Revision of the genus Buchneria (Bryozoa, Cheilostomata) from Japan

Abstract Buchneria dofleini (Buchner, 1924), type species of Buchneria Harmer, 1957,was first described from material collected in 1904–1905 from Sagami Bay, Japan, but the type specimens had not been reexamined since the original description. In this study, I examined specimens of Buchneria from historical collections and material recently collected near Akkeshi, Hokkaido, Japan. Three Buchneria species were detected, two from Sagami Bay that Ortmann (1890) had placed in Escharoides, and one from Akkeshi that Androsova (1958) had described as Porella variabilis. I concluded that Buchneria dofleini is a junior synonym of Escharoides teres Ortmann, 1890; selected a lectotype for Escharoides teres among Ortmann’s syntypes; and established the new combination Buchneria teres (Ortmann, 1890), which becomes the type species of Buchneria. I also established the new combination Buchneria rhomboidalis (Ortmann, 1890) and selected a lectotype among Ortmann’s syntypes. Porella variabilis is transferred to Buchneria establishing the new combination Buchneria variabilis (Androsova, 1958). Here the three new combinations are redescribed and a key to the Japanese Buchneria species is provided. Finally, I transferred Buchneria to Bryocryptellidae on the basis of ovicell and orifice morphology. Therefore, Buchneria now includes a total of three species; Buchneria sinuata Harmer, 1957, a species from Indonesia that has hitherto been placed in this genus, is almost certainly not congeneric with other Buchneria. As far as is now known, Buchneria is endemic to northern Japan and the northern Sea of Japan.

introduction Harmer (1957) introduced the cheilostome bryozoan genus Buchneria for species with erect colonies, and chose the deep sea species Palmicellaria dofleini as type, which was described by Buchner (1924) from Sagami Bay, Japan. According to Harmer's description, zooids in species of Buchneria have the proximal margin of the orifice with a broad sinus or nearly straight, with a small suboral avicularium at the edge of the peristome, and have only a few pores in the frontal shield. These characters, however, also match Escharoides teres and Escharoides rhomboidalis, both described by Ortmann (1890) from Sagami Bay, although Ortmann's (1890) limited descriptions and simplified illustrations are inadequate for taxonomic assessment.
The status of Buchneria has not been evaluated subsequent to Harmer (1957). In this study, I reexamined type specimens established by Buchner (1924) and Ortmann (1890) for species of Buchneria and morphologically similar genera. I examined relevant material collected in Japan over the past approximately 130 years, as well as material obtained through my own collecting efforts. I review, describe, and illustrate the known species of Buchneria.
The author collected additional specimens from Sagami Bay by dredge from RV Tansei-maru (Japan Agency for Marine-Earth Science and Technology, JAMSTEC) and research boat Rinkai-maru (Misaki Marine Biological Station, The University of Tokyo) in November 2007 and February 2012, and outside Akkeshi Bay, Hokkaido, in July 2010 and 2011, by dredge from research boat Misago-maru (Akkeshi Marine Station, Hokkaido University) (Fig. 1).

Preparation and observation of specimens
Specimens were observed by light microscope and scanning electron microscope (SEM). For SEM observation, part of each specimen was removed, soaked in a sodium hypochlorite solution to remove the soft tissue, rinsed in water, air dried, and mounted with double-sided adhesive tape or silver paste on an aluminum SEM stub. At Hokkaido University, mounted specimens were coated with Au in a Hitachi E-1030 sputter-coater and observed with a Hitachi S-3000N SEM at 15 kV accelerating voltage. At the SMF, specimens were coated with Pt-Pd and observed with a CamScan SEM. At ZSM, specimens were coated with Au in a POLARON SEM Coating System and observed with a LEO 1430VP SEM at 15 kV accelerating voltage. Fragments removed from specimens in the various collections, prepared and examined by SEM, and subsequently deposited in NMST are indicated in the text by the designation 'NSMT Te' (see Supplementary Table 1).
