Permian freshwater stromatolites associated with the conifer shoots Cassinisia orobica Kerp et al. — a very peculiar type of fossilization

REVIEW OF

PALAEOBOTANY AND

PALYNOLOGY E L S E V I E R Review of Palaeobotany and Palynology 91 (1996) 85-105

Permian freshwater stromatolites associated with the conifer shoots Cassinisia orobica Kerp et al. - - a very peculiar type

of fossilization

Pierre Freytet a,b, Hans Kerp c, Jean Broutin b a 41 rue des Vaux Mourants, F-91370 Verri~res le Buisson, France

b Laboratoire de Pal~obotanique et Pal~o~cologie, Universitd P. et M. Curie, 12 rue Cuvier, F-75005 Paris, France c Abteilung Palaobotanik am Geologisches-Palaontologisches Institut und Museum, Westfalische Wilhelms-Universitat,

Hindenburgplatz 57-59, D-48143 Mflnster, Germany

Received 27 October 1994; revised and accepted 18 May 1995

Abstract

Fofiated shoots of Cassinisia orobica Kerp et al., 1996 are fossilized by a combination of two processes, i.e. by tufaceous/stromatolitic encrustation, preserving their external morphology, and the formation of sparite crystal which include leaf tissues and vascular bundles. Encrustation took place in two phases, interrupted by a period of emersion. The arrangement of the stromatolitic laminations and the aspect of the crystals and their inclusions allow a comparison with some specific recent forms of Schizothrix. On the lake bottom, cavities in the axes and leaves were finally filled with internal sediment, gypsum and diagenetic sparite.

R~sum~

Les rameaux feuill6s de Cassinisia orobica Kerp et al., 1996 sont fossilis6s grace ~ un encro~tement tufac6/ stromatolitique (morphologie externe) et des cristaux de sparite (tissus et feuilles). L'encrofitement s'est effectu6 en deux 6pisodes s6par~s par une ~mersion. Les caract~res des laminations stromatolitiques (cristaux et inclusions) permettent de les rapprocher de certains Schizothrix actuels. La fossilisation s'est termin6e sur le fond d'un lac, avec envahissement d'une partie des axes et des feuilles par du s6diment encaissant, des cristaux de gypse et de sparite diag6n6tiques.

1. Introduction

This paper is concerned with the fossilization mode of three-dimensionally preserved conifer shoots f rom the Permian of the Orobic Alps, Southern Alps (Italy) which have been described as Cassinisia orobica by Kerp et al. (1996-this issue). For detailed information on the locality and source strata of the specimens and the system-

0034-6667/96/$15.00 © 1996 Elsevier Science B.V. All fights reserved SSDI 0034-6667 (95) 00078-X

atics of these conifers the reader is referred to that paper where additional specimens and thin sections are figured. All the thin sections are curated in the collection of the "Abteilung Pal~iobotanik", Westf'alische Wilhelms-Universit~tt MOnster, Germany.

The preservation of Cassinisia orobica shoots is the result of two interrelated processes which are often associated in fossil and modern tufas and

P. Freytet et al./Review of Palaeobotany and Palynology 91 (1996) 85-105 87

travertines. One is the inclusion of cellular structures in large sparite crystals known as "calcified wood". The other is the encrustation by a more or less strongly calcified algal film. The crystals in this algal film are either primary or diagenetically formed. In the latter case, they are recrystallization products of other forms of calcite (micrite and needle-like or fibrous calcite), or other related minerals (aragonite and calcium oxalates). The calcification is characterized by an alternation of light and dark layers, typical for stromatolites. Tufas and travertines are here considered as particular forms of stromatolites.

Laminations in recent tufas and travertines show two different appearances. The first is a more or less voluminous mass with planar, undulatory or columnar structures. This type is known, for instance, as stromatolite tufa (Buccino et al., 1978) and phytoherm boundstone (Pedley, 1990). The second type appears as encrustations formed around recognizable plant remains like mosses and vascular plants. This type is known as phytoherm tufa (Buccino et al., 1978) or phytoherm framestone (Pedley, 1990). These, and many previous authors, strongly emphasized the external shape of the carbonate build-ups and their macroscopically recognizable content. However, they did not pay much attention to the primary origin of these build-ups, i.e. the formation of an algal film. This algal film has also been described as biofilm, algal mat, cyanobacterial mat, microbial mat or periphyton.

Algal films are long known to science. As early as 1777 MQller demonstrated that algal films acted as sediment traps in littoral environments in the northern Baltic Sea (see Gerdes and Krumbein, 1987). Unfortunately, this was overlooked by later

authors until it was rediscovered by biologists (Great Britain: Carter, 1932; Bahamas: Black, 1933). The presence of algal films in saline lakes has been reviewed by Bauld (1981). Forel discov- ered algal films in a freshwater environment, on the bottom of Lake Geneva, in 1869, but he described the phenomenon only thirty years later in 1901 (p. 234). The actual role of algal films in the formation of recent tufas first was demonstrated by Meunier (1899) in the Swiss Alps. The role of algal films in the formation of lacustrine marls, particularly in the marshes in Florida, also has been recognized (Dachnowski-Stockes, in Gleason and Spackman, 1974). Many other authors have observed algae in decalcified tufas (first by Cohn, 1864) or oncolites (Roddy, 1915), however, without considering them of any special importance.

Physico-chemical crystallizations are emphasized in recent studies of freshwater carbonate deposits. Some authors noticed the systematic rela- tionship between organisms and crystallizations and they concluded that tufas and travertines are in fact stromatolites (Golubic, 1976; Pentecost, 1978, 1985; Casanova, 1981; Chafetz and Folk, 1983; Freytet and Plet, 1991; Pedley, 1990).

The encrusted foliated branches of Cassinisia orobica are of special interest because the first encrustation consists of "phytoherm tufa". This encrustation preserved the original shape of the twigs. The original outlines were later concealed by a second crust with small isolated and scattered bioherms having a mamillate or cauliflower-shaped surface. These structures are identical to the ones that are often found in recent fluviatile and lacustrine environments. They are somewhat rarer in the Mesozoic and Cenozoic (e.g. Monty and Mas,

PLATE I

General overview at low magnification. 1. Thin section PMti Cp 1245/PONT 5. Slightly enlarged general overview showing the leaves and crusts in longitudinal section.

See the explanatory drawing of Fig. 1. Scale bar = 1 cm. 2. Same thin section, detail of the laminations of sets II and III. A =zone with laminations forming adjacent columns; B=zone

with laminations showing only large undulations. Scale bar = 5 ram. 3. Thin section PMQ Cp 1245fPONT 6. Detail of the laminations. A =laminations with large undulations; B=adjacent columns

(in sets II and III). Scale b a r = 2 mm. 4. Thin section PMI1 Cp 1246/PONT 2. Detail of the laminations. A = a nice series of adjacent colunms. Scale b a r = 2 ram.

88 P. Freytet et al./Review r~f Palaeobotany and Palynology 91 (1996) 85-105

1981; Freytet and Plaziat, 1965, 1982) and much rarer in the Permian (Stapf, 1973; Bertrand-Sarfati and Fabre, 1974; Sch~ifer and Stapf, 1978; Szulc and Cwizewicz, 1989; Freytet et al., 1992).

2. Descriptive part

2.1. Description of the stromatolite encrustation around the foliated branches

All thin sections through foliated branches and leaves (Plate I; Fig. 1 ) show that they are covered by laminations which can be subdivided into three basal types or sets, here referred to with the roman numbers I, II and III. The first set (I) consists of laminations of various fabrics which form a thin cover around the substrate. The second set (II) primarily consists of sparitic laminations envelop- ing the leaves and almost completely filling the interfoliar spaces. The third set (III), which consists of sparitic laminations identical to those of set II, attenuates the irregularities of the underlying structures. The final cauliflower-like shape entirely masks the original foliated branchlet.

