Introduction

The idiosyncratic Devonian Ammonicrinus, a lecanocrinid flexible crinoid, was described by Springer (1926) and discussed subsequently in comparatively few publications (Krause 1927; Ehrenberg 1930; Wolburg 1938a, b; Wanner 1943, 1954; Ubaghs 1952; Yakovlev and Ivanov 1956; Kongiel 1958; Piotrowski 1977; Moore 1978; Haude 1981; Głuchowski 1993; Hotchkiss et al. 1999; Le Menn and Jaouen 2003; Hauser 2005, Hauser et al. 2009, and Prokop 2009, see “Remarks” below), mainly from the Devonian deposits of Germany (Rhenish Massif) and Poland (Holy Cross Mountains). Ammonicrinus is distinguished by the synarthrial (bifacial) articulation on columnals with fulcra aligned and unequal ligmentary areas on either side of each fulcrum, which produced a planispirally coiled proximal column. The stem is distinguished by the abrupt xenomorphic change between the distal barrel-shaped (dististele) and the middle and proximal columnals with lateral columnal enclosure extensions (mesistele, proxistele) (Fig. 1). The mesi- and proxistele could coil and show the ability to enclose the crown.

With the exception of two other Palaeozoic genera, Myelodactylus Hall, 1852 and Camptocrinus Wachsmuth and Springer, 1897, the partly enrolled Ammonicrinus (e.g., Fig. 1A) does not correspond to the erect model of most stalked crinoids, which were attached to the substrate by a diversely designed holdfast followed by an upright stem to elevate the food-gathering system, represented by the arms, above the sea floor (e.g., Hess et al. 1999). These modifications lead to the most atypical evolutionary model among crinoids by drastically changing a “normal” crinoid crown into a “plate-encased” individual. Accordingly, this genus is easily defined by the development of the spheroidal crown hidden in an enrolled stem, which was, according to new data, either lying on soft-bottoms with long mesi- and dististele, attached with its holdfast to hard objects like brachiopod valves, corals or bryozoans, or settled completely on hard objects (e.g., brachiopods) by strongly reducing the dististele.

Fig. 1.

A. Reconstruction of the Ammonicrinus life time position (modified after Piotrowski 1977: 208, fig. 2). B. Ammonicrinus plate diagram (not to scale).

Remarks

The privately published papers of Hauser (2005) and Hauser et al. (2009) discussing Ammonicrinus contained misinterpretations. Striking in this context is his reconstruction of “A. wanneri” from isolated mesistele columnals from different individuals as a “circular sphere” (2005: 34, 38–39, fig. 5a, b). They are given no further consideration herein (also see critical comments in Bohaty and Herbig 2007).

The isolated columnals described as “A. bulbosus sp. nov. (col.)” by Prokop (2009: 162) are very similar to the isolated Lower Devonian ossicle illustrated by Hotchkiss et al. (1999: 331, fig. 2.21). These elements could not be distinguished from juvenile ossicles of A. sulcatus and are in urgent need of further research based on more complete material in order to validate “A. bulbosus”. Therefore, this species could not be considered further herein (Rudolf J. Prokop, personal communication 2009).

Institutional abbreviations.—GIK, Institut für Geologie und Mineralogie der Universität zu Köln, Germany; LPB, Laboratoire de Paléontologie de Brest (Université de Bretagne Occidentale), France; MBE, Museum für Naturkunde der Humboldt-Universität zu Berlin, Germany; MZ, Muzeum Ziemi, Warsaw, Poland; SMF, Forschungsinstitut und Naturmuseum Senckenberg, Frankfurt am Main, Germany; USNM, National Museum of Natural History (Smithsonian Institution), Washington DC, USA.

Other abbreviations.—A ray, lateral-anterior radial plate of cup; AE interray, anterior part of cup between the anterior radial plate (A) and left antero-lateral radial plate (E); CD interray, posterior part of cup between the right posterior radial plate (C) and the left posterior radial plate (D); LCEE, lateral columnal enclosure extensions.

Material and methods

Type specimens are deposited in the GIK, the Geowissen-schaftliches Zentrum der Universität Göttingen, Germany (without repository numbers), the LPB, the MBE, the MZ, the SMF, and the USNM.

In addition to a detailed analysis of previously published data, this study focuses on new material, recently discovered within the Rhenish Massif. Specimens were cleaned and dissected using micro sand-streaming methods and studied with a binocular microscope. Photographs of NH₄Cl whitened crinoids were arranged using digital image editing software. Crinoid descriptive terms follow Moore and Teichert (1978) with the following exception: measurement terms follow Webster and Jell (1999). The systematics of the mentioned cupressocrinitids follows Bohatý and Herbig (2010).

Crinoid localities and stratigraphy are given in the Appendix 1. The terms subformation and member are not synonymised sensu Steininger and Piller (1999); they are used hierarchically (Ulrich Jansen, personal communication 2005; also see Bohatý 2005a). The capitalisation of the Givetian subdivisions follows Becker (2005, 2007).

Historical background

The first report (Springer 1926) of Ammonicrinus dealt with crowns, enrolled within the mesi- and proxistele and several isolated columnals of the mesistele (Fig. 2A–D). Ammonicrinus was recognised and classified as a true crinoid fossil from the Middle Devonian of the Prüm Syncline, in the vicinity of locality 3 (Eifel, Rhenish Massif, Rhineland Palatinate, Germany). Because the dististele and the attachment were not preserved, Springer's (1926) interpretation of this remarkable new genus was mainly based on comparison with other enrolled forms, like Myelodactylus or Camptocrinus (Springer 1926: 24). Springer (1926: 24) assigned his new genus to the Camerata and to the “Hexacrinidae” with its genus Arthroacantha Williams, 1883.

