Textures Follow - The Geology of Ore Deposits (Guilbert and Park)
Textures Follow - The Geology of Ore Deposits (Guilbert and Park)
Textures Follow - The Geology of Ore Deposits (Guilbert and Park)
Textures vary among ore deposits, depending upon the nature of the mineralizing fluids, the physical
and chemical characteristics of the host rocks, and the mode of emplacement. Textural interpretation
can assist greatly in determining the time relationships of successive mineral assemblages in a rock,
the overall environment of formation, and the manner of deposition. In epigenetic deposits, textures
can help define the sequence and nature of events as in the complex vein shown in Figure 3-11, and
they permit the distinction of mechanical and chemical controls as mentioned earlier in this chapter.
In syngenetic depositssuch as sedimentary, volcanic-hosted, and mafic igneous onestextures may
reveal accumulation rates and styles, cooling histories, and the nature of precipitation. Resorption of
early formed crystals may produce peculiar textures in magmatic segregation ores (Figures 4-2, 9-3,
and 9-4), and diagenetic changes may modify the primary
Page 133
Replacement
As defined by Lindgren (1S33), replacement, or metasomatism, is . . . the process of
practically simultaneous capillary solution and deposition by which a new mineral of partly or wholly
differing chemical composition may grow in the body of an old mineral or mineral aggregate. Replacement ordinarily implies little or no change in the volume of the replaced rock, although in some
rocks considerable shrinkage or expansion takes place. The process of metasomatism is of great
significance in the emplacement of epigenetic ore deposits; many ores are deposited almost entirely
in this manner, and nearly all ores show some evidence of replacement of preexisting materials. The
process is especially characteristic of deposits formed at high temperatures and high pressures where
open spaces are scarce and where communication with the surface is impeded.
The efficacy of replacement is often astounding. The intimate preservation of plant cell
or growth-ring morphology in petrified wood is well known, as is the fact that wooda fibrous
substance composed of carbon, hydrogen, oxygen, nitrogen, and minor elementscan be replaced
by silica, even though there is no apparent similarity between the two substances. The replacement
of one mineral or rock by another may be striking and clear; fossils, sedimentary textures, and folded
structures are commonly preserved in faithful molecule-by-molecule detail. A compilation of the minerals that can replace one another indicates that there is practically no limit to the direction of
metasomatism. As a bold generalization, it might be stated that given the proper conditions, any
mineral can replace any other mineral, though natural processes usually result in favored, commonly
recurring reactions. The important factor seems to be the chemical difference between the mineral
or rock being replaced and the replacement medium, be it liquid, gas, or a wave of diffusing ions.
Hence merely because quartz is stable at the Earths surface, we cannot conclude that quartz will resist
metasomatism. In fact, quartz and the silicates very commonly undergo replacement. In contrast, a
fluid that reacts with and replaces limestone may be inert to quartz, or vice versa. As a result, selective
replacement may be of the most detailed character. Bastin and his colleagues (1931) proposed the
following general rules for replacement: (1) sulfides, arsenides, tellundes, and sulfosalts can replace
any rock, gangue, or ore mineral, (2) gangue minerals replace rock and other gangue minerals, but do
not commonly replace sulfides, arsenides, tellurides, and sulfosalts, (3) oxides replace all rock and
gangue minerals but are rarely replaced by gangue minerals, and (4) oxides rarely replace sulfides,
arsenides, tellurides, and sulfosalts. These statements reflect the normal course of ore deposition,
Page 134
whereby fluids carrying ore components typically invade and replace rocks dominated by silicates.
Statements (2) and (4) are correct because the gan-gue and oxide minerals referred to are part of the
ore deposition event and are commingled with it rather than replacing it.
