Remarks

The systematic position of Callavia has been controversial. Although most authors have nested it with members of Holmiidae Hupé, Reference Hupé1953 (e.g., Harrington, Reference Harrington and Moore1959; Chernysheva, Reference Chernysheva1960; Repina, Reference Repina1979), Bergström (Reference Bergström1973) included Callaviinae Poulsen in Harrington, Reference Harrington and Moore1959 in Daguinaspididae Hupé, Reference Hupé1953 and Ahlberg et al. (Reference Ahlberg, Bergström and Johansson1986) preferred to treat it as an independent family, not related to holmiinids. Following the most consensual assignment, Palmer and Repina (Reference Palmer and Repina1993, Reference Palmer, Repina and Moore1997) maintained Callaviinae within Holmiidae, thus including Callavia within the superfamily Olenelloidea. Nevertheless, they assigned the family Judomiidae, morphologically similar to callaviinines, to the superfamily Fallotaspidoidea, revealing an inadequacy of this proposal for these taxa. In fact, following the concept of Palmer and Repina (Reference Palmer and Repina1993), Callavia cannot be included within Olenelloidea, and, consequently, in Holmiidae, because it lacks some of the diagnostic characters of the superfamily (frontal lobe [LA] enlarged and ocular lobe connected only to posterolateral part of LA, both absent in Callavia) and of the family (extraocular area [tr.] wider than twice the width of the interocular area, unlike in all holmiids). This was also stated by Lieberman (Reference Lieberman1998, Reference Lieberman1999, Reference Lieberman2001), who treated Callavia and a group of ‘fallotaspidoids’ (sensu Palmer and Repina, Reference Palmer and Repina1993) as representing an independent taxon of superfamiliar rank, the Judomioidea (not Nevadioidea, as mistakenly considered by Fletcher and Theokritoff, Reference Fletcher and Theokritoff2008).

One of the Callavia morphological characters that was misinterpreted by Palmer and Repina (Reference Palmer and Repina1993) and that were used to justify previous assignments to Olenelloidea/Holmiidae, is the relation between the ocular lobe and the LA. This relationship was considered the principal phylogenetic trend within the Olenellina by Palmer and Repina (Reference Palmer and Repina1993), with the earliest representatives having a glabella that is parallel-sided or tapering forward and an ocular lobe that is attached along the entire margin of the LA. Callavia shows this condition (see remarks on the genus). Nevertheless, it has been previously described as though the LA becomes inflated and expanded laterally and the ocular lobes connect only to its posterior part (like in Olenellidae and Holmiidae). Despite several errors in Lieberman's (Reference Lieberman1998, Reference Lieberman1999, Reference Lieberman2001) phylogenetic analyses (e.g., Webster, Reference Webster2007, Reference Webster2009), and the very limited and unrepresentative number of species coded, we herein prefer to assign Callavia to the (questionably monophyletic) ‘Judomioidea’ (sensu Lieberman, Reference Lieberman2001) instead of Olenelloidea, and we avoid family assignment within it.

Type species

Olenellus (Mesonacis) broeggeri Walcott, Reference Walcott1890 from the Brigus Formation, Branchian Series (Cambrian Stage 3/4), Newfoundland, Canada.

Other species

Olenellus (Holmia) callavei Lapworth, Reference Lapworth1891 from the Comley Limestone Formation, Branchian Series (Cambrian Series 2), Shropshire, UK; Paradoxides choffati Delgado, Reference Delgado1904 from the lower part of the Vila Boim Formation, Marianian (Cambrian Series 2), Vila Boim, Portugal (see Table 1).

Table 1. List of taxa previously assigned to Callavia Matthew, Reference Matthew1897 and their currently accepted generic assignment. Notes: 1—Fletcher and Theorokritoff (Reference Fletcher and Theokritoff2008) regarded Callavia broeggeri (Walcott, Reference Walcott1890) and Callavia crosbyi (Walcott, Reference Walcott1890) as different species; 2—Lieberman (Reference Lieberman1999) considered Paedeumias Walcott, Reference Walcott1910 a junior synonym of Olenellus Hall, Reference Hall1862. Webster (personal communication, 2021) regarded Paedeumias breviloba Poulsen, Reference Poulsen1927 as an indeterminate olenelline species.