With the present, large collections, it is herein recognised that Springer (1926) figured three different species; (i) A. wanneri (1926: pl. 6: 4–4b; refigured in Fig. 2A, C of the present work), (ii) a species with wider a diameter coiled stem, herein described as A. leunisseni sp. nov. (1926: pl. 6: 5–5b; refigured in Fig. 2B, D of the present work) and (iii) two isolated columnals from the mesistele of A. cf. sulcatus (1926: pl. 6: 6).

Also, the second note of an Ammonicrinus specimen (Krause 1927) was based on an enrolled crown, covered by the mesi- and proxistele. It was classified as “A. wanneri”, although the fossil differs from Springer's (1926) type material by its coiled, wide, barrel-shaped proxi- and mesistele (Fig. 2E). Krause (1927: 454) interpreted the then known individuals as crinoids with free, unstalked and possibly planktonic adult life habits.

The interpretation of a planktonic adult life style has to be rejected based on more complete specimens of the wider Ammonicrinus described by Krause (1927) as “A. wanneri” from the upper Eifelian of Sötenich (Sötenich Syncline, Eifel; locality 5). Another species, A. doliiformis Wolburg, 1938a (for 1937), from the Selscheider Formation of locality 11, was attached to brachiopod valves via an attachment disc, which, furthermore, has an attached dististele. This dististele is similar to a “normal” crinoid stem (Fig. 3A₁, A₂, B). Based on his discoveries, Wolburg (1938a: 238) correctly rejected the presumed planktonic mode of life and classified Ammonicrinus as a bottom-dweller that lived attached to hard objects. His reconstruction of A. doliiformis had the crown protruding toward the lateral-exterior, whereas the crinoid is lying exposed toward the assumed water current (Fig. 3B).

Fig. 2.

The first figures of Ammonicrinus from Springer (1926) and Krause (1927). A. Ammonicrinus wanneri Springer, 1926 (from Springer 1926: pl. 6: 4a, b). A₁, view of the extetrnal flanks of the coiled mesistele; A₂, coiled mesistele in lateral view. B. Ammonicrinus leunisseni sp. nov. (= “A. wanneri” in Springer 1926: pl. 6: 5, 5b). B₁, view of the extetrnal flanks of the coiled mesistele; B₂, coiled mesistele in lateral view. C. Photograph of the holotype of Ammonicrinus wanneri Springer, 1926, USNM-S2115; lateral view of coiled mesistele; connection between mesi- and dististele, dististele and attachment missing (see fracture surface at distal mesistele). D. Photograph of USNM-S2115, the Springer (1926) original of “Ammonicrinus wanneri” (= A. leunisseni sp. nov. herein); lateral view of coiled mesistele; connection between mesi- and dististele, dististele and attachment missing (see fracture surface at distal mesistele). E. Ammonicrinus doliiformis Wolburg, 1938a (for 1937) (= “A. wanneri” in Krause 1927: pl. 8: 4, 2). E₁, view of the extetrnal flanks of the coiled mesistele; E₂, coiled mesistele in lateral view. A–D from the Middle Devonian, Eifel Limestone; Prüm, Eifel, Germany (Springer 1926: 25); E from the Middle Devonian of Sötenich, Eifel (Krause 1927: 456). A, B, E not to scale; C, D scale bars 10 mm.

By carefully excavating a preserved crown of “A. wanneri” from locality 8 (= A. jankei sp. nov.), Ubaghs (1952) demonstrated that the crown remained enclosed within the proximal-most part of the mesistele and the proxistele and did not protrude toward the lateral exterior while feeding (Fig. 4B₂, B₆, B₇). As interpreted here, this applies solely to the stratigraphically younger ammonicrinids; but the oldest species, Ammonicrinu kerdreoletensis, is not covered entirely by the LCEE. That possibly implies feeding in the current. Ubaghs (1952) also proposed the plate diagram of the crown (Fig. 4B₅) and identified Ammonicrinus as a lecanocrinid flexibilian (1952: 204). It is confirmed herein that his second radianal plate (Ubaghs 1952: 205, fig. 1), or “supplementary plate” of Wanner (1954), is based on a frequent occuring anomaly, as already assumed by Wanner (1954: 235). This anomaly resembles the atypical gasterocomoid genera Nanocrinus Müller, 1856, Trapezocrinus Haude, 2007 and the rare Lecythocrinus Müller, 1858 that also show frequent occuring anomalies in the anal plate architecture (compare Bohatý 2006a).

Fig. 3.

The lecanocrinid species Ammonicrinus doliiformis Wolburg, 1938a (for 1937) from the Seischeid Formation of Ohle, Sauerland (Wolburg 1938a: 230). A. Casts of nearly complete specimen. Specimen attached to a brachiopod valve (right arrow) (A₁), showing the characteristic triangular connection between mesi- and dististele (left arrow) and slightly compressed mesistele (from Wolburg 1938a: pl. 17: 1); detail view of the attachment disc (arrow) (A₂), encrusting the brachiopod (from Wolburg 1938a: pl. 18: 8); detail view of the triangular connection between mesi- and dististele (arrow) (A₃) (from Wolburg 1938a: pl. 17: 6a); detail view of the coiled, slightly compressed mesistele (A₄) (from Wolburg 1938a: pl. 17: 4). B. Former assumed reconstruction of life mode, figured with a crown that protrudes toward the lateral-exterior (arrow) (from Wolburg 1938a: 240, fig. 5). C. Former assumed reconstruction of the crown (1938a: 233, fig. 4). Not to scale.

Combining the concepts of Ubaghs (1952) with the most complete specimens from Wolburg (1938a), Piotrowski (1977: 208–209, figs. 2, 3) provided the best interpretation of the mode of life of Ammonicrinus (Fig. 5). Piotrowski (1977: 208) assumed that the high specialisation of the stem provided a firm support in soft-bottom sediments and protection from water borne sediments. He (1977) also assumed that the crown was screened by an external cover so that the food could be supplied into it only by currents parallel to the bottom. “The water carrying food was introduced into the central part of the stem through a furrow formed by distal parts of external cover and the outflow proceeded through umbilical openings. During feeding the arms were presumably resting on stem plates. The contortion of crown in relation to symmetry plane of stem could facilitate water circulation inside the external cover as water current was directed by contorted crown to umbilical opening” (Piotrowski 1977: 209). Piotrowski (1977: 209) compared Ammonicrinus with the mode of life of other crinoids (e.g., calceocrinids, Meek and Worthen 1869), which were adapted to filter food from horizontal bottom-water currents.

