Genera: Haplotaxis Hoffmeister, 1843.

Included species: Basically following Brinkhurst (1988), we consider the following species as possible members of Haplotaxis / Haplotaxidae at this point: Haplotaxis gordioides (Hartmann in Oken, 1819) (type species), Haplotaxis aedeochaeta Brinkhurst & Marchese, 1987 (Argentina), Haplotaxis ascaridoides (Lake Baikal), Haplotaxis dubius (Lake Ohrid), Haplotaxis emissarius (Forbes, 1890) (eastern North America); Haplotaxis forbesi Smith, 1918 (eastern North America), Haplotaxis gastrochaetus (Hokkaido, Japan), Haplotaxis heterogyne Benham, 1905 (New Zealand), Haplotaxis ichthyophagous Gates, 1971 (western North America), Haplotaxis vermivorus (Michaelsen, 1932) (Sumatra). Species incertae sedis: Haplotaxis villiersi Omodeo, 1987 (Guinea, likely not Haplotaxis, with no ‘gizzard’ and chaetae paired with dorsals enlarged).

Material examined: Vouchers of the European Haplotaxis specimens (from Sweden, France, Italy, and Greece; see details in Table 1 and Supporting information, Table S1) used in this analysis are all sexually immature, but they agree with the general descriptions of Haplotaxis gordioides (e.g. in Brinkhurst and Jamieson 1971) in non-genital characters. Moreover, we have no access to material from the type locality of the generic type (Haplotaxis gordioides), and we cannot conclude that the latter species is present among our samples. The African specimens (Table 1; Supporting information, Table S1), although conforming to Haplotaxidae as defined above, will be described in a future contribution. Morphological vouchers were examined for most of the Japanese and North American Haplotaxis material. Haplotaxis spJP1: NSMT An-1915–1917: Japan, Shizuoka Prefecture, Kakida River, Doniwa, Shimizu-cho, Sunto-gun, 8 Jan 2010, collected by T. Torii; anterior end and slide mounted, part of tail analysed for DNA (CE 30885); two additional topotypic specimens from same date, whole mounted. Haplotaxis spJP2: NSMT An 1918–1922: Japan, Aomori Prefecture, Gudari-numa stream, Hakkoda Mountains, 17 Mar 2017, collected by A. Ohtaka; five topotypic specimens, whole mounted, part of tail from one specimen (CE31629) analysed for DNA. Haplotaxis spUS1: USNM 1716294: USA, Washington, Jefferson Co., Bull Creek near Clearwater River, 2 Apr 2016, collected by S. Fend; anterior end and slide mounted, part of tail analysed for DNA (CE30796). Haplotaxis spUS2: USNM 1716295–1716298: USA, Idaho, Blaine Co., Corral Creek 1 km above Trail Creek, 1 Jun 2016, collected by S. Fend; anterior end and slide mounted, part of tail analysed for DNA (CE30863); same collection data, three topotypic, whole mounted anterior ends (CE30864-66). S. Fend collection: 2 Jun 2008, seven whole mounted and five dissected specimens. Haplotaxis spUS3: SMNH 224406: USA, California, Guadalupe Creek above reservoir, 27 Mar 2007, collected by S. Fend, one of seven collected specimens, slide mounted, but part of tail analysed for DNA (CE5231); other slides in S. Fend collection. Haplotaxis spUS4: USNM 1716299–1716306: USA, Oregon, Jefferson Co., Metolius River above Camp Sherman, 4 Dec 2019, collected by S. Fend; anterior and posterior segments slide mounted, middle segments analysed for DNA (CE36035); same collection date, four topotypic, whole mounted anterior ends, (CE36036–39); 4 Jun 2018, three topotypic, whole mounted anterior ends (CE36014–16).

Additional Haplotaxis specimens were examined here for morphology and shown in Figure 3, but were not analysed for DNA: USA, Montana, Broadwater Co., spring near Brewer, 17 Mar 2007, collected by D.L. Gustafson; two sagittally sectioned specimens. California, Colusa Co., Stony Creek at Stonyford, 4 Nov 2002, collected by S. Fend; one whole mounted specimen. This material remains in S. Fend’s personal collection.

Remarks

This family (clade 1 in the present molecular analysis) is now defined by a unique combination of characters that have been associated with a predatory habit, and used to define the genus Haplotaxis by Brinkhurst (1988) (in italics, characters that are new in comparison with the Brinkhurst (1988) diagnosis, and are possible apomorphies): replacement of eversible pharynx and pharyngeal glands by a massive muscular pharynx in 1/2IV– V (VI); capacious, thin-walled oral cavity (possibly eversible); ventral chaetae large, single, sickle-shaped, dorsals small, straight, simple, often lost in many or all segments. Taxa with paired chaetae are no longer included in the family. However, as we have not been able to study more than a few of the included species, and as the available sexually mature worms are indeed few, our diagnosis, as well as the species list, should be seen merely as a hypothesis of monophyly of the family. The true identity of the type species of Haplotaxis, Haplotaxis gordioides, is not resolved in this study; see Discussion for more details.