Measurements were taken from SEM images with ImageJ 1.37v software (Image Processing and Analysis in Java, Wayne Rasband, National Institutes of Health, USA; http://rsb.info.nih.gov/ij/). Measurements in the text are presented in millimeters, as ranges followed in parentheses by the mean and standard deviation. Sample sizes for measurements were n = 4-82, generally from more than one colony. Abbreviations used for measurements are as follows; ZL, zooid length; ZW, zooid width; OrL, orifice length; OrW, orifice width; AvL, suboral avicularium length; AvW, suboral avicularium width; OvL, ovicell length; OvW, ovicell width. Diagnosis. Colony erect, rigid, dichotomously or irregularly branching: branches cylindrical, flattened, or plate-like, fan shaped. Zooidal frontal shield uniformly tessellated, with a few areolar pores near margin or offset centrally in secondarily calcified wall. Orifice deeply immersed, without teeth on distal periphery, without lyrula or condyles, slightly concave or straight proximally; oral spines absent. Secondary orifice at colony surface cormidial, formed by contributions of secondary calcification from distal and lateral zooids. Suboral avicularium lies at proximal margin of secondary orifice, directed proximally or laterally, sometimes enlarged and occupying about half of frontal shield; small, conical tooth associated with avicularium projecting into secondary orifice (Fig. 2). Mandible of the suboral avicularium semicircular or spatulate, but never acute. Vicarious and other frontal avicularia absent. Ovicell globose, acleithral, and is produced by the distal zooid (Fig. 3). Both the endooecium and ectooecium are calcified. Endooecium is completely calcified, whereas ectooecium is not completely covering the endooecium (Fig. 3A, B). Immediately after formation, the ectooecium is then partially covered by the secondary calcification that is coming from the distal and neighbour zooids (Fig. 3B). Finally, the secondary calcification covers most of the ectooecium in the old parts of the colony, but a small area of proximal margin remains uncovered (Fig. 3C). Small basal pore chambers present.
Remarks. Harmer (1957) defined Buchneria as follows: colony erect, not jointed; large spatulate avicularia present; zooids with a "sinuate" or nearly straight proximal margin of the orifice (the term 'sinuate' appears to be misapplied to the evenly concave proximal margin in Buchneria sinuata; perhaps Harmer intended the meaning as 'having a sinus'); small, acute suboral or lateral avicularia on the edge of the secondary orifice; few frontal pores; and a hyperstomial ovicell with an imperforate central tabula.
His generic diagnosis, however, largely derives from Buchneria sinuata Harmer, 1957 from Indonesia. This species is similar to the three Buchneria species treated herein in having erect colony form, few frontal or marginal pores, and a deeply immersed primary orifice. However, it differs substantially from them in having large, spatulate frontal avicularia; hyperstomial ovicells with an imperforate central tabula lacking secondary calcification; and a laterally placed, acute oral avicularium (Harmer, 1957: plate LIV, fig. 19). Buchneria sinuata, therefore, has currently been thougth to belong in another genus, perhaps Osthimosia Jullien, 1888 (Gordon, 1984). Gordon (1984) noted several similarities between Osthimosia virgula and nominal B. sinuata (e.g., broad orificial sinus, a lateral-oral avicularium on peristome, and spatulate frontal avicularia), and  showing smooth surface of endooecium and ectooecium with less secondary calcification B Ooecium started covered by tessellated secondary calcification from neighboring zooids C Ooecium almost covered by the secondary calcification with showing endooecium through the small proximal membranous window at ectooecium. Ec, ectooecium; En, endooecium; Sc, secondary calcification. Arrows indicate the proximal membranous window.