Fig. 4.

First illustration of the actual plate diagram and definition of genus Ammonicrinus as a lecanocrinid Flexibilia by Ubaghs (1952). A. Ammonicrinus doliiformis Wolburg, 1938a (for 1937), SMF-XXIII-165a from the “Rommersheim Formation” of the Auburg, Gerolstein, Eifel, Germany (Ubaghs 1952: 220). View of coiled mesistele (A₁); view of exposed proxistele (A₂) (taken from Ubaghs 1952: pl. 3: 1, 3). B. Anomalous crown of “Ammonicrinus wanneri” from the “Rommersheim Formation” of the Steineberg, N of Kerpen, Eifel, Germany (Ubaghs 1952: 220) (= holotype of A. jankei sp. nov., no. SMF-XXIII-167a) coiled by the mesistele. View of the coiled mesistele (B₁) (Ubaghs 1952: pl. 1:3); partly excavated crown (B₂), showing radiating ridges on radiais and one slightly lobe-like enlarged appendage that possibly could support the lateral water respectively faecal-ejection (arrow) (Ubaghs 1952: pl. 1:4); excavated crown in lateral view (B₃, B₄), the second “radianal plate” respectively “supplementary plate” (see arrows) is based on an anomaly (Ubaghs 1952: pl. 2: 3, 2); plate diagram (B₅), showing the two anomalous plates (arrows) (slightly modified after Ubaghs 1952: 205, fig. 1); schematic drawing of the coiled specimen (B₆); reconstruction of the assumed living feeding position (B₇) (Ubaghs 1952: 110, fig. 2; p. 223, fig. 5). Not to scale.

Carbonate microfacies analysis within several Ammonicrinus-localities of the Eifel (especially from locality 6) and the hydrodynamic interpretation of extremely fragile but perfectly preserved bryozoans (see Ernst 2008), lead to the recognition of low-intensity current water conditions close to the soft-bottoms, temporarily yielding a lack of the horizontal water currents assumed by Piotrowski (1977). Based on this interpretation, the exigencies of a feeding method that supplemented Piotrowski's (1977) interpretation is proposed.

Morphological variability

The best and nearly completely preserved Ammonicrinus-specimens from the Rhenish Massif came from the Eifel Synclines (localities 3, 6). These specimens and additional ammonicrinids from the Sauerland (locality 11; see Wolburg 1938a and Fig. 3 of the present work) and the Bergisches Land (locality 10) have substrate-controlled morphological variability of the dististele (distal column and holdfast). Together with the material from locality 12, three “morphological groups” are recognised:

The “exposed roller-type”.—These specimens predominantly have the general skeletal morphology, as illustrated in Fig. 6. This form is herein classified as an exposed roller-type and is recognised only in the oldest known ammonicrinid, Ammonicrinus kerdreoletensis. This type, characterised by a laterally unprotected crown, would allow feeding in the current. The newly recovered material indicates that the stem of A. kerdreoletensis tapers as it approaches the crown, not in quite as many columnals perhaps but similar to that of camptocrinids, and their crown could be elevated up above the substrate. This elevation is not much but puts them above the sediment and into a possible low velocity current for feeding (Gary D. Webster, personal communication 2009). Similarly, unpublished myelodactylid specimens from the Eifelian strata of the Eifel Synclines had a similar mode of life and were also attached on hard objects, like brachiopods (unpublished data).

Fig. 5.

Schematic illustrations of Ammonicrinus sulcatus Kongiel, 1958 after Piotrowski (1977). A. Lateral cross section through the feeding crinoid (Piotrowski 1977: 209, fig. 3). B. Former reconstruction of life time position (Piotrowski 1977: 208, fig. 2). Not to scale.

Fig. 6.

Reconstruction of a feeding “exposed runner-type” of Ammonicrinus kerdreoletensis Le Menn and Jaouen, 2003, attached to a tabulate coral (model). The crown is laterally not covered by the LCEE and implies feeding in the current. The stem tapers as it approaches the crown, which was obviously elevated up from the substrate into a low velocity current for feeding. Not to scale.

The “encased roller-type”.—These specimens predominantly have the general skeletal morphology, as illustrated in Figs. 1A, 5, 7, 14A, C. This standard form is herein classified as encased roller-type and is recognised in all known ammonicrinids, except of A. kerdreoletensis. The specimens are more or less enrolled, LCEE of the proxistele and mesistele are followed by several barrel-like columnals of the dististele. The proxi- and mesistele skeleton laid on the soft-bottom, whereas the holdfast was attached to hard objects, such as brachiopod valves (Fig. 3A₂, B; also see Haude 1981: 200, fig. 12A), tabulate corals (Fig. 7) or bryozoans (Figs. 9M, 12K₁, K₂). The hard object of attachment affected either the development of an attachment disc (Fig. 3A₂, B) or variously formed radices (see Figs. 7, 12K₁). Both modes of attachment were observed in one species.

Fig. 7.

Ammonicrinus leunisseni sp. nov. A. Reconstruction of a “encased runner-type” of A. leunisseni sp. nov. attached to a tabulate coral (model); the spined specimen dwelled enrolled on the muddy seafloor. B. The original (GIK-2102) from the Eifel (locality 6, Appendix 1), Germany, Lower Givetian (Middle Devonian); showing slightly compressed proximal mesistele. A not to scale, B scale bar 10 mm.

