CEPHALOSPYRIDIDAE Haeckel, 1882

Suzuki, Noritoshi, Caulet, Jean-Pierre & Dumitrica, Paulian, 2021, A new integrated morpho- and molecular systematic classification of Cenozoic radiolarians (Class Polycystinea) - suprageneric taxonomy and logical nomenclatorial acts, Geodiversitas 43 (15), pp. 405-573 : 488-492

Family CEPHALOSPYRIDIDAE Haeckel, 1882 n. stat.

Cephalospyrida Haeckel, 1882: 441 [as a tribe].

Archiphatnida Haeckel, 1882: 429 [nomen dubium, as a tribe].

Acrospyrida Haeckel, 1882: 441 [as a tribe]; 1887: 1085 [as a subfamily].

Brachiospyrida Haeckel, 1882: 441 [as a tribe].

Dipodospyrida Haeckel, 1882: 441 [nomen dubium, as a tribe].

Triospyrida Haeckel, 1882: 441 [nomen dubium, as a subfamily].

Taurospyrida Haeckel, 1882: 442 [as a tribe].

Aegospyrida Haeckel, 1882: 442 [nomen dubium, as a tribe].

Phormospyrida Haeckel, 1882: 442 [as a tribe]; Haeckel 1887: 1021 1084-1085 [as a family]. — Bütschli 1889: 1981 [as a family]. — Anderson 1983: 29 [as a family].

Polyspyrida Haeckel, 1882: 442 [nomen dubium, as a subfamily]; Haeckel 1887: 1024, 1059 [as a subfamily].

Tetraspyrida Haeckel, 1882: 442 [nomen dubium, as a subfamily]; Haeckel 1887: 1024, 1043 [as a subfamily].

Therospyrida Haeckel, 1882: 442 [nomen dubium, as a tribe]; Haeckel 1887: 1024, 1055 [as a subfamily].

Gorgospyrida Haeckel, 1882: 443 [as a tribe].

Petalospyrida Haeckel, 1882: 443 [as a tribe].

Zygostephanida Haeckel, 1882: 446 [nomen dubium, as a tribe]; Haeckel 1887: 970 [as a subfamily].

Semantiscida Haeckel, 1887: 956 [nomen dubium, as a subfamily].

Tholospyrida Haeckel, 1887: 1021, 1077-1078 [as a family]. — Bütschli 1889: 1981 [as a family]. —Anderson 1983: 29 [as a family].

Zygospyrida Haeckel, 1887: 1021, 1022-1024 [nomen dubium, as a family]. — Bütschli 1889: 1980 [as a family]. — Anderson 1983: 29 [as a family].

Dipospyrida Haeckel, 1887: 1024, 1035 [nomen dubium, as a subfamily].

Hexaspyrida Haeckel, 1887: 1024, 1046 [nomen dubium, as a subfamily].

Lophospyrida Haeckel, 1887: 1078 [as a subfamily].

Tiarospyrida Haeckel, 1887: 1078 [as a subfamily].

Pylospyrida Haeckel, 1887: 1078 [nomen dubium, as a subfamily].

Rhodospyrida Haeckel, 1887: 1085, 1087 [as a subfamily].

Androspyrida Haeckel, 1887: 1090-1092 [as a family]. — Bütschli 1889: 1982 [as a family]. — Anderson 1983: 29 [as a family].

Lamprospyrida Haeckel, 1887: 1092 [as a subfamily].

Archiphaenida Haeckel, 1887: 1133, 1158, 1173 [nomen dubium, as a subfamily].

Tholospyriden – Haecker 1907: 123-124 [as a family].

Phormospyriden – Haecker 1907: 124 [as a family].

Zygospyridae [sic] – Haecker 1908: 445 [nomen dubium] (= Zygospyrididae). — Popofsky 1908: 269; 1913: 304. — Schröder 1914: 142. — Clark & Campbell 1942: 53; 1945: 29. — Campbell & Clark 1944a: 33; 1944b: 21. — Chediya 1959: 177. — Tan & Su 1982: 164. — Chen & Tan 1996: 152. — Tan & Chen 1999: 272. — Tan & Su 2003: 86. — Chen et al. 2017: 167.

