Tulpa diverticulata Totton, 1930

Tulpa diverticulata Totton, 1930 View in CoL

( Figs 2c–d View FIGURE 2 , 4 View FIGURE 4 )

Tulpa diverticulata Totton, 1930: 145–146 View in CoL , figs 5a–c; Ralph, 1957: 844, fig. 7n (only material from Menzies Bay); Peña Cantero, 2024b: fig. 3h.

Not Tulpa diverticulata View in CoL — Ralph, 1957: 844, fig. 7m (material from Bligh Sound); Millard, 1977: 20, fig. 5G–H; Vervoort & Watson, 2003: 446–447, fig. 108E–G.

? Campanularia diverticulata — Naumov & Stepanjants, 1962: 72.

? Tulpa diverticulata — Stepanjants, 1979: 35, pl. 6 fig. 2.

Not Tulpa diverticulata — Watson, 2003: 173–174, fig. 24A–D (= T. tulipifera )

Material examined. TAN 1802/186, several stems up to 60 mm high, without gonothecae, on axis of dead gorgonian, basibiont of Symplectoscyphus nesioticus Blanco, 1977 ( NIWA 130060).

Description. Stems up to 60 mm high, monosiphonic, unbranched or with a lateral branch, with a series of up to 14 hydrothecae in alternate, almost unilateral arrangement; sometimes with distal hydrotheca.

Hydrothecae at distal end of pedicels of variable length, usually with regenerations. Hydrotheca tubiform, flaskshaped and with marked polygonal structure ( Fig. 2c View FIGURE 2 ). Hydrothecal diameter increasing from annular thickening to basal third, slightly decreasing to distal third and increasing again to aperture ( Fig. 2c–d View FIGURE 2 ). Hydrothecal rim even, distinctly everted and sinuous ( Fig. 4a–c, g View FIGURE 4 ). Sinuosity giving rise to alternate sequence of crests and troughs outside hydrotheca, with opposite sequence inside ( Fig. 4a–c View FIGURE 4 ), extending downward for a long distance. This resulting in a series of more or less marked, straight or slightly concave facets on external hydrothecal wall ( Fig. 2c View FIGURE 2 ), corresponding to external troughs, fading basally, and laterally delimited by longitudinal ridges coinciding with external crests; opposite situation on inner wall. Hydrothecal aperture might have few, short, everted renovations ( Fig. 2c View FIGURE 2 ). Hydrothecal perisarc smooth or with barely noticeable transverse striation. External troughs forming distal diverticula ( Figs 2d View FIGURE 2 , 4a–g View FIGURE 4 ).

Remarks. The material examined is undoubtedly conspecific with Totton’s species, matching perfectly in every detail, except for the presence of stems. These stems are monosiphonic, although some stolons may run alongside, giving the impression of polysiphony. The stems are typically unbranched, although some have a lateral branch. The basal part of the stems is devoid of hydrothecae for a significant extent.

The hydrotheca has a polygonal structure due to the straight or slightly concave facets, which usually extend downward along the distal three-fourths. The sinuosity of the rim creates the external facets, which correspond to the troughs between the crests on the outside. There are also alternating concave internal facets, although these are barely noticeable.

In this species, there is only a barely noticeable unevenness between the distal end of the rim external troughs and crests, and consequently there are no true cusps. This unevenness is merely an optical illusion caused by the structure of the hydrothecal rim. The rim is even and clearly curved outwards, but also sinuous. As mentioned earlier, the wave formed can be divided into alternating crests and troughs on both the inner and outer surfaces of the hydrotheca. The external troughs give the false impression of “cusps,” while the external crests correspond to the perceived “embayments.” At the rim, however, both crests and troughs are at the same level ( Fig. 4c, g View FIGURE 4 ), so that no real cusps are present. According to Stepanjants (1979), the longitudinal ridges along the hydrotheca’s wall and the deep troughs between them, extending to the edge of the aperture, create the appearance of rounded cusps, up to 12 in number.

