Clavodesia Harmelin & d’Hondt, 1992

Reverter-Gil, Oscar, Souto, Javier & Fernández-Pulpeiro, Eugenio, 2012, New and little known species of Bryozoa from Iberian Atlantic waters, Zoosystema 34 (1), pp. 157-170 : 159-162

Genus Clavodesia Harmelin & d’Hondt, 1992

TYPE SPECIES. — Nellia clavula Hayward, 1978 (senior subjective synonym of Clavodesia biradiculata Harmelin & d’Hondt, 1992 ).

CORRECTED DIAGNOSIS. — Colony erect, calcified, unjointed, with 4-6 series of membraniporiform autozooids. Gymnocyst reduced; opesia extensive; cryptocyst granular, reduced laterally and distally, more or less developed proximally. Small oral spines present. Distal autozooids of the colony prominent. Avicularia adventitious, lateral. Ovicell terminal, closed by the zooidal operculum. Ancestrula tubular, with two proximal calcified roots, inconstant.

Clavodesia clavula ( Hayward, 1978) n. comb. ( Fig. 1 ; Table 1)

Nellia clavula Hayward, 1978: 212 , fig. 2g.

Clavodesia biradiculata Harmelin & d’Hondt, 1992: 35 , fig. 1a, b.

MATERIAL EXAMINED. — Lectotype of Clavodesia biradiculata (designated here): MNHN 15493 (with other species). BALGIM stn DW11, 36°44.2’N, 9°31.4’W, 1505-1540 m, 29/V/1984.

DISTRIBUTION. — Clavodesia clavula n. comb. has been collected in the NW Bay of Biscay, 1550 m depth ( Hayward 1978 as Nellia clavula ) and in three stations in the Gulf of Cádiz between 1523 and 1530 m depth (Harmelin & d’Hondt 1992 as Clavodesia biradiculata ). DESCRIPTION

Colony erect, calcified;initially claviform with a single series of autozooids, later up to six series around the axis. Joined to the substratum by chitinous rhizoids projecting from two long proximal calcified roots, inconstant.Autozooids subrectangular with rounded extremes; basal zooids tapering proximally. Opesia oval, occupying almost the whole length of the autozooid, surrounded by a raised, crenellate rim. Gymnocyst reduced,smooth.Cryptocyst finely granular; reduced laterally and nearly absent distally; more or less developed proximally. Operculum terminal, semicircular. A pair of short, cylindrical spines in the distal angles, distally directed. Distal extreme of the colony irregularly shaped; autozooids with a free portion projecting distally. Two small adventitious avicularia, placed on the lateral gymnocyst one-half the distance down the margin; mandible short, oval, directed outwards. Ovicell terminal, globular, projecting distally in the tip of the colony, but not frontally; ectooecium membranous, entooecium uniformly granular; closed by the zooidal operculum. Ancestrula tubular, with two proximal calcified roots, inconstant; walls densely punctured, without opesia, and with a terminal square operculum. Polypide occupying almost all the length of the ancestrula.

This description is based on the examined colony as well as on previous descriptions of the species ( Hayward 1978 as Nellia clavula ; Harmelin & d’Hondt 1992 as Clavodesia biradiculata ). Original measurements are included in Table 2.

REMARKS

Hayward (1978) described a new species, named Nellia clavula , from a small colony 2.5 mm long,

Measurements Data

Zooidal length 0.80

Zooidal width 0.32-0.35 Opesia length 0.60

Lateral cryptocyst 0.07 Proximal cryptocyst 0.15-0.30

Spine 0.07-0.08 Avicularia length 0.07-0.08 Avicularia width 0.08

Ovicell length 0.20

Ovicell width 0.25 Ancestrula length 1.40-1.60 Ancestrula width 0.20 Ancestrular operculum width 0.13 Calcified roots 0.40

collected by the Thalassa expedition at 1550 m depth in the NW part of the Bay of Biscay (stn Z447, 48°47.4’N, 11°14.3’W, 27/X/1973). This colony, the holotype of the species, is conserved at the MNHN (no. 7914) ; we haven’t had the opportunity to revise ourselves this material, which was kindly revised for us by L. M. Vieira.

Later, Harmelin & d’Hondt (1992) described a new species, placed in a new genus, named Clavodesia biradiculata , collected by the BALGIM expedition in three stations at the Gulf of Cádiz at 1523-1530 m depth. The authors did not mention how many colonies of this species have been collected in each station ; only the holotype, coming from the station CP108, is referred in the original description ( MNHN 12742 ) but no registration number for other material is given. Moreover , it is not said which specimen is figured, though probably the holotype ; the two optical photographs included do not allow a detailed observation of some characters. However, the holotype is now so very badly preserved that no information can be obtained from its study. The only material we have been able to examine is a small colony, coming from the BALGIM stn DW11 (see Material examined), a locality that is referred in the original description of the species ; this sample is labelled, with more species, as “ Clavodesia biradiculata n g nov sp” (literally). Therefore , this colony belongs to the type series of the species, and is designated here as the lectotype of C. biradiculata ( ICZN 1999 : arts 74.1, 74.7) .