Phormospyridae [sic] – Haecker 1908: 446 (= Phormospyrididae). — Popofsky 1913: 310. — Chediya 1959: 183. — Cachon & Cachon 1985: 293. — Chen & Tan 1996: 152. — Tan & Su 2003: 97.

Rhodospyrinae [sic] – Haecker 1908: 446 (= Rhodospyridinae). — Chediya 1959: 183.

Androspyridae [sic] – Popofsky 1908: 270 (= Androspyrididae); Popofsky 1913: 311. — Chediya 1959: 184. — Tan & Su 1982: 166. — Cachon & Cachon 1985: 293. — Chen & Tan 1996: 152. — Tan & Chen 1999: 278. — Tan & Su 2003: 99.

Tholospyridae [sic] – Popofsky 1908: 270 (= Tholospyrididae); Popofsky 1913: 309. — Tan & Su 1982: 165. — Cachon & Cachon 1985: 293. — Chen & Tan 1996: 152. — Tan & Su 2003: 92. — Chen et al. 2017: 170.

Zygospyrididae – Poche 1913: 221 [nomen dubium].

Tholospyrididae – Poche 1913: 221. — Campbell 1954: D114.

Androspyrididae – Poche 1913: 221. — Campbell 1954: D116.

Cyrtostephanidae Popofsky, 1913: 288-289.— Campbell 1954: D106.

Dipospyrinae [sic] – Clark & Campbell 1942: 55 [nomen dubium] (= Dipospyridinae). — Chediya 1959: 177.

Tetraspyrinae [sic] – Clark & Campbell 1942: 55 [nomen dubium] (= Tetraspyridinae). — Chediya 1959: 178.

Hexaspyrinae [sic] – Clark & Campbell 1942: 56 [nomen dubium] (= Hexaspyridinae); Clark & Campbell 1945: 31. — Chediya 1959: 179.

Therospyrinae [sic] – Clark & Campbell 1942: 58 [nomen dubium] (= Theospyridinae). — Chediya 1959: 180.

Polyspyrinae [sic] – Clark & Campbell 1942: 59 (= Polyspyridinae). — Campbell & Clark 1944a: 36. — Clark & Campbell 1945: 33. — Chediya 1959: 181.

Semantiscinae – Clark & Campbell 1945: 29 [nomen dubium]. — Chediya 1959: 169.

Triospyridae [sic] – Frizzell & Middour 1951: 27-28 [nomen dubium] (= Triospyrididae ). — Petrushevskaya & Kozlova 1972: 529.

Dipodospyrinae [sic] – Frizzell & Middour 1951: 28 [nomen dubium] (= Dipospyridinae).

Zygostephaninae – Campbell 1954: D108 [nomen dubium]. — Chediya 1959: 171.

Dipodospyridinae [sic] – Campbell 1954: D112 [nomen dubium] (= Dipodospyridinae). — Petrushevskaya 1981: 341-342. — Afanasieva et al. 2005: S304. — Afanasieva & Amon 2006: 154.

Tetrarrhabdinae Campbell, 1954: D112 [nomen dubium].

Tripospyrididae – Campbell 1954: D112 [nomen dubium]. — Blueford 1988: 242.

Hexaspyridinae – Campbell 1954: D113 [nomen dubium].

Petalospyridinae Campbell, 1954: D114.

Therospyridinae – Campbell 1954: D114 [nomen dubium].

Androspyridinae – Campbell 1954: D116. — Petrushevskaya 1981: 350-351. — Afanasieva et al. 2005: S305. — Afanasieva & Amon 2006: 155.

Phormospyrididae – Campbell 1954: D116.

Phormospyridinae – Campbell 1954: D116.

Rhodospyridinae – Campbell 1954: D116.

Tiarospyridinae – Campbell 1954: D116.

Spyridobotryidinae Campbell, 1954: D116 [nomen dubium].

Archiphatninae – Campbell 1954: D119 [nomen dubium].

Tholocpyridae [sic] – Chediya 1959: 182 (= Tholospyrididae).

Lamprospyrinae [sic] – Chediya 1959: 184 (= Lamprospyridinae).

Archiphaeninae – Chediya 1959: 196 [nomen dubium].

