Facies (2005)DOI 10.1007/s10347-005-0042-0

ORIGINAL ARTICLE

L. P. Fernandez · M. Nose · E. Fernandez-Martınez ·I. Mendez-Bedia · St. Schroder · F. Soto

Reefal and mud mound facies development in the Lower DevonianLa Vid Group at the Colle outcrops (Leon province, Cantabrian Zone,NW Spain)

Received: 4 November 2004 / Accepted: 2 December 2005C© Springer-Verlag 2005

Abstract In the locality of Colle (Cantabrian Zone, NWSpain), the upper part of the Valporquero Shale Formation(Emsian, La Vid Group) contains an interval of shales andmarlstones (barren, greenish-grey shales and fossiliferous,greenish-grey or reddish shales/marlstones) with beds andpackages of homogeneous and cross-bedded skeletal lime-stones. Metre-scale mud mounds and coral biostromes oc-cur encased in the fossiliferous reddish and greenish-greyshale/marlstones, respectively, with the coral biostromesoverlying conspicuous skeletal limestone bodies. Theserocks were deposited on a carbonate ramp, ranging fromabove storm wave base for the cross-bedded skeletal lime-stones to below the storm wave base for the remainingdeposits, organic buildups included. The vertical stack-ing of these facies and the occurrence of the two typesof buildups are interpreted to reflect the interplay amongseveral (possibly 4th and 5th) orders of relative sea-levelvariations, during a 3rd-order highstand. Coral biostromesoccur in early 5th-order transgressive system tracts devel-oped within late 4th-order highstand, and are interpreted tohave thrived on a stable granular substrate (skeletal lime-stones) in non-turbid waters, being later aborted by the

L. P. Fernandez (�) · I. Mendez-Bedia · F. SotoDepartamento de Geologıa, Universidad de Oviedo,c/Arias de Velasco s/n,33005 Oviedo, Spaine-mail: lpedro@geol.uniovi.esTel.: +34-98-510-31-46Fax: +34-98-510-31-03

M. NoseState collection of Palaeontology and Geology, GeoBio-CenterLMU Richard-Wagner-Str. 10,80333 Munchen, Germany

E. Fernandez-MartınezDepartamento de Ingenierıa Minera, Universidad de Leon,c/Jesus Rubio, 2,24171 Leon, Spain

S. SchroderInstitute of Geology, University of Cologne,Zulpicher-Str. 49a,50674 Koln, Germany

onset of muddy sedimentation. Biostrome features sug-gest that they developed under environmental conditionsessentially different from those related to the sedimenta-tion of their granular substrate. Mud mounds occur in 5th-order transgressive and early highstand system tracts tiedto early 4th-order sea-level rise. Field relationships suggestthat mud mounds grew coevally with muddy sedimentation,with high-frequency variations in carbonate vs. terrigenousmud sedimentation influencing their development.

Keywords Cantabrian Zone . Lower Devonian .Carbonate ramp . Coral biostromes . Mud mounds .Allocyclic controls

Introduction

Since the 19th century, the village of Colle (CantabrianMountains, Leon province, NW Spain; Fig. 1) is wellknown for its fossiliferous Lower Devonian deposits, whichform an interval belonging to the Valporquero Shale Forma-tion (Vilas Minondo 1971; Vera de la Puente 1989) of the LaVid Group (Comte 1959; Garcıa-Alcalde and Racheboeuf1978). In literature, the fossils are described as “Sabero-fossils” named after a small mining town situated close tothe village of Colle (Fig. 1). More than 200 species fromdifferent groups (e.g., brachiopods, crinoids, blastoids, bry-ozoans, corals, stromatoporoids, trilobites, nautilids) hadbeen cited, but mostly without any relation to the faciesand sedimentology of the succession and mainly from non-reefal strata. Previous studies on the Colle section with spe-cial emphasis on its facies were carried out mainly by Stel(1975) and Fernandez et al. (1995). The study of Stel (1975)was focused on the origin of the rugose coral biostrome atthe base of the Colle section, whereas Fernandez et al.(1995) worked out a facies-oriented overview section ofthe complete calcareous succession in the La Vid Groupnear Colle. A brief overview of the biostromal reef de-velopment in the Colle section was published by Brouwer(1964) and Soto (1981, 1982). Finally, Schmid et al. (2001)summarized the features of the mud mounds that occur

some metres above the coral biostrome. Other papers dealwith this interval of the Valporquero Formation, either men-tioning the Colle section (Leweke 1982) or studying it inrelatively nearby areas of the same structural unit (EslaUnit; Fig. 1B; Ruhrmann 1971) or in the broader context ofthe southern part of the Fold and Nappe Province (Keller1988).

The aim of this paper is to provide for the first time adetailed and comprehensive data set on the faunal com-position, facies types and the development of the organicbuild-ups (coral biostromes and mud mounds) of this well-known locality of Colle. Moreover, in the light of these re-sults, we discuss the controls on the development of the de-posits, including the various reef types, in the Colle section,mainly in the light of a sequence stratigraphic framework.The vertical arrangement of facies suggests that severalorders of cyclicity modulated the sedimentation. Accord-ing to the literature and the available data, it seems thattectono-eustatic 3rd-order cycles interplayed with higherorder (4th–5th?) cycles, the latter being possibly the re-sult of a combination of regional tectonics and of climatefluctuations.

Geological setting

The studied area is located in the Esla Unit of theCantabrian Zone (Lotze 1945), which is the external zoneof the Variscan Iberian Massif in NW Spain (Fig. 1A).From a tectono-stratigraphic point of view, the CantabrianZone has been subdivided into several provinces (Julivert1971) and units (Perez-Estaun et al. 1988), one of whichis the Esla Unit (Fig. 1A and B). From Cambrian toDevonian and prior to the onset of the Variscan orogenyduring Carboniferous time, the Cantabrian Zone hosted ashallow-marine sedimentation with some alluvial depositson a passive continental margin (e.g. Ribeiro et al. 1990;Aramburu et al. 2004), which resulted in a rather completestratigraphic record, punctuated by hiatuses of variable du-ration and areal extent. The Devonian record comprises analternation of calcareous and siliciclastic formations, whichhave been grouped into two types of successions based ontheir facies. These are the so-called Palentine and Asturo-Leonese facies (Brouwer 1964; Fig. 2). The Palentinesuccession, whose features suggest deposition in distal anddeeper areas of the sedimentary basin, has an allochthonous

� Fig. 1 A Schematic geological map of the Cantabrian Zone show-ing the provinces and units of Julivert (1971) and Perez-Estaun et al.(1988). B Schematic geological map of the Esla Nappe showing thelocation of the village of Colle. White squares mark the location ofthe sections studied by Ruhrmann (1971). Map is based on works byLobato et al. (1984), Alvarez-Marron et al. (1990) and Heredia et al.(1991a, b). PF = Porma fault, LL = Leon Line, SGL = Sabero–Gordon Line. C Simplified geological map of the Esla Nappe nearColle showing the location of the study area. The Cambrian to De-vonian formations cropping out in the area, individualizing the LaVid Group, have been drawn. Based on Lobato et al. (1984). SGL =Sabero–Gordon Line

character being restricted to the Variscan Palentine nappes(Fig. 1A), which were emplaced gravitationally in thePisuerga-Carrion Province from the south (see Frankenfeld1983; Marquınez and Marcos 1984; Rodrıguez Fernandezand Heredia 1988). In contrast, the succession with Asturo-Leonese facies (Fig. 2) has a shallower and more proximalcharacter, and crops out in the Fold and Nappe Province. Itsbasal part comprises the Lochkovian-Emsian La Vid Groupand its lateral counterpart in the north, the Raneces Group,which crop out in the Somiedo-Correcillas and Esla units(Figs. 1 and 2).

The La Vid Group is dominantly formed of carbonates(limestones and dolostones), although a thick shaly unit ispresent in its middle and upper part (Fig. 2). The grouphas been subdivided into four formations: Felmın, La Pe-drosa, Valporquero and Coladilla (Vilas Minondo 1971;Vera de la Puente 1989), although other subdivisions havebeen later proposed (Leweke 1982; Keller 1988; Fig. 2).The Valporquero Formation (Emsian) comprises the shalyunit of the middle-upper part. This shaly unit is rathermonotonous, although its upper part contains a marly inter-val with limestone intercalations that constitutes the studiedinterval (Fig. 2). These limestones are mostly detrital, al-though some coral biostromes and mud mounds exist aswell. Keller (1988) named this interval of the ValporqueroFormation as the Saguera Member of his Esla Formation.It also corresponds to the Intermediate Limestone Memberof Leweke (1982).

The La Vid Group was deposited on a carbonate rampand its stratigraphic succession has been tied to two 3rd-order transgressive–regressive cycles (Vera de la Puente1988; Keller and Grotsch 1990; Keller 1997; Fig. 2).The shales of the Valporquero Formation are thoughtto record the highstand of the upper 3rd-order cycle.Within this shaly unit, the coarser-grained succession ofthe studied interval would record a minor (4th-order?)lowstand.

The Devonian limestone formations of the CantabrianZone contain a wealth of reefal and reef-related deposits.These deposits, which mainly appear in the Asturo-Leonesefacies, occur from the Pragian (Nieva Formation of theRaneces Group and Lebanza Formation of the Palen-tine nappes) until the Givetian (Candas-Portilla Forma-tion), and locally the Frasnian (Cremenes limestone ofthe Nocedo Formation), forming five main reefal episodes(Fig. 2; Mendez-Bedia et al. 1994; Fernandezet al. 1995; see a review in Garcıa-Alcalde et al.2002). The coral biostromes and mud mounds ofthe studied interval represent the second reefalepisode.

Methods

In the vicinity of Colle (Fig. 1C), the studied intervalcrops out patchily along a hill slope, where several gul-lies and footpaths provide sections of acceptable quality.The best exposed section was selected as the main sec-tion (Fig. 3), and two others, located within a distance of

Fig. 2 Chronostratigraphic chart showing the Devonian lithostrati-graphic units of the Asturo-Leonese facies that have been definedin the political regions of Asturias (northern part of the CantabrianZone) and Leon (southern part of the Cantabrian Zone) and the dis-tribution of the reefal episodes. Absolute ages based on Gradsteinet al. (2004). The stratigraphic subdivision of the La Vid Group isthat of Vilas Minondo (1971) and Vera de la Puente (1989), but sub-

divisions by Leweke (1982) and Keller (1988) are also shown. Notethat, contrary to other authors, Leweke (1982) treats La Vid Group asa formation made up of members. The log on the right depicts the in-terpreted relationships between the general stratigraphy of the La VidGroup and the sea-level curve (based on Keller and Grotsch 1990)and shows the location of the studied interval of the ValporqueroFormation

200 m from the latter, were additionally studied. The stud-ied interval approximately coincides with that discussedby Stel (1975, see his Fig. 1), and is bound above andbelow by poorly exposed shales with scarce limestone in-tercalations (Lower Shale and Upper Shale members ofLeweke 1982). Field features of the deposits were usedto separate a number of facies that are dealt with inthe following section. Rock slabs and stained thin sec-tions were studied for detailed features of sediments andfossils.