The “settler-type”.—In addition to the predominant roller-types, rare discoveries of ammonicrinids with a reduced column length and columnal number of the dististele require further classification (Fig. 8). These are attached primarily to empty brachiopod valves that laid on a soft-bottom. These ammonicrinids did not live partly enrolled on the seafloor with the column, as recognised in the roller-types. The proximal part of the crinoid larval stage settled on top of the hard object (Figs. 8, 12L). This form is herein classified as the rare settler-type and is recognised in A. leunisseni sp. nov., A. sulcatus, and A. wanneri. Elevated above the ground, this mode of life potentially allowed the animal to profit from a low water flow above the nearly still water condition at the bottom but below the “normal” tiering levels into which associated, “regular” crinoid groups (e.g., Halocrinites inflatus [Schultze, 1866]; H. sampelayoi [Almela and Revilla, 1950]; Arthroacantha sp.) lifted their crowns for feeding. A question is why every Ammonicrinus did not adopt this form, because of presumed saving of skeletal material and the hydrodynamically advantageous feeding position above the muddy seafloor. Perhaps, this is due to the instability of the soft-bottom and the continuous input of fine sediment. Most brachiopod valves partially sank in or, respectively, became buried postmortem by sediment.

By studying the connection of the barrel-shaped columnals of the dististele and the mesistele, an interspecific morphological difference between A. doliiformis and other species (A. sulcatus, A. wanneri, and A. leunisseni sp. nov.) is recognised. A. doliiformis, a form that is only known as a roller-type, developed an uniformly constructed connection in the form of an idealised triangular-shaped, wide columnal-plate between the columnals of the mesistele, with a LCEE and the barrel-like columnals of the dististele (Fig. 3A₁, A₃). In this connection, this species obviously has to be characterised as a relatively constant form, and it developed the most voluminous skeleton of all known ammonicrinids. The wide, triangular-shaped columnal-plate can be used for interspecific differentiation between A. doliiformis and the other species.

Fig. 8.

Reconstruction of a spined “settler-type” of Ammonicrinus leunisseni sp. nov., attached to a brachiopod brachial valve (Schizophoria sp.); the original (GIK-2103) from locality 6 is figured in Fig. 12L. Not to scale.

In contrast, A. sulcatus, A. wanneri, and A. leunisseni sp. nov. had variously developed connections of the dististele and the mesistele. The distal-most columnal of the mesistele may exhibit an abrupt connection between those ossicles, distinguished by LCEE and the barrel-shaped columnals of the dististele by developing an elongated triangular-shaped ossicle (rare) or a single barrel-like appendage toward the dististele (Fig. 10G–I). However, this barrel-like appendage can also be duplicated and directed both, to the dististele and the mesistele (Fig. 10J, K), and a sequence of intermediate shaped ossicles is possible.

The development of all morphologies obviously depends on the hardground on which the crinoids were attached. This intraspecific variability is recognised in A. sulcatus, A. wanneri, and A. leunisseni sp. nov.—all species with the ability to exhibit the encased roller- or the settler-type. That recognition affected Piotrowski's (1977: 214, table 3) interspecific separation of “A. kongieli” and A. sulcatus, which is mainly based on the development of either abrupt connection between columnals, distinguished by LCEE and barrel-like columnals or barrel-like plates with extensions. Therefore, and because of the recognised intraspecific variability of the ossicular sculpturing, “A. kongieli” is declared a subjective junior synonym of A. sulcatus.

Systematic palaeontology

Springer (1926: 23) originally classified Ammonicrinus with its type species A. wanneri as a possible member of the subclass Camerata Wachsmuth and Springer, 1885, family Hexacrinitidae Wachsmuth and Springer, 1885 (“Hexacrinidae” 1926: 23) and mentioned the similarities to Camptocrinus. Both assumptions were confirmed by Wolburg (1938a), who erected the species A. doliiformis. This assumption was rejected by Bassler (1938) and Moore and Laudon (1943), who placed Ammonicrinus in the “subclass Inadunata”, family “Heterocrinidae” (Bassler 1938) or “Iocrinidae” (Moore and Laudon 1943). Ubaghs (1952), who first dissected an A. wanneri crown from the surrounding stem and, therefore, was the first to demonstrate that Ammonicrinus is a true member of class Crinoidea Miller, 1821 (see Wanner 1954: 231). Ubaghs (1952) assigned the genus to the subclass Flexibilia Zittel, 1895, order Sagenocrinida Springer, 1913 and “family Lecanocrinidae Springer, 1913”, whereas Wanner (1954: 231) identified the exceptional position of Ammonicrinus within the subclass because of its bent crown and the atrophy of the two anterior basais and hypertrophy of the anterior and left anterolateral radial plate. Within the Crinoid Treatise (see Moore 1978), Ammonicrinus was finally assigned to the superfamily “Lecanocrinacea” (= Lecanocrinoidea Springer, 1913 sensu ICZN) and family Calycocrinidae Moore and Strimple, 1973, characterising lecanocrinids with bilateral symmetry in the plane bisecting the CD interray and the A ray or AE interray, as well as crowns distinctly bent on the stem or the stem coiled around the crown (Moore 1978: T783–T784).

Fig. 9.