Triospyrididae – Petrushevskaya 1971a: 243-251 [nomen dubium] ( sensu emend. ); 1971b: 990; 1981: 328-329. — Kozlova 1999: 162. — De Wever et al. 2001: 229-230. — Afanasieva et al. 2005: S304. — Afanasieva & Amon 2006: 154. — Matsuzaki et al. 2015: 39.

Triospyrididinae – Petrushevskaya 1981: 330 [nomen dubium].

Tholospyridinae – Petrushevskaya 1981: 347. — Afanasieva et al. 2005: S304. — Afanasieva & Amon 2006: 154.

Zygosmyridae [sic] – Cachon & Cachon 1985: 293 [nomen dubium] (= Zygospyrididae).

Triospyridinae – Afanasieva et al. 2005: S304 [nomen dubium]. — Afanasieva & Amon 2006: 154.

TYPE GENUS. — Cephalospyris Haeckel, 1882: 441 [type species by subsequent designation ( Campbell 1954: D112): Cephalospyris cancellata Haeckel, 1887: 1035 ] = junior subjective synonym of Platybursa Haeckel, 1882: 429 [type species by subsequent monotypy: Cantharospyris platybursa Haeckel, 1887: 1051 ].

INCLUDED GENERA. — Androspyris Haeckel, 1887: 1092 . — Ceratospyris Ehrenberg, 1846: 385 (= Liriocyrtis synonymized by Matsuzaki et al. 2015: 41). — Corythospyris Haeckel, 1882: 443 . — Dendrospyris Haeckel, 1882: 441 . — Desmospyris Haeckel, 1882: 443 (= Phormospyris synonymized by Caulet 1979: 136). — Dorcadospyris Haeckel, 1882: 441 (= Brachiospyris , Gamospyris synonymized by Petrushevskaya & Kozlova 1972: 532). — Elaphospyris Haeckel, 1882: 442 (= Giraffospyris with the same type species). — Gorgospyris Haeckel, 1882: 443 (= Gorgospyrium with the same type species). — Lamprospyris Haeckel, 1882: 441 (=? Eulophospyris n. syn.). — Liriospyris Haeckel, 1882: 443 (= Petalospyromma synonymized by Petrushevskaya 1981: 332). — Lophospyris Haeckel, 1882: 443 (nec Haeckel, 1887) (=? Semantrum n. syn.). — Pentaspyris Haeckel, 1882: 442 (= Taurospyris n. syn.). — Petalospyris Ehrenberg, 1846: 385 (= Anthospyris n. syn., Rhodospyris n. syn., Sepalospyris n. syn.; Patagospyris synonymized by Ling 1975: 272; Petalospyrantha , Petalospyrissa synonymized by Petrushevskaya 1981: 335; Petalospyrella synonymized by Petrushevskaya & Kozlova 1972: 532). — Platybursa Haeckel, 1882: 429 (= Cephalospyris, Cyrtostephanus synonymized by Petrushevskaya 1971a: 257; Clathrobursa synonymized by Haeckel 1887: 1045; Tessarospyris synonymized byPetrushevskaya 1971a: 259). — Thamnospyris Haeckel, 1882: 443 . — Tholospyris Haeckel, 1882: 441 (= Tholospyrium with the same type species; Tholospyridium n. syn.; Tristylospyris , Tristylospyrula , synonymized by De Wever et al. 2001: 230). — Tiarospyris Haeckel, 1882: 443 . — Triceraspyris Haeckel, 1882: 441 (= Acrospyris n. syn., Tripospyrella n. syn.; Triospyrium synonymized by Petrushevskaya 1971a: 248).

INVALID NAMES . — Archiphaena, Calpophaena, Coronophaena , Dipospyris , Pylospyris, Stephanophaena , Tripospyris .

NOMINA DUBIA. — Acrocorona , Aegospyris , Archiphatna , Cantharospyris , Calpocapsa , Cladocorona , Cladophatna , Clathrocircus , Clathrospyris , Coronophatna , Dipodospyris , Hexaspyridium , Hexaspyris , Polyspyris, Semantidium , Semantiscus, Spyridobotrys , Stephanophatna, Stephanospyris , Tetrarhabda, Tetraspyris , Therospyris , Triospyridium , Triospyris , Tripodospyris , Tripospyrantha, Tripospyrissa , Tripospyromma, Zygospyris.