Major lithological contacts were physically traced be-tween sections where possible and drawn on field photo-mosaics and panoramic photographs to produce a strati-graphic cross section that summarizes both physical andinferred correlations, and used to determine the geom-etry of sedimentary bodies and their mutual relation-ships. The hill slope trends approximately NW–SE. Thisdirection is oblique to the depositional dip of the De-vonian basin, whose distal and deeper parts would belocated to the south in this sector of the CantabrianZone (e.g., see Aramburu et al. 2004, and referencestherein).

Facies

The studied interval (Saguera Member of Keller 1988;see above) is composed of greenish and reddish shalesand marlstones with interleaved limestones (Fig. 3). Lime-stones are mainly detrital (bioclastic) packstones to grain-stones that form thick bodies and thin beds. Subordinately,some small but locally conspicuous coral biostromes andmud mounds exist. Based on their lithological features andfaunal content, the shale/marlstone and limestone depositsof the studied interval can be grouped into a number offacies, described and interpreted below. Special emphasisis given to the biostromal and mud mound facies.

Facies A: barren, greenish-grey shales

Description

It comprises greenish-grey shales to marly shales lackingany detectable macrofossil content (Fig. 4A). These de-posits form principally the lower and upper parts of the

studied interval (Fig. 3) and are also the major constituentof the remainder of the Valporquero Formation.

Interpretation

This facies is interpreted as deposited by clay falloutfrom suspensions in a deep and low-energy environmentwell away from the action of currents transporting par-ticulate sediment, as suggested by the lack of interleavedcoarser-grained deposits. Absence of macrobiota could bedue to several factors, not necessarily mutually exclu-sive: unfavourable, possibly somewhat stressed (oxygen-depleted?) bottom conditions (see also Stel 1975; Kellerand Grotsch 1990), and the lack of a stable, particulatesubstrate suitable for a fauna to attach. A high rate of claysupply could have also contributed to the absence of mac-robiota, although not being itself a determinant factor, sincethe mudrocks of facies B and C (see below) contain abun-dant fauna. In contrast to the other mud-rich facies (facies Band C), the absence of both coarse-grained sediment layersand benthos suggests that probably this facies representsthe most distal and deepest setting of the ramp, possiblybelow the pycnocline.

Facies B: fossiliferous, greenish-greyshales/marlstones with interleaved skeletal limestones

Description

This facies mainly consists of greenish-grey shales to shalymarlstones. Locally, skeletal packstones to wackestonesare present forming alternations with the shales/marlstonesin subequal proportions, although more commonly theyare a subordinate constituent (Figs. 3 and 4B–C). Theshales/marlstones contain abundant macrobiota, whichcomprises slightly disintegrated crinoids and blastoids, bra-chiopods (spiriferids and terebratulids), bryozoans (fen-estellids, ramose forms, mushroom-shaped fistuliporids),favositids (Crenulipora) and auloporids (Bainbridgia,Schlueterichonus). The skeletal wackestones to packstonesappear as cm-thick beds, which either are tabular or thinand pinch out laterally. Some beds laterally pass intothe surrounding muddy rock due to mixing by burrow-ing. Bioclasts are variably bioabraded and/or iron-stainedand essentially belong to the same taxa present in themuddy deposits. They comprise disintegrated echinodermplates (mainly crinoid and blastoid ossicles), although somewhole calyces exist. Fragments of bryozoans (fenestellids,fistuliporids and ramose forms) are generally subordinate,although they are dominant in some beds (Fig. 5A). Lesscommon are fragments of corals, brachiopods, trilobitesand ostracods. A few whole brachiopods were found. Inthese limestones, the matrix between grains is an argilla-ceous micrite to marlstone with comminute fossil debris(Fig. 5A).

Interpretation

The muddy rock that forms the bulk of this facies is muchsimilar, if not identical, to that of facies A and also in-dicates a low-energy environment with background sedi-mentation by clay fallout from suspensions. Nevertheless,the abundance of in situ fossils indicates that the muddywater did not prevent sea bottom colonization by benthicfauna. Based on their bioclastic composition, and their lat-eral continuity and distribution, the skeletal limestone bedsare thought to have been laid down by currents and to rep-resent the distal counterparts of the facies D (see below; cf.Ruhrmann 1971). The whole fossils (mainly echinodermsand brachiopods) that exist within the beds are interpretedas infauna and epifauna that colonized the granular sub-strate after currents laid down their bioclastic load (seeinterpretation of facies D). The related burrowing activitywould also account for the bedding destruction and mixingof the granular beds with the underlying muddy sediment.Nevertheless, it cannot be precluded that in some cases theskeletal limestone beds may represent “condensed” inter-vals, in which, a diminished rate of clay fallout, possiblytied to minor transgressive episodes, could have resulted ina deposit enriched in skeletal components and in lime mud.Thus, as a whole, this facies would represent an environ-ment close to or above the storm wave base, shallower thanthe setting of deposition of the former facies. It is conceiv-able that these shallower conditions, plus a more intensewater circulation, partly related to the currents that spo-radically swept the sea bottom and deposited the skeletalbeds, would have resulted in more oxygenated bottom con-ditions, favourable to host benthic communities (see alsoStel 1975).

Facies C: fossiliferous, reddish shales/marlstones withinterleaved skeletal limestones

Description

This facies is similar to the former (Fig. 4B–D),the main differences being the reddish colour of theshales/marlstones, and the generally higher fossil contentof facies C. In addition, the muddy rock has a higherfine-grained carbonate content, judging from its moremarly appearance. The fossil content is dominatedby echinoderms (crinoids and subordinate blastoids),bryozoans (fenestellids, mushroom-shaped fistuliporidsand occasional ramose forms), brachiopods (spiriferidsand terebratulids), diverse tabulate corals, such as ramosefavositids (Crenulipora, Thamnoptychia, Dendropora) andauloporids (Schlueterichonus, Cladochonus, Bainbridgia).Skeletal limestone beds are identical to those of the formerfacies, and bear the same relationships to the host muddysediment, containing the same types of bioclasts and wholefossils.

Interpretation

The setting of deposition of this facies seems to have beenessentially much the same as in the case of the facies Babove, as the faunal content and the types of skeletal lime-stones suggest. The most striking feature of this facies is thereddish colour of the shales/marlstones, which is likely asyn-sedimentary feature, generated by bacterial activity inthe marine environment (see Bourque and Boulvain 1993;Preat et al. 1999; Boulvain 2001). The slightly greater abun-dance of fauna in the background muddy sediment, com-pared to the facies B, could indicate that conditions weremore favourable for the development of benthic communi-ties. The more marly appearance of this rock and the factthat mud mounds are present only in the intervals of thisfacies are suggestive of a lowered clay input (see below).We speculate that this lowered clay input, probably relatedto an overall lower sedimentation rate, would have resultedin less turbid seawater favouring the growth of benthic or-ganisms. So, improving growth conditions together withreduced “dilution” of skeletal remains by mud depositionaccount for the high faunal content of this facies. The verti-cal relationships between reddish- and greenish-grey-shaleintervals suggest that the formation of these two facies wascontrolled by allocyclic, long-term factors, as it will bediscussed below.

Facies D: thick-bedded, skeletal limestones

Description

This facies is formed of skeletal pack- to grainstones. Lo-cally, they contain large fragments of corals and constitutemichelinid or Synaptophyllum float/rudstones (e.g., unit1 of Figs. 3 and 5B; see also Fig. 6). These limestonesform cm- to dm-thick, laterally continuous to wedge-shaped beds. Some beds, especially the coarsest-grainedfloat/rudstones, are structureless, although others displaycross bedding. Cross bedding is commonly of a low-angletype (Fig. 4C), and is frequently recognized as hummockycross-stratification (HCS). The beds of this facies arestacked and amalgamated to form dm- to m-thick packages(Fig. 4C). These packages may fine and thin upwards untilthey are formed of skeletal limestones with shale/marlstoneinterbeds. The uppermost limestone bed(s) of these pack-ages are generally richer in a muddy matrix, and this matrixis identical to the overlying sediment (facies B or C).These packages have an erosive base; they are lenticular inshape and extend laterally for at least some tens of metresbefore they pinch out or interfinger with shales/marlstonesuntil they finally pass into facies B or C intervals (e.g. unit12 of Figs. 3 and 4B–D). These packages may appear asrelatively thin packages capping a thickening-upward and

� Fig. 3 Simplified stratigraphic log of the main section describedin the locality of Colle, showing the facies distribution and thecoarsening-upward or fining-upward trends

coarsening-upward interval of alternations belonging toeither the facies B or C (Fig. 3). More commonly, theyform thicker packages that sharply overlie an interval ofthe facies B or C with a conspicuous erosional surface(Figs. 3 and 4C). In some cases, these erosional surfaceshave a relief of at least 1 m, against which the beds formingthe packages or their laterally equivalent alternations onlap(Fig. 4D). In places, the high relief of the erosional surfacesis caused by the presence of mound structures, which aremore resistant to erosion than the adjacent muddy sediment.The rare coral float/rudstones (e.g. Figs. 4D and 5B) arepresent at the base of the packages above these major ero-sional surfaces (e.g. base of unit 1 in Fig. 3; see also Fig. 6).Skeletal grains in this facies comprise bioclasts of echino-derms and bryozoans (fenestellids, fistuliporids and ramoseforms) as the main components. Less common elementsare coral fragments, brachiopods, trilobites, and ostracods.In some cases, whole brachiopods are found (Uncinulus,Atrypa and Xystostrophia). As in the case of the thin skele-tal beds of facies B and C, bioclasts display borings andiron-stained margins and, in some cases, they constitutelithoclasts, because their diagenetic history differs fromthat of the hosting limestone, suggesting they were par-tially lithified before deposition. For example, some coralsdisplay cavities connected to outside that are only filledwith cements, lacking the bioclastic muddy matrix that isfound surrounding the coral. In other cases, coral cavitiesdisplay overturned geopetal fills (Fig. 5B). Finally, theedges of some grains truncate the cements filling the graincavities.