Emended diagnosis

Posterior margin and posterior furrow of cephalon curved forward; base of genal spine lying slightly posterior to lateral margins of LO; genal spine broad-based; intergenal spine present, prominent to reduced; cephalic border developed as rounded ridge; anterior and lateral border furrows broad and deep; long tropidium-like structure extending across the lateral and anterior border furrows. Glabella subcylindrical, slightly tapered anteriorly; LA not contacting anterior border furrow, surrounded by a weak parafrontal band; LA not enlarged; preglabellar field very short, almost indistinct; occipital furrow (SO) not conjoined medially; occipital spine present; four preoccipital glabellar furrows (L1−L4) shallowing anteriorly, nontransglabellar, straight to slightly obliquely backward, when followed adaxially. Ocular lobe prominent; inner margin differentiated from a broad interocular area; extraocular area slightly narrower to slightly wider (tr.) than interocular area opposite S1; posterior tip of ocular lobe opposite SO. Intergenal ridge and posterior ocular line subparallel to converging toward the intergenal spine/swelling. Sculpture of reticulated pattern on external surface and terrace ridges along the abaxial limit of the anterior border.

Remarks

Callavia is one of those genera for which previous documentation and assigned species strongly exceed its currently accepted diversity (Table 1). In its most recent concepts, Callavia is extremely poorly diverse, ranging from monotypic (Lieberman, Reference Lieberman2001) to including only two species (e.g., Landing et al., Reference Landing, Westrop and Bowring2013b). Although describing Callavia as “the principal genus of the Olenellina from Avalonia,” Palmer and Repina (Reference Palmer and Repina1993, p. 14) considered a greater species diversity. This could also have led these authors to diagnose Callavia with characters that are not present in the type species, Callavia broeggeri. In fact, the figured material of this species is quite limited (Grabau, Reference Grabau1900; Walcott, Reference Walcott1910; Hutchinson, Reference Hutchinson1962; Landing et al., Reference Landing, Nowlan and Fletcher1980; Palmer and Repina, Reference Palmer and Repina1993; Lieberman, Reference Lieberman2001; Fletcher, Reference Fletcher2006), being mostly deformed or fragmented, including the type material (as stated by Hutchinson, Reference Hutchinson1962, p. 119).

In our opinion, several morphological characters have been misinterpreted: (1) presence or absence of the preglabellar field, (2) the tropidium-like ridge, and (3) the parafrontal band. Palmer and Repina (Reference Palmer and Repina1993) and Lieberman (Reference Lieberman2001) considered a preglabellar field as absent, with the frontal lobe directly contacting the anterior border furrow. Nevertheless, several illustrations of type and other material of Callavia broeggeri (and its possible junior synonym Callavia crosbyi Walcott, Reference Walcott1910) by Walcott (Reference Walcott1890, pl. 91, fig. 1, pl. 92, fig. 1, 1g; Reference Walcott1910, pl. 27, figs. 1, 4, pl. 28, fig. 4), as well as other figured specimens (e.g., Hutchinson, Reference Hutchinson1962, pl. 24, figs. 8–11; Palmer and Repina, Reference Palmer and Repina1993, fig. 6.8), clearly show a short but defined preglabellar field. One of the features that had contributed to this misinterpretation is the presence of a tropidium-like structure (e.g., Walcott, Reference Walcott1890, pl. 92, fig. 1b; Reference Walcott1910, pl. 28, figs. 1, 4; Hutchinson, Reference Hutchinson1962, pl. 24, figs. 7b, 8, 9; Palmer and Repina, Reference Palmer and Repina1993, fig. 6.8; Lieberman, Reference Lieberman2001, fig. 2.1) that is adaxial to the true anterior border furrow. Furthermore, and as previously stated by Fletcher and Theokritoff (Reference Fletcher and Theokritoff2008), a weak parafrontal band is present around the LA margins (e.g., Walcott, Reference Walcott1910, pl. 28, figs. 1, 4; Hutchinson, Reference Hutchinson1962, fig. 7a; Palmer and Repina, Reference Palmer and Repina1993, fig. 6.8), a character that led previous authors to consider the preglabellar field as absent in Callavia broeggeri. The parafrontal band is also observed in Callavia callavei (already stated by Lake, Reference Lake1937) and Callavia choffati (being clear only in better preserved specimens).

The parafrontal band was also described by Walcott (Reference Walcott1910), who erected the new species Callavia crosbyi based on this character, among others. We agree with Lieberman (Reference Lieberman2001) who considered Callavia crosbyi as a junior synonym of Callavia broeggeri. Fletcher and Theokritoff (Reference Fletcher and Theokritoff2008) argued that Callavia crosbyi is a valid species, differing from Callavia broeggeri in having a much narrower (tr.) extraocular area and a distinct pygidium and posteriormost thoracic segments. Nevertheless, the only Callavia crosbyi specimen preserving the thorax and the pygidium (Fletcher and Theokritoff, Reference Fletcher and Theokritoff2008, fig. 5.16) shows an extraocular area proportionally similar to that of Callavia broeggeri.