2.2. The laminations of set I

Set I is a very regular layer covering the leaves and axes. It always shows the same characteristics and lateral-vertical variations (Plate II; Fig. 2).

"Elementary" laminations "Elementary" laminations are the thinnest rec-

ognizable homogeneous layers. They are of several types: light homogeneous micrite, dark homogeneous micrite, grumelous micrite (= pelmicrite, clotted micrite, peloidal micrite), clots in the microsparite (=pelsparite, clotted microsparite, peloidal microsparite) and palisadie radial sparite (only recognizable in polarized light). These elementary types show some variations: ( 1 ) inclusions form 3-4 p.m thick microlaminations within the palisadic crystals (Plate II, 2); (2) spots of micrite or accumulations of peloids occur in the palisadic sparite: (3) small crystals of elongated sparite are found in the micritic laminations.

Laterally, the laminations are more or less con-

Fig. 1. Interpretation of thin section PMii Cp 1245/PONT 6 (Kerp et al., 1996, plate Iv, 6). Scale b a r = l cm. 1 =micrit ic and sparitic laminations of the first crust (set I); 2=spar i t ic laminations of the succeeding crust (sets II and III); 3 = internal sediment filling the voids in the interior made of micrite with some muscovite crystals, and in residual voids between the stromatolitic crusts coating the leaves; 4 = former gypsum crystals being replaced by sparite; 5 =plants remains, xylem, leaves; 6 = late sparite filling all the residual voids; with a variable number of dark organic matter inclusions; 7 = stylolitic contact; 8 =lower boundary of the crust cast of the original substrate (plant), thinly stippled areas: more complex zones (tangential sections of the leaves and crusts).

tinuous but they rarely enwrap a leaf completely. The upper and lower sides of the laminae can be parallel, planar or more or less undulated. Sometimes their base is flat or undulatory while the upper side is very irregular with pronounced protuberances, e.g. club-shaped structures (see Plate II, 1, 3, 4; Fig. 2, 1-9).


Fig. 2. The morphological variability of the laminations of the first encrustation (set I) and the internal stromatolite. All drawings are made from thin sections: (1, 2) thin section PMQ Cp 1246/PONT 2; (3-5, 10) thin section PMii Cp 1249/PONT 7; (7) thin section PMll 1246/PONT 9; (6, 8, 9) thin section PM~i Cp 1248/PONT 10. All scale bars=5 mm. All sections are oriented in the growth direction of the crusts: the laminations of the upper surfaces are covered with dome-like structures, columns, baculae and tufts, while those of the lower surface are regular or only undulatory. Legend: 1 = micritic laminations; 2 = radial palisadic sparite laminations; 3 = clearly peloidal laminations in microsparitic cement; 4 = laminations of set II; 5 = sparite with numerous inclusions of organic matter; 6=sparite that later filled void, foliar cavities and spaces between the columns and the baculae of set I; 7= distinct remains of plant tissues, a = sparite with inclusions corresponding with the lower part of the filling of the leaf after reworking of the stromatolite build-up (cf. Fig. 5); b =mechanical interruption of the first crust (set I). In a single case this is followed by the formation of an "internal stromatolite" (=c). Line d indicates the position of the section figured on Plate III, 7.


The vertical succession of the laminations On the basis of the more than 15 specimens

studied in detail, some general statements can be made. The stromatolites on the upper leaf surfaces grew in an upward direction; they show a very uniform succession of laminations and a very characteristic morphology. The stromatolites on the lower leaf surfaces grew downwards; they show a less regular succession than those on the upper leaf surfaces. The number of laminations can vary, in addition to their thickness and composition. The connection between the laminations of the lower and upper leaf surfaces is very abrupt and the transitional zones are very narrow.

The identical development of these two main types of layers on the upper and lower leaf surfaces, found on different leaves of a single branchlet, as well as on leaves preserved in all the studied

samples, is very remarkable. This regularity has never been seen in modem tufa and it does not seem to have been reported in the literature, at least as far as we know.

The following succession on the upper leaf surfaces can be schematically outlined from the inside to outside (Plate II, 1, 3): first the development of the base of set I; then a micritic lamination with a fiat bottom and a very irregular upper surface; culminating in palisadic radial sparite and then the base of set II. The base of the micritic lamination (Fig. 1, 1) molds the surface of the leaf cuticle; the upper surfaces are very irregular. This surface may be dissected, columnar, baculate, dome- or club-shaped. This micrite is more or less clotted, rarely peloidal; it can contain elongated sparite crystals (Plate II, 5, 6). The lamination with palisadic radial sparite (Plate II, 2C) often shows small

PLATE II

Detail of the laminations of the first crust (set I) 1. Thin section PMil Cp 1246/PONT 2. Series of laminations on a an upper leaf surface (set I). Scale bar = 500 lain. A = cavity

of the shrunken leaf filled with sparite; B = clotted micritic lamination, continuous in the lower part and in the undulatory upper part; C=palisadic radial sparite, with internal dark/light microlaminations; D = t h e same type of laminations of set II.

2. Detail of the central part of 1. Scale bar = 100 Inn. A = sparite fill of the foliar cavity; B = lower part of the micritic lamination showing its clotted microstructure; C=upper part of the micritic lamination showing the incipient development of radial palisadic sparite crystals with internal microlaminations; D = palisadic radial sparite lamination consisting of dark crystals rich in microlaminations, still containing micrite patches, and base of the laminations of set II.

3. Thin section PMtl Cp 1249fPONT 7. Scale bar=500 Inn. A =sparite fill the foliar cavity, set I; B=micritic lamination with the basal layer with tuft-like expansions; C=columnar palisadic radial sparite lamination, set II; D=base of the laminations; the columns are still adjacent.

4. Thin section PMtl Cp 1248/PONT 10. Lateral section through set I, intermediate between the typical development of the upper surface (set with two laminations and undulations, columns and tufts) and the lower surface (planar, set with 5-10 laminae). Scale bar=5000m. A=micritic grumelous lamination; B=radial palisadic sparitic lamination; C=micritic lamination; D = laminations of set II; E = void (fenestra) resulting from the growth irregularities of the set I laminations.

5, 6. Details of the contact zone between the laminations illustrated on 4 (indicated with A and B on 5). Scale ba r= 100 ~tm. 5. Normal light. A shows the juxtaposition of large clots (= peletoids or peloids), B shows the palisadic radial sparite which is

well visible in polarized light (cf. 6). Internal laminations are visible in normal as well as in polarized light. 6. Polarized light. 7. Thin section PMI1 Cp 1249/PONT 7. Scale bar = 500 Ixm. Complex encrustation on the lower leaf surface where the first crust

(set I) was cracked and where a crust of set II developed inside the foliar cavity and on the outer surface of set I (see also Fig. 2, 10). The arrows indicate the growth direction of the algal/stromatolitic film. A = "internal stromatolite" consisting of set II material; B---zone where the internal crust has come off, filled with sparite in an early stage; C=series of micritic laminations, homogeneous or clotted/peloidal and with a variable thickness; D =lamination of radial palisadic sparite with well visible internal microlaminations; E= uppermost lamination of light micrite.

8. Another example of the laminations of set I. Thin section Pont 10. Scale bar = 500 Ixrn. A = micritic laminations; B = lamination of radial palisadic sparite with internal microlaminations; C= light micrite lamination.

9. Detail of 8. Scale ba r= 100 lain. A =micritic lamination with coalescent clots and large irregular voids (fenestrae); B=micrite with coalescent clots and small irregular voids; C=lamination with isolated clots (peloids) in a microsparitic cement; D=dark micritic clotted lamination; E = lamination with radial palisadic sparite crystals, rich in dark inclusions and microlaminations; F=light micrite lamination; G=radia l palisadic lamination at the base of set II.


internal microlaminations. This lamination smooths the most extreme irregularities of the underlying lamina.

