54 Washington and Wright — Feldspar from Linosa. 



These variations are of considerable magnitude and the 

 values seem to differ from crystal to crystal rather than within 

 a single crystal. This phenomenon of variation in optic axial 

 angle might possibly be explained as a result of unequal cooling 

 of the different crystals, as is the case with orthoclase and sani- 

 dine ; but other evidence, particularly that of the extinction 

 angles, slrows that the composition probably varies slightly 

 from crystal to crystal. 



The extinction angles were measured both on cleavage flakes 

 after (001) and on (010). In each case the angles were 

 measured as accurately as possible, with the aid of the circularly 

 polarizing bi-quartz wedge,* and the probable error is less 

 than 15 7 ." 



On (001) the values for a^a ranged from — 0*6° to — 4*5°, 

 and on basal cleavage flakes from one crystal the angles averaged 

 — 2'0°, and from a second — 4*5°. Other pieces showing poly- 

 synthetic lamellae were observed occasionally with very high 

 extinction angles; but the examination in convergent light, 

 and also the determination of the ellipsoidal #xis, whether 

 a or c, proved that either the cleavage fragment was not 

 parallel to the base or that pericline lamellae were being 

 examined.*)' 



Extinction angles were also measured on the brachypinacoid 

 (010), both on the rhomb-shaped cleavage flakes and also on a 

 section ground parallel to the brachypinacoid. Different values 

 were obtained for different flakes, although in the larger 

 ground plates no marked indications of wavy extinction or 

 zonal structure were observed. The values ranged from 

 a^a= — 2'5° to —11°. Since, however, the cleavage parallel to 

 (010) is not perfect, it is possible that part of this variation was 

 due to the fact that the surfaces were not precisely parallel to 

 (010) at the place of measurement, but were inclined because 

 of minute irregularities of the cleavage. Flakes parallel to 

 (010) often show in white light the peculiar interference 

 phenomena characteristic of minerals with slight dispersion of 

 the bisectrices. 



In convergent polarized light the optic normal emerges near 

 the center of the field on plates parallel to (001) ; while the 

 bisectrix c is nearly normal to the brachypinaoid, and appears 

 near the center of the field on flakes parallel to (010). 



Taken collectively, these results indicate that the present 

 material is a plagioclase feldspar, of somewhat variable com- 

 position and with slightly modified characters. Taken alone, 



*Cf. F. E. Wright, this Journal, vol. xxvi, p 391, 1908. 



f A series of measurements on 12 different cleavage flakes from one of the 

 crystals was also made by Mr. E. S. Larsen, with the result a * a=2° 2' ±6', 

 a value well in accord with Wright's observations. 