Another misinterpreted character of Callavia broeggeri is the morphology of S1, which Lieberman (Reference Lieberman2001) considered conjoined medially and different from the condition observed in Callavia callavei. Based on this difference, he erected the new monotypic genus Callavalonia Lieberman, Reference Lieberman2001 for this latter species. Although the glabellar segmentation of Callavia callavei and Callavia broeggeri present some differences, namely the glabellar furrows are apparently more deeply incised and the anteriormost furrows longer (tr.) in the former, it is not possible to assure that S1 is conjoined medially in Callavia broeggeri. Fletcher and Theokritoff (Reference Fletcher and Theokritoff2008) also considered this character to be unrecognizable. In the studied material of Callavia choffati, the collapse due to the flattening of the glabella in some specimens created an artifact, with S0 or S1 appearing conjoined medially (e.g., Fig. 2.1, 2.5). Nevertheless, in specimens preserving glabellar convexity, either in Callavia choffati (e.g., Fig. 3.5, 3.16) or in Callavia broeggeri (see Palmer and Repina, Reference Palmer and Repina1993, fig. 6.8), it is clear that they are not transglabellar.

Figure 2. Callavia choffati (Delgado, Reference Delgado1904), Vila Boim Formation, middle Marianian, Vila Boim, Portugal: (1) MG 15781, lectotype; (2) MG 15782a, paralectotype; (3, 4) MG sn, paralectotype (latex): (3) dorsal view; (4) detail of terrace lines of the lateral cephalic border; (5) MG 15786, paralectotype; (6) MG 16684 (latex). Scale bars = 2 mm (4); 10 mm (1–3, 5, 6).

Figure 3. Callavia choffati (Delgado, Reference Delgado1904), ‘Herrerías shale,’ middle Marianian, Cañaveral de León (1−13, 15−20) and ‘Cumbres beds,’ middle Marianian, Cumbres de San Bartolomé (14), Spain: (1, 2) UHU-CVL 01: (1) dorsal view; (2) lateral view; (3) UHU-CVL 03; (4) UHU-CVL 05; (5, 6) UHU-CVL 09: (5) lateral view; (6) dorsal view; (7) UHU-CU 1/1/1; (8, 9) UHU-CVL 12: (8) dorsal view; (9) lateral view; (10) UHU-CVL 10; (11) UHU-CVL 17 (latex); (12) UHU-CVL 20; (13) UHU-CVL 40; (14−16) UHU-CVL 32: (14) dorsal view; (15) frontal view; (16) dorsal view; (17) UHU-CVL 29; (18) UHU-CVL 42 (latex); (19) UHU-CVL 46; (20) UHU-CVL 48. Scale bars = 2 mm (4); 5 mm (6, 7, 11, 13–16, 18); 10 mm (1–3, 5, 8–10, 12, 17). Arrows in 7, 8, 18, and 20 indicate intergenal spines.

Lieberman (Reference Lieberman2001) also differentiated Callavalonia from Callavia by the relative width of the thoracic pleural furrow, which he considered to extend approximately two-thirds of the width of the inner pleural region in Callavia callavei, being longer (tr.; approximately four-fifths) in Callavia broeggeri. This character seems to be dependent on the thoracic segment number as well as on preservation. On a complete specimen of Callavia choffati (Fig. 4.1), it is possible to verify pleural furrows extending to different widths of the inner pleura (e.g., compare the first, second, third, and sixth segments). On the other hand, complete specimens of Callavia broeggeri (see Palmer and Repina, Reference Palmer and Repina1993, fig. 6.5) show a pleural furrow extension similar to that of Callavia callavei (occupying only two-thirds of the inner pleura). For these reasons, we agree with Jell and Adrain (Reference Jell and Adrain2002), Fletcher (Reference Fletcher2006), Fletcher and Theokritoff (Reference Fletcher and Theokritoff2008), and Landing et al. (Reference Landing, Westrop and Bowring2013b), who treated Callavalonia as a junior synonym of Callavia. Nevertheless, we do not concur with Fletcher (Reference Fletcher2006) and Fletcher and Theokritoff (Reference Fletcher and Theokritoff2008), who treated Callavia broeggeri and Callavia callavei as synonyms. Landing et al. (Reference Landing, Westrop and Bowring2013b) criticized this synonymy based on different eye-lobe positions, and we add cephalic segmentation to the list of differences between the two species.

