I. — Introduction

Devonian dalmanitid trilobites with articulated hypostome represent exceptional findings worldwide. This is surprising, because these trilobites possess the conterminant type (Fortey & Owens 1999a) of hypostome, attached by the hypostomal suture to the doublure. In many trilobites showing this attachment type, the hypostome is even ankylosed with the doublure (see Fortey & Owens 1999a, b). Obviously, the large dimensions of these trilobites - some specimens exceeding 40 cm in length (see Šnajdr 1987; Budil et al. 2009a) - very probably caused certain taphonomic constraint and resulted in the predisposition to a relatively rapid decomposition of their exoskeletons. This explanation is indirectly supported by the rarity of articulated exoskeletons, as well as by the extremely rare finds of enrolled specimens (see Budil et al. 2008). Even though the dalmanitids are a common component of Devonian trilobite communities in the Prague Basin (see Fig. 1), only eight specimens with hypostome attached to the doublure and/or lying very close to the ventral surface of the cephalon in more or less “living position” have been described. Three of them - two cephala with parts of articulated thoraxes and one entire but very poorly preserved specimen supplemented by the cement often used in late 19^th century in Bohemia called the “Fritsch matter” - are stored in collections of the National Museum, Prague; next one poorly preserved cephalon with attached hypostome is stored at the Museum of Comparative Zoology, Harvard University, U.S.A. All these specimens belong to the generally common species Zlichovaspis (Zlichovaspis) rugosa rugosa (Hawle & Corda, 1847) of Lower Devonian Pragian age (Budil et al. 2008, 2009a).

Figure 1

In four other specimens of this species, disarticulated hypostomes occur close to the isolated cephalon. Such cephalon with hypostome preserved nearby, but clearly disarticulated is known also in Odontochile cristata Hawle & Corda, 1847 of Pragian age (see Budil 2005; see also Fig. 3 herein). Barrande (1852, Pl. 1, figs. 1-6; Pl. 2A, figs. 13-14; see also Fig. 2 herein) figured, however, exact and correct reconstruction of the cephalic ventral side with attached hypostome of Dalmania hausmanni Brongniart, 1822 (= Odontochile hausmanni). However, it is not proven if Barrande (1852) had at disposal another, nowadays unknown specimen of Odontochile hausmanni (perhaps from some private collection, such as the count of Sacher-Masoch, see “Observation” mentioned in explanation remarks to the plates of Barrande, 1852) and/or if his reconstruction is based solely on his interpretation. Only one specimen at disposal, NML 17012 (AD Barrande, 1852; Pl. 2A, fig. 13) is an incomplete dorsal exoskeleton with ventral side covered by the rock (see also Horný & Bastl 1970; Budil et al. 2009a; Figs. 4F and 4e herein). Following the Barrande's handwriting preserved on the rock sample (Tetin) and because of the remains of red ink below the locality name (used by Barrande to mark the figured specimens), it is not excluded that Barrande (1852) used the specimen NM L30261, Zlichovaspis (Z.) cf. rugosa for the reconstructions published on the Pl. 1, fig. 4 and Pl. 2A, fig. 14 (determined by Barrande as Odontochile hausmanni), respectively. The preservation of the specimen NM L30261 is so poor (Pl. IX, G-J herein) that this mismatch is quite understandable.

The newly discussed complete specimen has a great importance, being the only one entire, well-preserved specimen of the Lower Devonian representative with hypostome preserved in situ known from the Prague Basin. The unique sample is currently stored in the Musée d'Histoire Naturelle de Lille, France (MGL). This specimen MGL92636-1 (Pl. IX, L-P) belongs to Zlichovaspis (Zlichovaspis) rugosa rugosa (Hawle & Corda, 1847) - for the reconstruction of this species, see Fig 3. Zlichovaspis (Z.) rugosa rugosa is the most common dalmanitid species in the upper half of the thickness of the Praha Formation, especially in the micritic facies, e.g. in the Dvorce-Prokop Limestone (= lowermost Emsian, see Hladil et al. 2011). The specimen from Lille (Pl. IX, L-O) comes from the locality Tetín Hill near Beroun (Bohemia; Fig. 1) - wrongly transliterated as “Fetín” at the included label (Pl. IX, P). The sample was collected apparently in the second half of the 19^th century, in the time of the most intensive mining activities and thus the acme of fossil findings at the locality. Very probably, the specimen was unknown to local palaeontologists and was sold directly to some private collection and/or directly to the museum (Lille?). One can speculate that the specimen could have been collected after the death of Joachim Barrande (1883) and Otomar Pravoslav Novák (1894), when the quarryman lost the most important customers. For the short time, the quarryman tried to sell their findings directly abroad, before the onset of important decline of collecting activities in Bohemian Devonian (Budil et al. 2009b). It could not be excluded, that the part of palaeontological samples of the Barrandian provenience, stored at the Musée d'Histoire Naturelle de Lille, were originally a part of the collection of Ch. Barrois, gathered between 1900 and 1905. During the cataloguing in the French collection, probably thanks to the missing contacts with the native speaker, the wrong transliteration of the generally well-known locality Tetín (into the “Fetín”, see Pl. IX, P) had happened.