The peculiar diverticula of T. diverticulata are formed because the rim is strongly bent outwards over the deeply concave distal part of the external facets ( Fig. 4b, e View FIGURE 4 ), and these often project upwards beyond the rim, forming a pouch-like structure or diverticulum ( Fig. 2c–d View FIGURE 2 ). This was perfectly illustrated by Totton (1930: 145) ‘Distally the ends of these fluted facets appear as slight bosses that rise above the level of the entire, everted, slightly oblique margin’ and is represented in his figure 5c. Ralph (1957) also stated that the margin is entire, but everted, and that the ends of the fluted facets appear as bosses. In the present material, these protuberances are conspicuous when the hydrothecal aperture is viewed from above ( Fig. 4e–g View FIGURE 4 ), but also on the sides in lateral view ( Figs 2c–d View FIGURE 2 , 4d View FIGURE 4 ).

Ralph (1957) noted that T. diverticulata and T. tulipifera differed mainly in growth habit, the latter having sparsely branched stems with hydrothecae on pedicels arranged in a pinnate and alternate disposition, whereas T. diverticulata was considered to be stolonal. However, Ralph herself pointed out the possibility that new collections might reveal a change in habit, suggesting that T. diverticulata might be conspecific with T. tulipifera . As shown above, the material examined here, which is undoubtedly T. diverticulata , has branched stems. Therefore, the presence/absence of a stem is no longer a reliable character to distinguish between the two species. However, this does not mean that T. diverticulata and T. tulipifera are conspecific, as Ralph suggested. In my opinion, although they are closely related, they represent two distinct, well-defined species that differ in important morphological characters.

Apart from the presence of stems, both species share the size of the nematocysts and the general shape of the hydrotheca. However, T. tulipifera lacks the characteristic diverticula found in Totton’s species, usually has a distinct perisarc keel on the internal ridges, a much more everted rim, and a more pronounced sinuosity in the inner part of the hydrotheca. This results in more marked internal ridges on the inner hydrothecal wall. In Totton’s species, the sinuosity is similar both inside and outside the hydrotheca. Finally, whereas T. diverticulata lacks cups, T. tulipifera has a noticeable difference in height between the distal ends of the troughs and crests, resulting in small but conspicuous cusps, 40 µm high, visible in lateral view.

Ralph (1957) assigned specimens from two New Zealand localities (Menzies Bay in the north-east of the South Island, and Bligh Sound in the south-west) to T. diverticulata . I believe they belong to two different species, and her description contains details of both. In my opinion, the material from Menzies Bay belongs to T. diverticulata , and I believe Ralph’s figure 7n represents this species. The hydrotheca represented in this figure corresponds well to T. diverticulata , resembling those depicted by Totton (1930) and found in the present material, characterised by a distinctly everted aperture. According to Ralph’s (1957: 844) description of the hydrotheca, ‘margin entire, but everted and the ends of the fluted facets appear as bosses’, which clearly indicates that she was dealing with T. diverticulata due to the presence of diverticula and the everted, entire rim (she did not note the presence of cusps). However, her material also contains specimens with distinctly different hydrothecae, as shown in figure 7m. These hydrothecae are more robust and lack a distinctly everted rim. They certainly belong to the Bligh Sound material, as Ralph (1957: 844) stated that ‘Only the specimens from Bligh Sound, taken in May, had gonothecae’. This material undoubtedly belongs to a different species.

Gravier-Bonnet (1979) pointed out that while Totton (1930) characterised the hydrotheca of T. diverticulata as having an even aperture, Ralph (1957) depicted a hydrothecal rim with large, rounded cusps, like those described by Vervoort (1972) for T. tulipifera and those which she observed in her T. costata (see below). Gravier-Bonnet (1979) questioned whether Totton’s description reflected reality or whether Ralph’s material represented a different species. I believe Ralph’s material consists of two species, one being T. diverticulata and the other being clearly different. Gravier-Bonnet (1979) also noted that if the hydrothecal rim in T. diverticulata is as described by Totton, this would be enough to prevent T. diverticulata and T. tulipifera from being considered the same species, as Ralph (1957) had thought possible when she suggested that T. diverticulata could only represent the unbranched form of Allman’s species.

