226 The Ajnerican Geologist. April, 1900 
fied as basalts. It will be noted that this division is slightly 
inconsistent, for while they divided rocks containing biotite. 
amphibole or pyroxene on the basis of the feldspar, without 
reference to the ferromagnesian constituent, they classified 
rocks containing olivine according to this ferromagnesian 
mineral without reference to the feldspar. Moreover, curi- 
ously enough, they did not extend the subdivision of the 
Tertiary and post-Tertiary trachytoid soda-lime feldspar rocks, 
containing biotite, amphibole, or pyroxene, to the correspond- 
ing granular rocks, but defined diorytes as soda-lime feldspar 
rocks with amphibole, while soda-lime feldspar rocks with 
mica were called kersantytes, and those with pyroxene or 
olivine diabases, gabbros, or norytes.* In a later paper j 
however, Michel-Levy calls attention to the lack of signi- 
ficance of the ferromagnesian minerals and emphasizes the 
idea that the determination of the plagioclases is indispensable 
for rock classification. 
BroggerJ, in his classification, also abandons for the most 
part ferromagnesian minerals as a basis for division between 
diorytes, and diabases and gabbros, and between the correspond- 
ing classes of the eciuivalent dike and volcanic rocks, and 
makes his chief division according to the percentage of silica, 
the dioryte family containing from 48 to 62 per cent, and the 
gabbro family from 44 to 53 per cent. The chemical boundary 
between the two families is not sharp, because Brogger also 
takes into consideration the structure and composition of the 
rocks§. 
Turner very logically proposed to extend Michel-Levy's 
subdivision of volcanic rocks, according to their feldspars, to 
the granular rocks. "If," Mr. Turner said, "petrographers 
can be induced to call all oligoclase-andesine granular rocks 
diorytes. and all labradorite-anorthite granular rocks diabases 
and gabbros, the nomenclature of the granular plagioclase 
rocks will be greatly simplified." 
*Structures et Classification des Roches Eruptives, Paris, iSgg, p. 42. 
tPrincipes a suivre pour une classification universelle des Roches, 
p. 156, 155. 
JEruptionsfolge der Triadischen Eruptivgesteine bei Predazzo in 
Siidtyrol, Kristiania, 1895, P- 60. 
§bie Gesteine der Grorudit-Tinguait Serie, Kristiania, 1895, P- 93- 
ijSeventeenth Ann. Rept. U. S. Geol. Survey. Part I, 1896, p. 718. 