Interpretation

The features of this facies suggest that it resulted from depo-sition from traction currents, which, at least in some cases,were produced by storms (HCS). A similar storm-relatedinterpretation has been reached by Ruhrmann (1971), Stel(1975) and Keller (1988), for these Colle facies and for lat-erally equivalent deposits located elsewhere in the south-ern part of the Fold and Nappe Province. The variationof thickness and internal structuring of beds is thought toreflect proximal–distal trends or the magnitude of the cur-rents (cf. Aigner and Reineck 1982; Aigner 1985). Thedeposits of the weakest currents, or the most distal de-posits of the strongest currents, would be represented bythe thinnest beds of this facies and by the thin beds inter-leaved in the facies B and C, whereas the proximal depositsof the strongest flows would consist of HCS thickest beds.Structureless beds could be due to flows not being ableto exert enough traction on grains as they were being de-posited (see a detailed discussion on this topic in Chapter3 of Pickering et al. 1989; and references therein), or alsoto post-depositional homogenisation by burrowing. As awhole, this facies would have been deposited in an envi-ronment above storm wave base, which was episodicallyswept by high-energy events. The paucity and vertical dis-tribution of fine-grained sediment layers in between the

successive beds within packages suggest that calm condi-tions were of short duration and/or that the currents wereable to erode the muddy sediment deposited during the fairweather episodes. The higher frequency of the muddy in-tervals to the top of packages would indicate a progressivewaning of the successive storms, which, given the over-lying muddy facies, is interpreted to record the progres-sive deepening of a relatively shallow environment. The

types of grains suggest erosion, transport and depositionof relatively nearby bottom-communities by the currents.Nevertheless, in some cases, erosion was strong enough toaffect a buried and partly lithified sediment and rip it upas lithoclasts (skeletal grains with an advanced diageneticstage). Most of these lithoclastic grains (coral fragments)are not found in the neighbouring substrate, suggesting alonger-distance transport from shallower areas. Whole bra-

chiopods that are present within the beds are interpretedas in situ dwellers, which colonized the granular substrateafter its deposition. The related burrowing activity wouldalso account for the muddy matrix of the uppermost bedsof the skeletal deposits, which is interpreted as the productof mixing of a muddy sediment with the granular bed.

Facies E: Synaptophyllum biostromes

Description

This facies occurs in an interval of the lower part of the sec-tion (unit 1 of Fig. 3), which corresponds to unit C of Stel(1975). The deposits of this facies take place as a few thin(less than 30 cm thick) and laterally discontinuous (tensof m wide) biostromal horizons that alternate with skele-tal limestones (facies D) and fossiliferous, greenish-greyshales/marlstones (facies B) forming several dm-thick se-quences of facies. In each sequence of facies, a biostromalhorizon overlies a bed of skeletal limestones, which in somecases are Synaptophyllum rudstones, and is in turn over-lain by the fossiliferous, greenish-grey shales/marlstonesor by the basal limestone bed of the next sequence of facies(Figs. 3 and 4E). The matrix between the colonies of thebiostromes is a bioclastic and fossiliferous shale/marlstoneidentical to the overlying deposits (facies B). The organ-isms of the biostromes are mostly in situ and are domi-nated by branching colonial rugose corals represented bySynaptophyllum and by small numbers of massive colonies

� Fig. 4 Outcrop photographs showing the facies discussed in the text.The location of the photographs is shown in Fig. 6. A Barren greenish-grey shales of facies A. B Interval showing the sharp base of thefacies B (greenish-grey fossiliferous shales) over an interval of barrenshales of facies A. This sharp base represents sequence boundarynumber 9. On top of facies B, an interval of facies C with somesmall mud mounds (knobby rocks) gives way to an upper intervalof facies A. The three intervals overlying the sequence boundaryform a 5th-order transgressive system tract (sequence number 9).C Lower part of sequence number 2. Notice the low angle cross-bedding of the skeletal limestones of facies D and the offlappingattitude of the individual mud mounds. Encircled hammer for scale.D Detailed view of the sequence boundary number 4 showing itserosional relief downcutting towards the left. Notice the onlap ofthe bedding surfaces of the basal deposits of sequence number 4,consisting of fossiliferous reddish shales/marlstones with interleavedthin skeletal limestone beds, onto the discontinuity. Section dips tothe left; the palaeohorizontal is given by the stratal surfaces over theunconformity. E Coral-biostromal deposits of facies E overlying askeletal limestone interval (Synaptophyllum rudstone of facies D).Arrow points to the base of a colony. Notice the muddy matrix in theinterstices of the biostrome and the thin skeletal limestone bed (basemarked by white bars) that overlies the biostromal deposit and is inturn overlain by a new biostromal interval with stromatoporoids. Thearrow marks the vertical extent of the cycle comprised between theskeletal limestone beds and which is compared to a 6th-order cycle(see text for details). F Oblique view of a mud mound overlyinga facies C interval and overlain by facies A deposits. The sectiongently dips towards the observer, so the facies A deposits in thebackground can be traced into the same facies in the foreground. Theinterval depicted in this photograph belongs to sequence number 9and is laterally equivalent to that shown in B. Hammer for scale inall photographs is 33 cm long

of the genus Cantabriastraea (Schroder and Soto 2003)<10 cm in diameter. Tabulate corals, stromatoporoids,some of which are overturned, and subordinate fistuliporidbryozoans and atrypid brachiopods are also present. Tab-ulate corals, which usually grow on dead Synaptophyllumbranches and some of which are infected with tubes of thegenus Helicosalpinx, are represented by Favosites, Alveo-lites, Heliolites and Thamnopora. Stromatoporoids displaysymbiotic syringoporoids (Caunopora state) and belongto the genera Salairella, Stromatopora, Neosyringostromaand Stromatoporella.

Favositids and stromatoporoids mainly occur in thebiostromal horizons of the uppermost sequences of facies(Fig. 4E) and comprise rather flat (laminar to low-reliefdomical) and relatively large forms, up to 20–30 cm indiameter. Heliolitid corals form small spherical colonies,whereas alveolitids occur as irregular encrusting forms. Fi-nally, Thamnopora colonies are small and display a scarcenumber of branches. Alveolitid, heliolitid and favositidcorals show regular and continuous growth patterns, lack-ing the growth banding that is fairly usual in heliolitids,although in some cases it exists being very subtle. Also,stromatoporoids generally show regular latilamination.

Few and small necrotized areas affect the flanks of thecoral and stromatoporoid skeletons. Internally, massive tab-ulate corals and stromatoporoids display some thin sedi-ment laminae that veneered the upper surface of the skele-ton and were later overgrown. These sediment laminae aremainly of micrite, although some of them are clay-rich.

Generally, the internal cavities of the colonies are onlyfilled with cement. Nevertheless, the uppermost cavitiesof the colonies (e.g., see Fig. 5C) and those adjacent tothe necrotic portions of the skeleton flanks are completelyor partially filled with sediment. Three main generations ofcement can be distinguished. The first one consists of a thinisopachous rim of inclusion-rich fibrous calcite; the secondgeneration is formed of inclusion-rich columnar crystalsof calcite with undulose extinction (radiaxial-fibrous cal-cite?); and finally the third generation consists of equantcrystals of calcite, which in some cases are also inclusion-rich. These generations of cement were seen in thin sectionsmade for palaeontological determinations that had not beenstained, and thus ferroan vs. non-ferroan calcite could notbe distinguished. The sediment infilling the cavities can beof two types, of rather pure micrite or of shale/marlstone.Both types contain a variable amount of bioclasts. The up-permost internal cavities of the colonies only contain theclay-rich type of sediment, whereas those cavities close tothe necrotic portions of the flanks may display both the mi-critic and the clay-rich types. In some thin sections, the sed-iment that fills the uppermost cavities of the colonies post-dates the first generation of (isopachous) cement (Fig. 5C).

Interpretation

The occurrence of the Synaptophyllum biostromes atopskeletal beds reveals the need of a stable substrate forthe colonies to develop, a conclusion also reached by

Stel (1975). This is also proven by the alveolitids, whichbelong to forms with an encrusting habit, and by themorphology of the alveolitid and heliolitid colonies, whichis indicative of restriction to lateral growth. Thus, thegranular substrate constituted by the skeletal limestonebeds could be regarded as the stabilization stage of the“reefoid” episode. Nevertheless, this does not mean thatboth deposits, the granular bed and the biostromal interval,are related and form a continuum. Since the granularsediment of the substrate is not found between the coloniesand these occur clearly on top of the granular bed, weconclude that both deposits formed under very differentenvironmental conditions in two separate episodes ofsedimentation. On the other hand, the clay-rich sedimentthat surrounds the colonies and is also found as internalsediment in their cavities is very likely a later feature, notcoeval to the biostrome development either. Several databack this conclusion. The presence of clay-rich sedimentpost-dating cements in coral cavities clearly indicatesthat this sediment records an essentially later episode ofsedimentation. The few and small necrotic areas affectingthe flanks of the skeletons suggest that killing processes bysedimentation (“growth-interruption banding” of Youngand Kershaw 2005) were not frequent or important, and

� Fig. 5 Photomicrographs of the limestone facies. Stratigraphic toptowards the upper edge of the page in all the cases. A Skeletal pack-stone corresponding to a thin limestone bed of facies B. Note themicrite matrix and the clayey laminae. Skeletal grains mainly com-prise bryozoans (fistuliporids: Fi) and variably bored and iron-stainedpelmatozoan ossicles (Eq) and brachiopods (Br). Some auloporids(Au) and a rugosan (Ru) fragment are also present. B Detail of askeletal packstone-grainstone of facies D containing a large fragmentof Synaptophyllum (Sy) (Synaptophyllum float/rudstones). Coral isoverturned; note the geopetal structure, arrows point at the contactbetween micrite sediment (m) and cement (c). Other grains compriseiron-stained and bored pelmatozoan ossicles (Eq) and brachiopods(Br). C Detail of the upper surface of a Synaptophyllum of thecoral biostrome deposits (facies E) showing the time relationshipsbetween biostrome development and deposition of the fine-grainedsediment that surrounds the colonies. Note that a first generation ofisopachous cement growing on the uppermost tabula is postdated byclay-rich sediment containing scattered fossil hash. These relation-ships demonstrate that once the coral died, a phase of cementationtook place before fine-grained sediment was deposited. D Detailedview of a microbial boundstone (facies F) showing the complex re-lationships between sediment generations [microbial micrite (m1),locally peloidal (p); homogeneous mechanical micrite (m2); and mi-crosparitic material (m3)] and the three types of pores (vA, vB, andvC). Note the burrow-like aspect of the type C pores (vC) that cut themicrosparite and are in turn filled with the same type of sediment.Skeletal components include sponges (Sp), which are exceptionallyabundant in this sample, partially bored auloporids (Au) and smallbioclastic fragments that mainly correspond to sponge spicules (tinylight coloured spots). E Microbial boundstone showing the relationsbetween the three types of sediment (m1, m2 and m3) and of typesA and B pores (vA, vB). The centre of the photograph displays anexceptional example of burrows corresponding to type-B pores (vB)that cut into homogeneous mechanical micrite (m2) and are filled bymicrosparitic material. Note the encrusting attitude of most of bry-ozoans (fenestellids: Fn; and fistuliporids: Fi). Other grains comprisea pelmatozoan (Eq) and a rugosan (Ru). F Example of a type B cavity(vB) in microbial micrite (m1) and filled with microsparitic material(m3). The latter is in turn cut by burrows corresponding to the last(type C) generation of cavities