Figure 2

Figure 3

II. — Description of the specimen

The specimen corresponds well with the diagnosis of Z. (Zlichovaspis) rugosa rugosa (see Budil et al. 2009a), showing a moderately subtriangular cephalon of length/width ratio approximately 0.55; an anterior cephalic border with a prominent concave (sag.) precranial median process; thoracic and pygidial pleural furrows deep, flat-bottomed; a pygidium rounded subtriangular in outline, of length/width ratio approximately 0.7 (estimated in the studied sample); an axis with 16-18 rings plus subtriangular terminal piece; a postaxial ridge vaulted, crossing border furrow; 13 to 14 (?15) pleural furrows; anterior pleural bands strongly vaulted, inflated distally above the inner margin of doublure; posterior pleural bands vaulted; the whole exoskeleton excluding the apodemal pits covered by dense granules, coarsest on axial rings (especially in their posterolateral parts) and postaxial region; pleural furrows very densely granulated.

The internal mould with relics of the exoskeleton is preserved in gray biomicritic limestone corresponding to the Dvorce-Prokop Limestone Facies (Lower Devonian, Pragian). However the specimen is damaged, the genal spines, eyes similarly as the posterior part of the pygidium are broken off, in addition, both cephalon and pygidium have been slightly disarticulated from the thorax during the decomposition of the specimen. The anterior cephalic part of the specimen is broken; it has been glued by a glue of poor quality; thanks to this fact, the specimen fall apart into two pieces during the manipulation. This broken-off part shows the hypostome in situ, mostly covered, however, by the rock. The rock sample is penetrated by a series of numerous subparallel fractures. Any mechanical preparation would lead to its disintegration and thus we decided not to risk this irretrievable step. The hypostome is oriented sub-parallel with the dorsal exoskeleton. It is obvious that the hypostome is still positioned very close to the doublure and its orientation is original. The surface of the hypostome body is densely pitted.

III. — Conclusions

Because of the in situ hypostome preserved under the entire specimen, which is extremely rare exoskeleton configuration in Lower Devonian dalmanitids of the Bohemian facies development (= “odontochilids”), the sample MGL92636-1 plays an important role.

The only slightly post-mortem disarticulated exoskeleton most probably represents a carcass. Together with specimens MCZ 172840, NML 30261 and NM 22578, the sample provides a direct evidence of the type of hypostome articulation (Fig. 2); which was supposed already by Barrande (1852, 1872), Šnajdr (1987) and Budil et al. (2008, 2009a) in these trilobites.

Acknowledgments. — The study was supported by grant from the Contact Mobility – Barrande Program (Project N^o MEB 021122 – PHC Barrande 24465SH) and by the Grant Agency of the Czech Republic (Project N^o P210/12/2018), respectively. We would like to express many thanks to staff of the Palaeontological Department of the National Museum, Prague, namely V. Turek and M. Valent, for very enthusiastic help in searching the possible additional Barrande's figured specimens and comparative material, far beyond the curatorial responsibilities; and T. Oudoire, the curator of the Musée d'Histoire Naturelle de Lille (France).

Abbreviations: NML, National museum, Prague, Czech Republic; MGL, Musée d'Histoire Naturelle de Lille, France; MCZ, Museum of Comparative Zoology at Harvard University Boston, U.S.A.