Terra Rossa Genesis, Implications for Karst, and Eolian...

[The Journal of Geology, 2008, volume 116, p. 62–75] � 2008 by The University of Chicago.All rights reserved. 0022-1376/2008/11601-0004$15.00. DOI: 10.1086/524675

62

Terra Rossa Genesis, Implications for Karst, and Eolian Dust:A Geodynamic Thread

Enrique Merino and Amlan Banerjee1

Department of Geological Sciences, Indiana University, Bloomington, Indiana 47405, U.S.A.(e-mail: [email protected])

A B S T R A C T

Although terra rossa has long been thought to form by residual dissolution of limestone and/or by accumulation ofdetrital mud, ash, or dust on preexisting karst limestones, we present conclusive new field and petrographic evidencethat terra rossa forms by replacement of limestone by authigenic clay at a moving metasomatic front several cen-timeters wide. The red clay’s major chemical elements, Al, Fe, and Si, probably come from dissolved eolian dust.The replacement of calcite by clay exhibits a serrated, microstylolitic texture that helps prove that replacementhappens not by dissolution-precipitation, as conventional wisdom has it, but by pressure solution of calcite drivenby the crystallization stress generated by the growth of clay crystals. The acid produced by the isovolumetric re-placement of limestone by clay quickly dissolves out additional porosity/permeability in an adjacent slice of limestonewithin the front, triggering a reactive-infiltration instability that should, theoretically, convert the moving reactionfront into a set of wormholes, then funnels, then sinks—the very karst morphology that in nature does contain theterra rossa itself. This beautifully explains why terra rossa and karst are associated.

Introduction

Terra rossa clays are red claystones up to severalmeters thick and kilometers across that occur atthe earth’s surface and are associated with karstcarbonates. There are terra rossas across southernEurope and South Australia (where they supportvineyards), the Caribbean and surrounding lands,southern China, and elsewhere. The terra rossas inJamaica (Comer 1974) and southern France (Guen-don and Parron 1985) grade into bauxite, an alu-minum ore. The two existing theories about theirorigin, the residual and the detrital, are discussedin detail in “Previous Work on Terra Rossa.” Boththeories have problems, and both neglect to focuson the contact between terra rossa and the under-lying carbonate. According to the residual originhypothesis, terra rossa is the insoluble residuumleft by dissolution of limestone. Its main problem(pointed out by nearly every proponent of the de-trital theory) is that limestones contain little or noclay or other insoluble minerals (Ruhe et al. 1961;Ruhe 1975; Comer 1976; Mee et al. 2004), so un-realistically large thicknesses of limestone would

Manuscript received July 2, 2007; accepted October 12, 2007.1 E-mail: [email protected].

have to be dissolved to yield a significant thicknessof terra rossa. According to the detrital theory, terrarossa forms by accumulation of alluvial mud, vol-canic ash, or eolian dust on limestones. A basicproblem of the hypothesis is that it does not ac-count for the worldwide association of terra rossawith karst carbonate rocks.

The purpose of this article is to propose a newtheory of terra rossa formation, by authigenic re-placement of the underlying limestone at a narrowreaction front. The new theory is based on field andpetrographic evidence on the terra rossa at Bloom-ington, Indiana. Since the clay is authigenic, itsmajor elements—Al, Si, and Fe—must come to thefront as aqueous ions. We propose that these aque-ous ions probably result from dissolution of dustat the surface. Taking account of recent insightsinto the physics of mineral replacement (summa-rized in “Replacement Physics”) and into the dy-namics of moving reaction fronts, we then showthat the clay-for-carbonate replacement, because itgenerates acid, which dissolves out new porosity,should trigger a morphological reactive-infiltrationinstability, which theoretically “specializes” inproducing wormholes and funnels, precisely the

Journal of Geology T E R R A R O S S A G E N E S I S 63

morphology characteristic of karst. That is, terrarossa genesis and karst weathering appear to be cou-pled and mutually reinforcing phenomena that oc-cur at the base of the terra rossa, driven by theaqueous solutes resulting from dissolution of de-trital dust by rain and soil water at the surface ofthe terra rossa.

In the second article of this series (A. Banerjeeand E. Merino, unpub. manuscript), we will presenta quantitative reaction-transport model of terrarossa formation based on the qualitative geochem-ical dynamics discovered and described here. Thequantitative model yields predictions of the frontwidth, bizonality, and velocity; the predictionscompare reasonably well to field, petrographic, andpaleomagnetic observations. Paleomagnetic obser-vations made on the Bloomington terra rossa,which allow us to date it and to estimate its rateof formation, are presented in a third article (J. G.Meert, F. Pruett, and E. Merino, unpub. manu-script).

Previous Work on Terra Rossa

The Residual Origin. This theory was proposedearly (de Lapparent 1930; Thornbury 1954) and sur-vives in geomorphology textbooks. Of anecdotal in-terest is Bardossy’s uncritical description of terrarossa research since the nineteenth century (Bar-dossy 1982, p. 329–338, 350–351). Thornbury (1954,p. 319) simply took it for granted that terra rossais the residual product of limestone dissolution by“descending groundwater,” the very dissolutionthat is assumed to produce the depressions typicalof karst weathering. Moresi and Mongelli (1988),after comparing chemical analyses of the insolubleresidue left by dissolving Apulia limestones tochemical analyses of the terra rossa that occurs onthe limestones, thought that the terra rossa is prob-ably in part a residue from limestone dissolution.Other recent authors (Durn et al. 1999; Delgado etal. 2003; Durn 2003), also on the basis of comparingchemical analyses and comparisons of particle sizeanalyses, believe that only minor portions of spe-cific terra rossa formations in southeastern Spainand Istria (Croatia) are residual products of lime-stone dissolution, with the rest being detrital.

The residual theory of terra rossa origin has thepositive aspect of explicitly linking terra rossa gen-esis to karst formation, thus apparently explainingtheir association, but it has two problems. First,limestone dissolution can yield only a fraction ofthe observed thickness of terra rossa, because lime-stones contain little insoluble minerals to beginwith and, in particular, little Si, Al, and Fe, which

are the major elements of terra rossa clay. Thismass deficit, pointed out by Ruhe et al. (1961), Ruhe(1975, p. 17–19), Comer (1974), Hall (1976), Olsonet al. (1980), Herwitz et al. (1996), Donovan (2002),Foster and Chittleborough (2003), Mee et al. (2004),Muhs et al. (2007), and others, is what gave rise tothe detrital theory. Second, the dissolution of lime-stone into sets of funnels and sinkholes that char-acterize karst weathering is attributed to the fo-cusing of dissolution at supposed intersectingsubvertical fractures (e.g., Thornbury 1954; White1988, p. 111; Donovan 2002; Twidale 2004), but thisfails to explain why at least some holes (such asthe one in the “floater” of fig. 6C; see “Discus-sion”), funnels, or sinks have no associated frac-tures, and it does not help explain the associationof terra rossa with karst.

