Review of the dinoflagellate cyst Stephanelytron Sarjeant 1961 emend

The stratigraphic distribution of the Late Callovian to Early Oxfordian dinoflagellate cyst Stephanelytron Sarjeant 1961 emend provides new evidence pertaining to its evolution. Middle and Upper Callovian times favoured the development of speciations to a short-ranging Stephanelytron community with corona(s) in ventral–posterior position (Stephanelytron brontes, S. callovianum, S. ceto and S. tabulophorum) from eurytopic species with antapical coronas (S. caytonense, S. membranoidium, S. redcliffense and S. scarburghense). The former group of species (except S. tabulophorum) may represent an example of peripatric speciation from an unfavourable mutation. The reduced stratigraphic range gives the appearance of an endemic population. The genus Lagenadinium Piel, 1985 is a junior synonym of Stephanelytron Sarjeant, 1961. A new emendation of Stephanelytron, two new combinations (S. callovianum and S. membranoidium) and two new species (?S. brontes and S. ceto) are proposed.


INTRODUCTION
During an investigation of Jurassic dinoflagellate cyst assemblages in the Tethyan marine realm, the Middle and Upper Jurassic transition was studied in southeastern France (Fig. I).
The dinoflagellate cyst assemblages described here belong to the Compositosphaeridium polonicum-Sentusidinium pilosum association of Smelror & Leereveld (1989). However, significant differences in relative abundance are observed. For example, in the shales and muddy limestones collected from the Chenier and Rondette localities (Fig. I), dinoflagellate cyst assemblages are composed of relatively rare E. galeritum, G. jurussica, R. cladophora, R. aemula, S. rioultii, S. vestitum and Wanaea spp. M . groenlandicurn is abundant. Comparison of dinoflagellate cyst assemblages in the Tethyan and Boreal Realms requires more data before the latitudinal/environmental control of species (Smelror & Leereveld, 1989) can be fully appreciated.
Despite differences between the dinoflagellate cyst assemblages of Crussol, Chenier and Rondette, one element is common in Europe. It concerns the association of holocavate species of Chlamydophorella Cookson & Eisenack, 1958, Dingodinium Cookson & Eisenack, 1958and Stephanelytron Sarjeant 1961emend herein. Following Smelror & Leereveld (1989 such an association appears indicative of deep marine conditions. In the Chenier and Rondette sections, as well as Savournon (Fig. I), this association of holocavate species constitutes a consistent and significant assemblage, in association with specimens of Lagenadinium callovianum Piel, 1985. The objectives of the present study are to review the genus Stephanelytron and to discuss the evolutionary pathway proposed by Piel (1 985) for this group of species with corona(s). The genus Lagenudinium Piel, 1985 is considered a junior synonym of Stephanelytron Sarjeant, 1961 emend.