Figure 4. Callavia choffati (Delgado, Reference Delgado1904), Vila Boim Formation, middle Marianian, Vila Boim, Portugal (1, 4, 6–9) and ‘Herrerías Shale,’ middle Marianian, Cañaveral de León, Spain (2, 3, 5): (1, 4) MG 15787b, paralectotype (latex): (1) overview; (4) detail of pygidium; (2) UHU-CVL 49; (3) UHU-CVL 36; (5) UHU-CVL 37; (6) MG 15707b, paralectotype (latex); (7) MG 15764, paralectotype; (8) MG 15714, paralectotype; (9) MG 16658, paralectotype. Scale bars = 5 mm (2–5); 10 mm (1, 6–9).

Figure 5. Distribution of Callavia plotted on the Cambrian paleogeographic map (modified from Scotese and McKerrow, Reference Scotese and McKerrow1990; McKerrow et al., Reference McKerrow, Scotese and Brasier1992; Dalziel, Reference Dalziel1997; and Malinky and Geyer, Reference Malinky and Geyer2019).

Lieberman (Reference Lieberman2001) erected Sdzuyomia as a monotypic genus represented by Callavia? lotzei (herein revised). We consider Callavia? lotzei a junior synonym of Paradoxides choffati (see remarks on species) and assign it to the genus Callavia. Consequently, Sdzuyomia is treated as a junior synonym of Callavia. Lieberman (Reference Lieberman2001) carried out a phylogenetic analysis of the suborder Olenellina and erected the genus Sdzuyomia to incorporate solely the type species Callavia? lotzei, grouping it within the superfamily Judomioidea with Callavalonia (= Callavia), Bondonella Hupé, Reference Hupé1953, Neltneria Hupé, Reference Hupé1953, Callavia, and Judomia. Based on the new specimens from the Cumbres de San Bartolomé, Cañaveral de León, Sierra del Bujo, and Hinojales fossil sites, it is possible to verify that Lieberman's (Reference Lieberman2001) erection of Sdzuyomia was based on miscoded characters in both the Spanish species and Callavia broeggeri. Among other characters, he considered the intergenal spines as absent in Sdzuyomia (character 39) and that in Callavia, the LA contacts the anterior furrow (character 5), the S1 is conjoined medially (character 30), and the thoracic pleural furrows occupy almost all of the inner pleural region (character 51). In our opinion, these characters were misinterpreted by both Lieberman (Reference Lieberman2001) and Palmer and Repina (Reference Palmer and Repina1993). We do consider Callavia broeggeri, Callavia callavei, and Callavia choffati to be comparable in most of the significant olenelline features (cephalic outline, border structure, position and configuration of the ocular lobes and their relationship with the glabella, presence and position of the intergenal spines, glabellar outline and lobation, and thoracic structure). As Geyer (Reference Geyer2007) argued, the establishment of Sdzuyomia by Lieberman (Reference Lieberman2001) was premature, especially because to that date the known material of ‘Callavia? lotzei’ was poorly preserved and very limited. Currently, and with better knowledge of this Iberian species, we reinforce its assignment to Callavia.