JUNIOR HOMONYMS. — Lophospyris Haeckel, 1887 (= Elaphospyris ) nec Haeckel, 1882; Stephanospyris Haeckel, 1882 (= Dorcadospyris ) nec Haeckel, 1862.

DIAGNOSIS. — Skeleton having a sagittal ring with twin cupolas or twin set of body frames. Two or more feet, a coronal skirt and/ or a thorax are developed. In addition, the Lo-axis is parallel to the Lg-axis while the Sh-axis is parallel to the Sg-axis. The basal ring is marked by two to six basal pores. Protoplasm was identified in Lophospyris . A spherical endoplasm is located around the center of the sagittal ring while a brownish matter of unknown composition is aggregated below the spherical endoplasm and attached to the MB. Algal symbionts scattered in and out of the skeleton area. A gelatinous matter is found wrapping the endoplasm, the algal symbionts and the skeleton. The algal symbionts are scattered all over. Radiated pseudopodia are visible inside the gelatinous matter.

STRATIGRAPHIC OCCURRENCE. — Middle Paleocene-Living.

REMARKS

The family name “ Triospyridae ” has been used hitherto but cannot be upheld as a practical valid status. This is due to the absence of an illustrated type species of Triospyris . Amongst the potential candidate for family-names in Haeckel (1882) (Acrospyrida, Brachiospyrida, Cephalospyrida, Phormospyrida, and Gorgospyrida) we retain Cephalospyrididae as a valid family name, considering the best figured and described specimen representing the type genus ( Cephalospyris cancellata Haeckel, 1887 ). However, little is known about the basal pore patterns and the relationship of the feet with the cephalic initial spicular system. Thus, this family concept may be artificial. “Living” and protoplasm images are illustrated for Lophospyris ( KrabberØd et al. 2011: fig. 1.O; Matsuoka 2017: fig. 18; Zhang et al. 2018: 10, figs 2.12, 2.14-2.16) and Platybursa ( Aita et al. 2009: pl. 32, fig. 7; Zhang et al. 2018: 10, fig. 2.20). Ceratospyris is known to be infected with Marine Alveolata Group I ( Ikenoue et al. 2016). Evolution for Dorcadospyris was illustrated ( Kling 1978: 238-239; Riedel & Sanfilippo 1981: fig. 12.9).

VALIDITY OF GENERA

Ceratospyris

The reason for a synonymy with Liriocyrtis is written in Matsuzaki et al. (2015: 41). The basal ring structure comprises four basal pores: large, long rectangle, twin Ca -pores and small twin Cerv -pores divided by the V-rod ( Goll 1972a: pl. 50, figs 1-4). Twin J -pores are very large, developed on the lateral side of the lobate shell. This basal pore pattern is similar to that of Dictyospyris . Three feet are directly connected with D- and double L-rod. No downward rod below the V-rod.

Desmospyris

The basal ring of Desmospyris comprises five to six basal pores ( Goll 1968: pl. 173, figs 12, 13, 20; Goll 1972a: pl. 53, figs 3, 4). The J -pore is subdivided as twin J -pores by the D-rod ( Goll 1968: pl. 173, fig. 20) or a unified J -pore looks as two pores with a downwardly oriented D-rod ( Goll 1968: pl. 173, fig. 13; Goll 1969: pl. 53, figs 3, 4). Other basal pores are the large twin Ca -pores and relatively large twin Cerv -pores. The type-illustration of Phormospyris tricostata ( Haeckel 1887: pl. 83, fig. 15) looks “strange” but the most similar specimens to this species ( Caulet 1979: pl. 5, fig. 2) show identical skeletal structures. Desmospyris and Phormospyris were simultaneously published in Haeckel (1882: 443 for the former and 442 for the latter). Although Caulet (1979) did not mention the synonymy between Desmospyris and Phormospyris , this paper applied the name Desmospyris for them, implying the first reviser’s decision. The whole structure of Thamnospyris is similar to Desmospyris but they are not synonymized due to insufficient examination.