SYSTEMATIC BACKGROUND
The genus Lagenadinium Piel, 1985 contains the type species L . cullovianum Piel, 1985and questionably ?L. membranoidium (Vozzhennikova, 1967) Lentin & Vozzhennikova, 1990. According to Piel (1985 the genus Lagenadiniurn has a subspherical to ovoidal holocavate cyst. An autophragm and an ectophragm are separated by hollow, non-tabular processes. The most typical element is the presence of one or two coronas at or near the antapex and the archeopyle is apical. The type species L. callovianum Piel, 1985 is a proximochorate, holocavate cyst. The autophragm and ectophragm are clearly separated by smooth, non-tabular, tubiform, hollow processes. A short apical mamelon-like horn is present on both autophragm and ectophragm. The archeopyle is apical and the operculum, probably a compound one, remains attached. The cyst typically has two coronas that are located either on paraplates in the postcingular series, o r perhaps on the posterior intercalary and adjacent postcingular paraplate. The average size of the coronas is one-quarter of the width of the cyst (Piel, 1985: 112). When seen in optical cross-section, the elevation of coronas is low, about half the diameter. ?L. membranoidium (Vozzhennikova, 1967) Lentin & Vozzhennikova, 1990, possesses the main characters of the genus Lagenadinium (holocavate, numerous hollow processes, one corona which is approximately coincident with the 2"' paraplate and an apical archeopyle). Lentin & Vozzhennikova (1 990) questionably included this species in Lagenadinium because of its tendency to exhibit an intratabular process distribution.
At the time he erected the genus, Piel (1985) proposed the transfer of S. scarburghense Sarjeant 1961 to Lagenadinium. This transfer was not accepted by Riding (1987) because S. scarburghense has a single, prominent antapical corona and lacks an antapical horn, one of the principal generic criteria of Lagenadinium, despite the non-tabular arrangement of the processes in the ectocoel. Riding (1987) uses the criterion of an apical horn (antapical horn is probably a typographical error; p. 263) and one corona to refute the transfer because the original diagnosis or description did not mention formally an apical horn (Sarjeant, 1961: 111).
It is true that on S . scarburghense the processes are nontabular a n d so the paratabulation is vaguely expressed. When all the other species of Stephanelytron (S. caytonense Sarjeant, 1961, S. cretaceum Duxbury, 1983, S. redcliffense Sarjeant 1961, S. tabulophorum Stover et al., 1977 have processes frequently arranged in parasutural rows, both alternatives have merit (the transfert of S. scarburghense to Lagenadinium or retention in Stephanelytron; Wheeler & Sarjeant, 1990: 314).
These considerations show the difficulties in distinguishing these two genera. T h e difficulties are based on the distribution of autophragmal processes, the interpretation of the number of corona(s) (one or two) a n d the presence o r absence of an apical horn. Large number of specimens with corona(s) in southeastern France confirm the absence of objective criteria that would clearly separate Stephanelytron and Lagenadinium. Consequently, the two genera a r e considered synonymous, i.e.

Stephanelytron caytonense Sarjeant 1961 emend. Stover et a/. 1977
Stephanelytron ceto new species Stephanelytron cretaceum Duxbury 1983 5tephanelytron membranoidium (Vozzhennikova 1967 emend. Lentin and Vozzhennikova 1990) comb. nov. Dodekova 1975 Ch/amydophorel/a ectotabulata Smelror 1988  Original description Sarjeant (1961: 109): "Organic shells of spherical to ovoidal shape, bearing on one face (at one end in ovoidal forms) one or more structures consisting of a circular membrane rising upwards from the shell surface and everted, surrounding a matte of short hairs or spines; this structure is henceforth termed a 'corona'. Elsewhere the shell bears tubular processes of varied character and arrangement. " Stover et al. (1977: 331): "Cysts proximochorate with subspherical to ellipsoidal body composed of two wall layers. Inner wall, the autophragm, gives rise to usually short processes; outer wall, the ectophragm, thin and may be discontinuous. Processes frequently arranged in parasutural rows, less commonly non tabular; some forms have parasutural and scattered non tabular processes, penitabular processes being present one species. Processes cylindrical to tubiform, normally of uniform height or nearly so, but width may vary considerably. Paratabulation indicated by alignment of processes, on some species, formula: l'?, 5" X-6c, 5"', 0-lp, I"", 2s. Archeopyle apical, operculum free, rarely attached; exact archeopyle shape and number of paraplates in operculum uncertain; antapical area typically with one or occasionally with two coronas."

New species
Comparisons and remarks. ?S. brontes has an elongate horn (onequarter to one-third of the length of the central body); on s. callovianum the horn is short (not more than one-tenth the length of the central body). ?S. brontes is reported with doubt to the genus Stephanelytron because the archeopyle is unknown. The specimen figured in Plate 1, fig. 6 shows a tear that could be an apical archeopyle. On the holotype (Plate 1, fig. 3) and other specimens (e.g. Plate 1, fig. 4), a possible opening that may be an intercalary archeopyle is observable. If this proves to be the case, a new genus will be required. ?Stephanelytron brontes n. sp. Plate 1, figs 1 4 ; Fig. 2 Review of the dinoflagellate cyst Stephanelytron Sarjeant 1961 emend Stephanelytron ceto n. sp. Plate 1, figs 10-12; Fig. 3 Derivation of name. From the name of the daughter of Pontos (The Sea) and Gaea (The Earth) in classical Greek mythology. Diagnosis. Cysts subspherical to ellipsoidal, proximate, holocavate; incomplete ectophragm supported by a dense cover of small buccinate to tubiform and entire processes. Apical horn absent. Processes penitabular and intratabular, of small length. Paratabulation formula unknown; the paratabular features being five precingular paraplates, a paracingulum and an archeopyle. Archeopyle apical. One corona in ventral-posterior position is present probably on the ps paraplate; base of coronas relatively thick, laevigate; elevation of corona low (10% of diameter). Holotype. SAV 17; coordinate 60,3/104; England Finder N61/4; Plate 1, fig. 11; specimen housed in the University Claude-Bernard Lyon 1 Collection, France. Comparisons and remarks. S. ceto has probably one corona on the ps paraplate similar in position to S. callovianum, but the latter has two coronas. On ?S. brontes and S. tabulophorum the corona is on the 2"' paraplate (Table 2). S. ceto has processes similar to those of ?S. brontes, but has no horn. The arrangement of processes is intratabular and penitabular, while it is parasutural and penitabular on S. tabulophorum. Otherwise, S. ceto possesses a dense cover of processes, whereas S. tabulophorum has scattered processes.