The holotype of N. clavula and the lectotype of C. biradiculata are closely similar (see Fig. 1 and Hayward 1978: fig. 2g). Therefore, how the differences between the descriptions by Hayward (1978), Harmelin & d’Hondt (1992) and our own observations can be explained? We believe that they are due to some misinterpretation of several characters by the previous authors, caused perhaps by the scarcity of material and the lack of SEM observations, as explained bellow.

Hayward (1978) points out that the colony of N. clavula soon widens from its base with the growth of four lines of zooids forming two alternating groups of paired zooids with a quadrangular section. However, in his figure ( Hayward 1978: fig. 2g) can be seen that a fifth series of autozooids begins on the edge of the colony; this five series are clearly seen in the original material (L. M. Vieira pers. comm. December 2010). On the other hand, Harmelin & d’Hondt (1992) pointed out that C. biradiculata has up to six lines of autozooids; this is consistent with their figure of the holotype, with a quadrangular section (Harmelin & d’Hondt 1992: fig 1a, b) and also with the lectotype designated here.

After the original description and the colony revised, the autozooids of C. biradiculata bear a pair of small distal spines (see Fig. 1 ); these are not seen in the original figure (Harmelin & d’Hondt 1992: fig. 1a, b). Hayward (1978) described no spines in N. clavula , but in the holotype of this species there is at least one spine (L. M. Vieira pers. comm. December 2010); as the spines are small and not very conspicuous to the binocular, it would be necessary to take SEM photographs to confirm if more spines are present. Anyway, it is not impossible that the spines were really reduced in the material from the Bay of Biscay as a result of the geographic variability of the species.

Hayward (1978) assumed that N. clavula has jointed branches, but these are not present in the holotype of the species, neither in the lectotype of C. biradiculata , or the description by Harmelin & d’Hondt (1992), who have revised more material.

The measurements of the autozooids of the lectotype of C. biradiculata are similar to the zooids figured by Hayward (1978) but shorter than those measured by Harmelin & d’Hondt (1992) (see Tables 1; 2). However, these authors did not indicate if their measurements were made on different colonies or only on the holotype, which is a larger colony. Therefore, it is possible that the basal zooids in a colony were shorter than subsequent ones.

Harmelin & d’Hondt (1992) strangely defined the lateral avicularia as an inconstant vicarious one, 0.30 mm long. However, the avicularia are clearly adventitious; this interpretation is consistent with the description of Hayward (1978) and the lectotype of C. biradiculata , but also with the figure of its holotype.

Harmelin & d’Hondt (1992) pointed out that the colony of C. biradiculata anchors itself to the substrata through a pair of long roots coming from the ancestrula.This ancestrula does not appear in the colony of N. clavula , but neither in the lectotype of C. biradiculata , which have a similar size.Therefore,it is possible that the colony does not develop the long roots when it is too small or that it may have broken.

Finally, Hayward (1978) did not find ovicells in N. clavula . In the original description of C. biradiculata, Harmelin & d’Hondt (1992) reported the existence of a single terminal ovicell in the holotype, but the lectotype presents several terminal ovicells. It is possible that the ovicells get immersed in the colony as this grows, as stated by Harmelin & d’Hondt (1992), but the upper lip of the aperture, curved and granular, should be visible over the distal end of the bearing autozooid.

As a conclusion, in our opinion the descriptions and material of N. clavula and C. biradiculata refer to a single species, Nellia clavula Hayward, 1978 , according to the principle of priority (ICZN art. 23.1).

Hayward (1978) placed its species in the genus Nellia Busk, 1852 because of the shape of the internode, arrangement of zooids and the paired lateral avicularia; but especially in the assumption that the species has jointed branches. According to Gordon (1989) the genus Nellia is characterized by presenting colonies erect, jointed, dichotomously branched, the branches four-sided in cross section; the zooids present proximal gymnocyst and smooth cryptocyst, but no spines; the avicularia are adventitious and the ovicell is recumbent. However, as it was pointed out further above, in the description by Harmelin & d’Hondt (1992), who have revised more material, it is not included any branching or articular joints, and they do not even appear in the lectotype of C. biradiculata nor in the holotype of N. clavula . Moreover, as it has already been stated, the colonies can have more than four zooecial series, the gymnocyst is very reduced, the cryptocyst is granular and spines appear. Therefore, we believe N. clavula can not be placed in the genus Nellia .

Harmelin & d’Hondt (1992) have correctly erected a new genus for the species, and have discussed its position, but assuming the presence of a vicarious avicularia and that the ovicell opens via an independent pore. Gordon (2011) places the genus Clavodesia in the Calloporidae , position which seems correct, although this family is getting too large and complicated nowadays. The colony form of C. clavula n. comb. seems reminiscent of that in Bryocalyx Cook & Bock, 2000 , a calloporid genus. This taxon also has an elongated ancestrula anchored by rhizoids. However, this genus lacks cryptocyst, marginal spines and avicularia, and its ovicell is not closed by the zooidal operculum.

Family FLUSTRIDAE Fleming, 1828

Figures

FIG. 1. — Clavodesia clavula n. comb. (MNHN 15493): A, view of the colony; B, autozooid; C, ovicellate zooids.Scale bar: A, 400 µm.