Lecanocrinid Ammonicrinus species. A–J. Ammonicrinus wanneri Springer, 1926. A–I. From the Eifel (locality 3, Appendix 1), Germany, Lower Givetian (Middle Devonian). J. From the Eifel (locality 7, Appendix 1), Germany, Lower Givetian (Middle Devonian). A. Lateral view of a partly preserved specimen (GIK-2133) with coiled mesistele. B. Lateral view, respectively view of external columnal flanks of the coiled mesistele of a partly preserved specimen (GIK-2134) with one preserved, postulated cup ossicle (arrow). C. View of external columnal flanks of the mesistele of a partly preserved specimen (GIK-2135). D. Lateral view, respectively view of external columnal flanks of the coiled mesistele of a partly preserved specimen (GIK-2136), showing typical LCEE. E. View of external columnal flanks of a nearly uncoiled mesistele (“runner-type”) (GIK-2137). F. View of external columnal flanks and LCEE of a slightly compressed, coiled mesistele (GIK-2138). G. View of external columnal flanks of a nearly uncoiled mesistele (“runner-type”) (GIK-2139). H. View of external columnal flanks of the mesistele of a partly preserved specimen (GIK-2140). I. View of external columnal flanks of a nearly uncoiled mesistele (“runner-type”) (GIK-2141). J. View of external columnal flanks of the coiled mesistele of a weathered specimen (GIK-2142) on matrix. K–O. Ammonicrinus doliiformis Wolburg 1938a (for 1937). K, L. From the Eifel (locality 9, Appendix 1), Germany, upper Eifelian (Middle Devonian). M. From the Bergisches Land (locality 10, Appendix 1), Germany, Eifelian/Givetian threshold (Middle Devonian). N. From the Eifel (locality 4, Appendix 1), Germany, upper Eifelian (Middle Devonian). O. From the Eifel (locality 5, Appendix 1), Germany, upper Eifelian (Middle Devonian). K. Lateral view of a coiled specimen (GIK-2143) with lost dististele and cracked LCEE of the mesistele, exposing the coiled proxistele and several cup ossicles (arrow). L. Lateral view of a nearly completely coiled specimen (GIK-2144) with lost dististele and cracked LCEE of the mesistele, exposing distal-most part of the coiled proxistele and several cup ossicles (arrow). M View of external columnal flanks of a preserved, coiled mesistele (GIK-2145) on matrix; the imprint of the uncoiled distal mesistele (“runner-type”), of the dististele and of the holdfast, which is attached to a fenestrate bryozoan (imprint, see arrow), is traced by a dashed line. N. Facet view of a coiled, adult specimen (GIK-2146) with exposed distal part of the proxistele and disarticulated remains of the arms (arrows). O. Perfect, three dimensionally preserved, adult specimen (MB.E.-287, original of Krause 1927). Coiled mesistele in lateral view (O₁), dististele, attachment and spines missing, centres of tuberculated radiais partly visible (arrow); the specimen is infested by a (?)craniid brachiopod (arrow on the left); opposite lateral view (O₂), centres of radiais partly visible (arrow); oblique lateral view (O₃); view of the external flanks of the mesistele (O₄) (centre and upper part of the figure) and of the facet area of distal mesistele (below), showing wide barrel-shaped outline. Scale bars 10 mm.

Fig. 10.

The lecanocrinid species Ammonicrinus sulcatus from locality 1 (A–G, I–Q) and 2 (H) (see Appendix 1). A–D. Facet views of GIK-2104–2107, showing nodular tubercles and spine-tubercles on exterior flanks of the columnals of the mesistele. E. Facet view (E₁) and view of the exterior flank of a specimen (E₂) (GIK-2108), showing tubercles and spine-tubercles on exterior flank of the columnal of the mesistele. F. Facet view of a specimen (GIK-2109), showing tubercles and spine-tubercles on exterior flank of the columnal of the mesistele. G. Facet view of a strongly sculptured columnal (GIK-2110) of the distal-most mesistele, showing connection to the dististele. H. Facet view of a columnal of the distal-most mesistele (GIK-2111), showing long LCEE and connection to the dististele. I. Facet view of a columnal of the distal-most mesistele (GIK-2112), showing relatively long LCEE and connection to the dististele. J. Interior view of a distal-most, barrel-like columnal of the mesistele (GIK-2113) with LCEE. K. Interior view of a distal-most, barrel-like columnal of the mesistele (GIK-2114), with partly preserved LCEE. L. Facet view of a juvenile distal columnal of the mesistele (GIK-2115) with nodular tubercles on exterior flank and on LCEE. M–N. Iuvenile columnals of the proximal mesistele in facet views, showing well developed nodes on exterior flanks. M. GIK-2116. N. GIK-2117. O. Facet view of a juvenile distal columnal of the mesistele (GIK-2118) with nodular tubercles on exterior flank and on LCEE. P. Lateral view (P₁) and view of the exterior flank (P₂) of the partly preserved mesistele (GIK-2119); the specimen shows nodular tubercles, spine-tubercles and a few partly preserved spines (arrow). Q. Facet view (Q₁) and lateral view (Q₂) of a cracked, coiled mesistele (GIK-2120), showing several tuberculated and concave ossicles of the cup (arrows). Scale bars 10 mm.

Fig. 11.

The lecanocrinid species Ammonicrinus kerdreoletensis Le Menn and Jaouen, 2003 (GIK-2121) from Vireux-Molhain (locality 12, Appendix 1), France, lower Eifelian (Middle Devonian); lateral view of long mesistele, proxistele and huge cup (arrow) on matrix. Scale bar 10 mm.

Fig. 12.

