416 
CROSS. 
nitely expressed the opinion that microfelsite is simply a 
“ submicroscopic ” intergrowth of orthoclase and quartz. As 
will be seen later, the writer has been led to the same con¬ 
clusion for the microspherulites thought by Rosenbusch to 
represent microfelsite in its purest state and most character¬ 
istic development. As for scales and fibers of the supposed 
substance occurring in the groundmass of porphyries and 
rhyolites, it will be extremely difficult to prove that they are 
not simply minute particles of known minerals, as feldspar, 
whose properties cannot be accurately determined by exist¬ 
ing means under the circumstances of their development. 
Turning now to the classification of spherulites recently 
presented by Michel-Levy* and compared by him with that 
of the German author, we find here, too, a great confusion 
resulting from the introduction of the vaguely characterized 
element, petrosilex. The present writer has searched in 
vain through the publications of Fouque, Michel-Levy, and 
other French authors for a definition of petrosilex which 
should cover and explain the various usages of the term. 
In the pamphlet last cited, Michel-Levy states that petrosi¬ 
lex is the same thing as microfelsite, to which Rosenbusch 
gave the definition above mentioned, but then refers to it as 
“a partially amorphous magma impregnated with silica 
already individualized in the state of opal or chalcedony/’ 
a phrase which admirably illustrates the indefinite way in 
which the term is used. One finds in descriptions of spheru- 
litic rocks such expressions as: “ la matiere petrosiliceuse,” 
“ le substance petrosiliceux,” “ la structure petrosiliceuse,” 
“ l’aspect petrosiliceux,” “ la cassure petrosiliceuse,” “ glob¬ 
ules petrosiliceux a extinction,” “globules petrosiliceux 
promorphiques,” etc., etc. It is seldom that a French writer 
describes a dense, obscure groundmass of an acid eruptive 
rock, or a spherulitic growth, without applying this favorite 
adjective. The various uses of the term logically imply that 
petrosilex may possess either the properties of an amor¬ 
phous, of a crystallitic, or of a fully crystalline substance. 
* Michel Levy (A.). Structures et classification des roches eruptives. 
8°. Paris, 1889 , pp. 20 - 24 . 
