298 dr. h. s. Washington on the [June 19 1 4, 



found in the rare micrographic patches, as noted above. But the 

 total amount of these must be very much less than one half of 

 1 per cent., probably not more than one tenth of 1 per cent. They 

 certainly cannot be present in amount sufficient to account for the 

 more than 1*5 per cent, of zircon which the amount of zirconia 

 would furnish. We must, therefore, conclude that by far the 

 greater part of the zirconia exists in the pyroxenes. This seems 

 the more probable, as many of the sodic pyroxenes, such as 

 lavenite, hiortdahlite, etc., contain this element as an essential 

 constituent. This point will be adverted to again. 



The rare earths cannot be present as monazite or xenotime, as 

 the small amount of phosphorus pentoxide is needed for the apatite 

 known to be present, — a point also mentioned by Prof. Judd, — 

 and there is no escape from the conclusion that the cerium, like 

 nearly all the zirconium, must be present in the pyroxenes. It 

 may be mentioned that the amount of ceria in rockallite is the 

 highest known to me in any rock, among the many thousands of 

 analyses that I have collected and studied, with the single excep- 

 tion of a nepheline-syenite from Almunge (Sweden), which is 

 shortly to be described by Dr. P. Quensel. This rock contains 

 48*60 per cent, of silica, 8'74 of soda, and 059 of ceria, according 

 to an analysis by Dittrich kindly communicated to me by 

 Dr. Quensel. 



Mode. 



As the thin sections were small and the granularity relatively 

 coarse, measurement of the mode by Rosiwal's method was not 

 undertaken. But, since the mineral composition of the rock is very 

 simple, the mode may be readily ascertained by readjustment of 

 the norms, which are given on a later page. In these calculations 

 orthoclase is reckoned in with albite, the minute amounts of 

 magnetite and apatite are neglected, the ilmenite is assumed to 

 enter aegirite (for reasons which are given later), and the zirconia 

 to enter the pyroxenes. Data, such as would enable us to ascertain 

 the exact relative amounts of the green aegirite and brown acmite, 

 are lacking ; but the former is undoubtedly present in greater 

 quantity. The mode of I is taken from Prof. Judd's paper : — 



I. II. 



Quartz 38 30'4 



Albite 23 26'4 



^Egirite-acmite 39 43-2 



It will be seen that these figures confirm the conclusion drawn 

 from the first inspection of the analyses, that my portion of the 

 Inskip specimen contained somewhat less quartz and more albite 

 and pyroxene than that analysed by Makins. It will also be 

 evident, on comparing the modes and norms, that the mode is 

 essentially normati vc. 