The Detrital Origin. According to the detritaltheory, terra rossa is an accumulation of detritalmaterial—alluvial mud (Ruhe et al. 1961; Hall1976; Olson et al. 1980), volcanic ash (Comer 1974),or eolian dust (Yaalon and Ganor 1973; Herwitz etal. 1996; Yaalon 1997; Durn et al. 1999; Durn 2003;Foster and Chittleborough 2003; Frumkin and Stein2004; Muhs et al. 2007)—on an (assumed preexist-ing) karst limestone. Although photos in NationalGeographic (January 2004) of dust storms settlingSaharan dust on southern Europe, satellite photosof dust plumes traveling from the Sahara to theCaribbean, and extensive evidence obtained in thepast 25 years (e.g., Yaalon 1997; Prospero et al. 2001,2002; Muhs et al. 2007), plus evidence of eolian dustsupply to weathering profiles on silicate rocks (e.g.,Brimhall et al. 1988; Kurtz et al. 2001), leave nodoubt that eolian dust settles on all surface rocks,including carbonates, in many parts of the world,the idea that the terra rossa itself is a detrital ac-cumulation (with or without later alteration) alsohas problems. First, it implicitly assumes that dustaccumulates only on already karstified carbonates,which not only converts the terra rossa/karst as-sociation into a coincidence but also neglects thatGuendon and Parron (1985), based on excellentfield evidence, demonstrated that the karst underthe terra rossa/bauxite at the bauxite type localityof Les Baux, southern France, had developed si-multaneously with the development of the bauxitichorizons. (Their evidence was an isopach mapshowing that the greater thicknesses of bauxite oc-cur precisely above limestone sinks and the smallerthicknesses of it occur above the domes betweensinks.) Second, it does not account for the commonoccurrence of “floaters” in terra rossa—in situlimestone blocks of any size completely sur-rounded by, and “floating” in, the red clay: how

64 E . M E R I N O A N D A . B A N E R J E E

could detrital mud or dust get into a limestone andisolate one such block? (Floaters are uncovered fre-quently by developers in Bloomington. Many areexhibited in front yards of private houses for theirinteresting shapes, abounding in concavities; oneis shown in fig. 6C.)

Methodological Shortcomings. Aside from theproblems listed with both ideas of terra rossa for-mation, there are also methodological flaws in theevidence adduced to support them in previous re-search. Many researchers, trying to demonstratethat a terra rossa is a detrital deposit of eolian dustor volcanic ash, have compared chemical or iso-topic analyses (of Al/Fe, Zr, and Sr isotope ratios,U isotope ratios, or rare earths) of terra rossa, of theunderlying limestone, and of dust or ash and havedemonstrated that a number of chemical elementsin the terra rossa are indeed allochtonous (see Del-gado et al. 2003; Durn 2003; Mee et al. 2004; Muhset al. 2007), but this finding does not necessarilyimply that the mineral grains containing those el-ements are themselves detrital. They could be au-thigenic instead, but this possibility does not ap-pear to have been considered by previous authors.It seems to us that chemical analysis is by itselfblind to whether a rock’s mineral grains are detritalor authigenic, a crucial distinction ascertainablepetrographically.

This brings us to a second methodological prob-lem. To our knowledge, and with the single excep-tion of Ruhe et al. (1961), discussed in the nextparagraph, soil scientists appear to have neglectedthe study of terra rossa origin by polarized-lightpetrography, the only technique that could—withluck—enable an observer to establish whether min-eral grains in rocks are detrital or authigenic (Wil-liams et al. 1954; Pettijohn 1957, p. 111–112) and,if authigenic, whether they are cements, replace-ments, or displacive crystals (Folk 1965). It may berelevant to note in this regard that benchmark text-books of soil micromorphology, such as Brewer’s(1964) and FitzPatrick’s (1993), do not stress con-cepts and textures fundamental in understandingthe genesis of rocks, such as “authigenic,” “ce-ment,” and “replacement.” The excellent micro-morphological atlas by Delvigne (1998) goes a longway toward bridging the gap between the aims ofsoil micromorphology and the aims of petrographyof rocks, although the effort is marred by an opaqueterminology. An additional subtle problem may bethat terra rossa is often referred to as “terra rossasoils,” a term that, in effect, takes for granted thatthe genesis of terra rossa clays is pedological bydefinition. As we show in this article, however, itis more accurate to view terra rossa as a claystone

meters thick that grows authigenically at its baseby replacement of the underlying limestone. Ofcourse, the top portion of that claystone may laterbe—if it is not eroded first—pedologically alteredinto a soil.

As noted, Ruhe et al.’s (1961) study (summarizedby Ruhe [1975], p. 17–19) of the terra rossa of Ber-muda is, to our knowledge, the only one where thinsections of terra rossa samples taken from smallkarst funnels were examined with a polarized-lightmicroscope. Ruhe found what he called “sand-sizedgrains of calcium carbonate” surrounded by redclay and, perhaps unaware that those “grains”might be simply unreplaced bits of limestone—un-replaced small floaters—went on to call them “sandgrains” (no longer just sand-sized grains); on thissole basis, he concluded that the terra rossa “mustbe an accretionary layer in the soil,” that is, a de-trital sediment or pedisediment. But replacementmay be very difficult to detect (Pettijohn 1957, p.111–112), especially where it is complete andwhere preservation of internal details is poor (orwhere there were no details to preserve), even if thepetrographer is explicitly looking for it. It impliesno criticism of Ruhe et al. (1961) to say that theycould have missed it.

Another problem for previous workers has beenseeing the terra rossa/limestone boundary as“sharp” (Comer 1974; Hall 1976; R. V. Ruhe, pers.comm., 1978; Olson et al. 1980) and taking thatsharpness as evidence that the terra rossa is detrital.As our figure 1 demonstrates, however, the bound-ary is actually a reaction zone several centimetersthick, which we describe next. An unintended con-sequence of seeing the contact with the underlyinglimestone as sharp, combined with routinely re-ferring to terra rossa as a “soil,” has been to deflectthe attention of subsequent students of terra rossaorigin away from its bottom, which turns out to bethe best place to demonstrate petrographically thatthe red clay replaces the calcite.

New Evidence: Reaction Front

In 2005, we discovered a metasomatic reactionfront, shown in figure 1, between terra rossa andthe underlying Salem Limestone, on a wall of theexcavation for the foundation of a new sciencebuilding on the campus of Indiana University,Bloomington. The front is subvertical, cuts acrossbedding, and is 1–2 m below the surface. It consistsof a bleached zone A, 3 cm wide, and a replacementzone B, 6 cm wide. Zone B is the region where thereplacement of limestone by clay was last in pro-gress: white, still-unreplaced bits of calcite are vis-


Figure 1. Reaction front between terra rossa and Salem Limestone at the construction site for Simon Hall, IndianaUniversity campus, Bloomington. The Salem Limestone is a coarse-grained calcarenite, here massive and horizontal.The front thus cuts across bedding. It consists of a zone of bleached carbonate (zone A) 3 cm wide, containing a thinsubzone of opaques, and a zone of clay-for-calcite replacement (zone B) 6 cm wide. The labels “2a,” “2b,” “3a,” etc.,correspond to the approximate spots of photomicrographs in figures 2a, 2b, 3a, etc.

ible in zone B (fig. 1b). The bleached zone A con-tains a thin subzone of opaque grains of a Mn oxide.The Mississippian Salem Limestone is a massivecalcarenite consisting mostly of coarse crinoid andbryozoan fragments cemented by clear calcite;chemical analyses provided by the Indiana Geolog-ical Survey indicate that the unaltered Salem con-tains small amounts of Al, Mn, Si, and Fe, generallyless than 1 wt% of their oxides.