suggested the two species may be congeneric. Subsequently, Gordon (1989) observed and illustrated Japanese material of Buchneria present in the Natural History Museum London (NHMUK), and elucidated the umbonuloid frontal shield; he concluded Buchneria cannot be grouped together with Osthimosia and other lepraliomorphs. Unfortunately, I have never had a chance to check the type material of Harmer's B. sinuata, which is not in NHMUK (Mary Spencer Jones, pers. comm. 11 May 2012) and both institutes Zoological Museum Amsterdam (ZMA) and Naturalis Biodiversity Center in Leiden (Elly Beglinger, pers. comm. 19 September 2012). Although I have not checked the type material, I exclude Harmer's B. sinuata from the description of the genus in this paper based on the significant differences with the type species of the genus; B. sinuata is also different from other Buchneria species in mainly imperforate frontal shield, small colony size, and preference of unstable substrate which is unusual for Buchneria species. The status of B. sinuata is still unclear and should be clarified in future work. With the removal of nominal B. sinuata from Buchneria, the cormidial orifice and the rounded mandible of suboral avicularia may be considered diagnostic characters for Buchneria. Buchner (1924) described large frontal avicularia, but these are enlarged suboral avicularia; therefore, absence of a large frontal avicularium may also be considered diagnostic for Buchneria. Gordon (1989) regarded Buchneria close to Celleporaria in Lepraliellidae, based on the similarities between the type species of both genera in the umbonuloid and imperforate frontal shield with marginal areolar pores, broad orifice, and suboral avicularium. However, the ovicell of Buchneria is different from that of Lepraliellidae in having broader proximal window and deeper ooecium, and is more similar to that of Palmiskenea Bishop & Hayward, 1989 in Bryocryptellidae with a few small foramina close to the proximal margin. Buchneria also resembles Palmiskenea in the frontal shield having only marginal areolar pores, but differs from the latter in the orifice without condyles and in polymorphic avicularia. Buchneria resembles Marguetta Jullien, 1903 in having only marginal pores, an ovicell with a few small pores, and oval suboral avicularia, but differs from the latter in lacking frontal avicularia on margin of the frontal shield. Buchneria also resembles Porella Gray, 1848 and Porelloides Hayward, 1979 in having only marginal pores, an ovicell without or with a few small pores, and suboral avicularia, but differs from the latter two genera in lacking lyrula and condyles. Although some species of Porella also lack a lyrula, and species of Porelloides normally lack condyles, a small tooth on the distal margin of the suboral avicularium is characteristic of Buchneria. Considering the similarities of ovicell and orifice morphology between Buchneria and the four bryocryptellid genera, Porella, Palmiskenea, Marguetta and Porelloides, I conclude Buchneria is much better placed in Bryocryptellidae rather than Lepraliellidae. Harmer (1957) suggested that Haswellia auriculata Busk, 1884 be placed in Buchneria, on the basis of orifice morphology and the small lateral avicularia on the edge of the peristome. He also suggested that the species H. auriculata and Myriozoum marionense Busk, 1884 as described by Calvet (in Jullien and Calvet 1903) represent a single species referable to Buchneria, on the basis of very few frontal pores; a large, spatulate frontal avicularium; the form of the peristome with small avicularia; and im-mersed ovicells (Harmer, 1957). However, H. auriculata is currently regarded as a junior synonym of Galeopsis pentagonus (d'Orbigny, 1842), and the specimen described as M. marionense in Jullien and Calvet (1903) is also considered as a species of Galeopsis.
