THE ONYX MARBLES. 549 



various hues beiug sometimes constant throughout large masses, some- 

 times intermixed and blended, sometimes occurring in alternating par- 

 allel bands, and sometimes in distinct veins and spasmodic dashes. In 

 the majority of cases the coloring matter is supposedly iron in some of 

 its forms, aided in part by manganese; in any case the api^arent color 

 may be modified by conditions of crystallization and structure, a clear 

 translucent stone, by transmitted light, appearing much lighter and 

 of more delicate tints than one that is opaque,* Assuming that the 

 color constituents are only the two metals named, it yet remains to 

 account for their presence and explain the conditions under which they 

 give rise to such a variety of hues. It is easy and presumably correct 

 to assume that the coloring matters were deposited contemporaneously 

 with the calcite, but we must not in accepting this overlook the fact 

 that these substances are unequallj- soluble, and under proper con- 

 ditions would not be deposited together at all, but would undergo a 

 process of natural separation. As is stated by Bischof,t water which 

 contains carbonate of lime and i^rotoxide of iiou "may, when it passes 

 for long distances in contact with the air, finally deposit pure carbon- 

 ate of lime."' This, for the reason that the iron early becomes con- 

 verted into the condition of sesquioxide and is deposited almost at 

 the start. We can here account for their intimate association only on 

 the f-upposition that at the time of deposition the water was not flow- 

 ing, but lying in quiet pools where oxidation as well as loss of car- 

 bonic acid was retarded. The variation in color of the bands might 

 thus be in jiart accounted for on the suj)position that the waters, as 

 they issued, contained at times varying amounts of the oxides men- 

 tioned. (See pi. 3). I 



In fact, however, the varying hues are by no means due wholly to 

 the proportion of metallic oxides, but rather to the conditions under 

 which these oxides exist and to organic matter. In certain cases, as 

 in the slabs shown in jjl. 3, the bands are of alternating white, 

 green, and brown color, though all show practically the same percent- 

 ages of iron when calculated as i^rotoxide carbonate. As a matter of 

 fact, however, the iron exists in this state only in the white and faintly 



* A good illustration of the popular ideas regarding the cause of the color in these 

 rocks is given in the accompanying paragraph from the Engiueeriug and Mining 

 Journal (New York), vol. 49, p. 678. "Mexican onyx is a form of stalagmite, and 

 its colors are formed l»y oxides of metal in the earth over the caves through which 

 calcareous water passes. Gold is represented by purple, silver by yellow, iron by 

 red, copper by green, and arsenic and zinc by white." It is difficult to conceive of 

 any wording by which more errors could be comprised within the limits of a single 

 paragraph. 



1 Chemical and Physical Geology, p. 146. 



iThe higher or the lower the temperature of the water, the more or less rapid its 

 cooling, the greater or less abundance of bicarbomite of lime and jirotoxide of iron, 

 and the different proportions in which these compounds occur — all these circum- 

 stances may give rise to the most varied deposits of these substances as regards 

 their relative quality. — {Bischof, p. 147.) 