Petrography, Zone B. The field evidence of a re-placement-and-bleaching front is confirmed by pet-rographic analysis of six thin sections from samplesfrom the front. (Another 60 terra rossa samples notfrom the front have been examined petrographi-cally, and about 25 of them have been studied mag-netically.) Figure 2c shows a circular crinoid col-umnal from the replacement zone that is partlyreplaced by an orange clay aggregate that preservesthe shape and volume of the replaced portion of thecrinoid, a feature characteristic of replacement. Fig-

ure 3a shows another partial replacement of thecenter of a calcite cement plate by orange clay; thetwo unreplaced ends are still in optical continuity(fig. 3b), showing that they have not been displacedor rotated and that the replacement is therefore iso-volumetric. Figure 3c shows an incipient replace-ment of another crinoid fragment by orange clay;the incipient replacement is also clearly volume forvolume and exhibits a striking serrated microtex-ture also shown in figure 4a, 4b. We return to thatserrated texture in “Replacement Physics.”

The crossed-polars photomicrographs (fig. 3b, 3d)show that the orange clay aggregates replacing thecalcite cement and fossils have crystalline texture;that is, they consist of interlocking crystals andthus indicate authigenesis. The orange crystals areFe3�-bearing kaolinite; see “Mineralogy.” The factthat, in the replacement zone, the calcium carbon-ate fossils and cement are only partly replaced andoccur in all stages of replacement from incipient


Figure 2. Petrography of the leaching and replacement zones. The site of each photomicrograph is marked in figure1. a, Unaltered Salem Limestone consists of bryozoans and equinoderm fragments (black arrows) well cemented byclear calcite cement. Plane-polarized light. b, Dissolution voids (see also those in the unreplaced portion of the crinoidin c) are produced in the bleaching zone by leaching of calcite driven by the H� released by the replacement takingplace in zone B, reaction (1). Crossed polars. c, Circular crinoid columnal at center is partly replaced by orangekaolinite (black arrow) in the replacement zone (B) of figure 1. The circular shape of the replaced portion is roughlypreserved, indicating isovolumetric replacement, a fact used in adjusting reaction (1). The crinoid was slightly leachedwhen it was in the bleaching zone (A), before it started to be replaced. The leaching can be seen in the unreplacedportion (white arrow) of the calcitic columnal. Plane-polarized light. d, A fairly mature iron oxide pisolite from farbehind and above the front, surrounded by terra rossa clay. Both pisolite and clay contain many quartz silt grains,probably left undissolved from dissolution of eolian dust and worked into the terra rossa pedologically; see “ElementsMaking Up the Red Clay Probably Come from Dissolved Dust.” The pisolite grew slowly (from an incipient precursorlike the one in fig. 3a), replacing the surrounding clay and incorporating its quartz silt grains. Plane-polarized light.

to nearly complete makes the identification of theauthigenic replacement unmistakable. By contrast,in most thin sections of terra rossa samples takenfar behind the front, the carbonate is completelyreplaced, so that fossil shapes are no longer une-quivocally preserved or recognizable.

The authigenic kaolinite crystals may reach 1mm or more in size. Some display faint growthrings. We do not interpret the rings as evidence ofcutans formed by detrital illuviation because theyare contained within large, authigenic, single crys-tals, as seen by their sharp extinction under crossedpolars. Illuviation, if understood as detrital, can

happen in a soil, but it seems doubtful that it couldreach the bottom of a terra rossa claystone severalmeters thick. Also, detrital illuviation cutans dis-play sweeping (not sharp) extinction under crossedpolars. (Interestingly, Brewer [1964, p. 224] says thatilluviation cutans are “formed by movement of thecutanic material in solution or suspension [our ital-ics] and subsequent deposition.” Note that Brewerdid not see it as relevant whether a cutan is au-thigenic, detrital, or recrystallized-detrital [as is thematrix of graywackes; Williams et al. 1954, p. 297]and included all possibilities in the definition.)

Bleaching Zone, Zone A. The bleaching zone of


Figure 3. Clay-for-calcite replacements in zone B of the terra rossa front. Sites of the photomicrographs are markedin figure 1. Zone B is the zone of current replacement. a, Large orange clay aggregate (between arrows) partly replacesthe center of a plate of calcite cement in zone B, leaving two still-unreplaced portions (crosses), one on each side.The crinoid at center is also still unreplaced. The clay aggregate contains an incipient, fuzzy iron oxide pisolite (pi)that has crystallized recently, like the clay itself. b, Same as a, under crossed polars, showing that the two portionsof unreplaced calcite cement are in optical continuity and that the clay aggregate consists of interlocking crystals,indicating that the red kaolinite is authigenic. c, Brown clay starts (at arrow) to replace a crinoid fragment, producinga striking serrated texture (Pettijohn 1957, p. 674) also evident in figure 4a and throughout the replacement zone B;see “Petrography, Zone B.” Plane-polarized light. d, Same as c, under crossed polars, showing the large size of someof the authigenic clay crystals by the size of areas with one birefringence (arrow).

the front, zone A in figure 1, is white, contains athin belt of black opaque particles of a Mn oxide,and under a microscope is seen to contain manylarge dissolution pores, visible in figure 2b as theareas under extinction. This high-porosity/perme-ability region, occurring primarily in zone A (and toa lesser extent in zone B) but not in the fresh lime-stone ahead of the front or in the terra rossa leftbehind it, is crucial for the dynamics of the front.

Pisolites. The authigenic clay in the replace-ment zone contains, floating in it, a few small,fuzzy aggregates (such as the one marked “pi” infig. 3a) of very dark brown Fe oxides that are alsoauthigenic (Nahon 1991, p. 158 and fig. 4.14). Theyare the precursors of more mature pisolites, with

incipient concentric layering that can be seen inthin sections of terra rossa far behind the front (seefig. 2d), may reach 0.5 cm in diameter, and areeroded out of the terra rossa at many spots inBloomington. Most pisolites consist of goethite,but a few consist of maghemite, a strongly mag-netic polymorph of hematite (Pruett 1959; J. G.Meert, F. Pruett, and E. Merino, unpub. manu-script).