The lecanocrinid species Ammonicrinus leunisseni sp. nov. A–E, I, J, L. From the Eifel (locality 6, Appendix 1), Germany, Lower Givetian (Middle Devonian). F, G, K. From the Eifel (locality 3, Appendix 1), Germany, Lower Givetian (Middle Devonian). H. From the Eifel (locality 9, Appendix 1), Germany, upper Eifelian (Middle Devonian). A. Lateral view of a specimen (GIK-2122) with lost spines, showing complete coiled mesistele and one preserved columnal of the dististele (arrow). B. Lateral-facet view of a specimen with lost spines (GIK-2123), showing coiled mesistele and proxistele. C. View of the exterior columnal flanks of a slightly compressed specimen (GIK-2124) with lost spines, showing proxistele and mesistele with one distal-most, barrel-shaped columnal with LCEE (arrow). D. View of the exterior columnal flanks of a weathered and compressed specimen (GIK-2125) with lost spines, showing part of the mesistele and proxistele and rest of disarticulated ossicles of the cup preserved. E. Lateral view of a partly preserved specimen (GIK-2126) with lost spines and well preserved spine-tubercles on the coiled mesistele. F. View of the exterior columnal flanks of a partly preserved, coiled mesistele (GIK-2127) with lost spines and one radial plate preserved (arrow). G. View of the exterior columnal flanks of a partly preserved, uncoiled mesistele (GIK-2128) with lost spines. H. Interior view of a partly preserved, coiled specimen (GIK-2129), showing rest of cup and impressions of the lost arms (arrow). I. View of the exterior columnal flanks of an uncoiled specimen (GIK-2130) on matrix (“runner-type”), showing several preserved spines on partly preserved mesistele and dististele and developed radices on columnals of the dististele (arrow). J. A specimen on matrix with well preserved spines (GIK-2131). View of the exterior columnal flanks (J₁) with coiled proximal-most mesistele and proxistele and uncoiled distal column (“runner-type”) with one barrel-shaped columnal showing short LCEE (arrow on the right); the specimen shows numerous preserved spines on the mesistele; one radial plate is visible (arrow on the left); aboral view of proxistele and base of cup (J₂). K. Specimen GIK-2132. Isolated holdfast (K₁) attached to a fenestrate bryozoan (arrow); view of the exterior columnal flanks of uncoiled mesistele (K₂) on matrix (“runner-type”). L. Coiled specimen (GIK-2103), attached on a brachiopod brachial valve (Schizophoria sp.) (compare to reconstruction, figured in Fig. 8); the specimen strongly reduced the dististele and settled with an attachment disc on the brachiopod (“settler-type”). Scale bars 10 mm.

Postmortem epizonal encrusting

Articulated and isolated ossicles from the localities 1–2 have diverse, postmortem epifaunal encrustation, which infested nearly every hard object lying on or settling within the soft or moderately stabilised, muddy firmground. The following groups are identified:

Brachiopoda.—The specimen of A. doliiformis in Krause (1927; refigured in Figs. 2E, 9O of the present work) was infested by a (?)craniid brachiopod (Fig. 9O₁). The specimen settled on the exterior side of the former movable mesistele, on top of several spine-tubercles with lost spines. This is clear evidence of an immediate postmortem encrusting.

Bryozoa.—The following bryozoans were identified on skeletal remains of A. sulcatus:

Trepostomata: One pluricolumnal and one isolated columnal of the mesistele (GIK-2147, Fig. 13A and GIK-2149, Fig. 13C) were encrusted postmortem by the trepostome bryozoan Leptotrypella Vinassa de Regny, 1921. An additional pluricolumnal of the mesistele (GIK-2150, Fig. 13D) was also encrusted postmortem by the trepostomate bryozoan, Eostenopora Duncan, 1939. Trepostome bryozoans were also reported attached to the crown ossicles of the cladid crinoid family Cupressocrinitidae Roemer, 1854 (compare to Bohatý 2009). One brachial of a completely preserved Halocrinites nodosus crown (Sandberger and Sandberger, 1856) (Bohatý 2009: fig. 2.8), one cup of an also entire H. schreueri crown (Bohatý, 2006b) (see Bohaty 2009: fig. 11.4) and one theca of Procupressocrinus gracilis (Goldfuss, 1831) (Bohatý 2009: fig. 11.6) were encrusted postmortem by (?)Eostenopora sp. The boring trace of an affected arm of Robustocrinites cataphractus Bohatý, 2009 was also populated by (?)Eostenopora sp. (Bohatý 2009: figs. 6.3, 7.2).

Cystoporata: The erect pluricolumnal of the distal mesistele (GIK-2148, Fig. 13B) was encrusted by the cystoporate bryozoan Eridopora Ulrich, 1882. As strong evidence for a postmortem encrusting, the bryozoan encrusted the external and internal region of the ossicles. Another cystoporate bryozoan, Cyclotrypa Ulrich, 1896, is recognised on one columnal (GIK-2152, Fig. 13F) and one pluricolumnal (GIK-2153, Fig. 13G) of the mesistele.

Fenestrata: One isolated mesistele columnal (GIK-2155, Fig. 13I) was encrusted postmortem by a holdfast of an undetermined fenestrate bryozoan. Bohatý (2009: fig. 11.1) reported stems of Halocrinites geminatus (Bohatý, 2005b) and Procupressocrinus gracilis, which were encrusted by fenestrate bryozoans. The length of the overgrown pluricolumnals, as well as some observed embedding patterns of bryozoans located underneath the attached stem, allows the presumption of a premortem settlement (compare to Bohatý 2005b: fig. 3B). In contrast, some shorter stem fragments or other disarticulated cupressocrinid ossicles (see Bohatý 2009: fig. 11.2) were usually encrusted postmortem. This assumption is based on the entire enclosure of some skeletal elements. Similarly, holdfasts of probable rhomboporid bryozoans attached to the columnals of Schyschcatocrinus creber Dubatolova, 1975, as reported by Gluchowski (2005: fig. 3A, B). Gluchowski (2005) indicated that the bryozoans lived attached to the fragmented dead stems that lay horizontally on the sea floor. Strong evidence for the settlement of a living stem of Cupressocrinites hieroglyphicus (Schultze, 1866) is given by Bohatý (2009: fig. 11.16–18). The example is encrusted by the holdfast of a fenestrate bryozoan (Cyclopelta sp.) that grew all around the column without contact to the crenularium. The reticulate bryozoan colony surrounded the stem, whereas the dissepiments built concentric rings characteristic for this genus.