Although replacement has been generally recognized and described for many years, the
means by which the actual transfer of materials takes place has been the subject of debate. One
fundamental question is how the tremendous volumes of the old replaced materials are removed
by the same solution that deposits the new minerals. Presumably the dissolved materials are
transported back from a replacement front through the aqueous ore solutions by diffusion, as in wall
rock alteration. In other words, different ions may be diffusing down chemical gradients in quite
different directions. The fact that replacement generally appears to take place volume for volume
(without increasing porosity) raises another major problem: How can we write chemical equations
representing electrically and molec-ularly balanced reactions with equal volumes of solid materials on
each side? Ridge (1949, 1961) attempted to illustrate replacement with equations that balanced
molecularly, volumetrically, and electrically. He found that he could write equations acknowledging
near-surface chemistry and geology but could develop only rough approximations for deep-seated
reactions.
In fact, the budget approach to metasomatism requires that atomic structure, space,
and valence all be conserved, but there is no real evidence that all of those requirements must be met.
Metasomatism can be molecule for molecule, but it may also involve one structure growing while
another decays, with components being added and subtracted diffusionally from outside as required
by the relative rates of destruction and construction. In other words, there may be no molecular-level
structural inheritance at all. Barnes (1979) describes the replacement of limestone by sphalerite in a
brine medium as involving etching and dissolution of calcite, especially along grain boundaries, merely
with the concomitant precipitation of sphalerite and iron sulfides in the space thus created. The
reactions were probably
Page 135
with the H+ generated in Equations (4-6) and (4-7) aiding in the dissolution expressed by Equation
(4-5). Ridge (1961) showed the same reactions summarized as
(4-8)
(4-9)
Such careful treatment shows that volume for volume space can be conserved.
Limestone replacement at low temperature and pressure was studied by Garrels and
Dreyer (1952). They were able to produce replacement textures similar to natural ones under
controlled laboratory conditions. Many variables were examined, and it was concluded" that the
major control of replacement was the pH of the mineralizing fluid, which in turn controlled the
solubility of the carbonate host rock. Garrels and Dreyer suggested that dissolving limestone changes
ore-bearing solutions in the direction of ore-mineral precipitation. Accordingly, they consider the
solubility of the host to be the key factor in metasomatism. Ames (1961) proposed that the principal
factor is the solubility of the replacement product relative to the solubility of the host rock in the same
solution, rather than merely the solubility of the host. In any case, the importance of simultaneous
dissolving action and precipitation from a single solution is well established.
Garrels and Dreyer also found that numerous small, closely spaced openings and a
slightly higher secondary permeability create ideal conditions for replacement. They concluded that
the mineralizing solutions are carried along these zones of secondary permeability, but that the
solutions move from the channels to the replacement front mainly by diffusion rather than by
mechanical flow. It seems clear that replacement is far more important as a mechanism at higher
temperatures, and that high pressure inhibits replacement by closure of permeability.
Interpretations of mineral textures tc give unique and positive information are difficult.
In spite of the great amount of work done on mineral textures and structures, the causes of many
textural relationships are poorly understood. Although many textures are clear and unequivocal
enough for positive statements, others may have several interpretations. For example, textures such
as the one shown as Figure 3-5 have been attributed to the process of exsolution (Schwartz, 1931),
replacement (Loughlin and Kosch-mann, 1942), eutectic crystallization, or, if the two minerals are
chalcopyrite and bornite, the removal of part of the iron from chalcopyrite by hot water or steam
(Park, 1931). Brett (1964) found that many exsolution textures were deceptive because they look as if
they were formed by replacement.
Page 136
(a) Carbonate
(.b)
Figure 4-6. (ft) Pseudomorphism. Bementite (crosshatched) has replaced most of a rhomb of
preexisting carbonate. Bementite, a manganese silicate, is monoclinic and could not form a
rhombohedron by itself. Olympic Peninsula, Washington. 60 x. (6) Fluorite ha replaced a fossil
(circled) in Doctores Formation limestone at the Las Cuevas fluorite deposit, Mexico. The original form
of the fossil is obliterated where replaced by fluorite. Replacement is irrefutable since no CaF2 shells
are known. (From Ruiz et ah, 1980.)
and textures in sedimentary, igneous, or metamorphic rocks, as well as organic remains, may be
pseudomorphic. For example, preore fold structures or oolites now pseudomorphed by sulfides are
conclusive evidence of replacement.
