Geodiidae Gray 1867

Family Geodiidae Gray, 1867a (Subfamily Erylinae)

Although the taxonomic position of the genus Penares is still uncertain (at first different species of Penares were even placed in different families, e.g., Ancorinidae, Plakinidae; for more details see World Porifera Database), in this paper it is considered as belonging to the geodiid subfamily Erylinae.

Very characteristic triaenes with leaf-shaped clad branches are among the studied spicules (Figs. 4 U–X) that are almost identical with those of Recent erylinid species Penares sclerobesa Topsent, 1904. Today, this species inhabits only the North Atlantic Ocean (the Azores; van Soest et al. 2013). However, the spicules with only slightly less leaf-shaped cladome characterize also other present-day species— Penares cf. sphaera (Lendenfeld, 1907) that occurs today in South Africa, and they are also close to some spicules of the sponges from the family Calthropellidae (see van Soest & Hooper 2002b, fig. 1E). Moreover, similar spicules are encountered in lithistids but they are less leaf-shaped than the ones described here. From Australia, there are only three species of Penares recorded: Penares sollasi Thiele, 1900, P. schulzei (Dendy, 1905), and P. ob t u s u s (Lendenfeld, 1907; Atlas of Living Australia), but none of them has spicules similar to those discussed here. Considering this fact, the studied spicules are interpreted as belonging to P. cf. sclerobesa.

From the fossil record similar spicules were described by Mostler (1986, pl. 8, figs. 5–6) from the Rhaetian of the Alps but without precise taxonomical assignment. Similar spicules are also described from the Jurassic of Germany by Dunikowski (1882, pl. 2, fig. 20) but assigned to Stelletta sp. This morphotype is also known from the numerous Cenozoic strata (see Wiedenmayer 1994). This study is the first time when Penares cf. sclerobesa was described in the fossil record.

Apart from undoubted geodiid spicules, some characteristic spicules called pynakids were found in the studied material (Figs. 4 J–P). These flat, elliptical, disc-shaped spicules with various peripherally arranged gaps (that can be marginally open) are assigned usually to the fossil species Dactylocalycites callodiscus (described from the Cenomanian of England) by Carter (1871, pl. 9, fig. 40). It was proposed that pynakids are modified discotriaenes, for example those of the family Theonellidae (de Laubenfels 1953), or that they may be final evolutionary stage of dichotriaenes (Mostler 1986). The pynakids however, are much smaller than dichotriaenes, have no rhabds, and are not tetraxial (as should be the case of disco- or dichotriaenes) but are multiaxial. All these arguments allow me to reject the above speculations on theonellid affinity/relationship of pynakids.

There is no Recent sponge bearing this spicule morphotype and among living forms these spicules resemble most the aspidasters of geodiid Erylus monticularis Kirkpatrick, 1900 (pl. 14B) (order Astrophorida). The large variability of the spicules discussed here (for example see the Fig. 4 N and P) may be an effect of different developmental stages of spicules (for more details see Wiedenmayer 1994), or they may be different morphotypes indicating the presence of several different species, but this problem cannot be solved with the material discussed. The pynakids are quite common in the fossil record and were described by Carter (1871) from the Cenomanian of England, Mostler (1986, figs. 35.12–15) from the Late Jurassic of the Alps, and by Schrammen (1924, pl. 7.43) from the Cretaceous of NW Germany. From the Cretaceous the pynakids were also described by Rüst (1885, fig. 35) and Krüger (1978). They are also common in the Cenozoic (Eocene to Oligocene) of Barbados (see Wiedenmayer 1994). This morphological type of spicules was also already described from the Late Eocene of Australia (Pallinup Siltstone and Princess Royal/Norseman) by Hinde (1910, pl. 2, fig. 12) and from New Zealand by de Laubenfels (1953, fig. 6).

Among studied spicules, there are also some other morphotypes (Figs. 4 Q–T) that most likely belong to the family Geodiidae but their more precise taxonomic assignment is not possible due to the fact that they do not have enough characteristic features. For example, some of these spicules resemble plesiasters and may occur in other demosponge groups (Fig. 4 T) that are also similar with those of pachastrellid Thenea Gray, 1867a. In my opinion, however, their geodiid affinity is more probable.