A NEW STYLOPHORAN ECHINODERM 



77 



and decreases rapidly in width distally and the posterior end of the 

 posterior zygal plate is comparatively wider than in Juliaecarpus. 



Suranal plate and stylophoran orientation 



The occurrence of a well developed suranal plate in Juliaecarpus 

 and, perhaps, also in Beryllia and Domfrontia (see discussion above) 

 requires further comments on the orientation of comutes (including 

 comute-like ankyroids) with respect to mitrates. Workers supporting 

 the chordate interpretation of stylophorans (synthesis in Jefferies, 

 1986) homologize the plano-concave surface of the mitrate theca 

 with the dorsal side of the comute theca. Given this orientation, it 

 may be argued that the suranal plate is homologous with the subanal 

 plate observed in such early mitrates as Chinianocarpos thorall 

 Ubaghs, 1961, Peltocystis cornuta Thoral, 1935 and, perhaps, 

 Vizcainocarpus dentigerRuta, 1997 (see also Ubaghs, 1968, 1969). 

 In these mitrates, the subanal plate occupies the posterior part of the 

 plano-concave surface of the theca and floors the anal opening. As in 

 the case of the suranal plate, there are indications that, at least in 

 Chinianocarpos and Peltocystis, the subanal plate v/as flexibly ar- 

 ticulated with the theca (Ubaghs, 1969). 



In their discussion of the phylogenetic position of Reticulocarpos 

 and the origin of mitrates, Jefferies & Prokop ( 1972) proposed that 

 one of the supracentralia lying close to the mouth opening in derived, 

 mitrate-like comutes increased in size and gave rise to either the 

 large subanal of such primitive mitrates as Chinianocarpos and 

 Peltocystis (and, perhaps, Vizcainocarpus) or the rearmost element 

 of the plano-concave surface of other mitrates (e.g. mitrocystitids 

 and anomalocystitids; see also Ruta 1998, in press). 



However, as already explained in the introduction, the bulk of 

 anatomical evidence (arguments in Ubaghs, 1981,Kolataf'fa/., 1991, 

 Parsley, 1991, 1997, Sumrall, 1 997 and Lefebvre e/ «/. , 1998), based 

 mainly on a comparison between the aulacophore, auoral plates, 

 apophyses and zygal/septum complex of comutes and the like-named 

 stmctures of mitrates, strongly supports the convex-side-up orienta- 

 tion of the mitrates as well as the homology between the convex 

 surface of their thecae and the dorsal side of the comute thecae. 



Therefore, I regard the suranal plate oi Juliaecarpus as a modified 

 posterior dorsal integument plate, not homologous with the subanal 

 plate of certain primitive mitrates. The suranal plate was perhaps not 

 unique to Juliaecarpus. I tentatively suggest that its occurrence may 

 represent a shared derived feature of a group of ankyroids including 

 Juliaecarpus, Beryllia and Domfrontia. From the morphological 

 comparisons outlined above, it is clear that these taxa show certain 

 skeletal features in common. Enlargement of supracentralia on the 

 posterior part of the dorsal integument (forming or not an almost 

 regular spatial arrangement) and reduction or loss of the most 

 posterior pair of marginalia are other possible shared characters of 

 this group. A close relationship of these ankyroids with the clade 

 (Hanusia + {Reticulocarpos + Prokopicystis)) ( all these genera share 

 a dorsal bar; see Parsley, 1997, 1998) is possible, considering the 

 overall similar proportions and morphological characters of the 

 theca and aulacophore in Juliaecarpus and Reticulocarpos. 



However, I point out that, according to Parsley's (1997, 1998) 

 phylogenetic analysis, Beryllia is the sister taxon to a clade compris- 

 ing Chinianocarpos and the peltocystids (including Peltocystis as 

 the most basal representative of this group of mitrates; see also 

 Jefferies, 1973, 1986 and Jefferies & Lewis, 1978). In addition, 

 Parsley (1997. 1998) did not include in his analysis Domfrontia, 

 ranked among the poorly known or incomplete taxa. Clearly, the 

 mixture of primitive and derived features found in Juliaecarpus 

 requires a substantial reassessment of the polarity of a number of 

 characters. 