The main variations (Plate II, 3) are to be found in the total thickness of set I (1.8-3.2 mm) and in the shape of the micritic expansions on the basal layer. The palisadic radial sparite lamination can be fragmented and restricted to radial globular structures around small masses of micrite. This latter micrite barely touches the leaf by a small stalk. The spaces inbetween the globules are filled with sparite (Plate II, 4) and this picture illustrates the transitional zone between the upper and lower leaf surfaces. There is a characteristic layer of light micrite above the sparitic lamination. Fig. 2 illustrates the variability of these laminations.

On the lower surfaces of the leaves the encrustation develops downwards and the most complete succession is as follows (Plate II, 7-9); the base of the leaf is covered by:

(1) more or less homogeneous dark micrite, sometimes with large irregular voids,

(2) clotted micrite with more or less strongly developed voids,

(3) dark to very dark homogeneous micrite, (4) palisadic radial sparite; crystals with many

inclusions, (5) dark to very dark micrite, (6) clotted/peloidal microsparite, (7) dark homogeneous micrite,

(8) palisadic radial sparite with microlaminations,

(9) light homogeneous micrite, and finally the top of set I and base of set II which consists of palisadic radial sparite.

Lateral variations Laterally, the laminations are more regular and,

at most, they show undulations or fiat dome-like structures. They never form baculate or globose structures like the overlying laminations. They vary in thickness (2.4-3.3 mm), in the number and the shape of the voids, in the hypertrophy of some laminations (clotted/peloidal with microsparite), in the absence of one or more laminations, and particularly in the absence of the palisadic radial sparite layer (4). However, the palisadic radial sparite with microlaminations (8) and homogeneous light micrite (9) are always present; the homogeneous light micrite laterally passes into the lamination with palisadic radial sparite of the upper leaf surface (2). In summary, there is con- siderable variability in the laminations of set I on the upper and lower leaf surface (Fig. 2, 1-10).

2.3. The laminations of sets H and III

Laminations, marked by light and dark bands of variable thickness, are macroscopically visible in hand specimens and at low magnification in the

PLATE III

Detail of the laminations of set II and set III. All photographs are of thin section Pont 9, except for 3 which is thin section PMti Cp 1246/PONT 8. For the explanatory drawing of the photographs of thin section PMti Cp 1246/PONT 9 see Fig. 3, 3. 1, 2. Base of set II with the top of set I (A). Scale bar=500 lam. 1. Normal light. 2. Polarized light. 3. Dark microlaminations are well distinguishable (thicknesses of 3-4 ~tm) and the vertical threads correspond with bundles of

filament sheats enwraped by colonies of bacteria (see the modern equivalent on Plate IV, 5, 6). Normal light. Scale bar = 100 ~tm. 4. Another example. Polarized light. Scale bar = 500 ~tm. 5, 6. Another example. Notice the numerous spot-like inclusions. The form of the twin lamellae away the substrate resembles the

fascicular optic crystals of Kendall (1977). Scale bar = 500 ~n. 5. Normal light. 6. Polarized light. 7, 8. Another, more complex example. The laminations visible in normal light (7), i.e. variations in the intensity of the grey colour

and cleavages often differ from those that can be observed in polarized light (8). These discrepancies which are related to complex early diagenetic processes are often difficult to interpret. Scale bar = 500 ~tm.

94 P. Frevte! et al./Review (![Pulaeobotany and Palynologv 91 (1996) 85 105

a b

!

!

d

- 2 " I

e

f

Fig. 3. Laminations of sets II and 111. (l) Various types ol" sparitic laminations: a-with parallel boundaries: h=:with denticulate upper boundaries: c, d= interrupted: e-elongate short crystals: f = elongate long crystals: g-with lateral intercalary microsparite. (2) Composite lamination consisting of two single laminations forming a doublet: a isodiametric microsparite: b =lamination of elongate short crystals: c-layer with dark inclusions (in normal transmitted light): d-lateral intercalary microsparite: e-cleavages and twins: ! interrupted lamination; g= dark microlaminations. (3) An example of the lateral variability of the laminations of set II. Scale bar= 5ram. Lower part (stippled )= set I: middle part=adjacent columns: upper part=simple undulations: a h-section of which particular laminations, doublets and composite lalnin- ations, are figured in more detail in Fig. 4, 4. ( 4 ) Some doublets and composite laminations, some of them are figured on Plate II. Scale bar = 500 gm; a, b and c are on the same scale and d, e and ./are on the same scale. (a.h) Drawings of the central part illustrated on Plate Ill, 1.2. (a=normal light; h - polarized light) base=lamination of palisadic radial sparite; middle part-dark microsparite: upper part -= palisadic radial sparite. In normal light the inclusions and cleavages can be seen while the shape of the crystals is to be seen by using polarized light. (c). Drawing of a part of the thin section illustrated on Plate IlL 4: microsparitic laminations, palisadic

thin sections. A t higher magnif ica t ions they always look more complex and polar ized light reveals their crystall init ies.

The various O,pes o f laminations In normal light, very thin da rk mic ro lamina t ions

often occur within the large spari te crystals. They are 2 4 gm thick and very regular (Pla te III , 3, 4). In addi t ion , in no rma l light o ther l amina t ions can be seen that have very var iable thicknesses, ranging from 20 lam to a few mm. They are dis t inguished by their colour , which varies f rom dark - to light- grey to a lmos t white, by the appa ren t crystal l ini ty (microspar i te to large e longated spari te crystals) and by the closeness of da rk inclusions (Pla te III, 1, 5, 7). However , more in format ion is avai lable using polar ized light (Pla te III, 2, 4, 6, 8). The upper and lower boundar i e s o f the microspar i t i c laminae may be paral le l (Fig . 3, la), with s t rongly dent icula ted upper limits (Fig . 3, lb) . These laminae can be d i scont inuous (Fig . 3, lc ,d) and included in larger crystals . The spari t ic l amina t ions may consist of shor t e longated crystals, being 2 5 times longer than wide; they are or iented perpen- d icular or obl ique to the surface o f the subs t ra te (Fig. 3, le). They may also consist of e longated crystals, which are 10 35 t imes longer than wide, that are always pos i t ioned pe rpend icu la r to the subs t ra te (Fig . 3, l f ) . These long crystals show many cleavages, twins and inclusions o f da rk micro lamina t ions . They radia te because the surface of the subs t ra te is uneven or curved (Fig . 3, 1, 2, 4), and s imulate a black cross in polar ized light. We call this section palisadic radial sparite. We reject the term radiaxia l f ibrous calcite in t roduced by Bathurs t (1971) because the crystals are not fibrous. Small microspar i t e crystals can also be

radial with inclusions of microsparite (lamination discontinuous). laminations with elongate short crystals, palisadic radial crystals. (d,e) Drawings of the specimen figured on Plate I11, 5. 6 (d= normal light; e=polarized light). Spots with inclusions. microlaminations and sparite crystals forming lateral light traces can be seen in normal light. In polarized light the cleavage pattern can be seen. (f) Drawing of another example showing a lamination of microsparite, elongate short crystals and elongate long crystals with interrupted lamination. Drawing in normal light showing the numerous cleavages; very light, clear sparite without inclusions or microlaminations.