thus that sediment input during biostrome development wasneither frequent nor intense. In addition, the rather purecarbonate sediment that fills the internal cavities of theskeletons adjacent to these necrotic areas suggests thatthe rare sedimentation events related to killing of skeletonflanks were clay-poor, clearly different from the type ofsedimentation that is recorded in the clay-rich depositsfound among the colonies and in the cavities open tooutside. Summarizing, it seems more likely that threestages can be distinguished in biostrome development.During the initial stage of high energy, granular sedimentwas deposited forming the basal skeletal limestone bed.After the first stage, environmental conditions becamemore tranquil and biostromes developed during the secondstage. In this second stage, it is possible that episodiccurrents were able to rip up and rework the tabularskeletons of stromatoporoids that are found overturned.These relatively high-energy episodes would also accountfor the presence of isopachous rims of fibrous cement thatare in some cases seen to pre-date the sediment withincavities. This relatively early phase cementation and theradiaxial-fibrous calcite cement are assigned to a phreaticmarine diagenetic environment. Several criteria back thisinterpretation, their fibrous and radiaxial-fibrous crystalhabits, their occurrence forming the earliest pore-fillings,the comparison of the fibrous calcite with the non-ferroanfibrous calcite cement found in the mud mound facies. Anearly marine cementation is favoured by an active fluidcirculation through pores, and this pumping is particularlyactive where sea floor is swept by currents (see Chapter 7of Tucker and Wright 1990). During this second stage, thebiostrome developed under relatively non-turbid-water anduniform conditions (low, clay-poor sediment input). Theuniform growth patterns of the colonies, lacking markeddiscontinuities, suggest a monotonous environment, notprone to marked cyclical variations of environmentalparameters such as light, temperature (?) and waterturbidity (e.g., see Kershaw and Young 1997). Finally, inthe latest stage, environmental conditions changed and ledto the onset of clay-rich muddy sedimentation, which verylikely was responsible for the abortion of the biostrome andfinally buried it. This three-step evolution occurred severaltimes giving rise to the repetition of the sequence of faciesformed of the basal skeletal bed containing reworked coralcolonies, the biostrome and the capping muddy intervalplus the muddy matrix surrounding the biostrome.

Facies F: mud mounds

Description

This facies forms small (0.3–0.8 m thick and ∼1–4 mwide) mounds, and more rarely flat, bed-like bodies. Theydisplay ragged margins and are encased within intervalsformed of facies C (Figs. 3 and 4B, C and F); onlyone very small and low-relief mound appears within aninterval of facies B. The individual mounds are isolatedfrom each other (Fig. 4F) or grouped either as laterallyofflapping or as vertically stacked bodies (Fig. 4C). These

Fig. 6 A Detailed correlation pattern showing the inferred small-scale sequential architecture and the lateral and vertical facies distri-bution of the interval studied in the vicinity of Colle (note the verticaland lateral distances involved). See also Fig. 3. B Stratigraphic log

showing the vertical variation of thickness of the sequences and oftheir basal skeletal limestone intervals. The sequence composition interms of the different muddy facies as well as their vertical distribu-tion is also shown (see text for details)

mound structures display discontinuous and irregularreddish shale/marlstone partings that can be traced into theragged margins and connect with the surrounding marlyrock (facies C). The mounds mostly consist of a reddishand greenish micrite containing a relatively abundantmacrobiota (<25%), and displaying different types ofsubmillimetric to millimetric cavities (cf. Schmid et al.2001). The micrite is structurally heterogeneous, withseveral sediment generations (polymuds of Lees and Miller1985) revealed by differences in colour or texture. Underthe microscope, three types of carbonate sediment aredistinguished, namely types 1, 2 and 3 (Fig. 5D–F). Types1and 2 are both micrite, and their mutual relationships sug-gest that they are coeval. Type 1 is a dense and dark micrite,with a homogeneous appearance, although it is locallypeloidal (Bathurst 1975; Fig. 5D). It forms masses, whichcan have a positive relief with steep to vertical margins,surrounded by type 2 micrite. Type 2 micrite is a lightercoloured, homogeneous micrite. Sparse minute bioclastsare present in both types of micrite, although they are moreabundant in the lighter homogeneous type 2 micrite. Type

3 is a microsparitic material with scarce minute bioclasticfragments (Fig. 5D–F). Cross-cutting relationships showthat it is a later sediment than types 1 and 2 (see below).

The macrobiota of this facies does not differ significantlyfrom that of the reddish marlstones and shales (facies C)in which the mud mounds are encased, and, mainly in thecase of the corals, shows a variable degree of bioabra-sion (microborings). The most prominent organisms arefenestellids and platy fistuliporids (Fig. 5E). They are en-crusted by the dense and dark micrite masses (type 1 mi-crite) and, in turn, encrust the micrite types 1 and 2 andthe grains. Other organisms are branching bryozoans, dis-articulated plates of crinoids and blastoids, tabulate corals,rare brachiopods or their disarticulated valves and spongeremains (Fig. 5D–E). Tabulate corals, which markedly dif-fer from those of the coral biostrome facies, display ahigh diversity at both species and genus levels and mainlycomprise Crenulipora, a ramose favositid; other subordi-nate forms are Hamarilopora and reptant and erect aulo-porids (Schlueterichonus among others). Scarce spongespicules are present whereas some samples taken contain

whole sponges (Fig. 5D). Remarkably, sponges exhibit achambered structure indicative of a sphinctozoan nature.Spicules are partly preserved revealing hexactines arrangedin lattice formation. Although a detailed taxonomic descrip-tion of the sponge material from the Colle mounds has stillto be done, this is the first record of hexactinellid sphinc-tozoa from the Devonian (Krautter pers. comm.). Untilnow, only a few representatives of hexactinellid sphinc-tozoa are known, exclusively derived from Upper Triassicand Jurassic strata (Senowbari-Daryan and Garcıa-Bellido2002).

Apart from intraparticle porosity and very rare shelterporosity, the rock contains three different types of millimet-ric cavities, namely A, B and C (Fig. 5D–F). None of thesecavities can be considered as typical stromatactis porosity,although some cavities of type B share all characteristicsof stromatactis except for network and/or isopachouscements, and can be termed as stromatactis-like cavitiesaccording to terminology by Matyszkiewicz (1997) andSchmid et al. (2001). Type A consists of elongate pipe-likecavities found in the micrite of types 1 and 2 (Fig. 5D andE). This type of cavities is filled by the type 2 micrite. It isthought to result from burrowing in a soft sediment. Type Bcomprises elongate or more irregular pores in type 1 and 2micrites and is sealed by type 3 sediment (microsparitic ma-terial) which completely fills the pores or just floors themgiving rise to geopetal structures (Fig. 5D–F). The elongatecavities are burrow-like (e.g., see Fig. 5D and E), whereasthe irregular pores (e.g., see Fig. 5D and F) are larger (upto 1 cm) and display scalloped margins that truncate oldersediment (micrite types 1 and 2) suggesting an origin by,at least partially, dissolution (cf. Lees and Miller 1995).Nevertheless, scalloped margins have also been interpretedas indicative of sponge-boring activity (Schmid et al. 2001,see their Fig. 17). Type C cavities are elongate burrow-likepores in type 3 sediment filled with the same type of sedi-ment being only distinguished by subtle variations in colour(Fig. 5D and F).

Three generations of cement are found in the cavitiesof this facies. The first generation is a thin and poorlydeveloped isopachous crust of inclusion-rich fibrous cal-cite. This generation is found in intra-skeletal pores andin type B porosity. Cross-cutting relationships reveal thatthese cements started growing during the final stage ofthe microsparite sedimentation and continued after its end.The second generation is made of inclusion-rich columnarcrystals of calcite to slightly ferroan calcite with unduloseextinction and is found in some intraparticle pores and intype B porosity. The third generation is a mosaic of clean,slightly ferroan to ferroan calcite and is poorly developedoccluding the remnant voids in intraparticle and type Bporosity.

Interpretation

The textural features and geometry of these mounds arecomparable to those of mud mounds formed of microbialboundstones (see Lees and Miller 1995; Monty 1995; Pratt

1995). Following these authors, and despite the absenceof fossilized microbial filaments, which can be attributedto filament degradation or to the small size of microbes,the dark micrite of type 1 is interpreted to result fromcyanobacterial activity, judging from its dark colour, oc-casional peloidal texture, and the erect and steep-marginmorphology of the masses that it forms, which is sugges-tive of cohesiveness. The lighter-coloured micrite of type 2is thought to represent a sediment deposited mechanically.This type of sediment likely originated within the mudmound proper, given the terrigenous mud-rich environmentof the mounds. This interpretation of a local provenance ofthe carbonate mud in the mud mounds is a common themefound in the literature and is based on the same criteria(see Bosence and Bridges 1995: 5). It is considered thatdegradation of the organic masses of type 1 micrite and ofthe skeletal grains by grazing and boring organisms wouldhave led to this “mechanical” micrite. This interpretation isbacked by the conspicuous early burrowing activity that isrecorded by the type-A pores. The scattered sphinctozoansponges probably produced some additional mud (automi-crite) during degradation. Nevertheless, a partial externalsource of carbonate mud cannot be ruled out, since also car-bonate mud is found outside the mud mounds. Apart fromthe microbial communities, it seems that the organismsthat played a significantly active role in the mud moundstabilization were fenestellids and fistuliporids, by bindingone another, the microbial micritic masses, and the avail-able sediment (lime mud of type 2 micrite, clay particlesand skeletal grains). In some instances, these bryozoansare found to be roofing type-B cavities, which suggeststhat they might have encrusted a soft body that later dis-appeared, although, in some of these cases, it seems thatthe bryozoans could have grown downwards from the cav-ity roof. The other organisms, chiefly crinoids and blas-toids, are thought to have mainly played a passive role byproviding grains, i.e., their whole or disarticulated skele-tons, to the deposit. The small number of coral coloniessuggests that they did not exert a significant baffling orbinding role. The described biota are fairly similar to thoseof Devonian examples from Algeria (Wendt et al. 1997),Kess-Kess mounds of Morocco (Brachert et al. 1992) andthe suggested encrusting role of bryozoans has also beenclaimed in Early Devonian examples from the Clifton Sad-dle (west-central Tennessee, USA; Gibson et al. 1998).

