ITALIAN PETROLOGICAL SKETCHES 373 



rock was observed, that at the Asteria di Biagio, near Orvieto.^ 

 It may be noted here that in all these leucitic rocks the domi- 

 nant feldspars are orthoclase and a basic plagioclase, that augite 

 and diopside are the prevailing ferromagnesian minerals, that 

 both biotite and hornblende are rare, and that they are rich in 

 potash and also in lime. They thus resemble in their broad 

 features the non-leucitic rocks — apart from the presence of 

 leucite. 



Turning to Table V it will be observed that there is a break 

 in the silica percentages between 50 and 55. Below the former 

 are the leucitites, leucite-basanites, and most of the leucite- 

 tephrites ; between the two very few analyses are seen, while 

 between 55 and 56 there is a large number of analyses, and 

 above this last a much smaller number, with silica running up to 

 nearly 60, which is the extreme figure for leucitic rocks. This 

 feature is best seen when all the analyses available are tabulated. 

 Space considerations prevent this being done here, but the break 

 between 50 and 55 is very evident. The number of analyses 

 represented is so large, and covers so many varieties of rock, 

 that this clustering of the silica about 49 and 56 may be reason- 

 ably assumed to exist in fact and not to be due to chance in the 

 selection of analyzed material, as might well be the case with few 

 analyses. We may note also in the analyses of Table V that, 

 while alumina is somewhat variable, it is in general higher in the 

 basic group than in the more acid, as is also true of iron oxides 

 Lime and magnesia show the greatest differences, there being 

 more than double the amount of them in the basic than in the 

 acid group. There is less difference to be observed in the alka- 

 lies, they being very high in both groups, but perhaps more so 

 in the acid one. 



Coinparisoti of the trachydoleritcs and leucitic rocks. — The general 

 similarity of the two groups has already been noticed, and it is 

 further of great interest to observe, on referring to Tables I, II, 

 and III, that a clustering of the analyses about definite points 

 is to be found in the analyses of the trachydolerites as well as 

 'Jour. Geol., IV, 558, 1896. 