Mineralogy. X-ray diffractograms of Blooming-ton terra rossa show phyllosilicate peaks at ap-proximately 7, 10, and (expandable) 12–13 A, as wasreported by Olson et al. (1980). The 7-A peaks mustcorrespond to kaolinite, identified optically by itsdeep orange color, weak pleochroism, single cleav-


Figure 4. Mineral replacement via pressure solution. a,Serrated microtexture produced by the growing clay (up-per half) as it replaces the carbonate (lower half) in zoneB of the terra rossa front of figure 1. This microstylolitictexture is excellent evidence that the replacement of thehost calcite took place not by dissolution-precipitation,as is generally thought, but via pressure solution drivenby the crystallization pressure exerted by the growingclay; see “Replacement Physics.” Crossed polars. b, An-other microstylolitic texture, produced here by the re-placement of pentlandite by actinolite in an igneous orefrom South Africa; see text. (Backscattered-electron pho-tomicrograph by C. Li; from Li et al. 2004). c, Replace-ment of dolomite for sphalerite in Pb-Zn ores at Galmoy,Ireland, beautifully preserving the sphalerite layering(Merino et al. 2006). If the replacement had taken place

by dissolution-precipitation, the sphalerite layeringwould have been destroyed by sphalerite dissolution be-fore the dolomite grew and could not have been pre-served. Also, sphalerite dissolution could not have trig-gered the growth of dolomite because the two mineralshave no elements in common and sphalerite could nothave “known” how to dissolve leaving a void with theshape of the future dolomite idiomorph. But with re-placement viewed as resulting from pressure solution ofthe sphalerite by the growing dolomite, the petrographicfeatures of the replacement—preservation of host mor-phological details and of volume (and equality of volu-metric rates)—are easily accounted for; see “Replace-ment Physics.” (Photomicrograph by A. Canals,University of Barcelona.)

age, parallel extinction, and low birefringence(though higher than pure kaolinite’s). By compar-ison with the kaolinites typical of laterites (Nahon1986, p. 171; Muller et al. 1995), the orange colorand pleochroism undoubtedly reflect some substi-tution of Fe3� for Al in the kaolinite.

Replacement Physics

To interpret the clay-for-calcite replacement geo-chemically and to model terra rossa formation bymetasomatic replacement at a moving reactionfront (A. Banerjee and E. Merino, unpub. manu-script), it is essential to understand how a replace-ment takes place. In this section, we summarizethe new theory of replacement that emerges fromMaliva and Siever (1988), Merino et al. (1993), Na-hon and Merino (1997), Merino and Dewers (1998),and Fletcher and Merino (2001). The essence of thenew theory is that a replacement takes place be-cause the new mineral pressure-dissolves the hostby means of the induced stress, or crystallizationstress, that it exerts as it grows.

The term “replacement,” as long used by pe-trographers (e.g., Bastin et al. 1931, p. 586 ff; Wil-liams et al. 1954; Pettijohn 1957, p. 111), refers tothe occurrence of a guest mineral A just where thehost B used to be but preserving both the volumeand (often) some internal morphological details ofB. Replacement is thus detectable only visually; seefigure 4c. Because morphological details are pre-served, A growth and B dissolution must be si-multaneous. Because the volume is preserved, Agrowth and B dissolution must be coupled so as tomake their volumetric rates mutually equal, asthese could not be equal otherwise. But what is thecoupling factor?


Perhaps since the influential article by Weyl(1959), who—without mentioning replacement andapparently unaware of its uniqueness as grasped byBastin et al. (1931), Williams et al. (1954), Pettijohn(1957), and other petrographers—implicitly re-garded mineral replacement as dissolution plus pre-cipitation, mineral replacement has been identifiedwith its assumed mechanism, “dissolution-precip-itation” (e.g., Walker 1962; Carmichael 1969;Plummer 1975; Knauth 1979; Milliken 1989; Put-nis 2002; Putnis and Mezger 2004; Rendon-Angeleset al. 2006), a mechanism whereby the dissolutionof the host is assumed to precede and chemicallytrigger the precipitation of the guest. However, dis-solution-precipitation fails to account for thedolomite-for-sphalerite replacement shown in fig-ure 4c. (1) If the sphalerite had dissolved first, howcould the later dolomite have preserved its botrioi-dal layering at all? (2) If the sphalerite had dissolvedfirst, how could it have “known” to dissolve leav-ing a hollow with the euhedral shape of the futuredolomite? (3) How could sphalerite dissolution trig-ger instant dolomite growth if the two mineralshave no chemical elements in common? (4) Evenif there were elements in common between hostand guest (as there are indeed between calcite anddolomite in dolomitization), how could that chem-ical feedback ensure—except by chance or by ex-perimental design—that the guest mineral wouldprecipitate at the exact place the host had dissolvedand at the same volumetric rate (so as to preservevolume)? Other cases where the chemical couplingimplicitly taken for granted in dissolution-precip-itation also does not exist are the common replace-ment of limestone by chert (Maliva and Siever1988) and the replacement of pentlandite by actin-olite shown in figure 4b (Li et al. 2004).

Thus, Maliva and Siever (1988), after discuss-ing the weaknesses of dissolution-precipitationpointed out above and others, all of which invali-date it as accounting for a replacement texture inthe sense of Bastin et al. (1931), had the brilliantnew idea that a replacement happens when theguest mineral A starts to grow at a point in a rigidrock and, via the crystallization stress that it exertson host B, pressure-dissolves B as it grows. Malivaand Siever invoked not a chemical coupling but aphysical coupling between A growth and B disso-lution; that is why it works in all cases, regardlessof whether A and B have components in common.Dewers and Ortoleva (1989) demonstrated Malivaand Siever’s conjecture, using the Navier-Stokesequation for momentum conservation. Nahon andMerino (1997) showed how replacement by in-duced-stress–driven pressure solution automati-

cally forces the rates of A growth and B dissolutionto become mutually equal, thus preserving mineralvolume. Merino and Dewers (1998) showed thatmorphological details of the host can be preservedif the growth increments of the guest are smallerthan the details themselves (but are erased other-wise, much as the details of a photograph wouldbe erased if we tried to replace them with too-largetiles or pixels). Fletcher and Merino (2001) calcu-lated the growth-induced stress and the replace-ment rate from interacting rheology and stress-driven kinetics.

Another hint that replacement does involve pres-sure solution comes now from the clay-for-calcitereplacements in the Bloomington terra rossa. Asthe clay replaces the calcite, a striking serrated mi-crotexture develops, shown in figure 4a and seenthroughout the zone of current replacement, zoneB. Pettijohn (1957, p. 216, 674) discussed such ser-rated texture but for minerals different from clayand calcite. If the serrated texture in the terra rossameans what it means in stylolitization—namely,that it is produced by pressure solution—then itimplies that the replacement of calcite by clay doeshappen via pressure solution as well, lending un-expected support to Maliva and Siever’s (1988) con-jecture. The replacement in figure 4b of pentlandite(a nickel sulfide) by actinolite (a Ca-Mg amphibole)in an igneous ore from South Africa (Li et al. 2004)displays the same microstylolitic texture and isalso excellent independent evidence that pressuresolution was involved in the replacement. (Notethat the absence of serrated texture in a particularreplacement does not mean that pressure solutionwas not involved in it.) For the following sectionand for the second article of this series (A. Banerjeeand E. Merino, unpub. manuscript), we retain theidea that the replacement of calcite by clay is iso-volumetric and takes place by clay-growth–drivenpressure solution of the calcite.