Microconchida.—One isolated columnal of the mesistele of A. sulcatus (GIK-2155, Fig. 13H) was encrusted by two microconchid-valves, which settled postmortem at the facet region of the ossicle, below and above the crenularium. Microconchids with unstructured or sculptured valves frequently encrusted the ossicles of cupressocrinids from the Middle Devonian of the Eifel, as reported by Bohatý (2005b, 2006b, 2009). It is remarkable that larger individuals are rare and isolated (compare to Bohatý 2006b: pl. 5: 8), whereas numerous smaller microconchids encrusted the crinoids (see Bohaty 2009: figs. 2.6, 11.7, 8). As assumed for Ammonicrinus, the microconchid colonisation of the cupressocrinid remains occurred immediately postmortem. The single-species encrusting of microconchids on the columnals of Tantalocrinus scutellus Le Menn, 1985 and Schyschcatocrinus creber represent additional settlement examples (Głuchowski 2005: 323, fig. 5I–L).

Crinoidea.—The pluricolumnal of A. sulcatus (GIK-2151, Fig. 13E) was encrusted postmortem by a crinoid holdfast, which settled on several tubercles with lost spines. Another A. sulcatus pluricolumnal (GIK-2150, Fig. 13D) was encrusted postmortem by a trepostomate bryozoan that was then infested by a small crinoid attachment disc. Gluchowski (2005: 322) documented the postmortem encrustation of several small crinoid holdfasts attached to upper Eifelian crinoid columnals. Various attachments of crinoid juveniles to living or dead adults are known from the Silurian to the Mississippian (see Meyer and Ausich 1983). Coiling stems, modified discoid holdfasts on the columns of crinoid hosts, as well as dendritic holdfasts distributed on all sides of the column, were reported from Silurian strata by Franzen (1977) and Peters and Bork (1998). Furthermore, Bohatý (2009) reported crinoid holdfasts attached to the crown ossicles of different cupressocrinids. One cup of Halocrinites schlotheimii schlotheimii Steininger, 1831 (Bohatý 2009: fig. 11.9) and one isolated radial and arm plate of H. geminatus were encrusted by the hold-fasts of other cladid crinoids (?Procupressocrinus gracilis).

Chaetitida.—One weathered pluricolumnal of A. sulcatus was encrusted by Chaetitida indet. (unfigured material). The encrustation occurred postmortem, because the chaetitid settled on the external and internal regions of the ossicles. Bohaty (2009) mentioned H. s. schlotheimii cups, which were completely encrusted by indeterminable stromatoporoids. These encrustations were settled again by chaetetids.

Palaeoecological hypothesis

Two interpretations of the spinose endoskeleton

Exterior protection.—The distribution of the articulated spines on the skeleton could indicate the need for protection from vagile benthic “predators”. For example, numerous platyceratid gastropod conchs are preserved in associated faunas. Moreover, syn-vivo encrustation by epizoans was effectively prevented. In contrast, the ossicles of associated stalked crinoids are variously bored and pre- and postmortem infested by diverse organisms.

Interior protection.—The spinose pattern could also have protected the crown, when exposed by partial opening of the enrolled proximal stem. Fine articulated spines served as a skeletal micromesh or “bow net” sensu Haude (1981: 199, 200, fig. 12B). Nutrient particles could pass to the arms, whereas the penetration of potential “predators” or larger sediment particles was prevented from entering the vital crown elements.

As a soft-bottom dweller within non-turbulent muddy habitats, two further aspects need to be interpreted: (i) tolerance against sedimentary material, clogging the filtration fan; (ii) the question of the feeding mode under low-intensity current water conditions.

Fig. 13.

Postmortem epizoan encrusting on disarticulated columnals of the lecanocrinid crinoid species Ammonicrinus sulcatus Kongiel, 1958. A–G. From the Eifel (locality 1, Appendix 1), Germany, upper Eifelian (Middle Devonian). H, I. From the Eifel (locality 2, Appendix 1), Germany, upper Eifelian (Middle Dvonian). A. View of external flanks of a pluricolumnal of the mesistele (GIK-2147), encrusted by a trepostomate bryozoan (Leptrotrypella(?) sp.) (arrows). B. Internal view of a pluricolumnal of the distal-most mesistele (GIK-2148), encrusted by a cystoporate bryozoan (Eridopora(?) sp.) (arrows). C. Facet view of an isolated, distal-most columnal of the mesistele (GIK-2149), encrusted by a trepostomate bryozoan [Leptrotrypella(?) sp.] (arrows). D. External flanks of a pluricolumnal of the mesistele (GIK-2150), encrusted by a trepostomate bryozoan (Eostenopora(?) sp.) (see arrows below); the bryozoan is infested by a crinoid attachment disc (arrows in D₂); general view (D₁), detail view (D₂). E. View of external flanks of a pluricolumnal of the mesistele (GIK-2151), encrusted by a crinoid holdfast (arrow). F. Facet view of an isolated columnal of the mesistele (GIK-2152), encrusted by a cystoporate bryozoan (Cyclotrypa(?) sp.) (arrows). G. Facet view of a pluricolumnal of the mesistele (GIK-2153), encrusted by a cystoporate bryozoan (Cyclotrypa(?) sp.) (arrows). H. Facet view of an isolated columnal of the mesistele (GIK-2154) (H₁), encrusted by microconchid valves (see arrows in detail view (H₂). I. Facet view of an isolated columnal of the mesistele (GIK-2155), encrusted by a holdfast of a fenestrate bryozoan (arrow). Scale bars 10 mm.

Fig. 14.

