ROCKS FROM GOUGH ISLAND, SOUTH ATLANTIC. 403 



Comparison with Kocks from other Volcanic Islands in the 



South Atlantic. 



The wonderful uniformity in petrographical characters of the eruptive rocks of 

 the islands of the Mid- Atlantic rise has been emphasised by Prior.* In Ascension, 

 in St Helena, and in Tristan d'Acunha, for example, basalts are associated with 

 alkali-rich phonolitic rocks. In Gough Island the association is similar. The pebble 

 of trachytic obsidian (phleyrose) described and analysed by Pirsson f is strongly 

 alkaline, with soda in excess of potash. The alkali trachytes, with their predominant 

 anorthoclase, are decidedly soda-rich, and may be paralleled with the non-porphyritic 

 trachytes of Ponza type from Ascension described by Renard.J The trachydolerites 

 and essexite, too, as is shown by their high content of soda felspar, have been derived 

 from an alkali magma rich in soda. The trachydoleritic affinities of the basalt dykes 

 of Gough Island have already been pointed out, and very probably the glassy basalts 

 and tachylite are similar in composition. Renard § gives an analysis of a tachylite 

 from the Tristan d'Acunha group in which SiO 2 =48'09, Na 2 O = 5'06, and K 2 0=2"88. 

 The trachyandesitic character of some of the augite andesites || from the same group 

 is evidenced by the fact that their phenocrysts of plagioclase are accompanied by 

 others of sanidine. Nepheline-bearing rocks have not so far been found on Gough 

 Island, nor have soda rhyolites as is the case in Ascension, but it has to be remembered 

 that the specimens collected by Dr Pirie were all obtained from a small area. 

 Further collections will in all likelihood yield other rock types. There can be no 

 doubt, however, that the present collection proves that the rocks of Gough Island 

 have all been derived from a soda-rich alkali magma, and that in all probability they 

 have had a common origin with the rocks of the other volcanic islands in the Mid- 

 Atlantic rise. 



Acknowledgments. 



In conclusion, I wish to acknowledge my indebtedness to Dr JohnS. Flett, F.R.S., 

 for much valuable advice, and to Mr W. F. P. M'Lintock, B.Sc, who has very kindly 

 given me opportunities of studying similar rocks in the Royal Scottish Museum. 



I desire also to express my thanks to the Executive Committee of the Carnegie 

 Trust for defraying the expenses of illustrating this paper. 



* Min. Mag., vol. xiii. p. 261. t Loc. cit., p. 382. 



I Bull. Mus. Roy. d'hist. nat. Belgique, 1887, v. 5. § Loc. cit., p. 210. 



|| Ibid., p. 219. 



