250 /. H. L. VOGT 



upon a downward sinking of the crystals/ If a zonal plagioclase 

 sinks in a flow with more advanced crystallization in the upper 

 than in the lower part, consequently, to where there is relatively 

 more An than Ab, the new zone deposited on the crystal will contain 

 relatively less Ab than the outer zone of the original crystal. 

 Whether this explanation is right or not, may be decided by a study 

 of plagioclases and other mix-crystals from the upper as well as 

 from the lower part of the same flow. 



Turning now to deep-seated rocks: the plagioclases, pyroxenes, 

 etc., in most cases do not show zonal structure at all. This phenom- 

 enon is here exceptional, and even where shown the zonal structure 

 and the chemical difference between the "first" and the "second" 

 generations are not so prominent as in dikes and flows. 



I thus found in looking over many thin sections of norites 

 that only in a couple of cases did the outer zone of the hypersthene 

 show a higher, though only a httle higher, percentage of iron 

 (Fe SiOj) than the central part. 



As to the plagioclases we wiU mention two cases : The extremely 

 coarse-grained porphyritic labradorite-norite, with phenocrysts of 

 labradorite up to 15-18 cm. long and 6-8 cm. broad, described in 

 my treatise in Quarterly Journal of the Geological Society (1909), 

 and mentioned in this Journal for 192 1, page 439, shows the follow- 

 ing proportion between An and Ab, the smaU quantity of Or not 

 being taken into consideration. 



In the rock as a whole (or in the magma) 58 An: 42 Ab 



In the phenocrysts 65 An: 35 Ab 



In the ground mass 55 An: 45 Ab 



In a magma with the plagioclase components in the proportion 

 58 An: 42 Ab, the first crystallizing plagioclase, according to 

 Bowen's scheme, must have the composition 84 An: 16 Ab (Fig. 56). 



During the entire first stage of crystallization, from An84 Abie 

 to Anes Abjj, equilibrium between the sohd and the liquid phase 

 prevailed, i.e., the already segregated plagioclase at length trans- 

 formed its composition. Then the equilibrium in the actual case 

 ceased, probaby due to the fact that the crystals first segregated 



'See my treatise in Tscherm. Min. Petrogr. Mitt., XXIV (1905), 517, and the 

 therein cited previous treatises. 



