Partial taxonomic revision of Amphoriscus Haeckel, 1870 (Porifera: Calcarea) with description of A. decennis sp. nov. Author Chagas, Cléslei Universidade Federal da Bahia, Instituto de Biologia. Rua Barão de Jeremoabo s / n, Ondina, Salvador / Bahia, Brazil. 40170 - 115. Author Cavalcanti, Fernanda F. 0000-0002-1619-0860 Universidade Federal da Bahia, Instituto de Biologia. Rua Barão de Jeremoabo s / n, Ondina, Salvador / Bahia, Brazil. 40170 - 115. & fernanda. porifera @ gmail. com; https: // orcid. org / 0000 - 0002 - 1619 - 0860 fernanda.porifera@gmail.com text Zootaxa 2021 2021-11-03 5061 1 39 68 journal article 3616 10.11646/zootaxa.5061.1.2 0b7d66df-755c-4a1a-b925-3c71e1e8fab4 1175-5326 5642287 EFE2D12B-3F02-4831-A913-DEFE1F9A0C92 Amphoriscus gastrorhabdifera ( Burton, 1932 ) Leucaltis gastrorhabdifera Burton, 1932 Citations and Synonymies: Leucaltis gastrorhabdifera Burton, 1932: 259 , figs. 4-5; Amphoriscus gastrorhabdifera Burton, 1963: 136 , 147, 548; Klautau et al . 2017: 105-106 ; Cóndor-Luján et al. 2019: 1825 . Type material: BMNH 1928.2 .15.833 ( Holotype ; St. 6, Tristan da Cunha , South Atlantic ; 80–140 m deep) . Type locality: Tristan da Cunha , South Atlantic . Morphology: Colour is beige after fixation. The holotype is a tubular fragment ( Fig. 12A ), and a fringe of trichoxeas was present, according to Burton (1932) . The type of aquiferous system is unclear. Anatomy: The skeleton has no similarity with the other species of Amphoriscus . The cortical region is formed by triactines of varying sizes and subcortical giant tetractines (which are possibly the only typical character of Amphoriscidae ). However, there is no typical inarticulation ( Fig. 12B ). The presence of several broken spicules makes the visualisation of the triactines difficult ( Fig. 12C ). The apical actines of the subcortical tetractines cross the entire choanosome, often perforating the atrium ( Figs. 12D, E ). The subatrial/atrial region is comprised exclusively of large tangential diactines. They are abundant and line the atrial cavity ( Figs. 12E, F ). A specific atrial skeleton comprised of triactines or tetractines is absent. According to Burton (1932) , large diactines similar to those found in the inner part of the sponge were found projecting from the cortex, but they were not observed here. The aquiferous system could not be recognised and was not mentioned along the original description. FIGURE 12. Amphoriscus gastrorhabdifera (holotype BMNH 1928.2.15.833). A— Holotype in ethanol. B— General view of a skeletal section. C— A triactine (arrow) at the cortical region. D— Giant tetractines organised at the skeleton. E— Apical of the tetractines (arrows) perforating the atrial wall and layer of diactines surrounding the atrium. F— Subatrial/ atrial region composed exclusively of diactines (arrow). Abbreviations: at—Atrium and ct—cortex. Spicules : Cortical triactines: Regular, actines are slightly conical and sharp. The actines are straight (paired actines: 123.1– 150.7 ±21.6–204.4/ 8.4– 12.2 ± 1.7–15.3 μm; unpaired actines: 81.5– 133.6 ±23.2–180.0/ 8.9– 122.3 ±2.0–18.3 μm) ( Fig. 13A ). Subcortical tetractines: Giant, actines are slightly conical to conical and blunt. The paired actines are slightly curved (111.4– 146.6 ±21.8–192.1/ 10.8– 14.5 ± 2.0–19.2 μm). The unpaired actine is curved from the base to the tip (93.9– 116.7 ±21.5–156.6/ 14.1– 15.2 ±1.1–16.9 μm). The apical actine is long and conical (174.3– 228.3 ±29.1–285.4/ 16.9– 20.8 ±2.2–26.3 μm) ( Fig. 13B ). Diactines: Large, sinuous along their length. Most of them have one of the tips blunt while the other is sharp (330.0– 464.8 ±89.6–632.6/ 9.0– 18.9 ±4.0–25.7 μm) ( Fig. 13C ). FIGURE 13. Amphoriscus gastrorhabdifera (holotype BMNH 1928.2.15.833). A— Cortical triactine. B— Subcortical tetractine. C— Subatrial/atrial diactine. Remarks: This species was originally described as Leucaltis gastrorhabdifera (subclass Calcinea ) and was later transferred to Amphoriscus (subclass Calcaronea ) ( Burton 1932 , 1963 ). Such a