2!;2 



bounds the internal cavities of the bomb. Sections show 

 that this inner material is largely crystalline. Perhaps 

 one-third of its substance is composed of well-marked 

 crystals of triclinic felspar, olivine, and augite, the rest 

 being a black ground material, opaque in sections of 

 ordinary thickness, except where relieved by micro- 

 crystals of felspar. 



Foerstner' s Description of Speci7nens from Graham^ s Isle 

 {Ferdinandea) . 



With the above it is interesting to compare Foerstner's 

 account of the specimens from Graham's Isle which he 

 examined {Min. and Petr. Mitth., v., new series, 1883, 

 pp. 388-96). I suppose specimens are scarce. He ob- 

 tained three. That from Strasbourg consisted of light- 

 gray lapilli, apparently altered by acid vapours. It 

 differed much macroscopically from the other two. 

 These lapilli contained magnetite and plagioclase. 



The other two were as follows (I give an abstract of 

 Foerstner's description) : — 



(i) Specimen in Museum at Naples was pumiceous 

 and almost foamy, and had a brown ground mass with 

 glassy lustre, in which crystals of plagioclase, olivine, 

 and magnetite. 



(2) Specimen from Palermo (collected by G. Gemmel- 

 laro) was black and vesicular, but in sections was very 

 similar to (i). It appeared to be largely crystalline. 

 Plagioclase most conspicuous, but also augite, olivine, 

 and magnetite. 



I have not seen the specimens, but the descriptions 

 accord well with those given above for the outer and 

 inner material of the recent bombs. As will be seen 

 below, they also agree well enough chemically. 



Description of certain Basic Scorice from the Island of 

 Pantelleria. 



I did not obtain specimens of the plagioclase-basalts 

 from the Cuddie Monti and S. Marco, analyzed by 

 Foerstner (see below), but I have basic scoriae from 

 different localities in the north-west part of the island. 

 Specimens from Cuddie Brucciate, while differing in 

 structure, agree mineralogically with the recent ejecta, 

 having crystals of triclinic felspar, olivine, and augite, 

 set in a glassy or opaque black ground, as the case 

 may be. 



Further, some very black and rough specimens, from a 

 small patch of basic scoriae north-east of the Bagno del 

 Aqua, not only agree mineralogically, but are in texture 

 not unlike the intermediate portion of the bomb described 

 above. 



A Comparison, as to Chemical Constitution, of the Pro- 

 ducts of the Recent Eruption with those of Graham's 

 Island, and certain Lavas of Pantelleria, Etfia, and 

 Vesuvius. 



In the following table the iron oxides are taken to- 

 gether. In the case of the Etna lava of 1865, the 

 analyses by Fuchs and Silvestri disagree as to the pro- 

 portions of FeO and FeoOg. 



The results in the first four columns are from a table in 

 Foerstner's paper {loc. cit.). For the fifth column I have 

 to thank Mr. Geo. H. Perry, who has made an analysis 

 of part of the large mass above mentioned, and also Prof. 

 Thorpe, F.R.S., who kindly permitted the work to be 

 done in the laboratory of the Royal College of Science. 

 Mr. Perry's analysis will be found in full below. 



This table speaks for itself. There is in the recent 

 bomb a little less silica and a little more alumina than in 

 the specimens from Graham's Isle, Etna, and Pantel- 

 leria, while the percentages agree with those in the 

 specimen from Vesuvius. 



NO. I 159, VOL. 45] 



We may note that Foerstner ^ describes a decrease in 

 the amount of silica in the rocks erupted on Pantelleria 

 itself during the later part of its history. Thus, the 

 oldest rocks of the island are described by him as phono- 

 lites, liparites, and andesites, probably of Tertiary age, 

 containing from 60-73 per cent, of SiOs- Then, coming 

 to the group of rocks which, as containing the mineral 

 " cossyrite " (rather, cenigmaiite) and much soda and 

 iron, he distinguishes by the name of " Pantellerites " 

 (which rocks form the greater part of the surface of the 

 island), he finds that the older ones contain about 70 per 

 cent, of silica, and the younger only 67 per cent., thus 

 heralding the outburst of basic rocks (lavas and scoria) 

 containing some 49-50 per cent, of silica (see table 

 above), which form the most recent rocks of the island, 

 and are confined to the north-west part of it. 



That we should find a still further decrease of Si02 to 

 46-40 per cent, is, so far, interesting. I have not come 

 across an analysis of any very recent lava of Etna to 

 know whether the lava of that district shows any tend- 

 ency in the same direction. 



In conclusion, I must express my thanks to Prof. Judd, 

 to whom I am indebted for reference to the authors 

 quoted, and for loan of papers. 



December 26, 1891. Gerard W. Butler. 



Analysis of Volcanic Bomb, Pantelleria, October 1891. 

 Percentage Composition of Washed Powder, dried at 120° C. 



As the powder was magnetic, FegOs was probably 



combined with FeO to form Fe304. This would give — 



FeO 7-55 



FegOj 4-18- 



December 22, 1891. Geo. H. Perry. 



' " Nota preliminare sulla Geologia dell' Isola di Pantelleria' 

 geological map), Boll. Com. Geol. d'ltnl., 1881. 



(with 