Allspecimensattributedto Haplotaxis andtotheHaplotaxidae s.s. in the present analysis have the very elongate habitus of the type species, Haplotaxis gordioides. Other diagnostic characters also appear consistent with Haplotaxis s.s. descriptions, and can be easily seen in the available whole mounted specimens (Fig. 2A–C). All of these characters are unique to the family, or unusual among other oligochaete families. The peristomium is elongate, most often biannulate, with the first annulus anterior to the mouth, containing the brain; the posterior annulus surrounds the buccal region. Segments beginning in V also may be biannulate, with a narrow anterior annulus. Mouth gives into a thin-walled, capacious oral cavity with folded epithelium, joined to the body wall with numerous muscle bands. Foregut continues as a muscular pharynx whose walls contain radial and circular muscles; the latter may form anterior or median sphincters (cf. Benham 1905: fig. 1, Smith 1918: fig. 7, Yamaguchi 1937: fig. 2, Brinkhurst and Marchese 1987: fig. 9). Length of the pharynx, as well as thickness and orientation of musculature varies among populations. A thin cuticular layer is visible at the beginning of the mouth, then becomes gradually thinner but remains visible within the pharynx and seems to disappear afterwards. Pharyngeal glands have not been observed. Chaetae are always single (not in pairs), with ventrals much larger than dorsals (cf. Forbes 1890: figs 1–2, Smith 1918: figs 5–6, Hrabě 1931: fig. 8, Yamaguchi 1936: figs A–C). Other characters considered diagnostic in the literature, such as ‘Timm’s glands’ (Timm 1883: fig 12, Forbes 1890: figs 5–7, Hrabě 1931: fig. 9c) and nephridia (Forbes 1890: fig. 10, Benham 1905: fig. 8) are difficult to see in some preparations, and none of the present specimens is mature enough to observe the reproductive characters, other than rudimentary gonads.

Several described species would be attributed to Haplotaxidae s.s. as defined here, but the majority have been traditionally synonymized with Haplotaxis gordioides due to the unavailability of fully-mature specimens in most species descriptions, combined with a supposedly high intra-population variability in somatic characters, particularly the distribution of dorsal chaetae (Michaelsen 1899, Brinkhurst 1966, Brinkhurst and Jamieson 1971). Consequently, although the present analysis indicates multiple species having regional affinities, it is premature to assign these either to new or to previously-described species. It should be noted that a morphologically diverse group of Afrotropical species (to be described elsewhere) is well separated from the Holarctic species in the present analysis, and the Holarctic examples (representing up to 10 species) appear separated more by geography than by obvious morphological characters. For example, the two Japanese species in the DNA analysis are easily differentiated from each other by somatic characters, including extent of the muscular pharynx and morphology of the midventral (Timm’s) glands (Fig. 2B, C)—yet (aside from a somewhat shorter pharynx) these do not appear to be consistent synapomorphies separating them from Nearctic species.

It is important to note that some morphologically distinctive species recognized as haplotaxids by previous revisions (Brinkhurst and Jamieson 1971, Brinkhurst 1988) and conforming to Haplotaxidae as defined above, were unavailable for our molecular analysis. Among the more conspicuous examples, the Neotropical Haplotaxis aedeochaeta Brinkhurst & Marchese, 1987 has gonadal segments reduced to XI (testes) and XII (ovaries); there are narrow genital chaetae; dorsal chaetae are lacking; and the muscular pharynx is in V–VI (instead of the more common IV–V). Haplotaxis gastrochaetus Yamaguchi, 1953 (Japan) has some segments with remarkably enlarged ventral chaetae, without a nodulus (Yamaguchi 1953: fig. 14); in addition, dorsal chaetae are lacking, there is a single ovarian segment (progynous), and the ‘gizzard’ is described as ‘vestigial’. Haplotaxis forbesi Smith, 1918 (eastern North America) appears to have only one testicular segment; ovaries are unknown, but conspicuous organs associated with nephridia in XV–XVII resemble gonadal tissue (Brinkhurst, 1966). Haplotaxis heterogyne Benham, 1903 (New Zealand) has a single ovarian segment, and the muscular pharynx is reduced to IV only. In addition to these examples, rare observations of mature haplotaxid specimens have shown that they differ in reproductive characteristics, including genital chaetae (e.g. Haplotaxis dubius) and number of spermathecae (e.g. Haplotaxis ascaridoides). An examination of over 200 whole mounted (mostly immature, but many with developing gonads) Nearctic haplotaxid specimens by one of the authors (S.F.) suggests that populations have consistent differences in characters that include distribution of dorsal chaetae, thickness and presence of glandular cells in the epidermis, circular musculature of body wall, thickness of cuticle, extent and muscular development of the pharynx, chaetal morphology (Fig. 3I–J), and number of ovarian segments; rare mature specimens indicate that at least two populations have distinctive genital chaetae. This indicates that the species included in the present analysis are just a limited sample of the world fauna, and that taxonomy of the group will evolve as material becomes available.

The geographical distribution of Haplotaxis material in our analysis is Afrotropical and Holarctic. Based on published morphological descriptions, species conforming to our diagnosis of Haplotaxidae also occur in the Australasian (Haplotaxis heterogyne), Indomalayan (Haplotaxis vermivorus), and Neotropical (Haplotaxis aedeochaeta) regions.