A. Schematic reconstruction of Ammonicrinus (strongly modified after Piotrowski 1977: 209, fig. 3) of a feeding Ammonicrinus within low-intensity current water. Alternating water pressure was possibly generated in the interior of the enrolled proximal stem by slow, bellow-like partial opening and closing (red arrows) of the base of the central mass; due to the synarthrial (bifacial) articulation of the ammonicrinid mesistele that developed two huge ligamentary facets (orange), separated by the fulcrum, bellow-like partial opening could possibly enabled by stiffening of the outer ligaments (see orange bars in A₁); closing could be controlled by stiffening of the inner ligaments (see orange bars in A₂). A₁, suction during opening may result from low-pressure (P^-) and create an ingesting water flow (blue arrow); A₂, ejection during closure (red arrow) resulted from overpressure (P⁺); to minimise faecal recycling, the water ejection may have occurred laterally (blue arrows), feasibly at both lateral centres, which have “openings”. B. Lobe-like enlarged appendages (framed in red) could possibly support the lateral water faecal-ejection (modified from Ubaghs 1952: pl. 1: 4). C. Reconstruction of a feeding “encased runner-type” of A. leunisseni sp. nov., attached to a tabulate coral (model); the spined specimen dwelled enrolled on the muddy seafloor; alternating water pressure was obviously generated in the interior of the enrolled proximal stem globe by non-muscular, probably MCT-controlled, slow, bellow-like partial opening and closing of the oblate sphere at its bottom (dashed arrow); suction during opening created an ingesting water flow (see arrow on the left), which was funnelled in a “canal”, formed by the unspined interior of the columnals of the mesistele, whose U-shaped LCEE additionally formed a protection against immersive sediment; ejection during closure resulted from overpressure; to minimise faecal recycling, the water ejection occurred supposably laterally, feasibly at both lateral centres, which accordingly show “openings” (see arrows on the right). Not to scale.

Conclusions

Because of the high variability of the substrate-controlled dististele and attachment that strongly affected the overall form of the endoskeleton, Ammonicrinus has to be characterised as a lecanocrinid distinguished by high morphologic plasticity. This is mainly expressed by the two recognised main forms, the roller- and the settler-type. As bottom-dwellers on more or less muddy firmgrounds or, in particular, on mud substrates, ammonicrinids benefit from this constructional plasticity, which affords anchoring on different hard objects that are lying on the soft-bottom. Radices, observed in a few ammonicrinids, could additionally stabilise the individuals.

The younger ammonicrinids from the Rhenish Massif, the presumed soft-bottom dwellers, especially in in low-intensity current water, requires two main conditions: (i) It is apparently necessary to protect the crown by encasing it by the proximal mesistele. Furthermore, attacks from vagile benthic organisms presumably were anticipated with articulated spines. (ii) The hypothesis of a slow “stem pumping mechanism” could have possibly resulted in a self-generated water flow for feeding and out-pumping of excretory products as well as antagonising sedimentary material. This was possibly enabled by slowly stiffening and relaxation of mutable connective tissues of the mesi- and proxistele. Fossil indications for this hypothesis are: (i) the crown remained enclosed within the stem and did not protrude toward the lateral exterior; (ii) the interior of the columnals of the mesistele, whose U-shaped LCEE additionally formed a protection against immersive sediment, formed a “canal” for the incurrent water flow; (iii) the lateral-most radiais of several species have a lobe-like enlarged appendage that could possibly support the lateral water faecal-ejection; and, (iv), the mesistele developed two huge ligamentary facets, which are separated by the fulcrum. However, it is important to note that this assumed ability does not imply that every ammonicrinid imperatively feeds via “MCT-pumping”. In the same muddy still water habitats that were populated by the roller-type, the settler-type is recognised. This mode of life potentially profited from a low water flow above the nearly unmoved condition at the sediment water interface. Carbonate microfacies analysis within several Ammonicrinus-localities of the Eifel indicated former muddy firmgrounds and moving water conditions in which ammonicrinids could passively benefit from water current.

Fig. 15.

Schematic sketches of different LCEE of the mesistele in uncoiled (above) and coiled positions (below), indicating evolution of perfecting the crown-encasing in coiled position by modifying the extensions form Emsian to Givetian. A. Lateral view of Ammonicrinus kerdreoletensis Le Menn and Jaouen, 2003, showing similar shaped columnals with very short LCEE; thus, the crown is laterally nearly unprotected in coiled position. B. Lateral view of Ammonicrinus wanneri Springer, 1926, with lengthened LCEE of the similar shaped columnals, which lattice-like guarded the crown in coiled position. C. Lateral view of Ammonicrinus sulcatus Kongiel, 1958, showing smaller columnals of the mesistele, which are interconnected with longer ones and afford lateral density of the coiled stem. D. Lateral view of Ammonicrinus doliiformis Wolburg, 1938a (for 1937), showing regularly or irregularly arranged columnals with longer and shorter LCEE, which were interconnected with several columnals showing broadened convex and concave extensions that could interlock in coiled position.

Observations within the Eifel Synclines indicate that the Ammonicrinus morphology of the coiling of the stem, respectively encasing of the crown, was brought to perfection during the upper Eifelian. The oldest form, A. kerdreoletensis, has a relative huge crown in relation to the narrow mesistele, which is composed of narrow, similarly shaped columnals with very short extensions (Fig. 15A). Thus, the crown is nearly unprotected laterally in the resting position of the crinoid and, especially, in the feeding position, which implies feeding in the current and has similarities to the feeding position of camptocrinids and myelodactylocrinids. Younger ammonicrinids encased the crown with modified columnals of the mesistele in a resting- but, herein assumed, also in a feeding position; A. wanneri lengthened the LCEE of the similarly shaped columnals of the mesistele, which encased the crown in the coiled position (Fig. 15B). The developments of smaller columnals of the mesistele, which are interconnected with regular ones, are an advanced or evolved step to afford increase lateral density of the coiled stem. This morphology is recognised in A. sulcatus (Fig. 15C). In A. doliiformis, the LCEE of the mesistele is composed of characteristically regularly or irregularly arranged columnals with longer and shorter extensions, which were interconnected with several columnals showing broadened convex and concave extensions that could interlock in coiled position (Fig. 15D). Especially within the Eifel and the Holy Cross Mountains, the diversity and frequency of vagile benthic “predators” like platyceratid gastropods increases during the middle and upper Eifelian reaching a maximum toward the Eifelian/Givetian boundary (own unpublished data; see e.g., Gahn and Baumiller 2003 for Middle Devonian crinoid/platyceratid interactions). The necessity to increase the ammonicrinid crown protection could speculatively be linked to this ecological circumstance.