P. Freytet et al./Review of Palaeobotany and Palynology 91 (1996,) 85-105 95

I l ; . . . . . . .

~~~~'~-"~ ~ .. ......

Fig. 4. Laminations of sets II and III. (1) Drawing of thin section PMO Cp 1245/PONT 5. Scale ba r= 1 cm. Stippled area: laminations of set I; dashed and thicker lines: three successive macroscopically visible laminations. The lamination indicated by the dashed line left belongs to set II (encrusting of the leaves); in the right part it is the basis of set III (overall encrustation of the build-up). The laminations indicated by thicker lines belong to set III. (2-5) Theoretically possible lateral variations of laminations. (2) Starting from a dome at the top of set I, domes (a), hemispheres (b), columns (c) and clubs (d) can be formed. All elements are standing free. (3) Undulations forming adjacent columns with continuity of the laminations from one to another. (4) Columns on a planar substrate. (5) Parallel undulations on an undulatory substrate. (6-9) Drawings showing longitudinal variations on the upper and lower surfaces of the leaves. All drawings are made from thin sections. Scale bar= 1 cm. Legend: a=fol iar cavity with sparite with a variable amount of inclusions, gypsum and internal sediment, b =laminations of set I, just covering the leaves and branches; c = laminations of set II (coating of leaves and branches, incomplete filling of the interfoliar voids); d= laminations of set III (overall coating of the build-up). The dashes indicate the border between sets II and II1. The upper surfaces show many adjacent columns while the lower surface mainly shows coatings with simple undulations. (10) Scheme

present between two long elongated crystals (Fig. 3, lg).

The vertical succession of the laminations A succession of two different lamination types

forms a doublet. The most frequent doublets are those of a microsparitic (la,b) and a sparitic lamination (1d-f). However, repeating patterns can be more complex and may also consist of more than two laminations. The theoretically most complete succession is shown on Fig. 3, 2, which is, however, difficult to interpret. Several examples observed in our specimens are figured (Fig. 3, 4a-f; Plate III, 7, 8).

Relationships between the shape of the laminations and the substrate

Laminations are never completely rectilinear and always show more or less well-developed undulations. They can be continuous or discontinuous and form small, variably shaped build-ups. There is a direct correlation between the irregularities of the substrate and the more or less undulatory shape of the laminations, independently of the growth direction, either upwards, laterally or downwards (Fig. 4, 6-10).

Starting from an irregularity of the substrate (on Fig. 4, 2 at the top of set I) different types of structures may develop. These are dome-shaped (2a), hemispherical (2b), columnar (2c) or club- shaped (2d). They are distinct and separated from each other by voids (= fenestrae). However, within all these different structures the succession of the laminae is identical. On a slightly undulatory substrate, laminations may be continuous with adjacent columns (Fig. 4, 3). Juxtaposed columns can also develop on a planar substrate (Fig. 4, 4). A bent substrate leads to regular, almost concentric laminations as found near the leaf apices (Fig. 4, 5).

of the variations in shape and height of the coatings on the upper and lower leaf surfaces and the vertical part of the axis between two successive leaves. In reality, the remaining interfoliar voids are much smaller than on this scheme.

P. Freytet et al./Review of Palaeobotany and Palynology 91 (1996} 85-105 97

latory laminations of the lower leaf surface. The strongest growth is to be observed near the leaf apices. The interfoliar spaces are not always completely filled with stromatolitic material. We believe that these differences in growth are related to differences in light intensity.

Simultaneous horizontal and vertical variations Successions of laminae with isolated columns

can be covered by a set of continuous laminae. The same undulatory and continuous lamination can cross two into adjacent columns (Fig. 3, 3).

Growth anomalies: an example o f an "internal stromatolite'"

The specimen on Fig. 2, 10 shows the single example we have in approximately 15 sections that illustrates the development of set II laminations in the interior of a foliar cavity. For unknown reasons coating I was fissured after its early induration. However, it was still rigid enough to become encrusted with laminations of set II on its exterior side. A penetration into the foliar cavity resulted in the development of upwards growing laminations from the lower side of the leaf mould and downwards growing laminations on the lower surface of the remnant of the shrunken leaf (Fig. 2, 10c). This is a typical example of an "internal stromatolite", a feature that is known from some modern examples (Monty, 1976; Monty and Hardie, 1976). They are also known from recent insect larval tubes occurring in fluviatile tufas and travertines in France and Morocco (Freytet et al., in prep.).

2.4. Voids and alien bodies inside the stromatolite build-ups

The studied specimens do not consist entirely of plant materials and stromatolitic coatings. They contain a number of voids which can be of a very different origin: (1) voids that remained between the clots during compaction (packing) (Plate II, 5, 6, 9) were filled with microsparite; (2) voids between coalescent columns and clubs

may have been filled with sparite (Plate II, 4E) or peloids (Plate IV, 2); (3) residual voids may occur between the crusts coating the leaves (Fig. 1); (4) voids are found between the moulds represent- ing the original outlines of the leaves (base of set I) and the remnants of retracted leaf tissue; or (5) voids that were the result of a more or less strong putrefaction of leaves and branch tissues.

These two latter types of voids are the most interesting.

These voids can be filled with "internal sediment" that penetrated into the branch and leaf moulds. This internal sediment is identical to the type of sediment enclosing the stromatolites and it consists of micrite, rarely with muscovite crystals. The voids also contain pseudomorphs of lenticular crystals or sticks, interpreted as gypsum replaced by sparite mosaic crystals (Plate IV, 3, 4). These crystals are found in the internal micritic sediment and in the foliar cavities which exhibit a very complex filling. These voids may also be filled with sparite crystals that are always xenomorphous with many dark organic matter inclusions. The crystals contain remnants of plant tissue with well distinguishable cells, but more often they only show poorly preserved inclusions. Nevertheless, in particular cases, inclusions are very abundant in specific parts of the foliar cavity (Fig. 2, 1). At the moment of sparite crystallization, the organic detritus was, because of gravity, concentrated in the lower part of the foliar cavities (Fig. 5, type 6). The upper limit of this accumulated organic detritus can thus reflect the horizontal line at that time.

Spherical bodies with a diameter of 100-150 ~trn are very rarely found between two columns (Plate IV, 1). These have been interpreted as green algae (Chlorellopsis) or as insect eggs (cf. Freytet and Plaziat, 1982).

The relations between the authigenic-formed, biologic and detritic alien bodies, the various types of microsparite and sparite, and the three phases of encrustation by stromatolites (sets I, II and III) permit the reconstruction of the history of the encrustation of the shoots and their very complex diagenetic evolution (Fig. 5).

98 P. Freytet et al./Review ~['Palaeobotany and Palvnology 91 (1996) 85-105

Fig. 5. Diagenetic evolution of stromatolitic encrustations alter reworking on the lake bottom. The axis of the shoot is in horizontal position, while it was in vertical position during stromatolitic coating (cf. Fig. 6). Legend: 1 =bot tom sediments in which the stromatolitic build-up are partly buried: 2 = internal sediment, consisting of micrite and muscovite crystals, filling the basal parts of the mould that was formed after decay of the plant material: 3 =gypsum crystals; 4=p lan t debris, parts of vascular bundles and remnants of leaf tissues (partly shrunken and mummified); 5=sparite, light or with a few inclusions. 6=spari te with many dark inclusions in the basal parts (cf. Fig. 2, 1 ): 7 = successive stromatolitic crusts (sets 1, li and Ill ),