EVOLUTION AND SPECIATION EVENTS IN STEPHANELYTRON
The stratigraphic ranges of the species in Stephanelytron (Table  4) provide new evidence about the evolution of this genus. These considerations are speculative as the exact function of the corona is unknown. Piel (1985: 117-118, fig. 4) hypothesized that Chlamydophorella-like forms are ancestors of Stephanelytron. Wheeler & Sarjeant (1990: 315-3 16), assumed that the hypothesis of Piel (1985) was correct i.e. that the coronas of the Stephanelytraceae evolved from a single process of Chlamydophorella (e.g., C. ovula Wheeler & Sarjeant 1990). Such a scenario is conceivable, but it can also be proposed that a corona resulted from the coalescence of processes or the coalescence of processes and part of the ectophragm (see Plate 1, fig. 19). The oldest species of Stephanelytron ( S . scarburghense; Early Callovian) may have originated from the Chlamydophorella present at that time, i.e. C. raritubulae Dodekova, 1975or C. ectotabulata Smelror, 1989 fig. 9) or from another Chlamydophorella-like form. S. cretaceum, one of the two representatives of the genus in the Cretaceous, may have developed from S. membranoidium (the option shown in Fig.  4) or from the Chlamydophorella lineage. The common ancestor from Chlamydophorella-like forms theory explains the fact that S. cretaceum has strongest affinities with the Jurassic species ( S . caytonense, S. redclifSense and S. scarburghense) rather than ?Stephanelytron brontes, S. callovianum and S. ceto. It is conceivable that similar speciation mechanisms gave rise to comparable forms (i.e. S. cretaceum and S. scarburghense). Sarjeant (1962: 495) suggested that S. scarburghense gave rise to S. caytonense. These two species have similar ranges (Fig. 4). Some workers have commented on the gradation between S. caytonense and S. redcliffense (Stover et al., 1977: 333;Fensome, 1979: 25), but they have not been synonymized and their ranges differ. The extinction of the two species may be contemporaneous. S. caytonense first appears in the mid-Callovian, whereas S. redcliffense appears in the Lower Callovian (Table 4). Their close similarities and the time lag of their first occurrence argues for a relationship between S. redcliffense and S. caytonense in which S. redcliffense could be the ancestor of S. caytonense. They may represent two forms (?subspecies) of a polytypic species where forms (?subspecies) occupied distinctive geographical areas. This might explain why S. redclifSense and S. caytonense are apparently never found together, except in northwest Europe (England), where the two geographical regions are connected. S. tabulophorum, a short-ranging species, is characterized by parasutural and penitabular processes. These characters, only known on this species, may have evolved from a parasutural arrangement of processes like those of S. redcliffense.
The ?Stephanelytron brontes, S. ceto and S. callovianum complex, with distinctive coronas, is assumed to be derived from S . scarburghense because of the non-paratabular arrangement of processes. However, a Chlamydophorella ancestor, possibly C. ectotahulata, cannot be excluded.
The type species of S. membranoidium shows a corona, but numerous specimens have been found without this character. It is conceivable that this long-ranging species has some of the features of both Stephanelytron and Chlamydophorella.