The sediment forming the mud mounds soon seems tohave been semi-lithified as the features of the type-B poros-ity suggest. This second generation of porosity could havea variable origin (see a detailed discussion in Monty 1995).It seems to have resulted from dissolution and burrow-ing, maybe accompanied by current driven winnowingof unconsolidated sediment beneath a lithified crust (cf.Schmid et al. 2001). In other instances, it seems to repre-sent growth-framework or shelter cavities variably modi-fied by the former processes. Finally, it could also resultfrom the degradations of pre-existing soft-bodied organ-isms (cf. Lees and Miller 1995; see above the commentabout the bryozoan-roofed cavities). Type-B porosity isfilled by the microsparitic material (type 3 sediment). Bur-

rowing activity was a long lasting process over the moundlife and the last burrowing stages gave rise to the burrow-ing structures that cut and are filled by the type 3 sedi-ment. The absence of pore-filling clay suggests that thethree stages of burrowing took place before mud-moundabortion and burial by the clay-rich deposits of facies C.In addition, type-3 sediment pre-dating the isopachous fi-brous calcite cement indicates that it was deposited before asubstantial early marine-phreatic cementation took place. Itis worth mentioning that burrowing represented by type-Band, mainly, type-C porosity makes these examples differ-ent from other deep-water mud mounds, which are gen-erally devoid of infaunal soft-sediment dwellers (cf. Pratt1995: 107), but similar to the Silurian examples from An-ticosti Island, Canada, described by Schmid et al. (2001),which also show analogous infaunal organisms.

The ragged margins and the shaly partings of the mudmounds suggest that mound development was essentiallycoeval to sedimentation from fallout of clay and carbon-ate mud particles forming the surrounding deposits (faciesC), and they would fall within the “sediment-continuum”mound category of Schmid et al. (2001). These two featuresindicate that clay-fallout rate and/or carbonate productionrate varied in intensity, possibly in relation to some high-frequency external factor. It is interpreted that with highclay-fallout rates mound growth probably was somewhatinhibited and clay veneered the mound surface, resultingin the shale partings. On the contrary, during moments ofreduced clay-fallout, mounds would have grown faster en-croaching over the surrounding marly sea bottom.

Given the inferred temporal relationships between mudmound development and off-mound sedimentation, it fol-lows that the environmental conditions were the same forthe mud mounds and for the surrounding facies C sea-bottom, and thus, mud mounds formed in a low energy,relatively deep environment. The occasional presence ofthin skeletal beds suggests that this environment could havebeen close to the storm wave base. Several authors (e.g.,Wendt et al. 1997, 2001) have claimed a deep-water setting,in the lower part of the photic zone (some 100–200 m), forsimilar mud mounds with a comparable biota devoid ofshallow-water reef builders and of red and green algae.Nevertheless, we would like to be cautious with respect toabsolute depths, since a lowered light penetration due tosuspended sediment could modify the figures given. Re-markably, mud mounds nucleated and grew on a relativelymud-rich substratum. Strikingly, none of the mud moundsoverlies any conspicuous grain-rich layer. In contrast toSynaptophyllum biostromes, which strongly depended ona stable granular substrate to develop, mud mounds do notseem to be so strict about the mechanical features of thesea bottom to start developing, although the amount ofskeletal material present in the facies C indicates that thesubstrate was somehow granular. Another striking point,given the extremely low relief of the mud mounds, pertainsto the fact that microbially-induced carbonate sedimenta-tion not only started but also was maintained on patchyareas a few metres across (mud mounds), while sedimenta-tion out of these patches was so much different. It remains

obscure what factors led to that, although two main pro-cesses are most likely: first, low rates of clay-fallout, andsecond, clay sedimentation from thin muddy currents. Inthe first case, it can be argued that a generally low claysedimentation rate could have been outpaced by a higherrate of carbonate-mud production on the mud mounds andthat this would have resulted in clay dilution within themud mound deposits. In the second case, it could be spec-ulated that once mud mounds started forming, their subtlerelief prevented clay input. That would mean that clay didnot settle through the whole water column, but that it wastransported by and settled from very thin currents, thinnerthan the mound height, flowing directly on the sea bottom(hyperpycnal currents?). If this was the case, some sortof layering reflecting the alternation of event (hyperpycnalcurrent) and background sedimentation should be present,unless extensive bioturbation had homogenized the sed-iments. No solid evidence, either at regional or at localscale, exist that can favour the occurrence of hyperpycnalflows, unless these hyperpycnal flows were the distal coun-terparts of the flows that deposited the skeletal limestonebeds of facies D. The occurrence of hyperpycnal flows andof their resulting deposits in a shelfal environment is notsuch a strange phenomenon, and has been interpreted fromthe rock record linked either to storm activity (see Walker1984) or to catastrophic floods entering the basin throughriver mouths (Mutti et al. 1996).

Given the size of the mud mounds, and the lack of truecondensed features of the surrounding sediments, it seemsclear that each mud mound developed over a short timespan before final abortion. This could also account for an-other distinctive feature. In contrast to other examples ofmud mounds, not only larger but also of comparable size(e.g., Gibson et al. 1998), the cases here described display auniform composition lacking a vertical superposition of dif-ferent types of deposits as well as a differentiation into coreand flank facies. We consider that their short lives and smallsize would have made the mud mounds to develop under es-sentially uniform environmental conditions, both laterally(space) and vertically (time) (cf. Wendt et al. 2001).

Allocyclical controls on buildup development

Sequence stratigraphy framework

Individualization and identification of depositional se-quences and sequence boundaries must be taken with cau-tion, mainly in single well logs, cores and outcrops. Never-theless, even in these unfavourable cases some diagnosticcriteria can help in the discrimination of depositional se-quences. These are abrupt basinward shifts in facies, andchanges in the mode of vertical stacking of parasequencesor higher-order depositional sequences (see van Wagoneret al. 1990). These criteria can be easily applied in the Colleoutcrop where the vertical and lateral relationships amongthe facies above described permit to group the studied in-terval into a series of units bound by sharp surfaces (seeFigs. 3 and 6A), whose general features are summarized

in Fig. 7. Although, at small-outcrop scale, the basal sharpsurfaces are identifiable as erosional only in some instances(e.g., base of unit 4 in Fig. 3; see also Fig. 4D), when theyare traced over a larger area (see Fig. 6A), they clearlyshow truncation and onlap. Additionally, their erosionalcharacter is revealed by the presence of coral clasts witha diagenetic history different from that of the hosting sed-iment (see the description and interpretation of facies D).These erosive surfaces display a metric-scale relief (Figs.4D and 6A), show no evidence of subaerial erosion and arelocated at the inflection points between coarsening-upwardand fining-upward trends (Fig. 3). The units bound by thesesharp surfaces have a well-defined internal structuring andvertical stacking of facies (Fig. 7). Each unit sharply com-mences with a well-developed body of skeletal limestones(facies D; Fig. 4C) that laterally passes into a skeletal-bed-rich interval of facies B or C (Fig. 4D; see also Fig. 6A). Itis important to stress that, as discussed above, these skele-tal limestone bodies show evidence of resulting from theamalgamation of the deposits of successive storm events.Thus, they record a time span characterized by frequentstorms affecting the sea floor, and should not be consideredeach as the geologically instantaneous deposit of a singlestorm. This basal interval fines upwards and passes intoa middle interval composed of shales/marlstones belong-

ing mainly to facies B or C, although the middle part ofthis interval may be formed of facies A (Figs. 3, 6A and7). This middle interval forms the remaining of the unitor it passes into an upper interval of alternations as theshales/marlstones start containing skeletal limestone bedswith a coarsening and thickening upward trend (Figs. 3,6A and 7). The skeletal bodies of the basal interval wouldbe storm-related deposits laid down in a ramp environmentabove the storm wave base (see above). Their occurrence,abruptly overlying a mud-rich interval instead of form-ing the top of coarsening upward intervals (i.e., instead ofevolving from an underlying interval progressively richer inlimestone beds), indicates that the relatively shallow-waterand high-energy (storm influenced) environment sharplysucceeded the deeper and more tranquil environment ofdeposition of the mud-rich facies (essentially a deep rampbelow the storm wave base). That is, these deposits repre-sent a sharp basinward shift of facies. Consequently, theyare interpreted to record a relative sea-level drop accompa-nied by a significant submarine erosion of the substratum,judging from the relief of the basal erosional surfaces andthe lack of evidence of subaerial exposure, and representlowstand or early transgressive deposits of a depositionalsequence. The gradational upwards passage of these pack-ages into the middle interval of muddy facies (mainly fa-

3th

orde

r hi

ghst

and

stud

ied

inte

rval

?

?

?

?

A

A

B

Sequence boundary

Barren greenish grey shales

Fossiliferous greenish-grey shales/marls

Fossiliferous reddish shales/marls

Skeletal limestones

Mud mounds

Coral biostromes

low high

Sequence boundaryformation and early lowstand

Late lowstand

Transgressive

Highstand

4th order curveof sea level

Highstand

Transgressive

Late lowstand

4th -

orde

r se

quen

ce II

4th -

orde

r se

quen

ce I

low high

A

B

B

Higher (5th-6th)order cycles

Sequence boundary

Sequence boundary

transgression

regression

lowstand? to transgressivesystem tracts

highstand system tract

Sequence boundary

Sequence boundary

transgression

regression

lowstand? to transgressivesystem tracts

highstand system tract

Sequence boundary

Sequence boundary

transgression

regression

lowstand? to transgressivesystem tracts

highstand system tract

B

Fig. 7 Idealized schemeshowing the relationshipsbetween the facies distributionin high-order (5th-order)sequences and the position ofthe corresponding 5th-ordercycles within the 4th-order sealevel variations (see also Fig. 2).Location of the sequenceboundaries and system tracts ina complete sea-level cycle isaccording to van Wagoner et al.(1990). Letters A and B labelthe two main types of sequencesin terms of their faciescomposition. A depicts thosesequences with fossiliferousreddish shales and marlstones(facies C) and mud mounds(facies F). B illustrates thosesequences that lack fossiliferousreddish shales and marlstonesand mud mounds and thatcontain mainly barren greyshales (facies A) and, dependingon their location within4th-order cycles, coralbiostromes (facies E). Comparewith Figs. 3, 6 and 8

cies B or C; see Fig. 4C) indicates a gradual decrease ofenergy, very likely tied to a gradual deepening of the envi-ronment, and thus this middle interval would constitute thetransgressive system tract of the depositional sequence. Insome depositional sequences, this middle interval passesupwards into the coarsening-upward upper interval of al-ternations, reflecting a gradual increase in the energy of theenvironment likely due to shallowing. This upper intervalclearly corresponds to the highstand system tract, althoughit cannot be precluded that the upper part of the middle in-terval could include the distal counterparts of the highstandsystem tract. This vertical arrangement of facies, and theinterpretation that we propose, contrast with those givenby Ruhrmann (1971), who studied the same stratigraphicinterval in nearby localities of the Esla Unit (located inthe same thrust sheet but also on the relative autochthon;see Fig. 1B), and by Keller (1988). Both authors describedthis interval as formed of coarsening upward sequencescomprising mud-rich deposits, comparable to our facies Band C, capped by skeletal limestones (our facies D) andconcluded that these are shallowing upward sequences re-lated to tectonic uplift. In addition, Keller (1988) strikinglymentioned that frequently the skeletal limestones have asharp, channel-like base and interpreted these limestonebodies as channel fills. Because of the vertical organiza-tion that we have described and taking into account thatno channel s.s. features have been found (i.e. we havenot found evidence of erosional conduits acting as waysof transport for long periods of time during which onlycoarse lag deposits are left), we consider that, in the caseof the Colle deposits, our interpretation is more plausi-ble. On the other hand, Keller’s interpretation fails to ex-plain why channels would suddenly form in the relativelydeep and low energy settings represented by the underlyingsediments.