2. Widening of a fracture filling to an irregular mass where a fracture crosses a chemically reactive
mineral grain or rock (Figure 4-7). If a
Page 138
veinlet widens across only one variety of mineral, it suggests that this mineral was receptive to
replacement and that replacement occurred. Mineralization spreading laterally into a single limestone
unit in a sequence of varied sedimentary layers is a large-scale example.
3.
Islands of unreplaced host mineral or wall rock (Figure 4-9). Isolated, seemingly suspended
relict bits of one material in another, especially if
Figure 4-8. Vermicular irregular intergrowths at wide places along
fractures or at grain boundaries; not related to crystallographic
directions. Skutterudite (sk) and niccolite (ni) have replaced native
silver (Ag). The vermicular diffuse intergrowths of skutterudite
extending irregularly into the silver are evidence of replacement. The
niccolite is probably an intermediate reaction product. The entire field
once was presumably all silver. Cobalt, Ontario, Canada. 30 x. (After an
unpublished photo by D. E. Eberlein.)
Page 139
Figure 4-9. (a) Oriented islands of one mineral in another. In this texture, pyrite relict fragments form
an atoll structure, a common texture in Cu-Fe-S vein ores. An early pyrite cube, the outline of which is
preserved, has been almost completely replaced by bomite. Chalcopyrite is an intermediate reaction
product. Bisbee, Arizona. 40 x. (6) Photo-micrographic example of oriented relict islands of a replaced
mineral. Clusters of pyrite relicts (white) in chalcopyrite (light gray, within pyrite) and bomite (medium
gray, external to pyrite) clearly reflect earlier grain forms. Dark gray is quartz; black is holes in polished
surface. Butte, Montana. (Photo by Geological Research Laboratory, Anaconda Company.)
Page 140
an earlier form or common orientation is perceptible, require a replacement relationship, one that did
not carry to completion.
5. Concave surfaces into the host, the cusp and caries texture (Figure 4-10). The diffusion of ions at
the replacement front goes on at different rates, so that some parts of the front form concave
reentrants as if the replacing mineral had bitten into the host. The cusp and caries designation is a
dental analogycaries are cavities in teeth, clearly
Page 141
later than the enamel; the cusps are relict protuberances between cavities. The texture is a familiar
one to microscopists.
6. Nonmatching walls or borders of a fracture (Figure 4-11). If replacement works outward from a
central fissure, the opposite fronts of replacement will almost never match in detail and may differ
radically.
7. Rims of one mineral penetrating another along its crystallographic directions (Figure 4-12).
Replacement may work outward from any small fissures or from grain margins and boundaries, but
has advanced preferentially along cleavages. For example, galena may be replaced by covellite or
cerussite that penetrate it preferentially along directions that are obviously parallel to its cleavages
(see also Figure 3-3).
8. Oriented unsupported fragments (Figure 4-13). If a piece of one mineral is completely surrounded
by another mineral and still maintains its
Page 142
orientation with respect to more of the same material on the outside, the texture is practically
diagnostic of replacement at any scale. The fragment may be any size, and the orientation may be
established by crystallographic directions, cleavage, bedding, or foliation.
9. Selective association (Figure 4-14). Since replacement is a chemical process, specific selective
associations of pairs or combinations of minerals can be expected. If the chemistry of a system
changes, it will more likely reflect a changed ratio of related constituents than a complete switch to
different ones. Chalcopyrite is more likely to replace bornite by virtue of a changed Cu/Fe ratio or a
different /(S2) than it is to replace quartz. Thus consanguinity of assemblages is part of the diagnosis
of replacement.
10. A. younger mineral that transects older structures (Figure 4-15). The presence of a crystala
metablast, for examplethat interrupts bedding or foliation requires that the structure is older. If the
crystal had
Page 143
Figure 4-15. Metablasts of pyrite that cut foliation in schist from the Mother Lode district, California.
The pyrite cubes grew in the rock but do not bend or displace the foliation. They must, then, have
replaced the material that occupied their space. About 0.9 x.
grown by any process other than replacement, it would have displaced or deformed the structure.