Millard (1977) assigned material to T. diverticulata that I do not believe belongs to the species. She probably identified it as T. diverticulata on the basis of the stolonal structure of the colony, stating that ‘there is never any question of an erect branching stem with oblique pedicels as in T. tulipifera ’, but this is no longer a valid character for identifying Totton’s species. In addition, the shape of the hydrotheca in Millard’s material is completely different from that of T. diverticulata , and she neither mentioned nor illustrated the characteristic diverticula of Totton’s species.

The material from the Auckland Islands assigned to T. diverticulata by Naumov & Stepanjants (1962) and Stepanjants (1979) may belong to this species, although there is some uncertainty, as they did not mention or illustrate the diverticula. However, the shape of the hydrotheca is similar to that of T. diverticulata . Naumov & Stepanjants (1962) noted that the rim of the hydrothecal aperture is distinctly everted and wavy, and Stepanjants (1979) mentioned that the hydrothecal perisarc is finely striate and that the ridges and troughs of the hydrothecal wall extending to the edge of the aperture give the impression of up to 12 rounded cusps. Naumov & Stepanjants (1962) also mentioned that the gonotheca of T. diverticulata is characterised by a dense transverse striation, a feature clearly visible in the figure provided by Stepanjants (1979: pl. VI fig. 2). If confirmed, this would be the first and only record of gonothecae in Totton’s species.

Stepanjants (1979), who also identified material as T. tulipifera (see above), pointed out that both species are very similar in the shape and size of their hydrothecae, and that the main difference is that the colonies of T. diverticulata are stolonal. However, as mentioned above, this character is no longer reliable. Stepanjants (1979) also pointed out that T. diverticulata has longer pedicels and a less curved rim at the aperture.

In my opinion, the material assigned to T. diverticulata by Vervoort & Watson (2003), does not belong to this species. They did not mention or illustrate the characteristic diverticula of T. diverticulata , and the hydrotheca is clearly smaller and of different shape, more cylindrical and not clearly widening at the aperture. Apparently, they assigned their material to T. diverticulata on the basis of the absence of stems, a feature that is no longer valid for its identification. On the other hand, these authors noted that their material differed from that of Ralph (1957) in the lack of fine transverse striations on the neck of the gonothecae and the absence of an everted rim in the hydrothecae. However, as mentioned above, Ralph’s fertile material does not belong to T. diverticulata either. The material examined by Vervoort & Watson (2003) may represent a new, distinct species, but it would need to be re-examined to confirm this.

Peña Cantero (2024a) considered that Watson’s (2003) material of T. diverticulata from the Macquarie Island area actually belongs to T. tulipifera , an opinion shared here.

Ralph’s (1957) material from Menzies Bay was collected from an intertidal rock pool creeping over other hydroid stems. This clearly contrasts with the depths at which Totton’s and the present material were collected (546 m and 652–654 m, respectively). Consequently, either the species is eurybathic, ranging from intertidal to bathyal depths, or Ralph’s material was stranded and came from deeper waters. Although Ralph mentioned that her material was growing over other hydroid stems, she did not specify whether these other hydroids were attached to the substrate. It is worth noting that Stepanjants’ (1979) material came from shallower continental shelf waters (125 m).

Ecology and distribution. Tulpa diverticulata has been reported from the intertidal zone ( Ralph 1957) to depths of 654 m ( Peña Cantero, 2024b).

Tulpa diverticulata is mainly known from New Zealand waters: Three Kings Islands ( Totton, 1930), Menzies Bay, NE South Island ( Ralph 1957) and probably the Auckland Islands area ( Naumov & Stepanjants 1962; Stepanjants 1979). Peña Cantero (2024b) reported it from the Long Ridge area of Antarctica.