Previous generic assignment of Spanish material (‘Callavia? lotzei’ = Callavia choffati) to the genus Judomia by Sdzuy (Reference Sdzuy2001) is here rejected. After comparison with figured material from Siberia (Khomentovskii and Repina, Reference Khomentovskii and Repina1965; Repina et al., Reference Repina, Lazarenko, Meshkova, Korshunov, Nikiforov and Aksarina1974; Korobov, Reference Korobov1989; Pegel, Reference Pegel2000; Ponomarenko, Reference Ponomarenko2005; Rozanov and Varlamov, Reference Rozanov and Varlamov2008) and Laurentia (Fritz, Reference Fritz1973; McMenamin, Reference McMenamin1987; Gapp et al., Reference Gapp, Lieberman, Pope and Dilliard2011), we do not agree that Callavia choffati mainly differs from Judomia in ocular structure. The ocular lobes in Judomia are located very close to the glabella, thus the interocular area is almost absent. Furthermore, the inner margin of the ocular lobe is undifferentiated or only weakly differentiated from the interocular area in Judomia, which is a very narrow (tr.), depressed area (e.g., J. granulata Repina in Repina et al., Reference Repina, Lazarenko, Meshkova, Korshunov, Nikiforov and Aksarina1974, J. mattajensis Lazarenko, Reference Lazarenko1962, J. tera Lazarenko in Kryskov et al., Reference Kryskov, Lazarenko, Ogienko and Tchernysheva1960, and J. rasskasovae Korobov, Reference Korobov1963 as figured by Repina et al., Reference Repina, Lazarenko, Meshkova, Korshunov, Nikiforov and Aksarina1974, pls. 27, 28). On the contrary, in Callavia, the interocular area is broad and inflated, almost the same width as the extraocular area opposite S1, and the inner margin of the ocular lobes is well differentiated from it. Other characters are the straight and deep posterior border furrow in Judomia (e.g., J. tera by Repina et al., Reference Repina, Lazarenko, Meshkova, Korshunov, Nikiforov and Aksarina1974, pl. 28, fig. 1; Palmer and Repina, Reference Palmer and Repina1993, fig. 10.7), being faint and curved forward in Callavia due to a prominent intergenal ridge; and the posteriormost LA in direct contact with the ocular lobes in Judomia (e.g., J. rasskasovae as figured by Repina et al., Reference Repina, Lazarenko, Meshkova, Korshunov, Nikiforov and Aksarina1974, pl. 27, fig. 9). Other putative differences based on Palmer and Repina's (Reference Palmer and Repina1993) diagnosis of Judomia (e.g., the absence of intergenal spines) are not considered herein because some Judomia species do bear tiny intergenal spines very similar to those observed in Callavia choffati (e.g., J. rasskasovae as figured by Repina et al., Reference Repina, Lazarenko, Meshkova, Korshunov, Nikiforov and Aksarina1974, pl. 27, fig. 4.5), although in a more proximal position (adaxial to the adgenal angle). Even though we do not agree with Sdzuy's (Reference Sdzuy2001) assignment of ‘Callavia? lotzei’ to the genus Judomia, we do recognize several characters in common, not only in cephalic morphology but also in the thoracic structure (compare Fig. 4.1 with Palmer and Repina, Reference Palmer and Repina1993, fig. 10.7). A close relationship between these taxa was already suggested by Lieberman (Reference Lieberman2001) through a classification that is adopted herein (see remarks on the superfamily).

With regard to other documented occurrences of Callavia, and excluding those listed in Table 1, the genus was identified in the Purley Shales, Warwickshire, by Rushton (Reference Rushton1966) and Williams et al. (Reference Williams, Ruston, Cook, Zalasiewicz, Martin, Condon and Winrow2013), who described as Callavia? sp. a few fragments, mainly based on sculpture similar to that presented by Callavia callavei. Part of this material had been previously documented by Pringle (Reference Pringle1913), Illing (Reference Illing1913, Reference Illing1916), and Smith and White (Reference Smith and White1963). Based on the figured material (Rushton, Reference Rushton1966, pl. 4, figs. 27, 28; Williams et al., Reference Williams, Ruston, Cook, Zalasiewicz, Martin, Condon and Winrow2013, fig. 4), it is not possible to identify them to generic or suprageneric levels, and additional material from those beds is necessary to confirm the presence of Callavia in those levels.

After several misconceptions that indicated the presence of Callavia in the lower Cambrian of Morocco (Neltner and Poctey, Reference Neltner and Poctey1950; Hupé, Reference Hupé1959) these were later reassigned by Geyer and Palmer (Reference Geyer and Palmer1995) to other genera. Geyer and Landing (Reference Geyer and Landing2002) reported the presence of this genus in this region, namely a single cephalic fragment in the Bani section, Moroccan Anti-Atlas (Geyer, personal communication, 2020).