11.
Disparity in size of one mineral in another. Large crystals in a finegrained groundmass, or vice
versa, suggest that the two grew independently or by different processes that might include
replacement.
13.
A mineral deposited along what was clearly an advancing alteration- reaction front. If
deposition took place by open-space filling, ore minerals should stop abruptly at the wall of a fissure.
Conversely, replacement is indicated by the gradual enlargement and merging of meta- somes along
a replacement front. Such a zone should also be evident
Page 144
from the progressive increase in size of metacrysts and from the gradation of replacement from the
wall rock into the vein. Figure 3-4 is an excellent example.
14.
Doubly terminated crystals (Figure 4-18). If a crystal grows within an open cavity, it is
normally attached to a wall and develops crystal faces only at the free end. This restriction does not
affect crystals growing by replacement, so that doubly terminated crystals may indicate a
replacement origin. This criterion is limited because doubly terminated crystals may also develop
within magmas and, exceptionally, by unusual attachment in open spaces.
16.
Page 145
Figure 4-18. Doubly terminated quartz crystal from the Stibnite Vein, Wolf Creek, Fairbanks district,
Alaska, containing stibnite crystals. The quartz is said to be a replacement of gangue minerals in the
vein, and probably overgrew and enclosed the stibnite granules. 18x. (Collected by P. 0. Sandvik; photo
by W. J. Crook.)
17.
Residual resistant minerals. Some minerals are refractory to most mineralizing solutions and
may be preserved even after the surrounding minerals have been replaced. For example, zircon or
corundum found in the same proportions in the sulfides of an orebody as in a nearby schist would
support the argument that the ore had replaced some of the schist. The resistant minerals are special
types of islands or unreplaced fragments of host rock (criterion 4).
18.
No offset of an intersected linear fracture (Figure 4-19). When a vein is opened laterally, any
intersecting feature is also translated. But if a linear or planar feature projects straight across a vein
filling, there can have been no dilation, and filling must be the result of replacement.
19.
No offset along the intersection of fractures (Figure 4-20). Movement along a fissure offsets
any planar or linear feature that it intersects obliquely, but does not commonly offset an intersecting
fracture formed
Page 146
at the same time. Both fractures may be enlarged by replacement of their walls and tend to cross one
another without any change in course, but only a coincidence of displacement would permit a rematch
of intersecting veins that had been spread apart and open-space-filled.
Exsolution
A near neighbor to replacement is the process of exsolution. In both phenomena,
migration of one or more components, commonly involving diffusion, results in the supplanting of a
preexisting phase by a younger one. Exsolution textures and their evaluation are normally only seen
by the mineragrapher (the ore microscopist). It is true that exsolution is not strictly speaking a
depositional feature, and perhaps should not be included in this chapter. But there are enough
situations in which exsolution textures and distributions can be mistaken for those of replacement
that their description is justified.
Reexamine Figure 3-5. Were the process of exsolution to be carried further, the
miniscule blebs of chalcopyrite would migrate to cleavages and crystallographic planes and collect
there, leaving impoverished zones behind. That process can be seen to have begun especially on the
right side of Figure 3-5. Taken even further, the chalcopyritenow essentially all on planesmight
resemble a phenomenon of chalcopyrite from outside invading and replacing the sphalerite. Taken to
completion, the lowest free- energy state is for the exsolved phase to form separate grains occupying
texturally equivalent positions in a mosaic with the original host phase. For many years, veinlets and
grains of pentlandite t(Fe,Ni)9S8] in pyrrhotite (Fe^a-S) in the Sudbury Complex (Chapter 9) were
perplexing. Some considered them to be of replacement origin, with pentlandite invading from
outside the pyrrhotite. Others thought them to be the product of well- developed exsolution. A
careful textural study of scores of polished pieces of ore sulfides has verified the latter explanation. It
matters because it
Page 147
means that pentlandite, a principal contributor of nickel to the ores, can be expected
virtually wherever pyrrhotite is found, and pyrrhotite occurs extensively along the floor of the
complex. Were the pentlandite not integral to the pyrrhotite, its distribution would almost certainly
be more erratically subject to unknown external controls. In general, replacement veinlets widen at
intersections, and exsolution veinlets become narrower. In the former case, replacement attacks
corners formed by two veinlets from two sides, replacement is more complete, and the zone widens.