Remarks. Ortmann (1890) first described Buchneria teres (as Escharoides teres) based on Döderlein's specimens from eastern Sagami Bay; these specimens had not been reexamined until this study. Buchner (1924) subsequently described B. dofleini (as Palmicellaria dofleini) from Doflein's Sagami Bay specimens. Although Buchner's type specimen of B. dofleini was lost during WWII, I found other specimens he identified as this species in the Doflein collection at ZSM. In comparing these specimens with Ortmann's syntypes of E. teres, I found no diagnostic differences between the two; Buchner's specimens are simply the distal younger part of branches of E. teres. I thus consider Buchneria dofleini as a junior synonym of E. teres, which accordingly becomes the type species of Buchneria. Buchner (1924) reported large frontal avicularia in his species, but these are enlarged suboral avicularia. Although syntypes of B. teres in the Döderlein collection lack the colony base, I found an entirely kenozooidal colony base in complete colonies recently collected from Sagami Bay (Fig. 4D). (Ortmann, 1890) 7A). Branches flattened, multiserial, with zooids opening all around; 2.33−6.34 mm wide (3.34±0.88 mm, n=25), five to nine zooids across (Fig. 7B). Autozooids rhomboidal, arranged in quincunx (Fig. 7B), zooidal borders indistinct. Frontal shield convex, entirely tessellated with minute depressions, with two to four small areolar pores (Fig.  8A, B) offset from margin. Orifice subcircular, about as wide as long, smooth distally, proximal margin without sinus; lyrula and condyles absent (Fig. 8C). No oral spines.  Orifice deeply immersed; aperture at colony surface roughly semicircular in outline, without sinus proximally; cormidial, bounded by contributions of frontal calcification from distal and one or two lateral zooids, with suture lines sometimes evident between the sectors (Fig. 8C, D). Suboral avicularium small, proximal to orifice on the peristome periphery; circular, with complete pivot, rostrum slightly elevated, slightly denticulate, semicircular mandible directed proximolaterally (Fig. 8C, D, E); distal tooth of suboral avicularium small, rounded-conical (Fig. 8D). No other avicularia were observed. On both the edges of branches and in older part of colony, interzooidal kenozooids lacking orifice (Fig. 8F) are interspersed with autozooids; kenozooids especially numerous in the basal part of colony. Ooecium imperforate, smooth, completely immersed, not evident from colony surface; proximal margin almost straight or slightly curved, rarely obscuring the distal edge of primary orifice in oviccelate zooids (Fig. 7C).

Buchneria rhomboidalis
Distribution. Eastern part of Sagami Bay, and the Sagami Sea southwest of Jogashima and west of Ōshima, at depths of 200-493 m.
Remarks. Examination of Ortmann's (1890) type specimens revealed this species belongs not in Escharoides but in Buchneria, on the basis of the frontal shield with few pores, absence of oral spines, orifice without lyrula, immersed imperforate ooecium, and the suboral avicularia. Buchneria rhomboidalis is characterized by having rhomboidal zooids and flat branches. This species resembles B. teres, but differs in having flat rather than cylindrical branches, in lacking a peristomial labium and sinus, and in lacking a large avicularium at branch bifurcations. The depth distribution of Buchneria rhomboidalis (200-493 m) is shallower than that of B. teres (432-3660 m). (Androsova, 1958), comb. n. http://species-id.net/wiki/Buchneria_variabilis Figures 9, 10
Remarks. My material matches Androsova's (1958) description of Porella variabilis. She mentioned that the tooth on the proximal margin of the peristome is associated with the suboral avicularium, and that the tooth is present when the avicularium abuts the proximal margin of the peristome, but absent when the avicularium is offset proximally from the peristome. This is the case in all known species of Buchneria. An- drosova (1958) reported B. variabilis from 36 m depth, shallower than my specimen from 114-116 m near Akkeshi. This apparent difference in the bathymetric distribution between Sakhalin and Akkeshi may be related to water temperature, as Sakhalin is more northern and colder than Akkeshi.
Buchneria variabilis differs from B. teres and B. rhomboidalis in colony form and in having larger suboral avicularia.

Discussion
To date, species of Buchneria have been only reported from the northwestern Pacific, where they appear to have a cold-temperate distribution (Fig. 11). Most records are from northern Japanese waters; I have not detected Buchneria species in field surveys in southern Japan (e.g., near Okinawa). The southernmost record of living Buchneria in Japan is B. rhomboidalis and B. teres from Sagami Bay, of which B. teres has been considered as abyssal species (Buchner, 1924;Harmer, 1957) and was collected at depths of more than 400 m in Sagami Bay. Another, more northern Buchneria species in Japan showed shallower distribution; Buchneria variabilis occurred at 114-116 m depth near Akkeshi and at 36 m depth near Sakhalin (Androsova, 1958). While Buchneria might be endemic to this region, further sampling around the North Pacific rim, including deep-water sites, may expand the distributions of known species or detect additional species. Furthermore, as Buchneria resembles some other North Atlantic genera such as Porella, Palmiskenea, Marguetta and Porelloides, taxonomic studies of these other genera may also detect additional Buchneria species, and the close relationship may indicate a common history at times when there was a connection between the Pacific and Atlantic.
Taxonomic key to Japanese Buchneria species