Dorcadospyris

The genera synonymized herein belonged to the “Diplospyridinae” sensu Campbell (1954 : D112) which are defined by two lateral basal feet. Brachiospyris is marked by unbranched feet with lateral spines and no apical horn, Dorcadospyris by feet with lateral spines and a single apical horn and Gamospyris by two unbranched feet forming a ring and a single horn. Sanfilippo et al. (1985: fig. 10) illustrated these differences as an evolutionary change at species level. The basal ring structure shown in Ceratospyris ocellata , the type species of Brachiospyris , drawn by Ehrenberg (1876: pl. 20, fig. 5) is not precise when referred to the real sketch of this specimen in the Ehrenberg collection ( Ogane et al. 2009b: pl. 77, figs 4a-c). Petrushevskaya (1981: fig. 531) illustrated four basal pores in the basal ring. The basal pore consists of a unified J- pore, twin Ca -pores and a unified Cerv -pore. Two feet do not directly connect with any initial spicular system. Dorcadospyris in the sense of Sanfilippo et al. (1985) includes species with three feet. The basal ring of Dorcadospyris with three feet is also illustrated inPetrushevskaya (1981: fig. 533). Differing from the species with two feet, the basal ring comprises five pores: a tiny twin J -pore, a large twin Ca -pore and a relatively larger unified Cerv -pore ( Goll 1972a: pl. 62, fig. 3). The third foot is directly connected with the D-rod. It is unclear whether the difference of J -pore related with the presence of the third feet from the D-rod has a value as genus criteria or not. Dorcadospyris , Brachiospyris and Gamospyris were simultaneously published in Haeckel (1882: 441 for all genera). As Dorcadospyris dentata , the type species of Dorcadospyris , is commonly found, this genus name is selected as a valid name for them.

Elaphospyris

The basal ring is not part of the initial spicular system ( Goll 1972a: pl. 47, figs 2-4), because MB is obviously above the basal ring. D-, V-, double L- and double l-rods are downwardly oriented to connect the basal ring. Six “pores” are visible but they are not the true basal pores and, thus, there are no J-, Ca - and Cerv -pores. D-rod only is directly connected with a basal foot and all the remaining feet are not directly connected with any initial part of the spicular system. Many papers applied the genus name Giraffospyris to several species, but this is taxonomically problematic. Both genera were erected the same year by Haeckel (1882: 442). The first reviser for sure is Campbell (1954: D114). This revision is however, erroneous (Nigrini, personal com.). As written in the Atlas, Campbell (1954 : D114) erroneously considered Elaphospyris to be an objective synonym of Giraffospyris , and then he designated Ceratospyris heptaceros as the type species of Giraffospyris and only by inference of Elaphospyris . Therefore, we consider the type species designation of Ceratospyris heptaceros Ehrenberg to date from Chediya (1959: 180). It is clear if we check Haeckel (1887: 1056-1057) who erected the genus Elaphospyris with two subgenera Elaphospyris and Corythospyris .

Gorgospyris

The real specimens perfectly corresponding to the typeillustration of Gorgospyris medusa ( Haeckel 1887: pl. 87, fig. 1), the type species of Gorgospyris , are rare (the support image for Gorgospyris in the Atlas ). The basal ring structure in these specimens is different from that shown in Haeckel (1887: pl. 87, fig. 2). Gorgospyris is an available name older than Gorgospyrium .

Lamprospyris

The illustrated specimen of Lamprospyris darwinii , the type species of Lamprospyris , looks to have a free D-ring (sagittal ring) inside the cephalic lobe ( Haeckel 1887: pl. 89, fig. 13). Real specimens identifiable as this species, however, have three feet which are directly connected with D- and double L-rods (the support image for Lamprospyris ). Lamprospyris characteristically develops a latticed shell over the junction between the A-rod and the AV-arch. As this kind of lattice shell is rarely known in this family, Elaphospyris is synonymized with Lamprospyris herein. This synonymy needs to be confirmed in the future by an evolutionary connection between the type species of both these genera. Lamprospyris is an available name older than Eulophospyris .