3. Comparisons with modern analogues and discussion

3.1. Introduction

A lamination has been defined as the thinnest homogeneous layer of sediment that can be recognized on the basis of its colour and crystallinity (Monty, 1976). The laminations encountered in the Permian build-ups are more varied and complex than those of "classical" stromatolites (Kalkowski, 1908; Hofmann, 1969; Monty, 1976) but they resemble those of particular Oligocene lacustrine stromatolites that have been described by Bertrand-Sarfati et al. (1966) and Freytet and Plaziat (1982).

The difficulty is neither in the recognition of the laminations nor to establish their succession, but to interpret them in terms of their biology, systematics and palaeoecology. We will not discuss the laminations of intertidal stromatolites. Their interpretation is still controversial, because they may represent either annual, seasonal, daily or tidal accumulations. We restrict our discussion to laminations occurring in continental environments (tufas). Even for these, a number of different,

often contradictory, interpretations have been given. ( 1 ) The formation of a series of successive laminations has been attributed to variations in biological activity during the day and the night under favourable seasonal conditions (Meunier, 1899), or during the whole year (Doemel and Brock, 1974). (2) Other authors have argued that the formation of a series of successive laminations is caused by seasonal changes in the morphology of a predominant species, e.g. Phormidium incrusta- turn (Gomont, 1892, p. 170; Geurts, 1976; Monty, 1976, fig. 8). (3) The formation of a series of successive laminations has also been assigned to changes in the composition of the algal population, e.g. Phormidium incrustatum replacing Schizothrix sp. (Monty, 1972; 1976, fig. 9; Sch~ifer and Stapf, 1978, fig. 22). (4) The formation of crystalline laminations has also been attributed to variations in the activity of a single species independently of seasonal changes (Rivularia haematites: Pentecost, 1978; 1987, fig. 7; 1991, fig. 3). (5) Another interpretation is that the laminations are related to reduced and amplified activity of one of the constituents of a plurispecific biocoenosis consisting of bacteria, cyanophyceae, fungi and eucaryotic algae; this variation is independent of seasonal


changes. Within a build-up of Scytonema, Thunmark (1926) described layers of large sparitic crystals that were attributed to Desmidiaceae and coccoidal cyanophyceae. Winsborough and Golubic (1987) suggested that the proliferation of diatoms (Gomphonema), which are associated with Scytonema and Schizothrix, would result in the formation of sparite. Freytet and Plet (1991) described the association of Phormidium incrusta- turn, Gongrosira incrustans and various species of Schizothrix with sparitic/micritic laminations.

The algal film that covers emergent substrates always consists of several species as has been demonstrated for various types of environments such as springs, warm-water lakes, cold-water lakes and streams (see e.g. Penhallow, 1896; Schmidle, 1910; Roddy, 1915; Thunmark, 1926; Serpette, 1947; Fritsch, 1949; Blum, 1957; Wharton et al., 1982; Pentecost, 1985; Sabater, 1989).

Summarizing the available data and the com- plexity of phenomena in recent environments, it can be stated that some ten types of calcite crystallizations are known. They are associated with bacteria, cyanophyceae, fungi and eucaryotic algae. Some crystalline shapes are characteristic of some taxa; conversely, different taxa can produce the same form of calcite crystals. These organisms constitute biocoenoses in which the various species do not develop simultaneously. The optimal growth of each taxon occurs during a particular period of the season. Early diagenesis may weakly or strongly modify the primary crystallizations, e.g. the transformation of fibres or micrite into palisadic radial sparite (see

3.2. The individual laminations and their significance

Micritic, homogeneous, clotted and peloidal lamination

Recent micritic deposits often show small bodies with a diameter of 1 lam which are attributed to bacteria (e.g. Pedley, 1992). Fossil examples consist of masses with a completely clotted texture, e.g. the thrombolites of Aitken (1967). These clots or peloids are generally attributed to bacterial or coccoidal cyanobacterial activity such as that found in Entophysalis in Shark Bay (Australia)

and the Persian Gulf. Bradley (1963) described masses of microscopically visible bacteria, incorpo- rated in sparite crystals, from recent lacustrine sediments. Chafetz and Folk (1983) reported the same phenomenon from Pleistocene travertines from Tivoli (Italy) and Oklahoma, while Freytet et al. (1993) described the same from travertines with reeds and mosses from the Pleistocene of Egypt. Therefore, we have good reason to regard the micritic laminations of set I to be of bacterial origin. The differences in colour and the compact- ness of the peloids (Plate II, 9) may represent local meteorological variations within a number of days (cf. Freytet and Verrecchia, 1993).

Dark microlaminations Chafetz and Meredith (1983) and Chafetz and

Folk (1983) described sparitic tufts that contain very small (100-300 ~tm) thick laminations from Pleistocene travertines from Tivoli and Oklahoma. These were interpreted as being a result of day/night bacterial activity. In modem build-ups of Schizothrixfasciculata and S. penicillata we also have observed very thin laminations at intervals of 10-50 ~tm (Plate IV, 5, 6; Freytet, in prep.). We interpret these as the result of discontinuous bacterial activity. In Pleistocene and modern examples we have seen almost all intermediate cases between the columnar-type rich in inclusions (Plate IV, 5) and the microlaminations with elongated crystals. These features are similar to those of our Permian examples with dark microlaminations (Plate II, 2; Plate III, 3, 5) at intervals ranging between 5 and 20 ~tm.

We interpret the Permian microlaminations as the result of daily bacterial activity, reflecting a period of at least several weeks with intermittent preservation of the bacterial bodies. The question which still cannot be resolved, neither by us nor by Chafetz and Folk, is how a new lamination is formed exactly. This is particularly curious with respect to the fact that the dark laminations are included within the large crystals.

Sparitic laminations Certain species of cyanophyceae and eucaryotic

algae are surrounded by primary sparite crystals (Gongrosira, Vaucheria, Dichothrix). Large crystals

100 P. Freytet et aL / Review ~?]'Palaeobotany and Palynolog.v 91 (1996) 85 105

are formed all around filaments (Rivularia, Zygnema) or diatom stalks (Cymbella) (Wallner, 1934; Golubic, 1976; Casanova, 1981; Pentecost, 1991). However, none of these primary crystals resemble the palisadic radial sparite in the Permian samples. This particular form of calcite is most similar to what has been described in the literature as radiaxial fibrous mosaic crystals (Bathurst, 1959, p. 511, see Bathurst, 1971; Kendall and Tucker, 1973), or ray-crystals and composite crystals (see Mazzullo, 1980, with an excellent bibliog- raphy on this topic). They slightly differ from the fascicular optic crystals of Kendall (1977) because the twin lamellae "are convex away from substrate" in the radiaxial fibrous and concave in the fascicular optic type. These crystalline forms are all interpreted as having been formed by recrystallization of calcite or aragonite fibres in a littoral or deeper marine environment, often after (deep) burial.

In freshwater environments, from fluvial pisolites and laminated crystalline crusts of Great Britain, Braithwaite (1979) described and figured comparable crystals as coarse prismatic, coarse blade-shaped and coarse ragged crystals (Braithwaite, 1979, figs. 3, 4D,F). However, he interpreted these phenomena as recrystallizations of calcite fibres of purely physico-chemical origin, strongly resembling cave pearls and stalagmitic deposits, though "formed in streams on an open hillside" (p. 181). He did not try to dissolve specimens to isolate the organic matter which might have been present. Nevertheless, the flora of such concretions is well known (Penhallow, 1896: Roddy, 1915; Fritsch, 1949). Braithwaite's photos 48 and 4F show the same types of microlaminations and traces of filaments as in our Permian and Pleistocene specimens. Likewise, Irion and Miiller (1968, fig. 12) figured palisade-like crystals in a sinter crust on a moss tufa with the same microlaminations, and interpreted them as crystals having a purely physico-chemical origin.

