Agate Formation in Sierra del Gallego

The following is an excerpt from a scientific report on the formation of agates in the Sierra del Gallego area in Mexico:

As was noted above, the agates of the Sierra del Gallego area occur as amygdule fillings in the Rancho El Agate andesite. The amygdules were formed by the release of gases from the lava. The shape of the anygdule and ultimately of the agate nodule depended primarily on the viscosity of the lava when the gases were released. In a highly viscous lava, a great deal of energy would be required for spherical expansion, and for this reason anygdules formed in a highly viscous lava flow are commonly irregular. If the flow was still mobile during cooling, the cavities are stretched or lenticular parallel to the direction of flow. Conversely, if the flow had a low viscosity, relatively little energy would have been required for spherical expansion and the result is larger, rounder amygdules. The origin of the banding or layering found in the agates has been the subject of much controversy. A popular hypothesis for the formation of banding in agates has been formation by rhythmic or periodic precipitation in a silica gel and the development of Liesegang bands by the diffusion of various iron and manganese oxides and hydroxides through the colloid (Liesegang, 1910).

This process is easily reproduced in the laboratory, and the results are surprisingly similar to the fortification banding found in natural agates. There are, however, many problems with this hypothesis. These have been reviewed by Roedder (1968). He showed that each of 21 textural criteria proposed for colloidal deposition was invalid or ambiguous. The explanation that I prefer is that "the banding of agates is essentially a rhythmic deposition from an enclosed siliceous solution on the walls of the cavity" (Frondel, 1962). Each layer toward the interior of the cavity is younger than the preceding one. Bauer (1904) suggested that this rhythmic deposition was the result of hot, intermittent springs saturating the rocks periodically, dissolving out the silica and other constituents of the rock and filling the amygdaloidal cavities. then the waters receded, the silica-rich water trapped in the amygdule would eventually evaporate at the elevated temperatures proposed, precipitating a thin film of silica on the walls of the amygdule. This process would occur repeatedly, each time leaving a new film on the cavity walls. Frondel (1962) suggests a similar process in which the silica was introduced hydrothermally into the cavity while the volcanic rock was still hot. Roedder (1968) observed remarkably uniform microbands in "colloidal" sphalerite. He interpreted these "varves" as being the result of annual changes in the ore fluid and to dilution with surface waters. I agree that the banding is caused by the rhythmic deposition of silica resulting from the periodic flooding and evaporation of silica-rich water. I disagree, however, with the proposed temperatures at which this process took place. Oxygen isotope studies were carried out by L. Land on the agate nodules and geodes in the Sierra del Gallego area to determine the temperatures at which silica was deposited.

The oxygen studies are described in Appendix A, and Table 4 shows the temperature in the Sierra del Gallego area, suggesting that the agate formed as a weathering product of the Rancho El Agate andesite. The breakdown of alkali feldspar during weathering of the Liebres rhyolite would raise the pH (Silica solubility increases greatly at pH of about 9) (Krauskopf, 1956) sufficiently to leach the silica from the surrounding andesite and put it into solution, probably as monomeric silicic acid (H SiO). With the annual rains the water table would rise, flooding the amygdules with the silica-saturated solutions, exactly as in the model proposed by Bauer (1904) using water from rising hot springs. With the end of the rainy season, the water table would again drop, allowing the silica-rich water in the amygdules to evaporate and leave a thin film of silica on the amygdule walls. With the next rainy season and rise in the water table, the same process would take place and a second layer of silica would be deposited, and so on until the cavity is filled. Fractures in the andesite flows also provided open spaces that are also commonly filled with agate. This model is substantiated somewhat by the common association of the agate with beautifully zoned siderite rhombs that are best attributed to periodic changes in the composition of solutions as a result of fluctuations in the water table. The concept of a seasonally fluctuating water table for the formation of agate has also been suggested by Sim (1974).This model also explains the narrow canal commonly extending from the central part of many agate nodules to the surface. The bands or layers of the agate bend sharply outward along the course of the canal, exactly as would be expected for a conduit admitting silica-rich solutions for the deposition of silica. The various colors found in the agate bands can be explained by the inclusion of several impurities, predominantly iron and manganese oxides and hydroxides. Many agates from localities such as Brazil and Uruguay are naturally drab brown or gray and are later colored artificially. "Moss agate" is a very common variety of agate found throughout the study area. This variety is named for its inclusions of branching green, red, or black filaments. For many years it was thought that these filaments were actually organic plant material that became incorporated during silica deposition. Farrington (1927) pointed out, how-ever, that such filaments could be accounted for by totally inorganic means, exactly like the "chemical gardens" grown with children's chemistry sets. These gardens are produced simply by introducing metallic salts into a colloidal medium such as a silica gel. Farrington's description of chemical garden filaments as hollow tubes with concentric layers around them is very similar to the tubular filaments observed in thin sections of moss agate from the Sierra del Gallego area, suggesting a similar genesis. Brown (1957) described pseudo-algal chemical gardens in the "thunder eggs" from Oregon, concluding that the filaments had grown inorganically. Recently, however, Fairchild and others (1973) used the scanning electron microscope to discover filament structure resembling that of algal filaments in the Caballos novaculite from the Devonian of Texas.

So it appears that both organic and inorganic process may produce filaments in "moss agate." If the filaments in the Sierra del Gallego agate are of organic origin, however, residual carbon might be expected in the interior of the tubes forming the filaments. Extensive microscopic search of selected thin sections revealed no observable carbon. Finally, representative agate slabs were decomposed in HF in the hope that any carbonaceous material would be left behind. No residue was noted, though, and I feel it is reasonable to maintain that the filaments in the agates from the study area grew by inorganic processes. It is possible, however, that the carbon was totally replaced by silica as is the case in most petrified wood found in the American Southwest.