Liriospyris

Goll (1968) identified his specimens as “ Liriospyris clathrata (Ehrenberg) ” ( Goll 1968: pl. 175, figs 12, 13, 16, 17) but the specimen found in the Ehrenberg collection ( Ogane et al. 2009b: pl. 38, figs 5a-c) is quite different from the drawing in Ehrenberg (1854c: pl. 36, fig. 25). The specimen shown in Ogane et al. (2009b) follows the indication by Ehrenberg himself so it is not possible to ignore this specimen. But this specimen is obviously different from the current usage of “ Liriospyris clathrata (Ehrenberg) ” and is also impossible to use for determining taxonomic morphological features. “ Liriospyris clathrata (Ehrenberg) ” is the type species of Stephanospyris. If this specimen is regarded as a valid name, Stephanospyris must be validated instead of Liriospyris because the former was published in Haeckel (1862: 295) while the latter in Haeckel (1882: 443). Real structure of Liriospyris can be understood from Goll (1968: pl. 175, figs 12, 13, 16, 17; 1969: pl. 57, figs 1-4). The basal ring comprises four basal pores: large twin Ca -pores and relatively larger twin Cerv -pores. A unified J -pore has a large size and is placed on the lateral side of the bilobate test. Four of six basal feet are directly connected with the double L-rod, the D-rod and an un-coded rod below the V-rod. The remaining two basal feet are present near the double l-rods but are not directly connected with them. The basal structure of Petalospyromma is shown in Goll (1969: pl. 57, figs 11, 12, 15-17). Differing from the basal structure of Liriospyris , the basal pores consist of two very large twin pores and tiny twin pores. As it is not possible to specify the D-rod and double l -rod from this sketch, the relevant code of J -, Ca - and Cerv -pores is not specified for them. Petalospyromma is tentatively synonymized with Liriospyris for a simple practical identification. Liriospyris is similar to Ceratospyris , but the former has small twin Cerv -pores and no un-coded downward rod below the V-rod. Liriospyris is an older available name than Petalospyromma .

Lophospyris

The basal ring of Lophospyris has two pores which are very large, polygonal, twin Ca -pores ( Goll 1972a: pl. 58, figs 1-3). Both twin J -pores and twin Cerv -pores are also polygonal in shape, the former being placed on the dorsal lateral side of the test whereas the latter is placed on the ventral lateral side of the test. Goll (1976) has already proved that Semantrum quadrifore , the type species of Semantrum , is a young specimen of Lophospyris ( Goll 1976: pl. 13, figs 5, 6). Lophospyris is an older available name than Semantrum .

Pentaspyris

No exactly fit specimen of Pentaspyris has ever been so far reported. The morphospecies most similar to Pentaspyris was illustrated as “ Lophospyris pentagona hyperborea ” byGoll (1976: pl. 15, figs 1-12). As Goll (1976) classified this morphotype as a subspecies of “ Lophospyris pentagona ”, this morphotype has exactly the same basal ring structure as Ceratospyris pentagona , the type species of Lophospyris ( Goll 1976: pl. 15, figs 3, 7). No Taurospyris specimens are also so far reported. The most similar morphotypes of Taurospyris were illustrated as “ Phormospyris stabilis capoi ” by Goll (1976: pl. 7, fig. 5) but the number of basal feet is quite different. Under such suspect conditions, both these genera are synonymized as to reduce the number of “un-realistic genera” from the valid genus list. Pentaspyris and Taurospyris were simultaneously established by Haeckel (1882: 442 for both genera). Pentaspyris is validated among them in consideration of its type-illustration more realistic than that of Taurospyris .

Petalospyris

The genera listed here are artificially synonymized for simplicity of practical identification. The basal structure of Petalospyrissa and Petalospyrantha is a basal ring which is not directly connected with the initial spicular system below MB ( Goll 1968: pl. 174, figs 5-8, 10). They have six “openings” without any arches directly connected with any D-, V-, double L- and double l-rods, and subsequently have no J -, Ca - and Cerv -pores. This basal structure is similar to that of Elaphospyris . However, the basal structure in? Petalospyrella ( Goll 1969: pl. 56, figs 9-11) and Patagospyris ( Goll 1969: pl. 58, fig. 11) has a basal ring directly connected with the initial spicular system. Their basal structure has three “large” basal pores with nearly the same size: the twin Ca -pores and a unified Cerv -pore. Tiny twin J -pores are also visible. Although all these genera synonymized here have many feet, none of these feet are directly connected with the initial spicular system. Although the Petalospyrissa - type basal structure is different from the Patagospyris - type one at genus or family level, it is practically impossible to apply this difference for real specimens with their current knowledge. Due to this reason, we prioritize a practical usage based on the similarity of the whole appearance until they are better studied in the future.