On the other hand, other authors have noticed the presence of more or less well preserved filaments in sparitic zones. Sch~ifer and Stapf (1978, p. 99) stated that in oncolites from Lake Constance, the sparitization in laminations formed by a Phormidium-Calothrix/Diehothrix association

"may perhaps be initiated" by the micritic encrus- ration of the algal filaments. In Belgium, Monty and Mas (1981) figured laminations consisting of a radial sparite fabric which alternates with micritic laminations. These were attributed to the activity of Phormidium incrustatum and P. foveolarum. Love and Chafetz (1988) dissolved travertines from the Arbuckle Mountains, USA and isolated algal remains which "probably belong to the genus Phormidium". These remains are enclosed in the large sparite crystals that form the crystalline laminations alternating with micritic laminations. Recrystallization of this micrite would result in an accretional growth of neighbouring "coarse columnar crystals" (neomorphism, fig. 9). In specimens from Burgundy (France), Freytet and Plet (1991, plate Iv, 1, 2, 5) observed elongated vertical sparite crystals in micritic laminations formed by an association of Phormidium incrustatum, Schizothrix jasciculata, S. calcicola and Gongrosira incrustans, obtained after decalcification. Current investigations on material from France and Morocco have revealed that colonies of S. fasciculata, S. penicillata and S. pulvinata, which occur in almost pure populations (apart from bacteria), were first micritic but were soon overgrown by palisadic radial sparite with bacterial microlaminations (Plate IV, 5, 6).

Finally, it is not surprising that the equivalent of our palisadic radial sparite with microlaminations can be formed in freshwater environments (Braithwaite, 1979; Irion and Mtiller, 1968; Love and Chafetz, 1988) as well as in marine environments (Bathurst, 1971; Kendall and Tucker, 1973; Mazzulo, 1980). Kendall's (1977) fascicular optic calcite has been described from a marine environment; it also occurs in our freshwater samples, but we do not know whether it is primarily formed or whether it is the result of a later transformation of pre-existing micrite.

Both the radiaxial fibrous and fascicular optic sparite (Plate III, 6) occur in the thin sections we studied, however, the latter type is much rarer. Other terminologies were used by Irion and Mtiller (1968), Braithwaite (1979) and Mazzullo (1980). We retain here the descriptive term "palisadic radial sparite" and reject the term "radiaxial fibrous" because the crystals are not fibrous, but


large and individually visible. Unlike previous authors, we do not have a single argument to justify the formation of these palisadic crystals as having been derived from primary fibres. On the contrary, in modern and Pleistocene examples the progressive development of crystals can be observed in colonies of Schizothrixfasciculata and S. penicillata (Plate IV, 5, 6). Small clots of bacterial origin, distinguishable after decalcification, are randomly distributed in these build-ups or form dark laminations. Micrite crystals occur on filament sheaths and in bacterial colonies. Then the elongated crystals appear, first as small and isolated crystals, and later as larger and adjacent ones. These finally occupy the entire build-up and include remains of filaments and dark laminations. According to the literature (particularly Pentecost, 1991), the shape and the growth of the primary sparite crystals is determined by certain polysacha- rides found within the cyanobacterial sheaths, like in Rivularia haematites. It might be considered that the same is true for Schizothrix colonies.

Therefore, we believe that the palisadic radial sparitic laminations of the Permian stromatolites have an origin similar to that described above. This includes the recrystallization of micrite accumulations, largely of bacterial origin, in colonies of filamentous algae similar to some extant Schizothrix species. The laminations of sets II and III are almost completely recrystallised, but the recrystallization is less developed in laminations of set type I (Plate II, 2C, 5, 6). However, we cannot say that either S. penicilata or S. fasciculata already existed during the Permian. Therefore, we prefer to use the designation "diagenetic schizothrichoid laminations" when referring to our samples.

The other types of microsparitic and sparitic laminations with elongate short crystals in our thin sections have numerous equivalents in recent samples. These two lamination types can be interpreted as having resulted from casual growth forms related to microclimatic/meteorologic fluctuations. Continental stromatolites very easily reflect small variations in insolation, cloud cover, temperature, water level oscillations, desiccation, frost and salinity, whereas intertidal and shallow marine stromatolites have a monotonous fabric.

We explain the differences in thickness and shape

of the laminations of the upper and lower leaf surfaces simply as a reaction to the available light, which is the main growth factor. On the lower leaf surfaces and between the leaves, the growth conditions were optimal as on the upper leaf surfaces and on leaf apices. Similar differences in growth have been observed within Cretaceous-Eocene (Freytet and Plaziat, 1965, 1982), Oligocene (Leinfelder and Hartkopf-Frrder, 1990) and Pleistocene and Modern (Freytet, 1990) fluviatile stromatolitic encrustations.

5. Resemblances and differences with pre-Quaternary stromatolites

Fossil stromatolites are often more or less strongly recrystallized and dolomitized. Therefore, original microstructures are usually destroyed, although they are very well preserved when silici- fled (e.g. Freytet et al., 1992, fig. 7). Laminations described in the literature are very often diagenetic features. The systematics of fossil stromatolites is partly based on such features and other criteria such as the external shape and the ramifications (see e.g. Hofmann, 1969, fig. 2).

It is impossible to compare the shape of the Permian build-ups with those of build-ups formed on flat substrates. Binomial names have been given to the extremely diverse forms known from the Precambrian and Lower Palaeozoic. The younger ones, particularly those from the Permian, have very monotonous shapes. They have been described variously as heads, columns, nodules, oncolites etc. (Stapf, 1973, 1989; Bertrand-Sarfati and Fabre, 1974; Sch~ifer and Stapf, 1978; Szulc and Cwizewicz, 1989; Freytet et al., 1992).

On a smaller scale, in the order of a few laminations, we can recognize the classical Collenia/ Cryptozoon succession as it has been described and figured by Logan et al. (1964, fig. I). However, the classification of these authors is not suitable because the successions are very aleatory (e.g. Fig. 3, 3). We completely agree with previous authors that "a diversity of forms is produced by interaction of the algal film, detrital sediment, and physical environmental factors" (Logan et al., 1964, p. 69). Nevertheless, we would like add

102 P, Fro'let et al./Review ~/ Palaeobotany and Pul.vnology 91 (1996) 85-105

an important point. In marine and lake bottom stromatolites the actual growth direction is always upwards, while in our conifer encrustations some parts are covered with algal films that grew downwards. This accentuated the morphological anomalies (Figs. 2 and 4).

Non-marine Permian stromatolites often consist of dolomite or ankerite (Schfifer and Stapf, 1978: Freytet et al., 1992). Stromatolites with micritic laminations and fibrous palisadic laminations with filaments have been described by Bertrand-Sarfati and Fabre (1974). "Travertines" with thrombolitic laminations consisting of preserved filamentous bundles have been mentioned by Szulc and Cwizewicz (1989). A certain similarity exists between the crystallinity of Permian build-ups and those from younger deposits and, particularly,

