OF THE ROSS ARCHIPELAGO 



147 



the origin of the inclusion is beyond doubt. The following application to the present 

 collection * is therefore put forward tentatively : 



HOMCEOGENOUS INCLUSIONS 



Allomorphous, homologous 



antilogous 



Plesiomorphous, homologous. 



antilogous . 



Polygenous, endopolygcnous. 



exopolygenous . 



Pneumatogenous. 



Enallogenous Inclusions 



Enallogcnous inclusions 



Olivine nodules in limburgites. 

 Gabbroid nodules in basalts. 

 Sanidinites, etc., in trachytes and kenytes.f 

 Olivine and pyroxene nodules in basalts. 

 Hornblcndic inclusions in trachytes. 



Hornblende crystal in basalt. 



? Sanidinites in trachyte of Mt. Cis. 



Sandstones in the trachytes and kenytes. 

 Doleritc inclusions in the trachyte of Mt. Cis. 



The microsanidinite in the basic rock from Hut Point may almost be considered 

 an enallogenous inclusion, for although it is distantly related to its host, it may be 

 assumed that it arose from a trachyte differentiate of the primitive magma and not from 

 the primary basic differentiate. 



VIII. SUMMARY AND CONCLUSIONS 



The following types of inclusions have been described : 



Olivine Pyroxene and Gabbroid nodules in the basalts and limburgites of Hut Point. 



Olivine Sanidinites and microsanidinites in the phonolitic trachyte of Mount Cis. 



Biotite Sanidinites and microtinites in the kenytes. These are new to science. 



Hornblendic nodules in the trachytes. 



Doleritic inclusions in the trachyte of Mount Cis. 



Quartz- pyroxene inclusions in the trachyte of Mount Cis. 



Sandstone inclusions in the kenytes. 



Miscellaneous erratics, presumed to be inclusions. 



Only a limited number of the above types are of sedimentary origin, but those that 

 do occur have an important bearing on the question of the formation of the Ross Sea. 

 Since they are all metamorphosed sandstones, it follows that parts, at least, of Ross 

 Island are underlain by a sandstone formation. This can scarcely be anything else 

 than a down-faulted portion of the Beacon Sandstone. The presence of dolerites as well 

 as sandstones in the trachyte of Mount Cis strengthens this conclusion, since sills of 

 dolerites have been found to be ubiquitous in the Beacon Sandstones of the mainland. 



All the other inclusions are of igneous origin and genetically connected with the 

 alkaline rocks of Ross Island. Many of them are types not met with at the surface as 

 separate rocks, viz., peridotites, gabbros, pyroxenites, orthophyric trachytes, sanidinites, 

 microtinites, camptonites, and an orbicular augite syenite. Their study thus gives 

 some idea of the deep-seated rocks whose formation has accompanied the eruption of 

 the lavas. A fuller idea of the possible differentiates of such a magma as the funda- 

 mental alkaline magma of Ross Island is thus obtained, and more direct comparisons 

 are possible with areas of deep-seated rocks. 



* The inclusions of which the host is unknown cannot of course be classed. 

 f If the slight amount of differentiation be overlooked. 