Our interpretation is also backed by the mode of verticalstacking of the depositional sequences of the studied in-terval. The diagram of Fig. 6B shows that they form a setof sequences that display a systematic vertical variation of(1) sequence thickness, (2) importance of the basal skele-tal limestone packages and (3) distribution of and mutualrelationships among the three muddy facies (barren shalesof facies A, fossiliferous greenish-grey shales/marlstonesof facies B and fossiliferous reddish shales/marlstones offacies C). We speculate that these variations result from theinterplay between two or more orders of cyclicity. Follow-ing Keller and Grotsch (1990) and Keller (1997), the LaVid Group was deposited during two 3rd-order cycles, andthe highstand deposits of the upper cycle (Valporquero For-mation) record a minor lowstand (4th-order cycle?), whichis represented in the studied interval. It thus follows thatthe depositional sequences described in the studied intervalmay be attributed to 5th-order cycles.

The portions of the Valporquero Formation that underlieand overlie the studied interval crop out very patchily, solittle can be said about them. Nevertheless, Leweke (1982)found that they are mainly composed of shales, correspond-ing to his Lower Shale and Upper Shale members, respec-tively. The succession immediately below the studied inter-

val comprises several tens of metres of barren grey shales,that, mainly in the upper part, contain subordinate and thinintervals of skeletal limestones and of fossiliferous shales(see also Stel 1975). All these rocks are similar to the faciesA, B and D described above. Atop the studied interval, theValporquero Formation continues with barren grey shales(attributable to facies A), which occasionally host laterallydiscontinuous skeletal limestone beds and packages of beds(similar to facies D). Judging from these stratigraphic fea-tures and from the facies distribution across the 5th-ordersequences of the studied interval, it seems clear that thesecond 5th-order sequence (sequence #2 in Fig. 6A) rep-resents the maximum regression linked to the base of theupper lower-order (4th-order) cycle (depositional sequenceII in Fig. 6B). This second 5th-order sequence displays themost proximal character containing the thickest lowstanddeposits and the best-developed highstand tempestites. Thesuccessive overlying depositional sequences thin and fineupwards suggesting that they form a retrogradational se-quence set, which would belong to the transgressive systemtract of the lower-order (4th-order) depositional sequenceII. The overlying poorly exposed interval of the Valpor-quero Formation precludes drawing further conclusionsabout this topic. Nevertheless, this shaly interval couldrepresent at least a part of the highstand system tract ofthe 4th-order sequence, since a higher clay input has beensuggested to take place in the deeper basinal areas duringrising sea-level periods (Pasquier and Strasser 1997; Elrickand Snider 2002).

Within this scheme, the reddish marlstones/shales of thefacies C seem to occur in the 5th-order sequences associ-ated with the maximum regression/early transgression ofthe 4th-order sequences (see Figs. 6 and 7). Notably, a sim-ilar location within a sequence stratigraphy framework hasbeen postulated for marine red beds by Loydell (1998). Thisauthor points out that marine red beds appear to have beendeposited during a minor sea-level fall immediately after aperiod of very high sea-level. In our case, this would corre-spond with the 4th-order sea-level fall that takes place dur-ing the 3rd-order high sea-level. In turn, the barren shalesof the facies A occur preferentially in the lowermost anduppermost 5th-order sequences of the studied interval (Fig.6). These deposits, which are interpreted to represent thedeepest environment recorded in the area (Figs. 7 and 8),would appear in the transgressive and highstand systemtracts of those 5th-order sequences that belong to the latetransgressive and at least the early highstand parts of the4th-order sequences (Figs. 6B and 7; see also the commentsin the previous paragraph).

The sequential setting of the Synaptophyllumbiostromes and the mud mounds

Given the inferred sequence stratigraphy framework dis-cussed above, it seems clear that the Synaptophyllumbiostromal intervals and the mud mounds occur in spe-cific positions within depositional sequences and thus thattheir development was controlled, at least partially, by thefactors that determined the different orders of cyclicity and

their interplay (Figs. 7 and 8). Nevertheless, the few oc-currences of Synaptophyllum biostromes, restricted to thelower part of the section (Fig. 3), imply that conclusionsabout them must be taken with caution. On the contrary,mud mounds are relatively common in the studied suc-cession, where they show a close relationship to facies C(Fig. 3), and this makes conclusions about them to be morereliable. Synaptophyllum biostromes are probably relatedto the early transgression of a 5th-order cycle prior to themaximum regression in the boundary between two 4th-order cycles (Figs. 7 and 8A–C). On the contrary, mudmounds seem to be related to late transgressive or high-stand episodes of 5th-order cycles during the late lowstandand early transgression of a 4th-order cycle (Figs. 7 and 8Dand E).

During the 5th-order early transgression, progressivedeepening, coupled with relatively non-turbid-water con-ditions led to the onset of Synaptophyllum biostromes in anenvironment below and close to the storm wave base (Fig.8B). Later on, as clay supply was progressively re-activatedthe biostrome was killed (Fig. 8C). This process was re-peated several times through time (see unit 1 in Fig. 3).This repetition could be the result of a higher-order (6th-order?) cyclicity or merely the result of recurrent stormactivity during the early transgression. Finally, continuingtransgression brought the area to a deeper environment inwhich barren shales were deposited.

Later on a 5th-order cycle framework, during the sea-level rise and highstand (transgressive to highstand systemtracts), when the environment in the area of study wasdeeper and of lower energy compared to that existing dur-ing the development of the Synaptophyllum biostromes, andclay supply had been fully re-established, mud mounds de-veloped (Fig. 8D and E). This situation, which can clearlybe deduced from the sequential arrangement of the suc-cession, notably coincides with that inferred by Elrick andSnider (2002) for the deep-water Cambrian mud moundsof House Range (Utah, USA). Nevertheless, as said above,mud mounds do not occur in the same 5th-order sequencesas the Synaptophyllum biostromes, suggesting that tuningbetween 4th- and 5th-order cyclicity played a role in de-termining mud mound development (Fig. 7). The shalepartings of the mud mounds also record high-frequencyvariations in the environmental conditions. These varia-tions in the clay fallout rate could be of the same frequencyas the repetitions shown in the Synaptophyllum-bearing de-posits (unit 1, Fig. 3), composed of several packages, eachone consisting of a skeletal bed, overlain by a biostromeepisode, which in turn is followed by a muddy interval, andbe related to 6th-order cyclicity.

Controlling factors of cyclicity

The several orders or cyclicity here described are governedby different driving mechanisms. With respect to the lower-order cyclicity, as Johnson et al. (1985) pointed out, eustaticsea-level variations during greenhouse periods, like mostof Devonian times, must be of tectonic origin and relatedto variations in mid-ocean ridge volume or, in the case of

the third-order cycles here mentioned, to mid-plate thermaluplift and submarine volcanism. With respect to the higherorders of cyclicity, available data both on the stratigraphicinterval and on the time span involved are very limited.Classically, it has been interpreted that Palaeozoic sedi-mentation in the Cantabrian Zone and in the neighbouringWest Astur-Leonese Zone was largely governed by syn-sedimentary, extensional or strike-slip tectonics (e.g., see deSitter 1962; Evers 1967; Kullman and Schonenberg 1978;see also a review in Aramburu et al. 2004), frequently ac-companied by volcanic activity with a variable importance.Regional tectonics resulted in the individualization of up-lifted and subsiding blocks, the former acting temporarilyas shallow, starved areas with condensed sedimentation, oras emerged landmasses delivering sediments to the latter. Inthe case of the studied deposits of the Valporquero Forma-tion, several authors have interpreted that their occurrenceand location are the product of tectonics that led to the indi-vidualization of shallow areas where mainly storm-relatedprocesses were responsible of laying down high-energydeposits and also of taking skeletal debris and transportingthem elsewhere (Ruhrmann 1971; Stel 1975; Keller 1988;Keller and Grotsch 1990; Keller 1997). Specifically, Stel(1975) invoked a palaeohigh located in the South (NarceaHigh), where sediment was produced and then transportedas skeletal debris northwards to the sectors of the basinwhere the Colle section is nowadays. The other authors in-voked a southward transport of skeletal debris, with a roleof a palaeohigh located in the north (Pardomino High) and apattern of facies distribution following another palaeohighrelated to the so-called Sabero-Gordon Line.

Despite this, there is a broad agreement on the generaltectonic model for pre-Variscan times, some discrepanciesexist about several pre-Variscan features and their boundingfaults (e.g., Leon Line, Sabero-Gordon Line, Narcea High,Pardominos Ridge; see Fig. 1B and C) mainly during theDevonian. Disagreement concerns the extent, age, exact lo-cation and even existence of these features, as well as theirpotential role during Variscan (Carboniferous) times. Someauthors claim that some of them are actually Variscan orpost-Variscan features (e.g. see Marcos 1968; Aller 1986;see also the debate about the meaning of some Variscanstructures maintained by Nijman and Savage 1989, 1991;Alonso et al. 1991). Even if some of these controversialfeatures were inherited structures, they would have beensubsequently modified to such an extent that consideringthem the same structures with the same position as their po-tential precursors would be an extreme oversimplification.For instance, the so-called Sabero-Gordon Line, interpretedto be a pre-Variscan feature, strikingly cross-cuts severalVariscan thrust sheets (see Fig. 1B and C; see also Fig. 1of Keller 1988 and Fig. 2 of Keller 1997). Also, the Nar-cea and Pardominos highs fit the present-day outcrops oflower structural levels of some Variscan thrust sheets ex-humed after the emplacement of the Narcea Antiform andthe Esla Nappe, respectively (Fig. 1; see Perez-Estaun et al.1988; Alonso 1985 for a structural explanation of these twofeatures).