Type specimens

Lectotype (selected herein), MG 15781, an internal and external mold of a cephalon (Fig. 2.1), figured by Delgado (Reference Delgado1904, pl. 1, fig. 3) and Teixeira (Reference Teixeira1952, pl. 1). Paralectotypes of one incomplete exoskeleton, MG 15787a with counterpart 15787b (Fig. 4.1; Delgado, Reference Delgado1904, pl. 1, fig. 16, pl. 5, fig. 3; Teixeira, Reference Teixeira1952, pl. 4, figs. 1–3); MG 15782a with counterpart 15782b, five cephala (Fig. 2.2; Delgado, Reference Delgado1904, pl. 1., fig. 1; Teixeira, Reference Teixeira1952, pl. 2, figs. 1–3; MG 15785, external mold (Teixeira, Reference Teixeira1952, pl. 4, figs. 5, 6); MG 15786, internal mold (Fig. 2.5; Delgado, Reference Delgado1904, pl. 1, fig. 1; Teixeira, Reference Teixeira1952, pl. 12, figs. 1, 2); MG 16684, external mold; MG sn, external mold (Fig. 2.3; Teixeira, Reference Teixeira1952, pl. 3, figs. 3, 4); MG 15779a with counterpart 15779b, two cephalic fragments (Delgado, Reference Delgado1904, pl. 4, fig. 53; Teixeira, Reference Teixeira1952, pl. 3, figs. 5, 6; MG 15780a with counterpart 15780b (Teixeira, Reference Teixeira1952, pl. 3, fig. 7); MG 16658, three thoracic segments (all internal molds) (Fig. 4.9; Delgado, Reference Delgado1904, pl. 1, fig. 12); MG 15714 (Fig. 4.8; Delgado, Reference Delgado1904, pl. 1, fig. 8); MG 15764 (Fig. 4.7; Delgado, Reference Delgado1904, pl. 1, fig. 6); MG 15707a with counterpart 15707b, one pygidium (Fig. 4.6; Delgado, Reference Delgado1904, pl. 1, fig. 10, pl. 5, fig. 35).

Emended diagnosis

Intergenal spine reduced; S4 poorly defined and short (tr.); interocular area width approximately four-fifths width of the extraocular area opposite S1; intergenal ridge and posterior ocular line equally prominent, subparallel to the intergenal spine. Thorax of 18 or probably 19 segments, tapering posterior to fifth segment.

Occurrence

Vila Boim Formation (type locality), Marianian (Cambrian Series 2) of Vila Boim, Portugal; lower part of Cumbres beds, middle Marianian (Cambrian Series 2), Cumbres de San Bartolomé, Huelva, Spain; and lower part of Herrerías shale, middle Marianian (Cambrian Series 2), Sierra del Bujo (Arroyomolinos de León), Cañaveral de León and Hinojales, Huelva, Spain.

Description

Cephalon crescent-shaped; sagittal length ~40% of maximum width at posterior margin, with higher relief of the anterior border, the ocular ridges, and the glabella. Known cephala range 2.1–18.9 mm length and 5.2–42.4 mm width. Posterior margin of cephalon curved backward distally. Glabella hourglass-shaped to parallel-sided, with faint constriction at S1, inflated dorsally, exceeding the genal areas in lateral view; maximum glabellar elevation at half of glabellar length, sloping downward anteriorly with rounded profile. Glabella longer than wide; posterior glabellar width ~115–120% the anterior glabellar width; corresponding to 25% of maximum cephalic width; sagittal glabellar length ~75–80% sagittal cephalic length. Axial furrows deep, slightly sinuous (outlining glabellar lobes). Occipital furrow moderately deep to shallow medially, oblique backward when traced adaxially; occipital ring moderately convex dorsally, frequently preserved as two symmetrical subrectangular lobes, bearing a small axial node near posterior margin. Four glabellar furrows (excluding occipital furrow) shallowing anteriorly, nontransglabellar, nearly straight, oblique, inward and backward ~10–15° to a transverse line. S1 subparallel to SO, occupying approximately two-thirds of glabellar width; L1 subrectangular, moderately inflated dorsally; S2 slightly less oblique than S1; L2 similar to L1; S3 subparallel to S2 but shorter (tr.); L3 shorter (exsag.) and narrower (tr.) than L1 and L2; S4 poorly defined, very shallow and narrow (tr.), located immediately posterior to the parafrontal band; L4 poorly defined, typically merged with the frontal lobe (Fig. 3.17). Frontal lobe of glabella tapered to slightly pointed, surrounded by parafrontal band connecting ocular ridges. Preglabellar field short, < 10% of sagittal cephalic length. Some specimens showing shallow furrow connecting preglabellar furrow with anterior border furrow (Fig. 3.7). Lateral border inflated dorsally, widened posteriorly, reaching maximum width at base of genal spines, defining broad genal point, directed backward. Anterior and lateral border furrows broad, deep, merging into significantly shallower posterior border furrow, but deeper at genal areas. Interocular area dorsally arched, elongated (exsag.), trapezoidal in outline, with two differentiated interocular swellings. Width (tr.) of interocular area approximately four-fifths of extraocular area width at S1 level. Ocular lobe prominent, arc-shaped, located slightly closer to glabella than to lateral border; exsagittal length equivalent to 40% of sagittal cephalic length. Posterior tip of ocular lobe opposite SO; anterior tip opposite L3. Inner margin of ocular lobe well defined. Ocular lobes anteriorly connected by parafrontal band, posteriorly connected to posterior ocular line. Pronounced intergenal ridge, slightly curved to sigmoidal, extending into reduced, almost indistinct, intergenal spine (Fig. 3.8, 3.18. 3.20), giving rise to change of convexity of posterior margin (small dorsal swelling). Intergenal spine located in exsagittal line with inner edge of lateral border furrow. Genal angle greater than intergenal angle. Sculpture composed of reticulated pattern, with extraocular genal caeca, and terrace ridges along abaxial limit of anterior border (Figs. 2.4, 3.19).

