Podospongiidae

Family Podospongiidae de Laubenfels, 1936

In the studied material, there are numerous monaxial spicules with regularly arranged whorls of spines (that are of the same size along the whole length of the spicule), called sanidasters (Fig. 18 A–I). They vary from thin-shafted (Figs. 18 A–C) to more thick-shafted (Figs. 18 D–I). The slender version is identical with spicules of podospongiid Sceptrintus Topsent, 1898. This monotypic genus is represented by the species Sceptrintus richardi Topsent, 1898 (compare with Fig. 17). A wide range of size and length (250–500 µm) of the spicules is characteristic for both, Recent Sceptrintus and the spicules described here from the Eocene sediments. Today, sponges of this species are known only from the Azores, Canaries, and Madeira (Topsent 1989; van Soest et al. 2013). The spicules described here either belong to S. richardi that had a wider distribution in the geological past, or to another extinct species.

However, the first possibility seems more plausible because there are no other spicules in present-day sponges that are so similar to those described from the Late Eocene assemblage. The rest of the spicules (Figs. 18 D–I), which differ in thickness from those of Sceptrintus richardi, may belong to the different, extinct species of Sceptrintus as well.

Interestingly, Vosmaer & Vernhout (1902, pl. 4, figs. 5, 8) illustrated spicules (sterospiras sensu selenasters) resembling those described here from the Eocene as early developmental stage of selenasters of Placospongia melobesioides Gray, 1867b. One has to add that the other spicules characteristic for Placospongia were also present in the studied material (compare with the family Placospongiidae). But in my opinion the acanthorhabds described here are not the early developmental stage of placospongiid spicules but of Sceptrintus sanidasters.

The fossil record of these spicules is from the Late Jurassic of Germany (Schrammen 1936, pl. 9, fig. 18), from the Eocene of the western Atlantic (Burky 1978), and the Miocene of the Gulf of Mexico (Ehrenberg 1873, pl. 5, fig. 22).

Besides the podospongiid spicules mentioned above, there are numerous - about 60–100 µm long - dumb-bellshaped anisodiscorhabds (Figs. 19 A–E) found in the studied material. The upper, mushroom-shaped part of the spicule possess well-defined ornamentation, and the lower part is also clearly spined but without the articulated apex. In-between the apexes, around the short shaft, there is a serrated collar/fringe disc. The microscleres show a slight variability in the number and the size of the spines what could suggest that they belong to more than one species (e.g., compare Fig. 19 B with Fig. 19 E).

The podospongiid species Sigmosceptrella is characterized by possessing anisodiscorhabds and there are four species of Sigmosceptrella known: S. hospitalis Desqueyroux-Faúndez & van Soest, 1997, S. quadrilobata Dendy, 1922, S. carlinae (Boury-Esnault & van Beveren, 1982), and S. fibrosa (Dendy, 1897). However, the first species lacks spicules similar to those described here. On the other hand, comparison of the Late Eocene anisodiscorhabds with those of Recent Sigmosceptrella carlinae (Fig. 20 A; originally described as Latrunculia carlinae) or S. quadrilobata (Fig. 20 B) reveals great similarity to these spicules. Additionally, in the studied material there is one more spicule morphotype (Fig. 29 P) noted that resemble those of S. quadrilobata (compare with Dendy 1922, pl. 18., fig. 4B). S. carlinae was described only from the southern Indian Ocean (Kerguelen Islands; Boury-Esnault & van Beveren, 1982) and S. quadrilobata was noted from the Indian Ocean, Tulear, and Madagascar (van Soest et al.

2013). The size of the Late Eocene spicules is bigger (exceeding from about 60 µm to 100 µm) than the spicules of recent representatives (about 50 µm). However, the general decrease of Recent spicule morphotypes is clearly visible in most of the studied cases. Along the Australian continent only Sigmosceptrella fibrosa occur (Atlas of Living Australia), but still the spicules described here resembles those of S. quadrilobata the most. On the other hand, the separation of these two species (S. fibrosa and S. quadrilobata) is dubious (see Kelly & Samaai 2002, and the discussion therein) what is clearly supported by my observations. Taking into consideration all of these arguments, the spicule studied here may be assigned, in my opinion, to Sigmosceptrella quadrilobata.

It is worth mentioning that there are also some anisodiscorhabds of another podospongiid Podospongia virga Sim-Smith & Kelly, 2011 (compare with Sim-Smith & Kelly 2011, fig. 3C) and latrunculid Latrunculia crenulata (Lévi, 1993) also possess microscleres similar to spicules described here (compare with Lévi 1993, fig. 10B). The microscleres of L. crenulata look like more elongated forms of late Eocene scleres described here possessing at the same time all the features (i.e., ornamented heads with spined lower parts and serrated collar discs in the centre of the spicule) characteristic of sigmosceptellids’ spicules. It proves the close relationship of these two poecilosclerid families.

Anisodiscorhabds similar to those described here were already reported from the fossil state. For example, Schrammen (1924; pl. 7, fig. 41) noted identical “cricophalangasters” from the Late Cretaceous of Germany but without any particular assignment. As no fossil record of the genus Sigmosceptrella quadrilobata exists, this is the first fossil record and the oldest occurrence of this genus.