those found in the more than one meter large lacustrine Oligocene stromatolites from Limagne d'Allier (Freytet and Plaziat, 1982, plate 11).

6. Concluding remarks on the formation of the " C a s s i n i s i a stromatolites"

On the basis of our detailed observations, interpretations and comparisons with modern examples we can reconstruct the encrustation and fossilization processes of the conifer shoots. Stands of Cassinisia orobica apparently flourished along a lake side or a stream. Branches detached and fell into a shallow part of the lake or shallower quiet part of a stream. The branching systems were partly buried in the muddy sediment. Regarding

_....L @

,ace

Q

i>:: ' . . y," .'~,. '.~,)., '~!e].

Fig. 6. Fossilization of parallel oriented foliated shoots of the same branching system. (1) Uniformly buried shoots, attached perpendicularly to the main branch. Only those parts in contact with the water are encrusted; the parts buried in the sediment decay rapidly, (2) Shoots attached acutely; the main axis is not in horizontal position; the base lies deeper than the apex. (3) Shoots attached obliquely; the main axis is not in horizontal position; the apex lies deeper than the basis. (4) Detail of an encrusted shoot. The crust (a) reaches until the sediment surl;ace (e). In the encrusted part there remains a void at the position of the axis and the leaves (b) and an empty space (c) at the transition to the encrusted parts (d). The sediment later penetrates through this opening into the axial void and the foliar voids after reworking on the lake bottom (cf, Fig. 5).


the inclination of the axis three possible modes of deposition are possible (Fig. 6, 2, 3). The parts covered by sediment started to disintegrate slowly (cf. Gastaldo, 1992) while a first crust was formed on the parts in contact with the water. These first encrustation seem to be mainly bacterial rather than schizothrichoid. This could be related to the turbidity of the water. Initial encrustation took a rather short period of time, probably of a few weeks or a few months because only a thin crust was formed. Then the encrusted shoots became emersed. This required a change of the water level, probably of only a few decimeters. The leaves and tissues of the branches desiccated and shrunk inside the encasing formed by the first crust (set I). This coating itself well resisted the desiccation, although small cracks developed locally. Subsequent submersion led to the formation of the laminations of sets II and III and to the local development of internal stromatolites. During this second period of submersion the conditions were apparently more stable than during the first because the new encrustations are more homogeneous and their development is only influenced by small scale variations in light intensity (Fig. 4, 1, 6-10). The duration of this submersion must have been long enough to develop a thick crust, but not so long that the vascular bundles and certain leaf tissues would decompose completely. The duration of this second submersion may be estimated in the order of a few years. The encrusted parts of the branches were then separated from those parts buried and probably largely disintegrated in the sediment. A strong current displaced the stromatolites which were originally "standing" in a vertical position. They were reworked into a horizontal position and partly covered by muddy sediment (Fig. 5). It is important to note that the algal film did not develop on those parts of the axes situated at or just below the sediment surface (Fig. 6, 4); this means that the stromatolites were open at their bases. After reworking, the muddy sediment easily penetrated into the hollow axes and leaves.

Stromatolite formation took place in an alkaline medium, as is indicated by the precipitation of calcite, and under slightly reducing conditions as is evidenced by both the preservation of organic matter and the precipitation of gypsum (Krumbein

and Garrels, 1952, fig. 5). The position of the branches on the lake bottom can only be deduced from the inclusions of dark organic matter in particular parts of the leaves (Fig. 2, la; Fig. 5, 1). The gypsum finally dissolved and it was replaced by sparite, probably when the stylolitic structures developed (Fig. 1, 7).

Tufaceous and travertine coatings around plant remains (phytoherm tufa) are in a very broad sense always stromatolitic. They are formed by the calcification of an algal film. This algal film is a complex biocoenosis including many species of bacteria, cyanophyceae, fungi, eucaryotic algae and animals. Encrustations are often the only way to preserve mosses or even shells (ostracodes, molluscs). In our case, the encrustation by stromatolites has led to the preservation of the external morphology of the shoots, whereas early diagenetic phenomena have even resulted in the preservation of some of their tissues in the sparite crystals.

Acknowledgements

We gratefully acknowledge Dr. F. Penati (Morbegno) and Prof. G. BrambiUa (Pavia) who have drawn our attention to the material described in this paper. They made their material available for further investigations and they kindly guided one of us to the outcrop. We thank Dr. J.-C. Plaziat (Paris) for preparing the drawings.

References

Aitken, J.D., 1967. Classification and environmental significance of cryptalgal limestones and dolomites, with illustr- ations from the Cambrian and Ordovician of southwestern Alberta. J. Sediment. Petrol., 37:1163-1178.

Bathurst, R.G.C., 1959. The cavernous structure of some Mississippian Stromatactis reefs in Lancashire, England. J. Geol., 67: 506-521.

Bathurst, R.G.C., 1971. Carbonate Sediments and Their Diagenesis. Elsevier, Amsterdam, Dev. Sedimentol., 12, 620 pp.

Bauld, J., 1981. Occurrence of benthic microbial mats in saline lakes. Hydrobiologia, 81:87-111.

Bertrand-Sarfati, J. and Fabre, J., 1974. Les stromatolites nodulaires de l'Autunien lacustre du bassin d'Abadla-Brchar

104 P. Freytet et al./Review o) Palaeobotany and Palynology 91 (1996) 85-105

(Sahara occidental algdrien/. Bull. Soc. Hist. Nat. Afr. Nord, 65:176 206.

Bertrand-Sarfati, J., Freytet, P. and Plaziat, J.-C., 1966. Les calcaires concr~tionn6s de la limite Oligocene-Mioc6ne des environs de St. Pour~ain-sur-Sioule (Limagne d'Allier): r61e des algues dans leur 6dification, analogie avec les stromatol- ithes et rapports avec la s6dimentation. Bull. Soc. G6ol. Fr. 7, 8: 652-662.

Black, M., 1933. The algal sediments of Andros Island, Bahamas. Philos. Trans. R. Soc. London B, 222: 165-192.

Blum, J.L., 1957. An ecological study of algae of the Saline River, Michigan. Hydrobiologia, 9: 361---408.

Bradley, W.H., 1963. Unmineralized fossil bacteria. Science, 141: 919-921.

Braithwaite, C.J.R., 1979. Crystal textures in recent fluvial pisolites and laminated crystalline crusts in Dyfed, South Wales. J. Sediment. Petrol., 49: 181-194.

Buccino, G., d'Argenio, B., Ferreri, V., Brancaccio, L., Ferreri, M., Panichii, C. and Stanzione, D., 1978. ! travertini della bassa valle del Tanagro (Campania), studio geomorpholog- ico, sedimentologico e geochemico. Boll. Soc. Geol. Ital., 97: 617-646,

Carter, N., 1932. A comparative study of algal flora of two salt marshes, part I. J. Ecol., 20: 34l 370.

Casanova, J., 1981. Morphologie et biolithogen~se des barrages de travertins. M6m. Ass. Fr. Karstol., 3:45 54.

Chafetz, H.S. and Folk, R.L., [983. Travertines: depositional morphology and the bacterially constructed constituents. J. Sediment. Petrol., 54: 289-316.

Chafetz, H.S. and Meredith, J.C., 1983. Recent travertine pisolites (pisoids) from southern Idaho, U.S.A. In: T.M. Peryt (Editor), Coated Grains. Springer, Berlin, pp. 450-455.

Cohn, F., 1864. Ober die Entstehung des Travertin in den Wasserf'allen von Tivoli. Neues Jahrb. Mineral. Geol. Palgontol., 40: 580-610.

Doemel, W.N. and Brock, T.D., 1974. Bacterial stromatotites: origin of laminations. Science, 184: 1083-1085.

Forel, P.A., 1901. Le L6man, monographie limnologique. Rouge, Lausanne, 3, 717 pp.

Freytet, P., 1990. Contribution ~, l'etude des tufs du Bassin de Paris: typologie des 6difices tuffacds (stromatolitiques) des chenaux fluviatiles. Bull. Cent. G~omorphol. Caen, 38: 9-27 [macroscopic features]; 35-53 [microscopic features].