reallocation would, by the standards of modern-day research on Calcarea systematics, be considered out of the ordinary and require a detailed explanation, yet at that time Burton did not note the reasons supporting his decision in any detail. In his discussion of the genus Leucaltis , Burton (1963) argued that “ L. gastrorhabdifera is aberrant and is here doubtfully assigned to Amphoriscus ”, suggesting that the only certainty the author had was that it was not Leucaltis . He named the species “ Amphoriscus ? gastrorhabdifera ”, and although we suspect that he was influenced by the presence of the giant cortical tetractines, this cannot be confirmed.Also, there is no information or illustration in the literature about the aquiferous system of A. gastrorhabdifera , which would be essential to support its allocation in Amphoriscus . Whether or not it was one of the characters used by Burton to assign the species in Amphoriscus is a question that remains unanswered. We analysed the holotype (BMNH 1928.2.15.833) in this study. The microscopical slides contain sections of the skeleton that were probably not stained since it was not possible to unequivocally confirm whether A.gastrorhabdifera has the syconoid aquiferous system typical of the genus. We observed the main morphological characters reported in the original description, such as the presence of cortical triactines, subcortical giant tetractines with a long apical actine, and diactines. Burton (1932) mentioned diactines protruding through the cortex, but we did not observe this characteristic. The absence of a subatrial layer of triactines/tetractines was also confirmed and, consequently, the absence of inarticulate skeletal organisation. Therefore, the most important diagnostic characters typical of Amphoriscus either could not be confirmed (the syconoid aquiferous system) or are absent (the inarticulation formed by the apical actines of giant cortical tetractines and the unpaired actine of subatrial spicules). In order to assess the possibility of assigning A. gastrorhabdifera in another genus, a diagnosis of all genera under the order Leucosolenida was carried out. The presence of triactines and tetractines and of an atrial layer of diactines is a remarkable character found only in Sycodorus , although members of this genus are syconoid. Tangential triactines and tetractines also occur along the atrial cavity, suggesting that, even if A. gastrorhabdifera is syconoid, it could not be allocated in Sycodorus . An alternative decision would be to propose a new genus. The reasons why we did not choose this option to solve the problem of A. gastrorhabdifera were as follows: (i) the holotype is tiny, and no reports of additional specimens exist that could enrich our understanding of the morphology of the new genus; (ii) the lack of data on the aquiferous system could raise doubts on the family in which the genus should be inserted and also makes the description of a robust diagnosis difficult; (iii) the species is represented only by the holotype , found at Tristan da Cunha , Southern Atlantic Ocean, at a depth of 80 to 140 m . The lack of perspective in finding fresh samples to elucidate the questions mentioned earlier suggests that the problem with the species would persist (not as Amphoriscus but as a newly named genus); finally, (iv) we cannot undoubtedly rule out A. gastrorhabdifera actually belonging to Amphoriscus as the absence of an inarticulate organisation could be a secondary character caused by the loss of subatrial spicules along the species’ evolutionary history. The latter question will be resolved after the species is tested in phylogenetic analyses, though, as highlighted above, there is no fresh material available. Therefore, we decided to be conservative and avoid potentially increasing the number of open questions on the classification of this species. Amphoriscus gastrorhabdifera is thus maintained in the genus, but we indicate that it should be considered as incertae sedis until the discovery of additional samples.