Fig. 8 Schematic cartoonsshowing the inferred timing ofcoral biostrome andmud-mound development alongan idealized 5th-order cycle (seetext for details). Compare withFigs. 6A and 7

Given these facts, and thus being cautious about the iden-tification and location of any potential palaeohigh, we con-clude that regional tectonics must have played a signif-icant role in the facies distribution and the vertical or-ganization of the studied deposits. The relative sea levelvariations could reflect a varying balance between sub-sidence and sediment-input rates, although we disagreewith Ruhrmann (1971), who interpreted that the shallow-ing upward sequences purely resulted from tectonic up-lift. Apart from regional tectonics, other factors could haveplayed a role, by themselves or interacting with tectonics.Cyclicity could be the result of astronomically induced cli-matic variations within the Milankovich frequency bandor of shorter period (e.g., see Keller’s 1997 discussionon the onset of the carbonate sedimentation of the LaVid Group), involving or not the potential existence ofcontinental polar ice caps (see comments in Pasquier andStrasser 1997; Immenhauser 2005, and references therein).In this sense, Elrick (1995) and Elrick and Hinnov (1996)interpreted the high-frequency cyclicity in Middle Devo-nian rocks of the eastern Great Basin (North America)to be of glacio-eustatic origin and, in the case of thehighest-frequency cycles, due to millennial climate fluctu-ations not related necessarily to waning and waxing of icecaps.

Conclusions

The studied interval of the Valporquero Formation com-prises a set of detrital and organically bound deposits laiddown in a mixed, terrigenous/carbonate ramp. The detri-tal facies are skeletal lime pack- to grainstones depositedabove the storm wave base, and mud-rich deposits (bar-ren greenish grey shales, fossiliferous greenish-grey shalesand marlstones, and fossiliferous reddish shales and marl-stones) with some thin-bedded skeletal limestone beds de-posited in a low energy environment below storm wavebase. The organically bound facies are of two types: colo-nial rugose coral biostromes forming patches a few metresacross and mud mounds a few metres wide and up to ametre thick. Both were deposited below the storm wavebase, although the latter seem to be of deeper water thanthe former.

All those facies stack vertically in a predictable manner.Analysis of this vertical structuring of facies suggests thatit reflects a series of 5th-order sequences that developedduring a minor (4th-order) sea level fall occurring in themiddle of a major (3rd-order) highstand. The style of ver-tical stacking of the 5th-order sequences and their faciescomposition is thought to result from the interplay amongthese three (3rd, 4th and 5th) orders of cyclicity. This in-terplay seems to have been a major factor in determiningthe occurrence of the two types of organic buildups, thecolonial rugose coral biostromes and the mud mounds. Theformer seem to occur in 5th-order sequences formed duringthe late 4th-order highstand, whereas the latter seem to oc-cur in 5th-order sequences developed in the late lowstandto early transgressive 4th-order system tracts. In addition,

detailed facies relationships permit to decipher the timingof development of build ups with respect to the surroundingfacies.

The colonial rugose coral biostromes take place onskeletal limestone beds and the colonies are surroundedby fossiliferous grey shales/marlstones. Field relation-ships and detailed scale (thin section) studies suggest thatbiostromes developed on a stable granular substrate (skele-tal limestone beds) that was deposited during the low-stand and early transgressive system tracts. Before clayinput resumed later in the transgression, extensive colo-nization of the sea floor by colonial rugose corals, alve-olitids, favositids and stromatoporoids took place in non-turbid, possibly sporadically agitated waters. This resultedin low relief patches (biostromes) tens of metres wide,whose development was then aborted as clay sedimentationcommenced.

The mud mounds are encased in fossiliferous reddishshales/marlstones and have a uniform facies compositionthat suggests uniform ecological conditions over the mudmounds lifetime. They are dominated by microbial micrite,and fenestellid and fistuliporid bryozoans. Other metazoans(sponges, corals and crinoids) present seem to have playeda limited role, acting as grain suppliers. Distinctively, thesemud mounds display common burrowing structures andlack stromatactis cavities. In contrast to coral biostromes,mud mounds do not seem to have any particularly particle-rich substrate requirement to develop, although theirmuddy substrate contains an appreciable amount of skele-tal material. Mud mounds formed during transgression tostillstand and they are essentially coeval with sedimentationby settling from suspensions of the clay and carbonate mudthat form the encasing shale/marlstone deposits (facies C).Nevertheless, detailed structuring of mud mounds suggeststhat clay-fallout rate varied with time giving rise to periodsof mud mound growth and encroaching on the substrateand periods of mud mound inactivity and clay veneering ofits surface. In any case, it remains obscure what factors ledmud mounds to be so clay-free if compared to the adjacentsediments deposited some metres away. Carbonate-mudsedimentation rates outpacing low clay-fallout rates and,possibly, relief-induced sheltering of the mud moundsfrom thin hyperpycnal clay-laden flows are the most likelymechanisms.Acknowledgements This paper benefited from financial supportfrom the Spanish Ministerio de Educacion y Ciencia (Search ProjectsCAICYT PB98-1563 and CGL2005-03715/BTE), from the Ger-man Research Foundation (DFG - Le-580/10.1-2) and from theIGCP-499 Project. Constructive reviews by FACIES referees E.Gischler and M. Keller are much appreciated and resulted in bet-ter organization and exposition of data and interpretations. Ro-bustness of interpretations benefited from J.R. Bahamonde com-ments. S. Kershaw kindly revised the English version of thismanuscript.

References

Aigner T (1985) Storm Depositional Systems; Dynamic Stratigraphyin Modern and Ancient Shallow-Marine sequences. Lect NotEarth Sci 3:174

Aigner T, Reineck H-E (1982) Proximality trends in modernstorm sands from the Helgoland Bight (North Sea) and theirimplications for basin analysis. Senckenberg Marit 15:87–92

Aller A (1986) La estructura del sector meridional de las unidadesdel Aramo y Cuenca Carbonıfera Central. Serv Publ PrincipadoAsturias, Oviedo, 180 pp

Alonso JL (1985) Estructura y evolucion tectonoestratigrafica de laregion del Manto del Esla (Zona Cantabrica, NW. de Espana).Institucion Fray Bernardino de Sahagun, Diputacion Provincialde Leon, 276 pp

Alonso JL, Marcos A, Pulgar JA (1991) Persistent basementwrenching as controlling mechanism of Variscan thin-skinnedthrusting and sedimentation, Cantabrian Mountains, Spain:Comment Tectonophysics 194:171–176

Alvarez-Marron J, Perez-Estaun A, Aller Manrique J, Heredia N(1990) Mapa geologico de la Hoja n◦ 79 (Puebla de Lillo)del Mapa Geologico Nacional E. 1:50.000. Segunda SerieMAGNA-Primera Edicion, Inst Geol Min Espana, Madrid

Aramburu C, Mendez-Bedia I, Arbizu M, Garcıa-Lopez S (2004)Zona Cantabrica: Estratigrafıa: La secuencia preorogenica. In:Vera JA (ed) Geologıa de Espana. SGE-IGME, Madrid, , pp27–34

Bathurst RGC (1975) Carbonate sediments and their diagenesis.Dev Sedimentol 12:658

Bosence DWJ, Bridges PH (1995) Origin and evolution of carbonatemud-mounds. In: Monty CLV, Bosence DWJ, Bridges PH, PrattBR (eds) Carbonate mud-mounds: their origin and evolution.Int Ass Sed Spec Publ 23:3–9

Boulvain F (2001) Facies architecture and diagenesis of BelgianLate Frasnian carbonate mudmounds. Sed Geol 145:269–294

Bourque PA, Boulvain F (1993) A model for the origin andpetrogenesis of the red stromatactis limestone of Paleozoiccarbonate mounds. J Sed Petrol 63:607–619

Brachert TC, Buggisch W, Flugel E, Hussner HM, Joachimski MM,Tourneur F, Walliser OH (1992) Controls of mud mound for-mation: the Early Devonian Kess-Kess carbonates of the HamarLaghdad, Antiatlas, Morocco. Geol Rundschau 81:15–44

Brouwer A (1964) Devonian biostromes and bioherms of thesouthern Cantabrian Mountains, Northwestern Spain. In: vanStraaten LMJU (ed) Deltaic and Shallow Marine deposits. DevSedimentol 1:48–53

Comte P (1959) Recherches sur les terrains anciens de la CordillereCantabrique. Mem Inst Geol Min Espana, vol 60, Madrid, 440pp

de Sitter LU (1962) The structure of the southern slope of theCantabrian Mountains. Leidse Geol Meded 26:255–264

Elrick M (1995) Cyclostratigraphy of Middle Devonian carbonatesof the eastern Great Basin. J Sed Res B 65:61–79

Elrick M, Hinnov LA (1996) Millennial-scale climate origins forstratification in Cambrian and Devonian deep-water rhyth-mites, western USA. Palaeogeogr Palaeoclimatol Palaeoecol123:353–372

Elrick M, Snider AC (2002) Deep-water stratigraphic cyclicity andcarbonate mud mound development in the Middle CambrianMarjum Formation, House Range, Utah, USA. Sedimentology49:1021–1047

Evers HJ (1967) Geology of the Leonides between the Bernesga andPorma rivers, Cantabrian Mountains, NW Spain. Leidse GeolMeded 41:83–151

Fernandez LP, Fernandez-Martınez E, Mendez-Bedia I, Rodrıguez S,Soto F (1995) Field Trip A. Devonian and Carboniferous reefalfacies from the Cantabrian Zone (NW Spain). In: Perejon A (ed)VII International Symposium on Fossil Cnidaria and Porifera,Madrid, 1995. Field Trip A. Devonian and Carboniferous reefalfacies from the Cantabrian Zone (NW Spain), Graficas VaronaSalamanca, 76 pp

Frankenfeld H (1983) El manto de Monto-Arauz: Interpretacionestructural de la Region del Pisuerga-Carrion (Zona Cantabrica,Espana). Trab Geol Univ Oviedo 13:37–47

Garcıa-Alcalde JL, Racheboeuf PR (1978) Noveaux brachiopodschonetacea du Devonien de la Cordillere Cantabrique (NordOuest de l’Espagne). Geobios 11:835–865