In the latter, the host contributes the exsolved phase into the veinlet. Close to a comer the exsolved
ions are migrating both ways, and the intersection becomes starved. Indeed, some microscopic
exsolution patterns involve exsolved crosses the arms of which narrow to a pinpoint at the
intersection.
Excellent lists of criteria for distinguishing exsolution and replacement are available in
Edwards (1965), Ramdohr (1980), and Craig and Vaughan (1981), but need not be reviewed here.
Open-Space Filling
Open-space filling is common in shallow zones where brittle rocks yield by breaking rather than by
plastic flow. The openings in these zones tend to remain open because of low pressure transmitted
through surrounding rock. At shallow depths, ore-bearing fluids have relatively free circulation, and
their open connection with the surface permits deposition to be brought about by abrupt pressure
and temperature changesincluding retrograde boilingas opposed to prolonged contact with
surrounding rocks which undergo slow chemical changes in deeper environments. Although ores deposited in vugs and open cavities are generally readily distinguishable from replacement ores, criteria
associated with open-space deposition must nevertheless be used with caution because they are
sometimes inconclusive (Kutina and Sedlackova, 1961). As stated earlier, it should also be stressed
that many ore types show evidences of both replacement and open-space filling and thus an overlap
of the two processes. A list of common criteria by which open-space filling may be recognized is given
below.
Open-Space Filling Textures
1.
Many vugs and cavities (Figure 4-21). If veins, breccias, and other partly-filled openings
contain many vugs and cavities that can be interpreted as spaces left by incomplete filling of larger
open spaces, then open-space filling is indicated. The inward growth of ore and gangue minerals in an
open fissure stops when opposite walls meet. Since growth is not uniform along a fissure, the
progressively impeded circulation of ore-bearing solutions normally leaves unfilled pockets. These
crystal- lined centerline openings are called vugs.
2.
Page 148
Figure 4-21. Quartz-crystal-lined vugs left by incomplete vein filling. Locality unknown, but typical of
scores of epithermal veins around the world. 0.9 x. See also Figures 4-22 and 4-23.
in the center (Figure 4-22). The first crystals that form along the sides of an open vein are usually finegrained, probably because of heat loss to the wall rocks and consequent rapid crystallization.
Conversely, crystals that form in the centers of such veins are inclined to be coarse and probably form
from more dilute, cooler solutions late in the vein history. Whatever the reasons, the phenomenon is
common enough to be distinctive of open-space filling.
3.
Comb structure (Figure 4-23). Along the junction of crystals that have grown from opposite
walls of a fissure, there is generally an interdigi- tated vuggy zone due to the merging of euhedral
prismatic crystals. Because this jagged zone of juncture resembles the outline of a roosters comb, it
is known as comb structure.
5.
Symmetrical banding (Figure 4-24). Crystals deposited along a cavity or fissure ordinarily grow
symmetrically toward the center, in which case the orientations, morphologies, and compositions of
crystals on opposite sides of a vein are mirror images. As the ore fluid or environment changes,
minerals deposited will differ in composition, forming crustification in a bilaterally symmetrical
pattern inward from vein walls. Such symmetrical banding may involve mineralogical and color
changes, and can produce a banded vein diagnostic of open-space filling.
E. Matching walls. If an open fissure has been filled without replace-
Page 149
Figure 4-22. Fine-grained quartz grading into coarser crystals in a vug. Locality unknown. See also
Figure 4-23.
ment, the outlines of opposite walls should match, that is, if the vein material were removed, the
unreacted wall rocks on opposite sides should fit together like mated pieces of a jigsaw puzzle. The fit
across many veins is readily apparent.
