310 The American Geologist. May, 1901. 
paratus of especially desifable pattern have been figured in the text. 
Chemists will find the parts devoted to the estimation of small quanti- 
ties of the rarer rock-forming elements, titanium, zirconium, chromium, 
vanadium, molybdenum, barium, and strontium of particular interest ; 
also those devoted to ferrous iron. The author finds in the strong re- 
ducing action which sulphides even in small amounts have on ferric 
salts, an explanation of the fact that terrous iron estimations by the 
sealed tube method of Mitscherlick give higher results, as a rule, than 
those ODtained by other methods, this diflference becoming more marked 
as the percentage of iron increases. All rocks contain sulphur, at 
least in traces, and the more highly ferruginous ones not only com- 
monly carry higher percentages of ferric iron, but considerable quanti- 
ties of iron sulphides, conditions exactly calculated to produce the re- 
sults obtained. During the heating of the rock powder with sulphuric 
acid in the sealed tube the sulphur of the sulphides is oxidized to sul- 
phuric acid at the expense of the ferric iron. In this way as little as 
o.io^ of sulphur increases the ferrous iron by an error of 1.35,'^. This 
seems to furnish a correct and interesting explanation of a long un- 
solved problem. The wide experience and established reputation of the 
author make this carefully prepared bulletin a most valuable addition to 
the literature of analytical .chemistry and it will find wide use by geol- 
ogists and petrographers interested in rock analysis. w. o. c. 
Analyses of Rocks, Laboratory of the U. S. Geological Survey. Tab- 
ulated by F. W. Clark. (Bull. No. 168, U. S. Geol. Sur.) 
In this bulletin are tabulated the results of 1,404 analyses of rocks, 
minerals, clays, etc.. which have been made between the years 1880 
and 1889, by the chemists connected with the U. S. Geological Survey. 
The analyses are arranged in groups according to the geographical lo- 
cation of their respective rocks, and with each one are such notes and 
references as are necessary for the proper identification and study of 
the rocks. . A table, showing the average composition of the earth's 
crust as calculated by F. W. Clark from some eight hundred carefully 
selected rock analyses, is an interesting and valuable part of the intro- 
duction. Every geologist will appreciate the high value of this bul- 
letin, w. o. c. 
An Experimental Investigation into the Flow of Marble. By Frank 
Dawson Adams and John Thomas Nicholson. (Phil. Trans. 
Roy. Soc. London, Vol. 195, pp. 363-401, pis. 22-25. 1901.) 
The artificial production of structures found in nature is a depart- 
ment of investigation capable of yielding data on many problems of a 
theoretical nature. The folding and flowage of rocks in the deeper 
part of the earth's crust are connnonly supposed to be due to the inter- 
action of three factors ; — great pressure, high temperature, and perco- 
lating water. In the experiments here described carrara marble is 
placed in specially constructed cylinders and subjected to differential 
pressure under varying conditions. Dry marble at ordinary tempera- 
tures developed a cataclastic structure, exhibiting an anastomosing net- 
work of calcite granules with marked shearing lines. Macroscopically 
