40 WJ. P. Lddings—Nature and origin of Lithophyse 
presumably heated solutions, contemporaneous or nearly so 
with the final consolidation of the rock. The lithophysal cay- 
ities seem plainly caused by the expansive tendency of con- 
fined gases or vapors, while the shrinkage cracks in the walls 
and white masses of the Nathrop rock suggest the former 
presence of moisture.” 
It is seen from the foregoing that two distinct views of the 
origin of the cavities within the lithophysz have been taken; 
one, that the hollows were of primary origin, formed while the 
lava was still plastic, and were due to enclosed gases or vapors. 
Among those who held this opinion some considered the lith- 
ophyse as wholly distinct. from spherulites, while others 
thought them simply hollow varieties. The second view was 
that the hollows had been produced in solid spherulites by 
chemical decomposition and alteration and were subsequent to 
the solidification of the lavas in which they are found. With 
the latter view, which may apply to some hollow spherulites in 
particular cases of decomposed rocks, the lithophyse in the 
obsidian from the Yellowstone National Park have nothing to 
do. The extreme freshness of the whole rock and the absence 
of secondary alteration prevents the confusion which arises 
when these hollow forms are associated with decomposition pro- 
ducts or subsequent metamorphism, as is the case with many 
ancient lavas, or even with recent ones which have been attacked 
by solfataric or hot spring agencies. 
What, then, froma study of: this exceptionally fresh and 
beautiful material, seems to be the most probable origin of 
lithophyse, and how nearly the writer’s views accord with 
those of von Richthofen will appear from the following con- 
siderations. 
The association of fayalite, an iron-olivine, with abundant 
quartz and tridymite, and acid feldspar in a highly siliceous, 
igneous rock, containing less than two per cent of iron oxide, 
is quite contrary to ordinary experience and is not in accord 
with the generally accepted laws which appear to govern the 
crystallization of igneous magmas. Moreover, two of the ac- 
companying minerals, prismatic quartz and acid feldspar, have 
not been reproduced artificially by simple igneous fusion, as 
has been demonstrated by the repeated experiments of MM. 
Fouqué and Michel-Lévy,* who, though successful in obtaining 
most of the minerals found in igneous rocks, have failed to repro- 
duce by dry fusion quartz, orthoclase and albite in forms sim- 
ilar to those in which they are found in acid eruptive rocks. 
On the other hand quartz, tridymite, orthoclase and albite 
have all been obtained artificially by aqueo-igneous processes, 
on heating their chemical elements in the presence of water in 
* “ Synthése des Mineraux et des Roches,” Paris, 1882. 