7. Cockade structure (Figure 4-25). Mineralization within the open spaces of a breccia, or any other
fragmental rock, commonly produces a special pattern of symmetrical banding or erustincation known
as cockade structure. Each opening is a receptacle for sequential deposition. The overall rock pattern
is one of host-rock fragments coated with layers of inward-radiating crystals, with triangular patterns
predominating.
8. Offset oblique structures (Figure 4-26). If a planar or linear feature is cut obliquely by an open fissure,
it is offset at right angles to the vein walls because it was spread apart as the fissure opened.
Replacement along a fracture would cause no offset along such a preexisting structure; it would
project straight across. (See also Figure 4-19.)
9. Colloform structures. Colloform banding, composed of fine onionskinlike successively deposited
layers (Figures 4-27 and 4-28 and next section), can only form in open space.
Page 150
Figure 4-23. Crustification, comb structure, and centerline vugs, (a) Veinlet of aragonite through
limestone and shale. Locality unknown. (b) Vein from latite porphyry (top and bottom) through
amethyst (medium gray) to centerline inward-facing chalcedonic quartz. Commonwealth mine,
Pearce, Arizona. The bar scale is in centimeters. (Photo of b by W. K. Bilodeau.)
Page 151
Figure 4-25. Cockade structure of quartz infilling the cavities in a breccia. Locality unknown, but a
typical epithermal structure.
Page 152
Figure 4-26. (a) Offset of an oblique structure. A and B were adjacent before the vein opened; the
two segments no longer lie in a single plane, and the intervening material is open-space filling.
Lateral wall movement can of course complicate relationships. Compare with Figure 4-19. (6)
Specimen from Silver Bell, Arizona, in which the white quartz vein is 1 cm wide. (Photo by W. K.
Bilodeau.)
Colloidal-Colloform Textures
Amorphous minerals, such as opal, neotocite, copper pitch, wood tin, and some
garnierite, used to be thought to have been deposited from colloidal suspensions. Moreover, it was
believed that many cryptocrystal- line mineralschalcedony, some manganese oxides, pyrite,
marcasite, and pitchblende, and the oxidation products of copper, lead, and zinc, such as malachite,
azurite, chrysocolla, anglesite, cerussite, and smithsonitewere carried and deposited as colloids that
aggregated and crystallized shortly after deposition. Some geologists think that gold, especially in
shallow low- temperature deposits, travels short distances in the colloidal state. Most geochemists
now consider the existence of colloids in hydrothermal systems highly unlikely.
A colloidal system consists of two phases, of which one, the dispersed phase, is diffused
in the other, the dispersion medium.. Colloidal particles range in size between ions in true solution
and particles in coarse suspension, the general limits being defined at 10-7 and 10-3 cm. The colloidal
material may be solid, liquid, or gas, and may be dispersed in a medium of any of these same states.
In the study of ore transport, however, we are concerned essentially with solids dispersed in liquid
media. A colloidal system consisting of solid particles dispersed in a liquid is called a sol. A given kind
of colloidal particle may adsorb cations and behave as a positively charged body, or it may adsorb
anions and become negatively charged. Since the particles of a sol all have the same charge, they repel
each other and prevent coagulation. If an electrolyte is added to the sol, the colloidal particles become
neutralized and flocculate, or deposit. The result can be a
Page 153
banded or finely layered, typically botryoidal or reniform aggregate of the type that characterizes
African malachite, for example, and is common in some sphalerite from low-temperature deposits
(Figure 4-27). However, we saw in Chapters 2 and 3 that most hydrothermal fluids are indeed brines,
or at least saline enough to be the electrolyte that is anathema to colloids and colloidal transport. We
now perceive that at most temperatures and in most hydrothermal fluids, at least, colloidal transport
and deposition cannot be important. At low temperature and in nonsaline groundwater, colloids can
and do persist.