Morphological variation in the zygal bar 



Among the ankyroid stylophorans, the genera Beryllia, Domfrontia, 

 Prokopicystis and Reticulocarpos differ from other representatives 

 of this group in the morphology of their zygal bar. The unusual 

 configuration of this structure in the above-mentioned taxa deserves 

 further considerations. 



Reticulocarpos, Prokopicystis and Domfrontia were reconstructed 

 as having an incomplete zygal bar (Jefferies & Prokop, 1972; Cripps, 

 1989fl; Cripps & Daley, 1994). In particular, Reticulocarpos seems 

 to have had a small somatic plate (posterior zygal plate) attached to 

 the posterior zygal process of M',. In Cripps' ( 1989fl) reconstruction 

 of Prokopicystis, only the posterior zygal process of M\ seems to 

 have been present and no trace of a posterior plate attached to it has 

 been recorded. In their reinterpretation of Domfrontia, Cripps & 

 Daley ( 1 994) reconstructed the zygal bar of this stylophoran as being 

 similar to that of Reticulocarpos. 



However, with the possible exception of Reticulocarpos, the 

 incompleteness of the zygal bar in Domfrontia and Prokopicystis is 

 based on the interpretation of very poor material. None of the figured 

 specimens of these two ankyroid genera clearly shows the bounda- 

 ries of the posterior zygal plate. In a note published a few years 

 before the formal description of Reticulocarpos by Jefferies & 

 Prokop ( 1972) (Jefferies, pers. comm.), Ubaghs (1969) clearly stated 

 that the zygal bar in two of the specimens of Reticulocarpos exam- 

 ined by him was apparently incomplete. 



It is difficult to ascertain whether the incompleteness (if any) of 

 the zygal bar is evidence of its progressive reduction and final 

 disappearance in derived comutes and, therefore, of the fact that the 

 zygal/septum complex is not homologous in all stylophorans. 



The peculiar zygal bar of Beiyllia as reconstmcted by Cripps & 

 Daley ( 1 994) suggests that this structure had a separate somatic plate 

 in contact with a posterior right infracentral element, and that its 

 anterior two-thirds comprised, in tum, a posterior plate and a short, 

 stout process of M',. Cripps & Daley (1994: 108) state that the strut 

 of Beryllia is "... most clearly visible on NHM E63499a', but the 

 figured specimen in question does not provide unequivocal evidence 

 of a tripartite zygal bar. I re-examined the specimen in question and 

 the remaining material of Beryllia in the collections of the London 

 Natural History Museum. The specimens were cast using black- 

 stained latex. 



My own observations do not match Cripps & Daley's (1994). In 

 particular, the zygal bar in NHM E63499 is too poorly preserved for 

 its boundaries and general shape to be reconstructed. In particular, I 

 could find no evidence of tripartition and no clear delimitation of the 

 posterior zygal plate from the thecal frame. The impression left by 

 the zygal bar on the dorsal integument of another specimen (NHM 

 E63496b; Cripps & Daley, 1994: pi. 2. fig. 5) suggests that, in fact, 

 the zygal bar of Beryllia had a slender shape and a sinuous course and 

 that its rearmost end lay closer to the posterior right part of the thecal 

 frame than in the reconstmction provided by Cripps & Daley ( 1 994). 

 In these respects, the bar is similar to that of such ankyroids as 

 Amygdalotheca griffei Ubaghs, 1969 from the lower Ordovician of 

 southern France. 



That the zygal bar of Beryllia may have consisted of a posterior 

 process of M'j and of a posterior zygal plate is plausible, although not 

 certain. However, I favour the latter interpretation, based in particu- 

 lar on the visible course of the bar in specimen NHM E63496b. 



In summary, the specialized features of the zygal bar reported in 

 several ankyroid species ought to be confimied by discovery of 

 better preserved material. Juliaecarpus shows the primitive condi- 

 tion for the ankyroids in possessing a posterior somatic plate 

 articulated with the inner margins of the thecal frame. If my conclu- 



