyte of Observation Hill : " Under the microscope they 

 consist of a dense mesh of interlacing prisms of basaltic hornblende, similar to that in 

 the trachyte, with magnetite grains and only a little interstitial felspar. The hornblende 

 in these enclosures and in the trachytes has the characters of barkevicite." * 



This type is represented by only one specimen in the present collection. It agrees 

 in general characters with those described by Prior, but it may be noted that besides 

 the fine interlacing prisms of hornblende there are a few larger and proportionately 

 stouter prisms which stand to the former as phenocrysts to ground-mass, and thus give 

 the rock a lamprophyric character. The untwinned felspar has refractive indices 

 slightly lower than that of Canada Balsam, and probably belongs to anorthoclase. 



Inchision in the Trachyte of Cape Bird. A specimen of a dark green phonolitic 

 trachyte from the beach pebbles of Cape Bird proves to be rich in small hornblendic inclu- 

 sions. They are more coarsely crystalline than the Observation Hill type, and are 

 markedly vesicular with minute crystals of analcite in the vesicles. Under the microscope 

 hornblende, augite, olivine, felspar, magnetite, apatite, calcite, and analcite may be recog- 

 nised. The structure is distinctly porphyritic ; colourless olivine in prismatic forms with 

 acute pyramidal terminations and pale lilac titaniferous augite in long prisms with 

 obtuse pyramids terminated by the basal plane appear to have formed the earliest pheno- 

 crysts, since hornblende is occasionally moulded on both minerals. A considerable 

 amount of magmatic resorption has affected the augite in cases where it is enclosed with 



* hoc. cit., p. 118 and Fig. 66. 