Garcıa-Alcalde JL, Carls P, Pardo Alonso MV, Sanz Lopez J, SotoF, Truyols-Massoni M, Valenzuela-Rıos JM (2002) Devonian.In: Gibbons W, Moreno MT (eds) The geology of Spain. GeolSoc, London, pp 67–91

Gibson MA, Clement CR, Broadhead TW (1998) Bryozoan-dominated carbonate mudmounds in a cratonic setting fromthe basal Devonian of southeastern United States. In: McMillanNJ, Kembry AF, Glass DJ (eds) Devonian of the World. CanSoc Petrol Geol Mem 14:541–552

Gradstein F, Ogg J, Smith A (2004) A geologic time scale 2004.Cambridge University Press, New York, 384 pp

Heredia N, Rodrıguez Fernandez LR, Alonso JL (1991a) Mapageologico de la Hoja n◦ 105 (Riano) del Mapa Geologico Na-cional E. 1:50.000. Segunda Serie MAGNA-Primera Edicion,Inst Geol Min Espana, Madrid

Heredia N, Rodrıguez Fernandez LR, Suarez Rodrıguez A, Alvarez-Marron J (1991b) Mapa geologico de la Hoja n◦ 80 (Buron)del Mapa Geologico Nacional E. 1:50.000. Segunda SerieMAGNA-Primera Edicion, Inst Geol Min Espana, Madrid

Immenhauser A (2005) High-rate sea-level change during theMesozoic: new approaches to an old problem. Sediment Geol175:277–296

Johnson JG, Klapper G, Sandberg CA (1985) Devonian eustaticfluctuations in Euramerica. Geol Soc Am Bull 96:567–587

Julivert M (1971) Decollement tectonics in the Variscan Cordilleraof the northwest Spain. Am J Sci 270:1–29

Keller M (1988) Die La-Vid-Gruppe - Fazies, Palaogeographie undsynsedimentare Tektonik im Unterdevon des KantabrischenGebirges (NW-Spanien). Erlanger Geol Abh 115:77–154

Keller M (1997) Evolution and sequence stratigraphy of an EarlyDevonian carbonate ramp, Cantabrian Mountains, NorthernSpain. J Sed Res 67:638–652

Keller M, Grotsch J (1990) Depositional history and conodontbiostratigraphy of the Lower Devonian La Vid Group in theLuna area (Cantabrian Mountains, NW Spain). N Jb GeolPalaont Mh 3:141–164

Kershaw S, Young GA (1997) Palaeozoic stromatoporoid andtabulate growth bands and their palaeobiological implications.Palaeontological Association 41st Annual Meeting, Cardiff1997, Abstracts

Kullman J, Schonenberg R (1978) Facies differentiation caused bywrench deformation along a deep-seated fault system (Leonline, Cantabrian Mountains, North Spain). Tectonophysics48:T15–T22

Lees A, Miller J (1985) Facies variation in Waulsortian buildups. Part2. Mid-Dinantian buildups from Europe and North America.Geol J 20:159–180

Lees A, Miller J (1995) Waulsortian banks. In: Monty CLV, BosenceDWJ, Bridges PH, Pratt BR (eds) Carbonate Mud-Mounds:their origin and evolution. Int Ass Sed Spec Publ 23:191–271

Leweke B (1982) The transition from sandy to carbonate sedimenta-tion in the Lower Devonian of the southern Cantabrian Moun-tains (La Vid Formation). N Jb Geol Palaont Abh 163:188–192

Lobato L, Garcıa-Alcalde JL, Sanchez de Posada L, Truyols J (1984)Mapa geologico de la Hoja n◦ 104 (Bonar) del Mapa GeologicoNacional E. 1:50.000. Segunda Serie MAGNA-PrimeraEdicion, Inst Geol Min Espana, Madrid

Lotze F (1945) Zur Gliederung der Varisziden in der IberischenMeseta. Geotekt Forsch 6:78–92

Loydell DK (1998) Early Silurian sea-level changes. Geol Mag135:447–471

Marcos A (1968) Nota sobre el significado de la “Leon Line”. BrevGeol Asturica 7:1–5

Marquınez J, Marcos A (1984) La estructura de la UnidadGildar-Monto (Cordillera Cantabrica). Trab Geol Univ Oviedo14:53–64

Matyszkiewicz J (1997) Stromatactis cavities and stromatactis-likecavities in the Upper Jurassic buildups at Mlynka and Zabierzow(Oxfordian, southern Poland). Ann Soc Geol Polon 67:45–55

Mendez-Bedia I, Soto F, Fernandez-Martınez E (1994) Devonian reeftypes in the Cantabrian Mountains (NW Spain) and their faunalcomposition. Courier Forsch inst Senckenberg 172:161–183

Monty CLV (1995) The rise and nature of carbonate mud-mounds:an introductory actualistic approach. In: Monty CLV, BosenceDWJ, Bridges PH, Pratt BR (eds) Carbonate Mud-Mounds:their origin and evolution. Int Ass Sed Spec Publ 23:11–48

Mutti E, Davoli G, Tinterri R, Zavala C (1996) The importance offluvio-deltaic systems dominated by catastrophic flooding intectonically active basins. Mem Sci Geol 48:233–291

Nijman W, Savage JF (1989) Persistent basement wrenching ascontrolling mechanism of Variscan thin-skinned thrusting andsedimentation, Cantabrian Mountains Spain. Tectonophysics169:281–302

Nijman W, Savage JF (1991) Persistent basement wrenching ascontrolling mechanism of Variscan thin-skinned thrustingand sedimentation, Cantabrian Mountains, Spain: reply.Tectonophysics 194:177–182

Pasquier JB, Strasser A (1997) Platform-to-basin correlation by high-resolution sequence stratigraphy and cyclostratigraphy (Berri-asian, Switzerland and France). Sedimentology 44:1071–1092

Perez-Estaun A, Bastida F, Alonso JL, Marquınez J, Aller J, Alvarez-Marron J, Marcos A, Pulgar JA (1988) A thin-skinned tectonicsmodel for an arcuate fold and thrust belt: The Cantabrian Zone(Variscan Ibero-Armorican Arc). Tectonics 7:517–537

Pickering KT, Hiscott RN, Hein FJ (1989) Deep marine environ-ments; clastic sedimentation and tectonics. Unwin Hyman,London, 416 pp

Pratt BR (1995) Origin, biota and evolution of deep-water mud-mounds. In: Monty CLV, Bosence DWJ, Bridges PH, Pratt BR(eds) Carbonate Mud-Mounds: their origin and evolution. IntAss Sed Spec Publ 23:49–123

Preat A, Mamet B, Bernard A, Gillan D (1999) Bacterial mediation,red matrices diagenesis, Devonian, Montagne Noire (southernFrance). Sed Geol 126:223–242

Ribeiro A, Quesada C, Dallmeyer RD (1990) Geodynamic evolutionof the Iberian Massif. In: Dallmeyer RD, Martınez Garcıa E(eds) Pre-Mesozoic geology of Iberia. Springer Verlag, BerlinHeidelberg New York, pp 399–409

Rodrıguez Fernandez LR, Heredia N (1988) Evolucion tectonose-dimentaria de una cuenca de antepaıs ligada a una cadenaarqueada; el ejemplo de la Unidad de Pisuerga-Carrion (ZonaCantabrica, NO de Espana). II Congr Geol Espana SGE,Granada, Simposios, pp 65–74

Ruhrmann G (1971) Riff-ferne Sedimentation unterdevonischerKrinoidenkalke im Kantabrischen Gebirge (Spanien). N JbGeol Palaont Mh 4:231–248

Schmid DU, Leinfelder RR, Nose M (2001) Growth dynamicsand ecology of Upper Jurassic mounds with comparison toMid-Palaeozoic mounds. Sed Geol 145:343–376

Schroder S, Soto F (2003) Lower Devonian (Emsian) rugose coralsfrom the Cantabrian Mountains, northern Spain. Acta PalaeontPol 48:547–558

Senowbari-Daryan B, Garcıa-Bellido DC (2002) Fossil “Sphinc-tozoa”: Chambered sponges (Polyphyletic). In: Hooper JNA,van Soest RWM (eds) Systema Porifera: A guide to theclassification of sponges. Kluwer Academic, Plenum, NewYork, pp 1511–1533

Soto F (1981) Synaptophyllum aus dem Unterdevon desKantabrischen Gebirges (Nordspanien). N Jb Geol Palaont Abh2:91–99

Soto F (1982) Synaptophyllum (Rugosa) aus dem Unterdevon desKantabrischen Gebirges (Colle, Prov. Leon). N Jb Geol PalaontAbh 163:236–238

Stel JH (1975) The influence of hurricanes upon the quiet deposi-tional conditions in the Lower Emsian La Vid Shales of Colle(NW Spain). Leidse Geol Meded 49:475–486

Tucker ME, Wright VP (1990) Carbonate Sedimentology. Blackwell,Oxford, 482 pp

Van Wagoner JC, Mitchum RM, Campion KM, Rahmanian VD(1990) Siliciclastic sequence stratigraphy in well logs, cores,and outcrops. AAPG Methods in Exploration Series No 7, 55 pp

Vera de la Puente C (1988) Estratigrafıa, Sedimentologıa yPaleogeografıa de los grupos Raneces y La Vid en la CordilleraCantabrica (Asturias y Leon). Unpublished PhD Thesis,University of Oviedo, 653 pp

Vera de la Puente C (1989) Revision litoestratigrafica y correlacionde los Grupos Raneces y La Vid (Devonico Inferior de laCuenca Astur-leonesa). Trab Geol Univ Oviedo 18:53–65

Vilas Minondo L (1971) El Paleozoico Inferior y medio de laCordillera Cantabrica entre los rıos Porma y Bernesga (Leon).Mem Inst Geol Min Espana 80, Madrid, 169 pp

Walker RG (1984) Shelf and shallow marine sands. In: Walker RG(ed) Facies models. Geol Assoc Can, Geosci Can Reprint Ser1:141–170

Wendt J, Belka Z, Kaufmann B, Kostrewa R, Hayer J (1997) Theworld’s most spectacular carbonate mud mounds (MiddleDevonian, Algerian Sahara). J Sed Res 67:424–436

Wendt J, Kaufmann B, Belka Z (2001) An exhumed Palaeozoicunderwater scenery: the Visean mud mounds of the easternAnti-Atlas (Morocco). Sed Geol 145:215–233

Young GA, Kershaw S (2005) Classification and controls of internalbanding in Palaeozoic stromatoporoids and colonial corals.Palaeontology 48:1–29