Freytet, P. and Plaziat, J.-C., 1965. Importance des construc- tions algaires dues ~ des Cyanophyc6es dans les formations continentales du Cr~tac~ sup6rieur et de l'l~oc6ne du Languedoc. Bull. Soc. G~ol. Fr. 7, 7: 679-694.

Freytet, P. and Plaziat, J.-C., 1982. Continental carbonate sedimentation and pedogenesis-Late Cretaceous and Early Tertiary of southern France. Contrib. Sedimentol., 12:1-213.

Freytet, P. and Plet, A., 1991. Les formations stromatolitiques (tufs calcaires) r~centes de la r~gion de Tournus (Sa6ne et Loire). Geobios, 24: 123-139.

Freytet, P. and Verrecchia, E., 1993. Complex calcitic crystallizations in Nostoc parmelioides KtRz. (freshwater cyanobacterium): rhombs around trichomes inside Nostoc

colonies and epiphytic bacterial microstromatolites. Geomicrobiol, J., 11: 77-84.

Freytet, P., Baltzer, F., Conchon, O., Plaziat, J.-C. and Purser, B.H., 1993. Signification hydrologique des carbonates conti- nentaux quaternaires de la bordure du d6sert oriental 6gyptien (c6te de la Mer rouge). Coll. G. Lucas, "Carbonates intertropicaux". Bull. Soc. G6ol. Fr., 165(6): 593 601.

Freytet, P., Lebreton, M.-L. and Paquette, Y., 1992. The carbonates of the Permian lake of the North Massif Central, France. Carbonates Evaporites, 7:140 149.

Fritsch, F.E., 1949. The lime encrusted Phormidium communi- ties of British streams. Int. Ver. Theor. Angew. Limnol. Verh., 10: 141-144.

Gastaldo, R.A., 1992. Plant taphonomic character of the Late Carboniferous Hamilton Quarry, Kansas, USA: preserva- tional modes of walchian conifers and implied relationships for residency time in aquatic environments. In: J. Kovar- Eder (Editor), Palaeovegetational Development in Europe and Regions Relevant to Its Palaeofloristic Evolution. Proc. Pan-European Palaeobot. Conf., Vienna 19-23 September 199l, pp. 393 399.

Gerdes, G. and Krumbein, W.E,, 1987. Biolaminated Deposits. Springer, Berlin, 183 pp.

Geurts, A.M., 1976. Gen~se et stratigraphie des travertins de fond de vallee en Belgique. Acta Geogr. Lovan., 16: 1-66.

Gleason, P.I. and Spackman, W., 1974. Calcareous periphyton and water geochemistry in the Everglades. Miami Geol. Soc. Mere., 2:146 187.

Golubic, S., 1976. Organisms that build stromatolites. In: M.R. Walter (Editor), Stromatolites. Elsevier, Amsterdam, Dev. Sedimentol., 20, pp. 113-126.

Gomont, M., 1892. Monographie des Oscillari6es. Ann. Sci. Nat., 7, Bot. 15:263 368; 16:91 264 [reprinted 1962 in Historia Naturalis Classica, Cramer Verlag].

Hofmann, H.J., 1969. Attributes of stromatolites. Geol, Surv. Can. Pap., 69-39, 58 pp.

lrion, G. and M~iller, G., 1968. Mineralogy, petrology and chemical composition of some calcareous tufa from the Schw/ibische Alb, Germany. In: G. MOller and G.M. Friedman (Editors), Recent Developments in Carbonate Sedimentology in Central Europe. Springer, Berlin, pp. 157-171.

Kalkowsky, E., 1908. Oolith und Stromatolith im nord- deutschen Buntsandstein. Z. Dtsch. Geol. Ges., 60: 68-125.

Kendall, A.C., 1977. Fascicular optic calcite: a replacement of bundled acieular carbonate cement. J. Sediment. Petrol., 47: 1056- 1062.

Kendall, A.C. and Tucker, M.E., 1973. Radiaxial fibrous calcite: a replacement after acicular carbonate. Sedimentology, 20: 365-389.

Kerp, H., Penati, F., Brambilla, G., Clement-Westerhof, J.A. and Van Bergen, P.F., 1996. Aspects of Permian palaeobotany and palynology. XVI. Three-dimensionally preserved stromatolite-incrusted conifers from the Permian of the western Orobic Alps (northern Italy). Rev. Palaeobot. Palynol., 91: 63-84, this issue.

Krumbein, W.E. and Garrels, R.M., 1952. Origin and classifi-


cation of chemical sediments in terms of pH and oxidation- reduction potentials. J. Geol., 60: 1-33.

Leinfelder, P.R. and Hartkopf-Fr/Sder, Ch., 1990. In situ accretion mechanism of concavo-convex lacustrine oncoides ("swallow-nests") from the Oligocene of the Mainz Basin, Rhineland, F.R.G. Sedimentology, 37: 287-301.

Logan, B.W., Rezak, R. and Ginsburg, R.N., 1964. Classification and environmental significance of algal stromatolites. J. Geol., 72: 68-83.

Love, K.M. and Chafetz, H.S., 1988. Diagenesis of laminated travertines crusts, Arbuckle Mountains, Oklahoma. J. Sediment. Petrol., 58: 441-445.

Mazzullo, S.J., 1980. Calcite pseudospar replacive of marine acicular aragonite, and implications for aragonite cement diagenesis. J. Sediment. Petrol., 50: 409-422.

Meunier, S., 1899. Observations relatives au dfpft de certains travertins calcaires. C.R. Acad. Sci. Paris, 129: 659-660.

Monty, C.L.V., 1972. Recent algal stromatolitic deposits, Andros Island, Bahamas, Preliminary report. Geol. Rundsch., 61(2): 742-783.

Monty, C.L.V., 1976. The origin and development ofcryptalgal fabrics. In: M.R. Walter (Editor), Stromatolites. Elsevier, Amsterdam, Dev. Sedimentol., 20, pp. 193-249.

Monty, C.L.V. and Hardie, L.A., 1976. The geological significance of the freshwater blue-green algal calcareous marsh. In: M.R. Walther (Editor), Stromatolites. Elsevier, Amsterdam, Dev. Sedimentol., 20, pp. 447-477.

Monty, C.L.V. and Mas, J.R., 1981. Lower Cretaceous (Wealdian) blue-green algal deposits of the Province of Valencia, Eastern Spain. In: C.L.V. Monty (Editor), Phanerozoic Stromatolites. Springer, Berlin, pp. 85-120.

Mtlller, O.F., 1777. Flora danica, IV. Kopenhagen [non vidi, cited in G. Gerdes and W.E. Krumbein, 1987].

Pedley, H.M., 1990. Classification and environmental models of cool freshwater tufa. Sediment. Geol., 68: 143-154.

Pedley, H.M., 1992. Freshwater (phytoherm) reefs: the role of biofilms and their bearing on a marine reef cementation. Sediment. Geol., 79: 225-234.

Penhallow, D.P., 1896. Note on calcareous algae from Michigan. Bot. Gaz., 21: 215-217.

Pentecost, A., 1978. Blue-green algae and freshwater carbonate deposits. Proc. R. Soc. London B, 200: 43-61.

Pentecost, A., 1985. Association of cyanobacteria with tufa deposits: identity, enumeration, and nature of the sheath

material reevaluated by histochemistry. Geomicrobiol. J., 4: 285-289.

Pentecost, A., 1987. Growth and calcification of the freshwater cyanobacterium Rivularia haematites. Proc. R. Soc. London B, 232: 125-136.

Pentecost, A., 1991. Calcification processes in algae and cyanobacteria. In: R. Riding (Editor), Calcareous Algae and Stromatolites. Springer, Berlin, pp. 3-20.

Roddy, H.J., 1915. Concretions in streams formed by the agency of blue-green algae and related plants. Am. Philos. Soc. Proc. Philadelphia, 54: 246-258.

Sabater, S., 1989. Encrusting algal assemblages in a Mediterranean river basin. Arch. Hydrobiol., 114: 555-573.

Sch/ifer, A. and Stapf, K.R.G., 1978. Permian Saar-Nahe Basin and recent Lake Constance (Germany): two environments of lacustrine algal carbonates. In: A. Matter and M.E. Tucker (Editors), Modern and Ancient Lake Sediments. Spec. Publ. IAS, 2. Blackwell, London, pp. 83-107.

Schmidle, W., 1910. Postglaziale Ablagerungen im nordw- estlichen Bodenseegebiet. Neues Jahrb. Mineral. Geol. Pal~ontol,, 1910: 104-122.

Serpette, M., 1947. Observations 6cologiques et systfmatiques sur quelques Cyanophycfes de Tunisie. Bull. Soc. Bot. Fr., 94: 306-309.

Stapf, K.R.G., 1973. Limnische Stromatolithen aus dem pf'alzischen Rotliegenden. Mitt. Pollichia, 3, 20:103-112.

Stapf, K.R.G., 1989. Biogene fiuvio-lakustrine Sedimentation im Rotliegend des permokarbonen Saar-Nahe-Beckens (SW-Deutschland). Facies, 20: 169-198.

Szulc, J. and Cwizewicz, M., 1989. The Lower Permian freshwater carbonates of the Slawkow Graben, southern Poland: sedimentary facies, context and stable isotopes study. Palaeogeogr. Palaeoclimatol. Palaeoecol., 70:107-120.

Thunmark, S., 1926. Bidrag till k/innedomen om recenta kalktuffer. Geol. F6rderhandl., 48: 541-583.

Wallner, J., 1934. Ueber die Beteiligung kalkablagernder Pflanzen bei der Bildung stidbayerischer Tuffe. Bibl. Bot., 110: 1-29.

Wharton, R.A., Parker, G.M., Simmons, G.M., Seaburg, K.G. and Love, F.G., 1982. Biogenic calcite structures forming in Lake FryxeU, Antarctica. Nature, 295: 403-405.

Winsborough, B.M. and Golubic, S., 1987. The role of diatoms in stromatolite growth: two examples from modern freshwater settings. J. Phycol., 23: 195-201.

Permian freshwater stromatolites associated with the conifer shoots Cassinisia orobica Kerp et al. — a very peculiar type of fossilization

Documents