As a result of his study of colloform sphalerite specimens from many of the worlds
major geological districts, Roedder (1968) decided that most if not all of these ores grew directly as
minute druses of continuously eu-hedral crystals that projected into an ore fluid that was a true
solution, not a colloidal dispersion. He found that all of the textural features believed to be diagnostic
of colloidal deposition were ambiguous, inapplicable, and therefore invalid as criteria for the samples
he studied, and perhaps of most other colloform mineral samples as well. However, Roedder did not
explain the fact that such colloform minerals as the manganese oxides, garnierite, and opal at times
do not reveal any structural pattern even upon X-ray study; they are truly noncrystalline.
In a thoughtful discussion of Roedders paper, Haranczyk (1969) concluded from his
extensive studies in the Silesian-Cracovian lead and zinc deposits of Poland that colloform textures
might form from both true solutions and sols. He proposed the term hemicolloids for solutions that
are intermediate between true and colloidal ones. The cryptocrystalline, white, claylike form of
sphalerite known as brunciie (Figures 20-12 and 20-13) was almost certainly deposited as a colloid.
Haranczyk later (1971) clearly showed that some of the second-generation minerals in the SilesianCracovian ore deposits of Poland could be of colloidal origin. He states that most of the secondgeneration sulfides show hemicolloidal textures. This conclusion clashes
Page 154
with the modern concept that these deposits were formed from warm electrolyte brines; clearly, we
need more experimentation and morphological data.
The terms colloidal texture and colloform texture should be used carefully. The former
should be used only where deposition from a true colloidal system is demonstrable; colloform
suggests a texture that resembles, and could be, colloidal in origin, but cannot be determined to be
so. A truly colloidal origin can be postulated if the following are observed: (1) infinitesimal grain sizes
in banded textures as in some malachite (Figure 4-28) and most wad, (2) shrinkage or dehydration
cracks, as in opal and some limonites, (3) diffusion bands, or Liesegang rings (Figure 4-29), (4) chaotic,
amorphous noncrystalline structure of component compounds, (5) the presence of unpredicted,
possibly originally adsorbed ions, such as barium in psilomelane, and (6) continuous spherulitic
textures demonstrably controlled by surface tension. Interrupted bands normally deny the role of
Figure 4-28. Colloform banding in malachite from Kolwezi, Zaire. Some of the bands can be traced
through the whole specimen, but some lobes interfere with others, perhaps as stalagmite-stalactite
masses were filled in. Deposition can have been either truly colloidal or as banded microcrystalline
accumulations. The bar scale is in centimeters. (Photo by W. K. Bilodeau.)
Page 155
Figure 4-29. Probable diffusion banding in an agate, quite possibly developed while the silica was a
colloidal silica gel. Note also the cockade pattern of the later quartz filling in the central vug. Locality
unknown. 1.7 x. (Photo by W. J. Crook.)
colloids (Figures 20-9 and 20-10). Clearly, all colloform textures are open-space fillings; they may or
may not have involved colloids.
Even though the part played by colloids in ore deposition may be trivial, it seems likely
that under near-surface conditions and in dilute watery solutions at comparatively low temperatures
the role may be appreciable. Many deposits have been described for which a colloidal origin is
probable or in which colloids have played a supporting role.
An unusual deposit of colloform magnetite was found in an igneous metamorphic
environment at the northern end of Vancouver Island, Canada (Stevenson and Jeffery, 1964). The
colloform magnetite possessed the textural features usually ascribed to colloidal deposition, and the
deposits ap-
Page 156
peaned to have been formed by a type of gel metasomatism of the limestone. The iron, carried in acid
solution, replaced limestone and appears to have been precipitated colloidally during an intermediate
state of aggregation between ionic solution and the final precipitate.
Hosking (1964) pointed out the presence of opaline gels that are apparently forming in
the deposits of Cornwall, England (Chapter 6). He described the tin lodes of St. Agnes where almost
perfect concentrically zoned spherules of wood tinultrafine-grained colloform cassiterite are
suspended in quartz. These spherules are covered at places with acicular crystals of cassiterite whose
long axes are normal to the spherule surfaces. He considers these spherules to have developed as the
result of centripetal migration of tin-bearing fluids through a silica gel.
Page 157