Slike strani



That pore water exerts a distinct solvent action upon many natural silicates has long been known. As far back as 1848 the Rogers brothers published a series of observations upon this subject,' and showed that some species of minerals would give an alkaline reaction to test paper. They did not, however, give details concerning the individual minerals thus investigated. The more recent researches of Daubrée and of Cossa are well known.

By the use of phenolphthalein as an indicator the alkalinity of many silicates can be demonstrated with the utmost ease, and the experiments described below serve to bring out very clearly the relative decomposability of certain minerals and rocks by pure water. The method adopted was as follows: A series of glass-stoppered bottles was placed against a white background. In each bottle half a gram of finely pulverized mineral was put, and then 50 cubic centimeters of distilled water, containing a very little alcoholic phenolphthalein, was added. As the indicator was mixed, once for all, with the total amount of water taken for the entire series, the 22 samples examined were treated exactly alike. Two of the bottles were filled with the water and indicator in blank, in order that possible action upon the glass itself might be detected if it occurred. The two blanks, however, remained colorless during the two weeks through which the experiments lasted. The results obtained were as follows:

Muscovite.- A doubtful trace of coloration, which soon disappeared. Lepidolite.--Like muscovite.

Phlogopite.—The peculiar nonfluoriferons variety from Edwards, New York. Gave a very distinct, permanent pink coloration,

Orthoclase.-A trace of coloration which increased for a few days and then faded.

Oligoclase.The transparent variety from Bakersville, North Carolina. Distinct and permanent, but pale coloration.

Albite.From Amelia County, Virginia. Gave a good, permanent, alkaline reaction.

Leucite.A slight reaction at first, which faded in a few days.

Nephelite.The elæolite from Litchfield, Maine. Good coloration, but partly fading in time.

Am. Jour. Sci., 2nd series, Vol. V, p. 401.

Cancrinite.—Litchfield, Maine. Gives a deep rose coloration, which is permanent.

Sodalite.-From Canada. A deep, permanent rose color.

Spodumene.—The transparent, yellow variety from Brazil. A good reaction, but gradually fading.

Scapolite.-The wernerite from St. Lawrence County, New York. Gave a faint, evanescent trace of coloration.

Laumontite.-A doubtful trace of coloration.
Stilbite.-Faint, evanescent coloration.
Chabazite.--Like stilbite.
Heulandite.-Slight reaction, but distinct.

Thomsonite.—Variety lintonite. A fairly strong reaction, fading in time.

Analcite,-Good alkaline reaction.
Natrolite. From Bohemia. Strong coloration, permanent.
Pectolite.-From Bergen Hill. Gave a very deep rose color.
Apophyllite.-From Bergen Hill. A very deep rose color.

In nearly every case the reaction was obtained at once, showing a more rapid action of water upon the silicate than had been anticipated. In some instances fading is noted. This is doubtless due, in general, to the action of light, but in certain cases the colored solution separated into two layers, the color being wholly in the lower. Here the color was really held as a coating upon the fine solid particles, and as they subsided the appearance of stratification was produced. Toward the end of the experiments the mineral aegirite was added to the series. This also gave a strong alkaline reaction and a fairly deep rose color,

A neat method of demonstrating the reactions described above is the following: Place a little of the mineral to be tested in a watch glass upon a sheet of white paper. Add a drop of alcoholic phenolphthalein solution, and then a few drops of pure water; in most cases the reaction is given instantaneously. Orthoclase gave no coloration, leucite a trace, and scapolite a trace; albite, nephelite, and phlogopite furnished distinct reactions. Under the same circumstances thomsonite, aegirite, natrolite, cancrinite, sodalite, pectolite, and apophyllite gave immediately a deep, rich, rose color. The strongest alkaline reactions seemed to be given by pectolite and apophyllite.

In general the order of intensity of the color produced was what might have been expected. Among the micas, muscovite and lepidolite showed little or no solubility, while phlogopite was distinctly attacked. In nature the magnesian micas are far more easily alterable than muscovite, a fact which is reiterated by these experiments. Again, orthoclase was slightly dissolved, albite much more so, and oligoclase gave a reaction between the two; that is, more than the one, less than the other. In other words, the plagioclase feldspars alter more easily than orthoclase, as is apparent in the study of the rocks themselves.

In order to bring out the latter point more clearly, a series of rocks which had been analyzed in the laboratory of the United States Geological Survey was placed in a row of bottles and treated, just as the mineral species had been, with water and phenolphthalein. A granite and an amphibole-gabbro gave no alkaline reaction. A rhyolite, trachyte, leucite-basalt, feldspar-basalt, and diorite gave faint traces of color. Granitite, gneiss, phonolite, diabase, and camptonite yielded distinct alkaline colorations.

In all of these instances the production of color is doubtless due to the solution from the mineral or rock of alkaline silicates. The noteworthy point is the quickness with which the reaction can be obtained. With minerals like cancrinite, sodalite, natrolite, pectolite, and apophyllite, the reaction is striking enough to be used as a lecturetable experiment.



The results cited in the preceding paper are purely qualitative in character. The following experiments are analogous in kind, but are aimed at putting the subject on a quantitative basis. In both investigations the same samples of material were employed, so that the data are fairly comparable.

The work was carried out as follows: One-half gram of each of the finely ground minerals was weighed out, and placed in a 2-ounce bottle with 50 cubic centimeters of water. These bottles were set aside where the temperature remained about 700 F. for one month, and were skaken from time to time. At the end of the period all were filtered, and the solutions were titrated with a standard hydrochloric acid solution, being used for an indicator.

What has gone into solution by this treatment I can not say-sometimes soda, sometimes potash, possibly sometimes lime, but for the sake of comparison the results in the following table have been calculated in terms of Na,0; although the percentage of sodium is very small in some of the specimens. I have given also in another column the percentage of the combined alkalies as shown by analysis of specimens from the same localities.

[blocks in formation]
[blocks in formation]

Pectolite, Bergen Hill, N.J ..Ca(SiO3)2NaH

Natrolite, New Jersey

Al2(SiO4)3Na2H4 Lintonite, Lake Superior.... Al(Si04)(CaNa,)3.7H2O Phlogopite, Edwards, N. Y... Al(SiO4)3Mg,KH, Laumonite


Lepidolite, Maine

Elæolite, Litchfield, Me Al(SiO),Nas
Henlandite, Nova Scotia Alb(Si308)6(CaNa2)3.16H,0....


[blocks in formation]


Oligoclase, Bakersville, N. C.


Wernerite, St. Lawrence
County, N. Y.

Leucite, Vesuvius, Italy
Stilbite, Nova Scotia
Chabazite, Nova Scotia..


NaAl(SiO3)2. H2O..





Na4A1 SiO24C1....


Al SiO,SiO(CaNa,).6H2O

[blocks in formation]





It is worth noting in comparing the depth of color produced by phenolphthalein solution, as shown in Professor Clarke's paper, with the percentage of alkali in solution as shown in this work, that some of the minerals which give a deep coloration with the former, show in the above table a comparatively small amount of alkali in solution, and vice versa. Muscovite, for example, while giving only a faint coloration with phenolphthalein, contains in solution alkaline compounds equivalent to 0.49 per cent KO; pectolite, with 0.57 per cent Na2O, being the only one of the series showing a larger amount. The cause of these discrepancies remains to be ascertained.

« PrejšnjaNaprej »