140 THE VOYAGE OF H.M.S. CHALLENGER. 



veyed to the place where they were found, by icebergs, or brought up from great depths 

 as enclosures by neo-volcanic eruptive masses. The former hypothesis seems very 

 probable, and is confirmed by taking into account the changes of level which the land 

 has undergone. During the periods of submergence the ice-packs detached from the 

 Antarctic continent and driven towards the north, as at present, may have dropped the 

 rock fragments which they carried. This is not a mere supposition ; the Challenger 

 dredgings between Kerguelen and Heard Island have brought up blocks of consider- 

 able size, which belong to the crystalline and schisto-crystalline series : granite, diorite, 

 gneiss, &c. No one can doubt that these rocks have been carried by floating ice 

 to the place where they were found, and we may add parenthetically, that they prove 

 the existence of an Antarctic mass of continental land, to which Mr. John Murray has 

 recently directed the attention of geographers. It may possibly be, however, that the 

 fragments viewed as ancient rocks have been carried up by the trachyte and basaltic 

 masses in their passage through the underlying strata. The typical volcanic outflows 

 show many examples of similar facts, but nothing in Professor Eoth's description 

 enables one to decide this question. While fully recognising the care which he has 

 taken to establish his diagnoses of the rocks in question, we may yet insist upon the 

 great difficulties in the way of precise differentiation between the ancient and modern 

 crystalline series. These difficulties increase with the amount of alteration of the rocks, 

 and very often it becomes impossible to solve all doubts even with the microscope. 

 Of this no further proof is required than the discussion still going on as to the true 

 basis for a classification of eruptive rocks. This is not the place to carry ou a contro- 

 versy, but, confining ourselves to the subject in hand, we may remark that it is just 

 in the case of rocks like those of Kerguelen, classed as of the ancient series, that the 

 difficulties are greatest. In this way certain granular eruptive masses, which we have 

 described, from Foul House Bay may be equally well classed as peridotic diabases or as 

 dolerites, but their association with basalts gives greater probability to the deter- 

 mination we have thought it right to adopt. However it may be, we must acknowledge 

 that in all the specimens from Kerguelen we have examined, there is not one which 

 can be certainly referred to massive rocks of the ancient type. 1 



The superposition of basalt sheets and their scoriaceous surfaces show plainly that 

 they have accumulated like lava in successive flows. They must have been spread 

 one over another at intervals, this periodicity of the eruptions being shown by the 

 alveolar structure of the surface of the beds. It is evident that if these basalts 



1 Amongst the rocks from Kerguelen submitted to us there was one without any indication of locality, collected 

 by Mr. Moseley. This at first sight resembles those of the ancient type. The microscope shows a greyish ground-mass 

 very like that of porphyries. Silica predominates in irregular grains, and some sections are similar to altered felspar. 

 It cannot, however, be classed with porphyry, for microscopic examination shows a section of vegetable origin filled 

 with quartz and micaceous substance. Hence we believe this to be a trachytic tufa, the constituent elements of which 

 were bound up with vegetable remains by an infiltration of silica, such as the amygdaloidal rocks of Kerguelen and the 

 fossil woods exhibit so abundantly. 