Thorax composed of 18, probably 19 thoracic segments in only complete specimen (Fig. 4.1, 4.2). Wide (sag.), convex, little-pronounced axial rings, narrowing toward back. Axial ring width (tr.) occupying 20% of total thoracic width anteriorly, 35% posteriorly. Axial ring furrows deep. Axial nodes on each thoracic segment. Lateral extension of pleural region not surpassing extension of genal spine of cephalon. Pleural region subtly widened (tr.) to third pleura, then progressively narrowing (tr.) toward posterior end. Pleurae thin (sag.), knife-shaped, slightly curving backward; curvature progressively increasing posteriorly. Pleural furrow deep, wide, extending 35% of whole pleura (tr.), slightly oblique. Curvature located at approximate midlength of pleural width, with pleural spine widened (exsag.) at that point. Last segments of thorax fused with pygidium.

Pygidium very reduced, relatively narrow. Pygidial rachis well-defined, with triangular outline, composed of three faint axial rings. Pleural regions poorly preserved, posteriorly extended.

Remarks

Callavia choffati differs from the type species Callavia broeggeri and Callavia callavei in having shorter and narrower intergenal spines, shorter (tr.) interocular areas than in Callavia broeggeri, and a different glabellar lobation (fainter glabellar furrows and shorter S3 and S4) than in Callavia callavei.

A group of specimens described by Delgado (Reference Delgado1904) under different species names are conspecific with ‘Paradoxides’ choffati, as previously stated (e.g., Richter and Richter, Reference Richter and Richter1941; Teixeira, Reference Teixeira1952; see synonymy for further information). Callavia choffati was originally described by Delgado (Reference Delgado1904) as a species of Paradoxides and quickly after his publication, Charles Schuchert (in Dana, Reference Dana1905, p. 159) stated that “Paradoxides choffati is clearly an Olenellus.” In fact, Delgado (Reference Delgado1904, p. 320) expressed the same opinion in the original publication: “… ces exemplaires pourraient être exclus du genre Paradoxides (s. str.) et réunis plutôt à Olenellus, sous-genre Holmia …” It was the observation of ‘clear’ facial sutures in Delgado's (Reference Delgado1904) opinion, especially in the specimen herein selected as lectotype (Fig. 2.1), which justified the assignment to Paradoxides. Today, it is widely agreed that these cephalic ‘lines’ observed in many specimens of Olenellina are in fact fractures (Whittington, Reference Whittington1989, Reference Whittington1997), and that the suborder Olenellina is characterized by lacking dorsal sutures throughout its entire ontogeny. But during several decades, the regular presence in olenellines and the curious configuration of these lines (which mimic the anterior branches of a facial suture) led to much discussion about their significance (e.g., Størmer, Reference Størmer1942; Hupé, Reference Hupé1953; Bergström, Reference Bergström1973). As Geyer (Reference Geyer1996) stated, they could represent favored loci for fracturing due to thinner or less calcified cuticle. In the studied material of Callavia choffati, these fractures are, in fact very common (Fig. 3.7, 3.10, 3.11, 3.13–3.19).