Dynamics of the Reaction Front

The dynamics of the replacement-and-leaching frontis determined by the reactions happening in it andtheir feedbacks, and it has two interrelated aspects.The chemical dynamics discussed in this sectiongives rise to, and interacts with, a morphologicaldynamics of the front, discussed in “The Reactive-Infiltration Instability and the Origin of Karst.”

The chemical dynamics is schematically shownin figure 5: aqueous Al, Si, and Fe reach the frontfrom the terra rossa side and combine at the frontto produce clay crystals that replace calcite grainsand fossils by reaction (1) below, releasing acid that


Figure 5. Dynamics of the replacement-and-leachingreaction front of figure 1. Under advective supply of aque-ous Fe, Al, and Si from the right, clay crystals grow andpartly replace calcite fossils and cement in the currentreplacement zone, releasing H� ions according to reac-tion (1). This H� leaches more limestone, creating theadjacent bleached zone. In the near future, when replace-ment in the current replacement zone is complete, thelocus of limestone replacement will shift to today’sbleached zone, and the acid generated there will leachnow still-fresh limestone. The front travels toward thelimestone, leaving behind a trail of terra rossa clay, theage of which is increasingly older the farther it is behindthe front (or increasingly older the higher it is above thefront, if the front is moving downward).

immediately bleaches and dissolves voids in an ad-ditional slice of limestone, zone A. When the un-replaced calcite in today’s replacement zone B iscompletely replaced in the near future, replacementwill start in what is today the bleached zone, A, andthe acid generated there will start leaching andbleaching what is today still fresh limestone just tothe left of zone A in figure 1b. This is how the meta-somatic front moves across the limestone, leavingbehind (and/or above) it a trail of terra rossa that isincreasingly older the farther it is from the front.

Local Mass Balance upon Replacement. The con-stant-volume replacement of calcite by clay (hereassumed to be pure kaolinite for simplicity) can bewritten as

3�2.7CaCO � 2Al � 2SiO3(calc) 2

� 5H O p Al Si O (OH)2 2 2 5 4(kaol)

�� 2.7Ca � 2.7HCO � 3.3H , (1)3

where the 2.7 (ratio of formula volumes of kaoliniteand ) ensures constant volume. Ascalcite p 99/37implied in the previous section, the mass balance(eq. [1]) is really the sum of two stress-coupled re-actions: clay growth from its aqueous ions and crys-tallization-stress–driven pressure solution of anequal volume of calcite.

Choice of Species in Equation (1). Before we dis-cuss the consequences of the release of acid by re-action (1), a word is in order about the choice ofaqueous and mineral species in it. All re-��CO3

leased by calcite is assumed to pair with H� to be-come the released . The aqueous aluminum�HCO3

in reaction (1) is in the form of Al3� because itwould probably be the predominant Al species inthe pore fluid (because meteoric and soil waters inauthigenic clays are bound to be weakly acid) andbecause, as an already octahedrally coordinatedspecies, it is the ion that actually enters into theoctahedral sheet of the kaolinite structure (Merinoet al. 1989).

If, instead of pure kaolinite, we had chosen Fe3�-bearing kaolinite or illite in reaction (1), whichwould be more realistic than pure kaolinite, the leftside of equation (1) would contain species such asFe3�, Mg2�, and K�, in addition to Al3�. Again, Fe3�

would be the appropriate and perhaps predominantaqueous iron species in the acid environment ofauthigenic terra rossa clays. In all cases, the re-placement mass balance would still release H�.

Consequences of Replacement (Eq. [1]). Reaction(1) has an important by-product, H�. The H� shouldleach an adjacent slice of limestone, giving rise toa bleached zone, which is indeed what happens; seezone A in figure 1. It should also liberate the traceof Mn�2 contained in the limestone (the Salem con-tains up to 0.06 wt% Mn; Indiana Geological Sur-vey, pers. comm.), which would immediately oxi-dize to Mn�3 and/or Mn�4 and reprecipitate as tinyblack Mn oxide particles. This is also what hap-pens, in fact; see the thin black subzone of zone A


in figure 1b. But most important, the released H�

should dissolve out voids in zone A by the reaction

� �� CaCO � H p Ca � HCO , (2)3(calc) 3

and this is also confirmed by the dissolution voidsshown in figure 2b. The voids increase the porosity/permeability of zones A and (to some extent) B andappear to be mostly plugged later by clay growth thatreplaces the bleached calcite. In the language of dy-namics, the moving bleached zone becomes a po-rosity soliton, or a single porosity wave. The porositywave traveling across the limestone should triggera reactive flow instability that makes the replace-ment self-accelerating and the front spontaneouslyfingered, as discussed in detail in “The Reactive-Infiltration Instability and the Origin of Karst.”

The occurrence of the predicted bleached zoneadjacent to the replacement zone, with its thinblack belt and its dissolution pores, is excellentevidence that the observation of replacement, theconjecture that it occurs via pressure solution, andits representation by reaction (1), are correct. Thisis an important point. Note that the conventionaldissolution-precipitation process—namely, the cal-cite host dissolves first, followed by growth of guestkaolinite; see “Replacement Physics”—could notpossibly preserve mineral volume, a basic propertyof replacement and one we observe petrographi-cally (fig. 2c; fig. 3a, 3b), since there is only at besta weak coupling (via H�) between the the calciteand the kaolinite and their rate constants differ byfive or more orders of magnitude. In addition, if thecalcite host had dissolved first, there would be noway for the growth of the guest kaolinite to pre-serve morphological features of the calcite, such asthe circular shape of the crinoid in figure 2c, theother basic feature of replacement. (These problemsof the dissolution-precipitation mechanism be-come evident in our quantitative reaction-transportmodel of terra rossa genesis [A. Banerjeee and E.Merino, unpub. manuscript].)

Cementation. As the unleached calcite of zoneA starts to be replaced by kaolinite, not only doesreaction (1) take place but also the previouslydissolved-out pores may become cemented or half-cemented by kaolinite, according to

3�2Al � 2SiO � 5H O p2 2

�Al Si O (OH) � 6H , (3)2 2 5 4(kaol)

which also produces acid that would help that pro-duced by reaction (1) to generate the next leachingzone. As noted in “Discussion,” this cement may

interfere with the operation of the reactive-infil-tration instability (see “The Reactive InfiltrationInstability and the Origin of Karst”), by pluggingsome of the porosity/permeability created by theleaching.

Elements Making Up the Red Clay ProbablyCome from Dissolved Dust

Many investigators, cited in “Previous Work onTerra Rossa,” have shown that specific terra rossaformations contain chemical elements that can betraced to elements in eolian dust, confirming thefact that eolian dust must settle on the earth’s sur-face. But since the red clay making up terra rossais authigenic, as shown in figures 1, 3b, and 3d, theAl, Fe, and Si needed to make the clay crystals mustbe supplied to zone B as aqueous ions. We thereforesuggest here that the aqueous Si, Fe, and Al sup-plied to the front come from the dust fraction thatis dissolved at the surface of the existing terra rossaand then delivered, by infiltration, to the front afew meters below. Acid rain and soil water wouldquickly dissolve the finest fraction of the dust,which presumably is clay rich. In support of thisidea is the following petrographic evidence.