In the sense of Campbell (1954), all the genera with the exception of Sepalospyris belonged to the “ Triospyrididae ” whose shell is composed of a cephalis and its apophyses and no thorax ( Petalospyris , Petalospyrantha , Petalospyrella , Petalospyrissa and Anthospyris ) ( Campbell 1954: D112) or to the “Phormospyrididae” which have a thorax ( Rhodospyris and Patagospyris ) ( Campbell 1954: D116). These two “families” are differentiated by the presence/absence of a thorax. As there are many intermediate forms between them, this family criterion is not applicable for these genera. Regardless of different “families”, the definition of subfamilies is the same among them ( Campbell 1954: D112 for the “Petalospyridinae” and D116 for the “Rhodospyridinae”). The criterion for establishing a subfamily rank is also meaningless. The differences among Petalospyris , Petalospyrella and Petalospyrissa ( Campbell 1954: D114) are based on the number of large basal pores on the basal ring. Since the number of basal pores has not yet been examined in the type species of all these genera, it is difficult to positively apply this criterion for the current taxonomy. The remaining character written in Campbell (1954) is the number of apical horns. One apical horn characterizes Petalospyris , Petalospyrantha , Petalospyrella , Petalospyrissa and Patagospyris . Three apical horns are found in Anthospyris and Rhodospyris . These genera can be divided into two groups but not in seven genera. It is not necessary to separate them by the number of apical horns. Differing from these genera, Sepalospyris has an apical cupola ( Campbell 1954: D116). Although the presence of a cupola is not a sure criterion for genus or species definition, we synonymized this genus with Petalospyris because no real specimens identifiable as Sepalospyris have been found so far. The oldest available name among them is Petalospyris .

Platybursa

Clathrobursa has the same type species as Tessarospyris . Cyrtostephanus was classified into the “Cyrtostephanidae” of the superfamily “Stephaniicae” ( Campbell 1954: D105-106). Following this classification, Cyrtostephanus is characterized by an incomplete ring, a latticed sagittal ring (D-ring) or a netlike fan of repeated anastomosed spines and a skeleton formed of a sagittal ring without a basal tripod. This definition, however, does not partly correspond to the type-illustration and the support image for Cyrtostephanus in the Atlas . The type-illustration ( Popofsky 1913: pl. 28, figs 4, 5) is a basal view (a view from the bottom of the test) as the supra view (the apical view) is not known. The right photography of the support images for Cyrtostephanus in the Atlas appears to show an AV arch free in the latticed cephalic lobe. The remaining three genera belong to the “ Triospyrididae ” of the “superfamily Triospyridicae ” whose shell is composed of a cephalis and its apophyses, a binocular cephalis with a sagittal constriction, the absence of an apical cupola or dome or thorax. All these characters, however, are not recognized as superfamily criteria by the molecular phylogeny results ( Sandin et al. 2019). The three genera were classified into different three “subfamilies”: the “ Triospyridinae ” with three basal feet for Cephalospyris ( Campbell 1954: D112), the “Tetrarrhabdinae” with two lateral and two sagittal feet for Tessarospyris ( Campbell 1954: D112), and the “Hexaspyridinae” with six basal feet for Platybursa ( Campbell 1954: D114). These subfamily criteria are suspect because of the indistinguishable robustness of the feet and lengths among them. Real specimens of these type species (support images for Platybursa , Cephalospyris and Clathrobursa in the Atlas ) seem to have less numbers or more numbers of “basal feet”. If this subfamily criterion would be accepted for these genera, more genera and subfamilies must be established. Following ignorance of these “subfamily” differences, Cephalospyris is characterized by an apex with an instum and sinistram apical hole (a right and left apical hole), Tessarospyris by the absence of apical horns, and Platybursa by the absence of apical horns, according to Campbell (1954: D112, 114). There are no reasons to separate Tessarospyris and Platybursa anymore. No distinguishing marker for Cephalospyris exists in any real specimens. These observations conclude that they should be regarded as a same genus. Some concerns remain about this synonymy. The basal rings are confirmed in Platybursa , Cephalospyris and Tessarospyris , but not in Cyrtostephanus. A complete ring is present in only Cyrtostephanus. These concerns should be solved in the future. All the genera, except Cyrtostephanus, were simultaneously published in Haeckel (1887: 429 for Clathrobursa and Platybursa , 441 for Cephalospyris , 442 for Tessarospyris ). In respect to the first reviser, Platybursa is regarded as a valid genus among them.