Some decades after Delgado's (Reference Delgado1904) publication, Richter and Richter (Reference Richter and Richter1941) erected a new species, Callavia? lotzei, from the northern Huelva province (Spain). These authors suggested for the first time that ‘Paradoxides’ choffati should be also assigned to the genus Callavia. Nevertheless, and although they considered it very similar to the newly erected Callavia? lotzei, they did not specify the morphological features that justified the erection of a new species. Furthermore, Richter and Richter (Reference Richter and Richter1941) considered that Portuguese and Spanish Callavia-bearing assemblages are coeval and share some taxa at specific levels, thus correlation would benefit from such discussion and eventual synonymy. While revising Delgado's (Reference Delgado1904) material, Teixeira (Reference Teixeira1952) followed Richter and Richter's (Reference Richter and Richter1941) suggestion and maintained the Portuguese species assigned to the genus Callavia. A few years later, Hupé (Reference Hupé1960) suggested that Callavia? lotzei and Callavia choffati might be synonyms, but Sdzuy (Reference Sdzuy1962) rejected this hypothesis based on deeper glabellar furrows and larger cephala of Callavia choffati. Furthermore, Sdzuy (Reference Sdzuy1962) considered that both Iberian species bear significant differences when comparing with Callavia, namely the lack of intergenal spines and the absence of a large occipital spine (both structures are indeed present but based on the poorly preserved material that Sdzuy had at his disposal, it was not possible to verify them). For these reasons, Sdzuy (Reference Sdzuy1962) suggested that Callavia choffati was more closely related to Kjerulfia Kiær, Reference Kiær1917, whereas Callavia lotzei should be assigned to Judomia, with which it shares several morphological characters, although he maintained both species as Callavia?. This proposed assignment was later reinforced by the author (Sdzuy, Reference Sdzuy2001), who definitely transferred the Spanish species to the genus Judomia, stating that it was unexpected because Judomia is characteristic of the Siberian domain. Among the characters that Sdzuy (Reference Sdzuy2001) listed in common, are the absence of intergenal spines and the pygidium configuration, which he considered to be much more similar to those of Judomia than Callavia. Nevertheless, the intergenal spines are present in Callavia lotzei, and Whittington (Reference Whittington1989, p. 134) stated that the “pygidium attributed to Callavia by Raw (Reference Raw1936, pl. 21, figs. 3a–c) is composed largely of one pair of spines similar in form to those of” Judomia tera. In that work, Sdzuy (Reference Sdzuy2001) also reinforced the independence of Callavia choffati and Callavia lotzei through photographic retrodeformation of Portuguese figured specimens, but the results are not reliable (Sdzuy, Reference Sdzuy2001, figs. 13, 14).

The new data described in this work suggest that Callavia choffati and Callavia lotzei are synonyms and support the assignment of the Iberian species to the genus Callavia. The previous poor documentation of Callavia choffati from the Vila Boim Formation (Portugal), namely short descriptions and low resolution photographs (Delgado, Reference Delgado1904; Teixeira, Reference Teixeira1952), together with the large size of the available specimens, certainly hinder proper comparison with ‘Callavia? lotzei’ from the Herrerías shale (Spain). On the other hand, in erecting ‘Callavia? lotzei,’ Richter and Richter (Reference Richter and Richter1941) included specimens in this species belonging to Gigantopygus cf. G. bondoni Hupé, Reference Hupé1953, a redlichiine, and Sdzuy (Reference Sdzuy1962, pl. 19, figs. 7–11) included some additional material of meraspides. These compromised comparison with the Portuguese material. The great number of newly collected, well-preserved specimens from the Herrerías shale allowed clarification of the morphology of ‘Callavia? lotzei’ and verification that it agrees with Callavia choffati in all characters. Reduced intergenal spines are present in both sets of specimens, located in similar positions (e.g., Fig. 3.1–3.17). This structure seems to be progressively reduced to a tiny spine or node on the posterior cephalic border through ontogeny, being larger and more evident in small specimens (Fig. 3). Due to the deformation and large size of the type specimens (Fig. 2), the intergenal spine is not evident, being expressed as a swelling on the posterior border, and more evident in only one specimen (Fig. 2.3). The inflated cephalic border, widening posteriorly, is one of the characters that Sdzuy (Reference Sdzuy2001) considered to separate Portuguese and Spanish forms; it has the same configuration in all of the studied specimens. The apparent relatively shorter cephalic border in the Portuguese material (Fig. 2.1–2.5) could be attributable to differences in size due to ontogenetic allometry. Another important comparative character is the glabellar morphology and lobation; both sets of specimens show a constriction opposite S1 (Figs. 2.1, 3.17, 3.19), resulting in a faint hourglass-shaped glabella in specimens preserving part of the original convexity. The outline and position of the ocular lobes are indistinguishable in Portuguese and Spanish specimens, including the typical swellings of the interocular areas (e.g., Figs. 2.1, 2.2, 3.11, 3.17). Furthermore, both bear an entirely comparable, faint parafrontal band surrounding the LA (Figs. 2.1–2.3, 3.7, 3.13–3.18) and a well-marked posterior ocular line subparallel to the intergenal ridge (Figs. 2.1, 3.18, 3.19). The few articulated specimens also show the same thoracic structure (compare Sdzuy, Reference Sdzuy1962, pl. 9, fig. 6 with Fig. 4.1). For these reasons, ‘Callavia? lotzei’ is herein considered a junior synonym of Callavia choffati.