If the finest dust is dissolved, the undissolvedfraction, in general mainly quartz (and feldspar) silt,should be left on the terra rossa and would beworked into it pedologically. (The quartz silt wouldbe insoluble in the soil pore water, enriched in sil-ica by dissolution of the finest dust fraction.) In-deed, we have seen (fig. 2d) scattered quartz silt inmany thin sections of Bloomington terra rossa.

The Reactive-Infiltration Instabilityand the Origin of Karst

Where a reactive fluid both flows through a porousrock and partly dissolves it, a porosity-making frontis established that was quantitatively predicted inthe 1980s and 1990s to become fingered (Chadamet al. 1986; Ortoleva et al. 1987; Aharonov et al.1997); see figure 6A. The dissolution increases po-rosity/permeability, which accelerates advection ofreactive water. In turn, the faster advection accel-erates further dissolution. This is the reactive-infiltration instability. It works as follows. Anyhigher-than-average porosity/permeability in a vol-ume element of the front, such as at point a infigure 6A, captures flux from the neighbor ele-ments. Dissolution thus accelerates at a, increasingits porosity/permeability even more, drawing stillmore flux from the neighbor elements and produc-ing a high-permeability finger. Simultaneously, the


Figure 6. The reactive infiltration instability; see text.A, Schematic representation of the spontaneous fingeringof a dissolution front moving through a porous rock, pro-duced by the reactive infiltration instability. Any slightexcess porosity/permeability at a point of the front, pointa, captures reactive water (arrows) from its neighborregions, b, increasing the dissolution rate at its tip, a′,and simultaneously starving its neighbor zones, b, whichare left behind (b′) as the front advances, creating newfingers at c, which themselves grow faster than b andbecome new fingers. See the predicted fingering in Ahar-onov et al. (1997; their fig. 3). B, Schematically, jumpsin the scale of finger spacing predicted to take place spon-taneously (Szymczak and Ladd 2006). C, Actual fingersor wormholes in a “floater” from Bloomington.

neighbor volume elements b, deprived of reactiveflux, get left behind by a and also by volume ele-ments farther out, c, which now start to becomenew fingers or funnels of higher permeability them-selves and later produce still others farther away.Soon, the front becomes an advancing set of po-

rosity/permeability fingers. (This competitive mor-phological dynamics also takes place in the com-pletely different context of quartz growth withinagates in basalts; see Wang and Merino 1995. Inthat case, the fingers of the instability are thequartz fibers themselves, which make agates in-variably fibrous.) After the terra rossa front has be-come fingered, the competition for reactive fluxcontinues among the fingers themselves, leading tosuccessive jumps in the spacing and size of the fin-gers (Szymczak and Ladd 2006). The predicted “cas-cade” of scales is shown schematically in figure 6B.From being a set of advancing fingers, the frontpasses to being a set of funnels one order of mag-nitude larger than the fingers. In turn, the funnelslater jump to sinks an order of magnitude greaterin size and spacing than the funnels.

The reaction front we have discovered betweenterra rossa and the underlying Salem Limestone ap-pears to be a natural case of the kind of movingporosity-making front whose dynamics is describedabove and in figure 6. The instability-triggering dis-solution is the leaching (eq. [2]) carried out in thebleached zone by the acid released by reaction (1)in the replacement zone. The replacement-plus-bleaching front should therefore become fingeredas it advances into limestone, and the fingersshould jump to funnels and these to sinkholes. Butthese predicted forms coincide with the morpho-logical features that are characteristic of karst car-bonates (e.g., Thornbury 1954; White 1988; Don-ovan 2002; Twidale 2004). We thus arrive at thesurprising realization that the terra rossa’s authi-genic clay indirectly makes the very karst lime-stone morphology that contains it, which explainswhy the two are associated.

The singularity and separation between karstfunnels or sinks are traditionally attributed to thefocusing of descending acid water by intersectingfractures, which are conveniently assumed to havethe spacing required to produce the observed sink-hole spacing (e.g., White 1988; Twidale 2004). Ad-mittedly, limestone dissolution would be faster atthe intersection of two subvertical fractures, givingrise to a funnel or sink, but intersecting fracturesare an accidental feature, one that cannot becounted on systematically to determine the loca-tion of every sink of every karst, and furthermore,the assumed intersecting fractures have to have therequired spacing to produce the observed spacingbetween karst features. (Most concavities in manyfloaters in Bloomington’s front yards do not exhibitintersecting fractures; see fig. 6C.) On the otherhand, the reactive-infiltration instability is a mech-anism triggered internally, and it automatically


would go through a cascade of scales, which seemsbest able to account for the spectrum of scales ob-served in karst limestones.

Discussion

Perhaps the most striking aspect of the new geo-dynamic theory presented here is the predictionthat terra rossa genesis causes the karst morphol-ogy, a prediction confirmed by the worldwide as-sociation of terra rossa with karst limestones. Thenew picture of terra rossa genesis starts with thediscovery of a 10-cm metasomatic front at the baseof the Bloomington terra rossa. The petrographicevidence of replacement of limestone by authigenicred clay is conclusive. The field evidence, figure 1,of a bizonal replacement-plus-leaching front also isconclusive. The solutes needed to make the claymust come from dissolution, by rain and soil water,of the finest fraction of the eolian dust supplied tothe top of the terra rossa. The top of the terra rossais commonly altered pedologically. The terra rossacan be easily eroded, especially in advanced karst.A particular terra rossa formation, if still uneroded,may now be viewed as the authigenic claystonerecording the finest dust fraction fallen at the sitein question.

The new picture generates predictions that areconfirmed by observations:

1. The prediction that the replacement reaction(eq. [1]), via the acid it releases, should leach ad-ditional limestone is confirmed by the occurrenceof the leaching zone A just ahead of the replace-ment zone B (fig. 1).

2. Maliva and Siever’s (1988) conjecture of themechanism of replacement, which calls for pres-sure solution of host driven by the induced stressgenerated by the guest, is confirmed by our dis-covery of microstylolitic, serrated replacementcontacts between clay and calcite (figs. 3, 4), to ourknowledge not reported before.

3. The new porosity created in the front’s leach-ing zone qualitatively should trigger the reactive-infiltration instability (but see “Future Work”about the need to check this quantitatively). Thisinstability theoretically must deform the movingfront into a set of wormholes (and then funnels andthen sinks). This predicted morphology is con-firmed by the occurrence of karst limestones as-sociated with terra rossa everywhere. (There aremany karst limestones, especially mature karsttowers, without terra rossa on them, but it mustbe remembered that the red clays can be erodedand/or washed off easily from the underlyinglimestone.)

4. We suggest that the major elements needed tomake authigenic clay at the front must come inaqueous form from the dissolution of the finestdust fraction at the surface. This suggestion findstentative confirmation in the widespread occur-rence of undissolved quartz silt in many thin sec-tions of older terra rossa far behind and above thefront (see fig. 2d).