Tholospyris

The combination of Tholospyris and Tholospyrium and that of Tristylospyris and Tristylospyrula are respectively based on the same type species. Under the scheme ofCampbell (1954: D112, 114), Tristylospyris and Tholospyris have in common three unbranched basal feet and no thorax. The “difference” among these genera relies on the definition of the family. Tristylospyris belonged to the “ Triospyrididae ” whose shell is characterized by the presence of a cephalis and its apophyses and the absence of an apical cupola ( Campbell 1954: D112) whereas Tholospyris belonged to the “Tholospyrididae” whose shell includes a cephalis with an apical cupola ( Campbell 1954: D114). The main difference is the presence/absence of “an apical cupola” but type images as well as support images for Tholospyris and Tristylospyris in the Atlas show the presence of “an apical cupola” in both genera. According to Campbell (1954), Tristylospyris lacks an apical horn and apical holes and Tholospyris has an apical horn. This difference depends on the development stage of the apical horn so it is not useful for genus criterion. “Apical holes” exist in both genera.

The basal ring ( Goll 1969: pl. 56, figs 3-6, 8) comprises four pores: very large, Japanese rice spatula-shaped twin Ca - pores and large twin Cerv -pores.Twin J -pores are visible from a base view, but they are obliquely oriented on the dorsal side (A-rod side) of the test. Three basal feet are directly connected with D- and double L-rods. The size of the twin Cerv -pores is obviously larger than in Ceratospyris , but the independency of both these genera has to be re-evaluated. Tholospyris and Tristylospyris were simultaneously published in Haeckel (1882: 441 for both genera). Tholospyris has already been selected as a valid genus by the first reviser ( De Wever et al. 2001: 230) although no explanations were given.

Triceraspyris

Following Campbell’s concept ( Campbell 1954: D112, D116), the genera listed here can be placed into a group with one apical or simple horn ( Triospyrium , Tripospyrella , Acrospyris ) and a group with three apical horns ( Triceraspyris ). Other differences given are absence of thorax for the “ Triospyrididae ” for Triceraspyris , Triospyrium and Tripospyrella , and presence of thorax for the “Phormospyrididae” as Acrospyris . The “thorax” of the type-illustration of Acrospyris ( Haeckel 1887: pl. 95, fig. 17) is represented by supplementary meshes connecting the adjacent feet and, thus, this “genus” has no true thorax. The lectotypes of Ceratospyris didiceros ( Ogane et al. 2009b: pl. 39, figs 1a-c) and Ceratospyris furcata ( Ogane et al. 2009b: pl. 39, figs 5a-d), the typical species of Triceraspyris and type species of Triospyrium , show nearly the same morphology except for the developmental stage of the three apical horns and the distal ends of the three feet. These differences are not significant to separate them into two genera. The basal structure was only documented for Tripospyrella , which comprises three “large” basal pores ( Campbell 1954: D112). The support image for Tripospyrella cited from Haeckel (1887: pl. 95, fig. 2) shows twin Ca -pores (coded as k in Haeckel 1887) and a unified Cerv -pore (i in Haeckel 1887). A very tiny twin J -pore is as well drawn in this figure. If this is correct, this structure is identical to that of Dictyospyris and Dorcadospyris at a family level. Triceraspyris and Acrospyris were simultaneously published in Haeckel (1882: 441 for these two genera). Real specimens identifiable as Triceraspyris are found so that this genus is selected as a valid genus.