Implications for Limestone Weathering. The geo-dynamic model suggests that limestone weatheringtakes place—see figures 1, 5, and 6—partly by pres-sure solution in the replacement zone and partlyby chemical dissolution in the leaching zone byreaction (2), driven by the acid released by the re-placement (eq. [1]). Note that this acid comes fromclay formation, not from meteoric carbonic acid. Itis because the two dissolutions—pressure solutionand chemical dissolution—are mutually acceler-ating that carbonate weathering probably proceeds(at the wormholes, funnels, and sinks) at much fas-ter rates than hitherto thought, if the dust supplypermits. (In contrast, according to the conventionalview in geomorphology, limestone weathers bychemical dissolution alone, with the dissolutiondriven at roughly constant kinetics by meteoric car-bonic acid.)

Future Work. We hope that other investigatorswill undertake petrographic analysis of the narrowreaction zone between other terra rossas and theirunderlying carbonates to confirm and refine ourpetrography and to establish (by petrophysicalstudy) the porosity and permeability ranges of redclaystones. Also, the new model must be checked(1) by quantitative reaction-transport modeling ofthe formation of terra rossa clays (A. Banerjee andE. Merino, unpub. manuscript) and comparison ofits predictions to observations; (2) by isotopic andchemical analyses aiming to establish that (orwhether) the major elements do reach the front asaqueous species from dust dissolution; and espe-cially (3) by quantitative modeling of the reactive-infiltration instability in the specific case of thebizonal front of figure 1. Aharonov et al.’s 1997impressive three-dimensional modeling of the re-active-infiltration instability referred to a movingfront at which only a feed-dissolution ↔ advectionback takes place. But for the terra rossa–makingfront of figure 1, the feedback that must be studiedby instability analysis is that between advectionand several coupled reactions, namely, growth ofclay with simultaneous pressure solution of calciteby reaction (1), cementation of previous voids byreaction (3), and simultaneous leaching of voids byreaction (2).


A C K N O W L E D G M E N T S

Ideas and petrography were discussed with R. L.Hay, professor emeritus of the Universities of Illi-nois and California at Berkeley, in the fall of 2005.He died on February 10, 2006, in Tucson. This ar-ticle is dedicated to his memory. Thanks to ourfriends A. Canals of the University of Barcelona forthe photomicrograph in figure 4c and A. Basu, C.

Li, J. Schieber, and J. R. Dodd of Indiana Universityfor comments on a draft, the photomicrograph infigure 4b, the use of a Zeiss polarizing microscope,and advice on Bloomington’s Mississippian carbon-ates, respectively. We appreciate critical reviews bythree reviewers for the journal. We thank IndianaUniversity for a grant-in-aid to cover thin section-ing and travel.

R E F E R E N C E S C I T E D

Aharonov, E.; Spiegelman, M.; and Kelemen, P. 1997.Three-dimensional flow and reaction in porous media.J. Geophys. Res. 102:14,821–14,834.

Bardossy, G. 1982. Karst bauxites. Amsterdam, Elsevier,441 p.

Bastin, E. S.; Graton, L. C.; Lindgren, W.; Newhouse, W.H.; Schwartz, G. M.; and Short, M. N. 1931. Criteriaof age relations of minerals, with especial referenceto polished sections of ores. Econ. Geol. 26:561–610.

Brewer, R. 1964. Fabric and mineral analysis of soils.New York, Wiley, 470 p.

Brimhall, G. H.; Lewis, C. J.; Ague, J. J.; Dietrich, W. E.;Hampel, J.; Teague, T.; and Rix, P. 1988. Metal en-richment in bauxites by deposition of chemically ma-ture aeolian dust. Nature 333:819–824.

Carmichael, D. M. 1969. On the mechanism of progrademetamorphic reactions in quartz-bearing peliticrocks. Contrib. Mineral. Petrol. 20:244–267.

Chadam, J.; Hoff, D.; Merino, E.; Ortoleva, P.; and Sen,A. 1986. Reactive infiltration instability. IMA J. Appl.Math. 36:207–221.

Comer, J. B. 1974. Genesis of Jamaican bauxite. Econ.Geol. 69:1251–1264.

———. 1976. Genesis of Jamaican bauxite; a reply. Econ.Geol. 71:822–823.

de Lapparent, J. 1930. Les bauxites de la France meri-dionale. Paris, Imprimerie Nationale, 187 p.

Delgado, R.; Martın-Garcıa, J. M.; Oyonarte, C.; and Del-gado, G. 2003. Genesis of the terra rossa of Sierra deGador, Almerıa, Spain. Eur. J. Soil Sci. 54:1–16.

Delvigne, J. E. 1998. Atlas of micromorphology of min-eral alteration and weathering. Quebec, Min. Assoc.Can.; and Paris, ORSTOM, 494 p.

Dewers, T., and Ortoleva, P. 1989. Mechano-chemicalcoupling in stressed rocks. Geochim. Cosmochim.Acta 53:1243–1258.

Donovan, S. K. 2002. A karst of thousands: Jamaica’slimestone scenery. Geol. Today 18:143–151.

Durn, G. 2003. Terra rossa in the Mediterranean region:parent materials, composition and origin. Geol. Croat.56:83–100.

Durn, G.; Ottner, F.; and Slovenec, D. 1999. Mineralog-ical and geochemical indicators of the polygenetic na-ture of terra rossa in Istria, Croatia. Geoderma 91:125–150.

FitzPatrick, E. A. 1993. Soil microscopy and micromor-phology. Chichester, Wiley, 304 p.

Fletcher, R., and Merino, E. 2001. Mineral growth inrocks: kinetic-rheological models of replacement, veinformation, and syntectonic crystallization. Geochim.Cosmochim. Acta 65:3733–3748.

Folk, R. L. 1965. Some aspects of recrystallization in an-cient limestones. In Pray, L. C., and Murray, R. D.,eds. Dolomitization and limestone diagenesis. Soc.Econ. Paleontol. Mineral. Spec. Publ. 13:14–48.

Foster, J., and Chittleborough, D. 2003. Soil developmenton dolomites of the Cambrian Normanville Group atDelamere, South Australia. In Roach, I. C., ed. Ad-vances in regolith. Bentley, Cooperative ResearchCentre for Landscape Environments and Mineral Ex-ploration (CRC LEME), p. 131–132.

Frumkin, A., and Stein, M. 2004. The Sahara-east Med-iterranean dust and climate connection revealed bystrontium and uranium isotopes in a Jerusalem spe-leothem. Earth Planet. Sci. Lett. 217:451–464.

Guendon, J. L., and Parron, C. 1985. Les phenomeneskarstiques dans les processus de la bauxitization sursubstrat carbonate: examples de gisements du sud-estde la France. Ann. Soc. Geol. Belg. 108:85–92.

Hall, R. D. 1976. Stratigraphy and origin of surficial de-posits in sinkholes in south-central Indiana. Geology4:507–509.

Herwitz, S. R.; Muhs, D. R.; Prospero, J. M.; Mahan, S.;and Vaughn, B. 1996. Origin of Bermuda’s clay-richQuaternary paleosols and their paleoclimatic signifi-cance. J. Geophys. Res. 101:23,389–23,400.

Knauth, L. P. 1979. A model for the origin of chert inlimestone. Geology 7:274–277.

Kurtz, A. C.; Derry, L. A.; and Chadwick, O. A. 2001.Accretion of Asian dust to Hawaiian soils: isotopic,elemental, and mineral mass balances. Geochim. Cos-mochim. Acta 65:1971–1983.

Li, C.; Ripley, E. M.; Merino, E.; and Maier, W. D. 2004.Replacement of base metal sulfides by actinolite, ep-idote, calcite, and magnetite in the UG2 and Meren-sky Reef of the Bushveld Complex, South Africa. Econ.Geol. 99:173–184.

Maliva, R. G., and Siever, R. 1988. Diagenetic replace-ment controlled by force of crystallization. Geology16:688–691.

Mee, A. C.; Bestland, E. A.; and Spooner, N. A. 2004. Ageand origin of terra rossa soils in the Coonawarra areaof South Australia. Geomorphology 58:1–25.

Merino, E.; Canals, A.; and Fletcher, R. 2006. Genesis of


self-organized zebra textures in burial dolomites: dis-placive veins, induced stress, and dolomitization.Geol. Acta 4:383–393.

Merino, E., and Dewers, T. 1998. Implications of replace-ment for reaction-transport modeling. J. Hydrol. 209:137–146.

Merino, E.; Harvey, C.; and Murray, H. 1989. Aqueouschemical control of the tetrahedral aluminum contentof quartz, halloysite and other low-temperature alu-minosilicates. Clays Clay Miner. 37:135–142.

Merino, E.; Nahon, D.; and Wang, Y. 1993. Kinetics andmass transfer of replacement: application to replace-ment of parent minerals and kaolinite by Al, Fe andMn oxides during weathering. Am. J. Sci. 293:135–155.

Milliken, K. L. 1989. Petrography and composition ofauthigenic feldspars, Oligocene Frio Formation, southTexas. J. Sediment. Petrol. 59:361–374.

Moresi, M., and Mongelli, G. 1988. The relation betweenthe terra rossa and the carbonate-free residue of theunderlying limestones and dolostones of Apulia, Italy.Clay Miner. 23:439–446.

Muhs, D. R.; Budahn, J.; Prospero, J. M.; and Carey, S.N. 2007. Geochemical evidence for African dust in-puts to soils of western Atlantic islands: Barbados, theBahamas and Florida. J. Geophys. Res. 112:F02009,doi:10.1029/2005JF000445.

Muller, J. P.; Mainceau, A.; Calas, G.; Allard, T.; Ildefonse,P.; and Hazemann, J. L. 1995. Crystal chemistry ofkaolinite and Fe-Mn oxides: relation with formationconditions of low temperature systems. Am. J. Sci.295:1115–1155.

Nahon, D. 1986. Evolution of iron crusts in tropical land-scapes. In Colman, S. H., and Dethier, D. P., eds. Ratesof chemical weathering of rocks and minerals. SanDiego, CA, Academic Press, p. 169–191.

———. 1991. Introduction to the petrology of soils andchemical weathering. New York, Wiley, 313 p.

Nahon, D., and Merino, E. 1997. Pseudomorphic replace-ment in tropical weathering: evidence, geochemicalconsequences, and kinetic-rheological origin. Am. J.Sci. 297:393–417.

Olson, C. G.; Ruhe, R. V.; and Mausbach, M. J. 1980. Theterra rossa limestone contact phenomena in karst,southern Indiana. Soil Sci. Soc. Am. J. 44:1075–1079.

Ortoleva, P.; Chadam, J.; Merino, E.; and Sen, A. 1987.Geochemical self-organization. II. The reactive-infil-tration instability in water-rock systems. Am. J. Sci.287:1008–1040.

Pettijohn, F. J. 1957. Sedimentary rocks (2nd ed.). NewYork, Harper & Row, 718 p.

Plummer, L. N. 1975. Mixing of sea water with calciumcarbonate ground waters. Geol. Soc. Am. Mem. 42:219–236.

Prospero, J. M.; Ginoux, P.; Torres, O.; Nicholson, S. E.;and Gill, T. E. 2002. Environmental characterizationof global sources of atmospheric soil dust identifiedwith the nimbus 7 total ozone mapping spectrometer(TOMS) absorbing aerosol product. Rev. Geophys. 40:1002, doi:10.1029/2000RG00009.

Prospero, J. M.; Olmez, I.; and Ames, M. 2001. Al and Fein PM2.5 and PM 10 suspended particles in south-central Florida: impact of long-range transport of Af-rican mineral dust. Water Air Soil Pollut. 125:291–317.

Pruett, F. D. 1959. A study of the magnetic properties ofsome residual soils. MA thesis, Indiana University,Bloomington, 58 p.

Putnis, A. 2002. Mineral replacement reactions: frommacroscopic observations to microscopic mecha-nisms. Miner. Mag. 66:689–708.

Putnis, C. V., and Mezger, K. 2004. A mechanism of min-eral replacement: isotope tracing in the model systemKCl-KBr-H2O. Geochim. Cosmochim. Acta 68:2839–2848.

Rendon-Angeles, J. C.; Pech-Canul, M. I.; Lopez-Cuevas,J.; Matamoros-Veloza, Z.; and Yanagisawa, K. 2006.Differences on the conversion of celestite in solutionsbearing monovalent ions under hydrothermal condi-tions. J. Solid State Chem. 179:3645–3652.

Ruhe, R. V. 1975. Geomorphology. Boston, Houghton-Mifflin, 246 p.

Ruhe, R. V.; Cady, J. G.; and Gomez, R. S. 1961. Paleosolsof Bermuda. Geol. Soc. Am. Bull. 72:1121–1142.

Szymczak, P., and Ladd, A. J. C. 2006. A network modelof channel competition in fracture dissolution. Geo-phys. Res. Lett. 33:L05401.

Thornbury, W. D. 1954. Principles of geomorphology.New York, Wiley, 618 p.

Twidale, C. R. 2004. Karst towers: some questions andanswers. Stud. Geol. Salmanticensia 40:55–83.

Walker, T. R. 1962. Reversible nature of chert-carbonatereplacement in sedimentary rocks. Geol. Soc. Am.Bull. 73:237–242.

Wang, Y., and Merino, E. 1995. Origin of fibrosity andbanding in agates from flood basalts. Am. J. Sci. 295:49–77.

Weyl, P. K. 1959. Pressure solution and the force of crys-tallization: a phenomenological theory. J. Geophys.Res. 64:2001–2025.

White, W. B. 1988. Geomorphology and hydrology ofkarst terrains. New York, Oxford University Press.

Williams, H.; Turner, F. J.; and Gilbert, C. M. 1954. Pe-trography. San Francisco, W. H. Freeman, 406 p.

Yaalon, D. H. 1997. Soils in the Mediterranean region:what makes them different? Catena 28:157–169.

Yaalon, D. H., and Ganor, E. 1973. The influence of duston soils during the Quaternary. Soil Sci. 116:146–155.

Terra Rossa Genesis, Implications for Karst, and Eolian...

Documents

Transcript of Terra Rossa Genesis, Implications for Karst, and Eolian...